cardiovascular hemodynamics for the clinician

CardiovascularHemodynamicsfor the Clinician

Edited by

George A. Stouffer, MDDirector of Interventional CardiologyDirector of the Cardiac Catheterization LaboratoriesVice-Chief of CardiologyUniversity of North Carolina at Chapel HillChapel Hill, NC, USA

C© 2008 by Blackwell PublishingBlackwell Futura is an imprint of Blackwell Publishing

Blackwell Publishing, Inc., 350 Main Street, Malden, Massachusetts 02148-5020, USABlackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UKBlackwell Science Asia Pty Ltd, 550 Swanston Street, Carlton, Victoria 3053, Australia

All rights reserved. No part of this publication may be reproduced in any form or by anyelectronic or mechanical means, including information storage and retrieval systems, withoutpermission in writing from the publisher, except by a reviewer who may quote brief passagesin a review.

First published 2008

1 2008

ISBN: 978-1-4051-6917-2

Library of Congress Cataloging-in-Publication Data

Cardiovascular hemodynamics for the clinician / edited by George A. Stouffer.p. ; cm.

Includes bibliographical references and index.ISBN-13: 978-1-4051-6917-2 (alk. paper)ISBN-10: 1-4051-6917-6 (alk. paper)

1. Hemodynamics. 2. Heart–Diseases–Diagnosis. 3. Heart–Diseases–Treatment.I. Stouffer, George A.

[DNLM: 1. Heart Diseases–diagnosis. 2. Heart Diseases–therapy.3. Coronary Arteriosclerosis. 4. Hemodynamic Processes. WG 210 C26845 2007]

QP105.C39 2007616.1′2075–dc22 2007022606

A catalogue record for this title is available from the British Library

Commissioning Editor: Gina AlmondDevelopment Editor: Beckie BrandEditorial Assistant: Victoria PittmanProduction Controller: Debbie Wyer

Set in Palatino 9.5/12 by Aptara, New Delhi, IndiaPrinted and bound in Singapore by Fabulous Printers Pte Ltd

For further information on Blackwell Publishing, visit our website:www.blackwellcardiology.com

The publisher’s policy is to use permanent paper from mills that operate a sustainable forestrypolicy, and which has been manufactured from pulp processed using acid-free and elementarychlorine-free practices. Furthermore, the publisher ensures that the text paper and cover boardused have met acceptable environmental accreditation standards.

Blackwell Publishing makes no representation, express or implied, that the drug dosages in thisbook are correct. Readers must therefore always check that any product mentioned in thispublication is used in accordance with the prescribing information prepared by themanufacturers. The author and the publishers do not accept responsibility or legal liability forany errors in the text or for the misuse or misapplication of material in this book.

Contents

List of Contributors, vii

Preface, ix

Part I Basics of hemodynamics

1 Introduction to basic hemodynamic principles, 3James E. Faber PhD, George A. Stouffer MD

2 The nuts and bolts of right heart catheterizationand PA catheter placement, 17Vickie Strang RN

3 Normal hemodynamics, 37Alison Keenon BS, Eron D. Crouch MD, James E. Faber PhD,George A. Stouffer MD

4 Arterial pressure, 57George A. Stouffer MD

5 The atrial waveform, 67David P. McLaughlin MD, George A. Stouffer MD

6 Cardiac output, 81Frederick M. Costello MD, George A. Stouffer MD

7 Detection, localization, and quantification of intracardiac shunts, 89Frederick M. Costello MD, George A. Stouffer MD

Part II Valvular heart disease

8 Aortic stenosis, 101David P. McLaughlin MD, George A. Stouffer MD

9 Mitral stenosis, 113Robert V. Kelly MD, Chadwick Huggins MD

10 Aortic regurgitation, 125George A. Stouffer MD

iii

iv Contents

11 Mitral regurgitation, 135Robert V. Kelly MD, Mauricio G. Cohen MD

12 The tricuspid valve, 145David A. Tate MD

Part III Cardiomyopathies

13 Hypertrophic cardiomyopathy, 155Jayadeep S. Varanasi MD, George A. Stouffer MD

14 Heart failure, 169Steven Filby MD, Patricia P. Chang MD

15 Restrictive cardiomyopathy, 179David P. McLaughlin MD, George A. Stouffer MD

Part IV Pericardial disease

16 Constrictive pericarditis, 185David P. McLaughlin MD, George A. Stouffer MD

17 Cardiac tamponade, 197Siva B. Mohan MD, George A. Stouffer MD

18 Effusive—constrictive pericarditis, 209Eric M. Crespo MD, Sidney C. Smith MD

Part V Miscellaneous

19 Right ventricular myocardial infarction, 215Robert V. Kelly MD, Mauricio G. Cohen MD

20 Pulmonary hypertension, 223Daniel Fox BS, David P. McLaughlin MD, George A. Stouffer MD

21 Coronary hemodynamics, 233David P. McLaughlin MD, Samuel S. Wu MD, George A. Stouffer MD

22 Hemodynamics of intra-aortic balloon counterpulsation, 245Richard A. Santa-Cruz MD

23 Arrhythmias, 255Lukas Jantac MD, George A. Stouffer MD

Contents v

24 Hemodynamics of pacemakers, 265Rodrigo Bolanos MD, Kimberly A. Selzman MD

25 Unknowns, 275George A. Stouffer MD

Appendices, 289

Index, 295

Contributors

Rodrigo Bolanos, MDFellow in Cardiovascular MedicineUniversity of North Carolina at Chapel HillChapel Hill, NC, USA

Patricia P. Chang, MD, MHS, FACCAssistant Professor of Medicine (Cardiology)Director, Heart Failure and Transplant ServiceDivision of CardiologyThe University of North Carolina at Chapel HillChapel Hill, NC, USA

Mauricio G. Cohen, MDAssistant Professor of MedicineAssociate DirectorCardiac Catheterization LaboratoryDivision of CardiologyDepartment of MedicineUniversity of North Carolina at Chapel HillChapel Hill, NC, USA

Frederick M. Costello, MDCardiologistUniversity of North Carolina at Chapel HillDivision of CardiologyChapel Hill, NC, USA

Eric M. Crespo, MD, MPHCardiologistUniversity of North Carolina at Chapel HillDivision of CardiologyChapel Hill, NC, USA

Eron D. Crouch, MDCardiovascular Medicine(Invasive, Non-interventional)Trinity Clinic – CorsicanaTrinity Mother Frances HospitalTyler, TX, USA

James E. Faber, PhDProfessorDepartment of Cell and Molecular PhysiologyUniversity of North CarolinaChapel Hill, NC, USA

Steven Filby, MDUniversity of North Carolina at Chapel HillSchool of MedicineChapel Hill, NC, USA

Daniel Fox, MDUniversity of North Carolina at Chapel HillSchool of MedicineChapel Hill, NC, USA

Chadwick Huggins, MDUniversity of North Carolina at Chapel HillSchool of MedicineChapel Hill, NC, USA

Lukas Jantac, MDUniversity of North Carolina at Chapel HillSchool of MedicineChapel Hill, NC, USA

Alison Keenan, BSUniversity of North Carolina at Chapel HillSchool of MedicineChapel Hill, NC, USA

Robert V. Kelly, MD, FACC, FESC, MRCPIAssistant Professor of MedicineAttending CardiologistDivision of CardiologyUniversity of North Carolina at Chapel HillChapel Hill, NC, USA

David P. McLaughlin, MDFellow, Interventional CardiologyUniversity of North Carolina at Chapel HillChapel Hill, NC, USA

Siva B. Mohan, MDUniversity of North Carolina at Chapel HillSchool of MedicineChapel Hill, NC, USA

Richard A. Santa-Cruz, MDInterventional CardiologyAgnesian HealthCareFond du Lac, WI, USA

vii

viii Contributors

Kimberly A. Selzman, MD, FACCAssistant Professor of MedicineCardiac Electrophysiology and PacingMedical Director, EKG ServicesUniversity of North Carolina at Chapel HillChapel Hill, NC, USA

Sidney C. Smith, MDUniversity of North Carolina at Chapel HillChapel Hill, NC, USA

Vickie Strang, RN, BS, CCRN, CNIIIClinical NurseUniversity of North Carolina at Chapel HillUNC HospitalChapel Hill, NC, USA

David A. Tate, MDAssociate Professor of MedicineUniversity of North Carolina at Chapel HillChapel Hill, NC, USA

Jayadeep S. Varanasi, MDUniversity of North Carolina at Chapel HillChapel Hill, NC, USA

Samuel S. Wu, MDClinical FellowUniversity of North Carolina at Chapel HillSchool of MedicineChapel Hill, NC, USA

Preface

This book, a collaborative effort of the Cardiology faculty and fellows at theUniversity of North Carolina, is intended to be a reference for understand-ing the practical application of hemodynamics in clinical medicine. We haveattempted to provide a comprehensive resource for the clinician to use inextracting maximum information from pressure and blood flow in patientswith cardiovascular disease. A basic overview of circulatory physiology andcardiac function is followed by detailed discussion of pathophysiologicalchanges in various disease states. This book is not intended for the researcherin the field but rather concentrates on providing clinically useful informationin a disease-based framework.

The initial section reviews the principles underlying hemodynamics andincludes a brief description of the basic formulas necessary to understandhemodynamics, a discussion of the nuts and bolts of right heart catheteriza-tion, an overview of normal hemodynamics and then primers on interpretingarterial pressure tracings, atrial pressure tracings and cardiac output measure-ment. The remainder of the book is divided into sections on various clinicalentities including valvular heart disease, cardiomyopathies, pericardial dis-ease, intracardiac shunt detection, coronary blood flow, intra-aortic ballooncounterpulsation, arrhythmias and cardiac pacing. The emphasis of these chap-ters is on clinically useful and accurate interpretation of hemodynamic datain various disease states based on an understanding of the underlying patho-physiology.

Dedication

This book is dedicated to my wife Meg and children (Mark, Jeanie, Joy andAnna) for all the joy they bring into my life.

George A. Stouffer, MD

ix

PART I

Basics of hemodynamics

CHAPTER 1

Introduction to basic hemodynamicprinciples

James E. Faber, George A. Stouffer

Hemodynamics is concerned with the mechanical and physiologic propertiescontrolling blood pressure and flow through the body. A full discussion ofhemodynamic principles is beyond the scope of this book. In this chapter,we present an overview of basic principles that are helpful in understandinghemodynamics.

1. Energy in the blood stream exists in threeinterchangeable forms: pressure arising from cardiacoutput and vascular resistance, hydrostatic pressurefrom gravitational forces, and kinetic energy of blood flow

Daniel Bernoulli was a physician and mathematician who lived in the eigh-teenth century. He had wide-ranging scientific interests and won the GrandPrize of the Paris Academy 10 times for advances in areas ranging from astron-omy to physics. One of his insights was that the energy of an ideal fluid (ahypothetical concept referring to a fluid that is not subject to viscous or fric-tional energy losses) in a straight tube can exist in three interchangeable forms:perpendicular pressure (force exerted on the walls of the tube perpendicularto flow; a form of potential energy), kinetic energy of the flowing fluid, andpressure due to gravitational forces. Perpendicular pressure is transferred tothe blood by cardiac pump function and vascular elasticity and is a functionof cardiac output and vascular resistance.

Total energy (TE) = potential energy + kinetic energyTE = (perpendicular pressure + gravitational pressure) + kinetic energyTE = (PPer + Pgrav) + 1/2 �V2

where V is velocity and � is blood density (approximately 1060 kg/m3)

TE = PPer + (� × h × g) + 1/2 �V2

where g is gravitational constant and h is height of fluid.Although blood is not an “ideal fluid” (in the engineering sense), Bernoulli’s

insight is helpful. Blood pressure is the summation of three components—lateral pressure, kinetic energy (also known as the impact pressure or the

3

4 Part 1 Basics of hemodynamics

pressure required to cause flow to stop), and gravitational forces. Kineticenergy is greatest in the ascending aorta where velocity is highest but eventhere it contributes less than 5 mm Hg of equivalent pressure.

Gravitational forces are important in a standing person. Arterial pressure inthe foot will greatly exceed thoracic aortic pressure due to gravitational pullon a column of blood. Likewise, arterial pressure in the head will be less thanthoracic aortic pressure. Similarly, gravitational forces are important in thevenous system as blood will pool in the legs when an individual is standing.This lowers cardiac output and explains why a person will feel lightheadedif rising abruptly from a sitting or supine position. In contrast, gravity hasminimal impact when a person is lying flat. Gravitational pressure equals theheight of a column of blood × the gravitational constant × the fluid density. Tocalculate hydrostatic pressure at the bedside (in mm Hg), measure the distancein millimeters between the points of interest and divide by 13 (mercury is 13times denser than water).

Pressure is the force applied per unit area of a surface. In blood vesselsor in the heart, the transmural pressure (i.e. pressure across the vessel wall orchamber wall) is equal to the intravascular pressure minus the pressure outsidethe vessel. The intravascular pressure is responsible for transmural pressure(i.e. vessel distention) and for longitudinal transport of blood.

2. Blood flow is a function of pressure gradientand resistance

One of the properties of a fluid is that it will flow from a region of higherpressure (e.g. the left ventricle) toward a region of lower pressure (e.g. theright atrium) (Figure 1.1). In clinical practice, the patient is assumed to besupine (negating the gravitational component of pressure) and at rest (kineticenergy is negligible compared to blood pressure at normal cardiac output) andthus blood flow is estimated using pressure gradient and resistance.

The primary parameter used in clinical medicine to describe blood flow iscardiac output, which is the total volume of blood pumped by the ventricle

valve

r

bloodpressure

Flow

Figure 1.1 A simple hydraulic system demonstrating fluid flow from a high-pressure reservoir to alow-pressure reservoir. Note that the volume of flow can be controlled by a focal resistance (i.e.the valve).

Chapter 1 Introduction to basic hemodynamic principles 5

per minute (generally expressed in L/min). Cardiac output is equal to the totalvolume of blood ejected into the aorta from the left ventricle (LV) per cardiaccycle (i.e. stroke volume) multiplied by heart rate. This formula is importantexperimentally but of limited used clinically because stroke volume is diffi-cult to measure. Cardiac output in the catheterization laboratory is generallymeasured using the Fick equation or via thermodilution techniques, which arediscussed in Chapter 6.

To compare cardiac output among individuals of different size, cardiac index(cardiac output divided by body surface area) is used. Normalization of cardiacoutput for body surface area is important as it enables proper interpretation ofdata independent of the patient’s size (e.g. cardiac output will obviously differwidely between a 260-pound man and a 100-pound woman). Indexing to bodysurface area is also used for other measurements such as aortic valve area.

The relationship between blood flow, resistance, and pressure can be deter-mined using a modification of Ohm’s law for the flow of electrons in an elec-trical circuit:

Flow(Q) = pressure gradient (�P)/resistance(R)

where �P is the difference in pressure between proximal and distal points inthe system and R is the hydraulic resistance to blood flow.

A useful clinical equation based on Ohm’s law is

Mean arterial pressure (MAP) − central venous pressure (CVP)

= cardiac output (CO) × systemic vascular resistance (SVR)

Using this equation, we can calculate systemic vascular resistance knowingcardiac output, CVP, and arterial pressure. MAP is the average arterial pres-sure over time and is generally estimated using MAP = (1/3 × aortic systolicpressure) + (2/3 × aortic diastolic pressure). This formula was developed forheart rate = 60 bpm (at heart rate = 60 bpm, diastole is twice as long as systole)but becomes progressively more inaccurate as heart rate increases. In a patientin shock (i.e. low blood pressure and impaired tissue perfusion), measurementof CO and calculation of SVR can help identify the etiology (e.g. septic shockwith high CO + low SVR or cardiogenic shock with low CO + high SVR).

3. Resistance to flow can be estimated usingPoiseuille’s law

Blood is not an “ideal fluid” and energy (and pressure) is lost as flowing bloodovercomes resistance. Resistance to blood flow is a function of viscosity, ves-sel radius, and vessel length. The relationship is known as Poiseuille’s law(sometimes referred to as the Poiseuille–Hagen law) and is described by thefollowing equation:

Resistance = 8 × viscosity × length/� × radius4


or since flow = difference in pressure/resistance:

Blood flow = � × radius4 × difference in pressure/8 × viscosity × length

Since radius is raised to the fourth power, its importance in determining resis-tance is paramount. A 20% increase in radius leads to a doubling in flow ifall other variables are constant. Or as another example, resistance is 16 timesgreater in a coronary artery with a diameter of 2 mm (e.g. a distal obtusemarginal) than in a coronary artery with a diameter of 4 mm (e.g. the proximalleft anterior descending).

Viscosity is also important in determining resistance (commonly abbreviatedas � and has units of poise = dyne s/cm2). It is difficult to measure directlyand thus is commonly reported as relative to water. The viscosity of plasma is1.7 × viscosity of water and viscosity of blood is 3–4 × viscosity of water, thedifference being due to blood cells and particularly hematocrit.

It is important to note that Poiseuille’s law only provides an approximationof resistance when used in blood vessels. The four important assumptionsunderlying the derivation of this equation are: (1) the viscosity of the fluid isunchanging over time or space; (2) the tube is rigid and cylindrical; (3) lengthof the tube greatly exceeds diameter; and (4) flow is steady, nonpulsatile, andnonturbulent. Many of these assumptions are violated when this equation isapplied to blood flow in the body. Poiseuille’s law is important, however, as itindicates the variables that are determinants of resistance to flow.

In the mammalian circulation, resistance is greatest at the level of the arteri-oles. While radius of a typical capillary is smaller than the radius of an arteriole,the number of capillaries greatly exceeds the number of arterioles, and thus theeffective area is much larger. Also of importance is that arteriolar resistance canbe regulated (capillaries have no smooth muscle and thus resistance cannot beregulated at that level). This enables rapid changes in vascular resistance tomaintain blood pressure (e.g. in hypovolemic shock) and also enables regula-tion of blood flow to various organs (i.e. autoregulation). A general principleto remember is that reduction of arteriolar resistance decreases SVR result-ing in an increased cardiac output while decreasing pressure proximal to thearterioles and increasing pressure distal to the arterioles.

4. Reynold’s number can be used to determine whetherflow is laminar or turbulent

Flow in blood vessels, as in any hydraulic system, is usually smooth and orderlybecause the fluid separates into an infinite number of concentric layers withdifferent velocities. When a fluid (such as blood) flows past a solid surface(such as the vascular wall), a thin layer develops adjacent to the surface wherefrictional forces retard the motion of the fluid (Figure 1.2). There is a gradientof frictional resistance (and thus velocity) between fluid in contact with thesolid surface and fluid in the center of the stream. If the fluid particles travel


Laminar(streamline)

Turbulent

Figure 1.2 Schematic of laminar flow. Flow in straight, nonbranching tubes is usually smooth andorderly because the fluid separates into an infinite number of concentric layers with differentvelocities. When a fluid (such as blood) flows past a solid surface (such as the vascular wall), athin layer develops adjacent to the surface where frictional forces tend to retard the motion of thefluid. There is a gradient of frictional resistance (and thus velocity) between fluid in contact with thesolid surface and fluid in the center of the stream. If the fluid particles travel along well-orderednonintersecting layers this is termed laminar flow.

along well-ordered nonintersecting layers this is termed laminar flow. The flowresistance in laminar flow is due entirely to viscous resistance. In laminar flow,the average velocity of a fluid is one half of the maximum velocity observedin the center of the stream.

In contrast, turbulent flow is where fluid particles from adjacent layersbecome mixed. Turbulent flow is chaotic and less efficient because of energylosses (these losses are termed inertial resistance). In turbulent flow, the rela-tionship between pressure difference and flow is no longer linear since resis-tance increases with flow. Thus, larger pressure differences are required tomaintain flow. Turbulence, and associated loss of energy, is one of the causesof a drop in pressure distal to a severe stenosis.

Turbulence is important for several reasons, one of which is that it createsnoise, which is the cause of some cardiac murmurs and the Korotkoff sounds(used when measuring blood pressure; Figure 1.3). Another is that turbu-lence alters the relationship between flow and perfusion pressure. Because of

160

140

120

100

80

60

40

20

0

Cuff pressure (mm Hg)

Art

eria

l pre

ssur

e (m

m H

g)

Sounds (from turbulent blood flow)heard distal to the cuff

Cuff pressure > Systolic pressureNo flow distal to cuff

No sounds

Cuff pressure < diastolic pressure Flow distal to cuff is laminar

No sounds

Systolic > cuff > diastolic pressuresFlow distal to cuff is turbulent,

producing sounds

Figure 1.3 Schematic illustrating the use of transient transition from laminar to turbulent flow inmeasuring blood pressure.


35

30

25

20

15

10

5

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10

Δp(m

m H

g)

Q(L/min)

TurbulentPoiseuille

,s law

Figure 1.4 Transition from laminar to turbulent flow. Note that the pressure gradient required toincrease flow increases markedly when flow transitions from laminar to turbulent.

increased energy losses associated with turbulence, the relationship betweenperfusion pressure and blood flow is no longer linear (as described by thePoiseuille relationship) but rather greater pressure is required to maintain ade-quate flow (Figure 1.4).

The transition from laminar to turbulent flow can be predicted by calculatingthe Reynold’s number, which is the ratio of inertial forces (V�) to viscous forces(� /L):

R = diameter × velocity × density/viscosity

where viscosity (�) of blood is 0.004 Pa s, density (� ) of blood is approximately1050 kg/m3, velocity (V) of blood is in m/s, and the diameter of the tube is inmeters. Reynold’s number is dimensionless.

In a given hydraulic system there exists a critical Reynold’s number belowwhich flow is laminar. At Reynold’s numbers near this critical number, atransitional zone exists where flow is neither completely laminar nor turbu-lent. Higher Reynold’s numbers are associated with turbulent flow. In a long,straight, nonbranching pipe with nonpulsatile flow, flow is generally laminarif R < 2000 and turbulent if R > 2000. It is important to note that the Reynold’snumber depends on the exact flow configuration and must be determinedexperimentally.

In the aorta, transition from laminar to turbulent flow generally occurs at aReynold’s number between 2000 and 2500. In atherosclerotic arteries and/orat branch points, the critical Reynold’s number is much lower and there canbe turbulence even at normal physiological flow velocities. In severe stenoses,turbulence can be initiated at Reynold’s numbers an order of magnitude lessthan in the theoretical, straight pipe.


The Reynold’s equation is important for demonstrating variables impor-tant in determining whether flow is laminar or turbulent. As a simpleapproximation from this equation we see that laminar flow is difficult to main-tain in conditions of high velocity (e.g. stenotic artery) and large diameter.Vessel diameter is doubly important for not only is it a direct variable in theequation, it also influences velocity. Because of the continuity equation (seeitem 9 below) we know that velocity increases as diameter decreases. Thus,effects of blood vessel diameter on Reynold’s number are magnified. Becauseboth velocity and diameter decrease in the microcirculation, the flow theretends to be laminar.

5. Force developed by the ventricles is a function of preloador stretch---the Frank -- Starling law

The three most important factors in the regulation of ventricular function (andthus cardiac output) are preload, afterload, and contractility. Preload for theventricles is defined as amount of passive tension or stretch exerted on theventricular walls (i.e. intraventricular pressure) just prior to the initiation ofsystole. This load determines end-diastolic sarcomere length and thus the forceof contraction. The Frank–Starling law states that the passive length to whichthe myocardial cells are stretched at the end of diastole determines the activetension they develop when stimulated to contract. The Frank–Starling law isan intrinsic property of myocytes and is not dependent upon extrinsic nervesor hormones. The general principle is that increased preload causes increasedforce of contraction, which increases stroke volume and thus cardiac output(Figure 1.5). The Frank–Starling law (or mechanism) helps the heart matchcardiac output to venous return.

The Frank–Starling law was derived from independent work by thesetwo investigators. In the 1890s, Otto Frank measured pressure developedby isolated beating frog ventricles against an occluded aorta under vary-ing preloading conditions. He found that as end-diastolic volume increased,ventricular systolic pressure and the maximum rate of pressure develop-ment (dP/dTmax) increased. Approximately 20 years later, Ernst Starling found

SV

SV

Preload Afterload

Normal

Normal

CTY

CTY CTY

CTY

Figure 1.5 Frank–Starling principle. In panel A, note that stroke volume (SV) increases aspreload increases on any given line of contractility (CTY). Panel B demonstrates that strokevolume increases as afterload decreases.


similar results using a heart–lung preparation in an anesthetized dog in whichhe controlled heart rate, venous pressure, venous return, and arterial resistance.

For any given heart, there is not just a single Frank–Starling curve. Ratherthere is a family of curves, each of which is determined by the afterload andinotropic state (i.e. contractility) of the heart. While changes in venous returncause a ventricle to move along a single Frank–Starling curve, changes incontractility and afterload cause the heart to shift to a different Frank–Starlingcurve.

6. Wall tension is a function of pressure and radius dividedby wall thickness---Laplace’s relationship

Laplace’s law describes the relationship between the transmural pressure dif-ference and the tension, radius, and thickness of the vessel wall or ventricularchamber (Figure 1.6). Simply stated, Laplace’s law is

Wall tension = pressure × radius/wall thickness

The pressure inside a blood vessel or ventricle exerts a distending force (ten-sion) on the walls proportional to the magnitude of the pressure and radius.Thus, wall tension on the aorta is high. In chronic hypertension, aortic wall

Figure 1.6 MRI image of a right ventricle and a left ventricle. The left ventricle is labeled todemonstrate the law of Laplace: t = (P × r)/W, where P is pressure; r is radius; W is wallthickness; and t is wall tension.


thickness increases as an adaptation to normalize wall tension. Similarly, LVhypertrophy develops in response to chronic pressure elevations and/or dila-tion, again as an adaptation.

There are several important implications of this relationship. One is thatlarger arteries must have stronger walls since an artery of twice the radius mustbe able to withstand twice the wall tension at a given blood pressure. Simi-larly, the increased wall tension is thought to contribute to the developmentof aneurysms (and possibly to predict aneurysm rupture) in larger arteries.Another implication is that as the radius of the LV increases (e.g. in dilated car-diomyopathy), increased active wall tension must be developed during systoleby the myocytes in order to create the same pressure. Thus, a dilated ventriclemust use more ATP and oxygen to generate the same stroke volume.

7. The normal venous system is a low pressure, largevolume reservoir of blood which enables cardiacoutput to increase rapidly

Approximately 50% of blood in the normal individual is in the venous system(with 15% being within the heart, 20% within the pulmonary circulation, 10%in arteries, and 5% in arterioles and capillaries). There is a large capacitance inthe normal venous system, and veins are partially collapsed. Thus, the venoussystem can absorb a large amount of volume with minimal increase in pressure.Once veins become fully distended, however, the pressure–volume relation-ship changes significantly (Figure 1.7). Veins have limited elasticity once theyare fully distended and at this point pressure increases rapidly with increasedvolume.

Pressure

Artery

Vein

Volume

Figure 1.7 Pressure–volume relationship for arteries and veins.


Since the cardiovascular system is a closed loop, venous return and cardiacoutput are closely coupled, and increased venous return to the heart is oneof the primary mechanisms by which cardiac output is increased rapidly. Forexample, venous return (and cardiac output) can be rapidly augmented byincreased sympathetic tone, which causes the smooth muscle in veins to con-tract. Similarly, skeletal muscular contractions during exercise reduce venouscapacitance in the muscle beds by rhythmically compressing the veins and canmarkedly increase venous return.

8. The pressure and velocity of a fluid in a closed systemare related

As we saw in item 1, Bernoulli derived the formula (now known as the Bernoulliequation) that relates the pressure, velocity, and height (i.e. gravitational forces)in the steady motion of an ideal fluid (i.e. a fluid without any viscosity and inwhich there are no frictional losses during flow). The usual form is TE = PPer +(� × h × g) + 1/2 �V2, where TE is the total energy, V is the blood velocity, PPer

the perpendicular pressure, � the blood density, g the gravitational constant,and h the height above an arbitrary reference level. It is based upon the law ofconservation of energy and states that the sum of potential and kinetic energyis the same at every point throughout a rigid tube.

The Bernoulli equation provides the theoretical foundation for the use ofpulse wave and continuous wave Doppler to estimate pressures. While theactual derivation is more complex, for practical use in Doppler echocardiogra-phy the Bernoulli equation is simplified to P1 − P2 = 4V2, where P is pressureand V is velocity (labeled the modified Bernoulli equation). Because of therelationship between velocity and pressure, Doppler-determined blood veloc-ity can be used to estimate pressures within the heart and vasculature (e.g.estimating pulmonary artery pressures at the time of echocardiography bymeasuring the velocity of tricuspid regurgitation).

Derivation of modified Bernoulli equation

Ignoring gravitational forces, the Bernoulli equation predicts that the relation-ship between pressures at two points (P1 and P2) within a system with a flowingfluid would be:

P1 + 1/2 �V2 = P2

Or stated another way, the difference in pressure between the two points wouldbe:

P2 − P1 = 1/2 �V2

Inserting units and blood density (� = 1050 kg/m3):

�P(kg/m s2) = 1/2(1050 kg/m3)V2(m2/s2)


Since 1 mm Hg = 133.3 Pa = 133.3 kg/m2

�P × 133.3 = 1/2(1050)V2

and thus we arrive at the formula we recognize:

�P(mm Hg) = 3.938 V2

commonly abbreviated to:

�P(mm Hg) = 4V2

9. The velocity of blood increases and pressure decreasesas cross-sectional area of the blood vessel decreases

An important hemodynamic concept is the continuity equation, which isderived from the law of conservation of mass. This equation is based on theprinciple that flow at any given point in a closed hydraulic system will be equalto flow at any other point. Thus, since flow is constant, velocity is inverselyproportional to the cross-sectional area (Figure 1.8).

Q = A1V1 = A2V2 = A3V3

(where A is area and V is velocity at any given point within the system).An implication of this equation is that velocity increases as cross-sectionalarea decreases (e.g. at the site of an arterial stenosis). Similarly, blood velocitydecreases as it flows from the aorta into the capillary system but then increasesagain as it coalesces from venules to veins to the vena cava. An average velocityof blood at any given point within the circulation can be calculated knowingthe volume of blood flowing past a given cross-sectional area of blood vessel.

The continuity equation is used in the echocardiography laboratory to esti-mate aortic valve area. The cross-sectional area of the LV outflow tract ismeasured along with blood velocity at that point (using pulse wave Doppler)

Qin = Qout

Relative velocity

Relative Areas

VV

VV

AA

A A

Qi Qo

Figure 1.8 Relationship between blood velocity and cross-sectional area of blood vessels.Because of the law of conservation of mass, flow at any given point in a closed hydraulic systemwill be equal to flow at any other point. Thus, velocity is inversely proportional to thecross-sectional area.


and then volumetric blood flow in the outflow tract is calculated. Using thecontinuity equation, the cross-sectional area of the valve can be calculated bydividing volumetric blood flow by the measured velocity at the valve.

10. Resistance increases when blood vessels areconnected in series and decreases when blood vesselsare connected in parallel

Poiseuille’s equation estimates resistance to flow in a single vessel. The humancardiovascular system, however, includes complex circuitry with distinct bloodvessels in series (connected one after another) and in parallel (arising from thedivision of a larger vessel). Blood ejected from the heart moves from aorta →large arteries → small arteries → arterioles → capillaries → venous system→ heart. While the aorta is a single vessel, the rest of the circulatory systeminvolves multiple vessels connected in parallel (e.g. the carotid arteries, renalarteries, and other major branches from the aorta form a parallel circuit).

For blood vessels connected in series, the total resistance of the system isequal to the sum of resistance in each vessel (Figure 1.9).

Rtotal = R1 + R2 + R3

Thus, the resistance of the system is always greater than resistance in anyone vessel. Thus, for three vessels in series, each with resistance R, the totalresistance of the system is 3R.

For blood vessels connected in parallel, resistance is equal to the sum of thereciprocal of the resistance in each vessel.

1/Rtotal = 1/R1 + 1/R2 + 1/R3

Thus, resistance of the system is always less than resistance in any vessel.For three vessels in parallel, each with resistance R, the total resistance of thesystem is R/3.

R3

R2

R1

A

R1R2

R3

Figure 1.9 Circuits (or blood vessels) in series and in parallel.


3000

2000

1000

0

120

80

Pre

ssur

e(m

m H

g)

40

0

20

10

2

0

Tota

l agg

rega

tecr

oss-

sect

iona

l are

a (c

m2 )

Area Velocity

Velocity (cm

/s)

Aor

ta

Art

erie

s

Art

erio

les

Cap

illar

ies

Ven

ules

Vei

ns

Ven

a ca

va

Relative

resistance

Relativeresistance

Psystolic

Pdiastolic

pressure(P )

Figure 1.10 Pressure, velocity, and flow in various parts of circulation.

An important principle is that more than 60% of the resistance to flowoccurs within the arterioles. The diameter of the arterial system progressivelydecreases from aorta → arteriole. Energy losses (pressure) are minimized inthe larger arteries, despite decreases in diameter, by having many arteriesin parallel. The large pressure loss in arterioles is due to a dramatic decrease indiameter (Figure 1.10) without a compensatory increase in the number of arte-rioles in parallel. It goes without saying that the energy loss in arterioles servesmany important functions such as lowering pressure and velocity in capillar-ies to allow optimal transit time for red blood cells for diffusion of O2 andCO2.

Since the greatest resistance occurs primarily within the arterioles, systemicvascular resistance is sensitive to any change in arteriolar constriction. Assum-ing constant cardiac output (keep in mind that cardiac output is a dynamicprocess and may increase or decrease by mechanisms other than changes invascular resistance), agents that dilate arterioles in a tissue bed will decreaseblood pressure (i.e. pressure proximal to the arterioles) but also increase pres-sures in the capillary bed. Similarly, vasoconstrictors increase blood pressurebut reduce pressure within capillaries. In real life, cardiac output does notremain constant. The use of vasopressors to maintain blood pressure willincrease vascular resistance but at a potential cost of decreased cardiac out-put (e.g. in the patient with cardiogenic shock), organ failure due to increasedvascular resistance (e.g. worsening renal function in patients on vasopressors),and increased afterload on the heart.

CHAPTER 2

The nuts and bolts of right heartcatheterization and PA catheterplacement

Vickie Strang

The pulmonary artery catheter

The pulmonary artery (PA) catheter (also known as the Swan–Ganz catheteror right heart catheter) was developed in the 1970s by Dr Harold Swan, DrWilliam Ganz, and colleagues [1]. The typical PA catheter is 100–110 cm longand has either 3 or 4 lumens (one lumen is used to inflate the balloon and thusthere is one less port than the number of lumens). There is a proximal portapproximately 30 cm from the tip that generally lies in the right atrium (RA)and can be used to transducer pressure or as an infusion port. The distal portis at the tip of the catheter and is used to measure PA pressure and pulmonarycapillary wedge pressure (PCWP). Near the tip is a balloon that can be inflatedand a thermistor to measure temperature (Figure 2.1).

The PA catheter provides information about ventricular preload (e.g. RApressure is a reflection of right ventricular (RV) preload and PCWP is a reflec-tion of left ventricular (LV) preload), afterload (systemic vascular resistanceand pulmonary vascular resistance), and cardiac output. The PA catheter canbe helpful in various clinical scenarios including valvular heart disease, con-gestive heart failure, cardiomyopathy, pericardial tamponade, shock, renalfailure, pulmonary edema, pulmonary hypertension, or cardiac structuralabnormalities.

PA catheters are used primarily in three different settings—in the cardiaccatheterization laboratory, in intensive care units, and in the operating room.Despite the tremendous theoretical benefits that could accrue from having theinformation obtainable from PA catheters, these catheters have never beenshown to improve patient outcomes in either the operating room or intensivecare unit (potential benefit has never been studied in the cardiac catheteri-zation laboratory). In numerous studies examining various groups such aspatients with heart failure, patients undergoing high-risk noncardiac surgery,and patients with acute respiratory distress syndrome (ARDS), PA cathetershave in general had no beneficial effects on survival with increased rates ofcomplications (Table 2.1) [2]. These studies have been criticized for several

17


Distal port

Balloon

Thermistor

VIP venousinfusion port (RV)

Proximalinjectate port

Balloonsyringe

Balloon

inflationstopcock

Opticalconnector

Thermistorconnector

VIP

venousinfusionport

CVPproximal

injectate

PA distal port

Figure 2.1 Schematic of PA catheter.

reasons including improper patient selection (e.g. including low-risk patientswho would not be expected to benefit), study design (e.g. expecting a moni-toring tool to affect outcomes without specified treatment protocols), and notcontrolling for operator experience in either insertion of the catheter or ininterpretation of data. Currently, there is no clear consensus on whether PAcatheters are beneficial or harmful with articulate proponents on both sides.PA catheters remain in widespread use presumably because they provide ben-efit to individual patients but the studies do sound a note of caution againstindiscriminate use of these catheters.

The effectiveness of a PA catheter depends on accurate assessment of theinformation provided. Waveform analysis is crucial for proper interpretationof data from a PA catheter and is discussed in detail in Chapters 3, 4, and 5. Inthis chapter, we will concentrate on the following topics:� Brief review of physiology relevant to right heart and PA catheterization� Vascular access� Right heart catheterization and placement of a PA catheter� Ensuring that accurate data are obtained from a PA catheter� Cardiac output� Calculating vascular resistance� SvO2 monitoring� Complications of PA catheters

Brief review of physiology relevant to right heart and pulmonaryartery catheterizationLet us begin by looking at one mechanical cardiac cycle. While the ventriclesare in systole the atria are filling as blood flows continually from the venoussystem with the inferior and superior vena cava emptying into the RA and thepulmonary veins emptying into the left atrium (LA). Meanwhile the tricuspid

Chapter 2 PA catheter placement 19

Tab

le2.

1S

ampl

eof

tria

lsth

atha

veex

amin

edth

eef

fect

ofPA

cath

eter

use

onm

orta

lity

incr

itica

llyill

patie

nts

Stu

dyP

opul

atio

nN

Type

ofst

udy

Mor

talit

yw

ithPA

Cvs

with

outP

AC

SU

PP

OR

T=

[3]

Med

ical

and

surg

ical

ICU

patie

nts

5735

Pro

spec

tive

coho

rtst

udy

Incr

ease

dm

orta

lity

with

PAC

:odd

sra

tio,1

.24;

95%

confi

denc

ein

terv

al,1

.03–

1.49

San

dham

JD[4

]H

igh

risk

surg

ical

patie

nts

1994

Ran

dom

ized

cont

rolle

dtr

ial

No

diffe

renc

ein

mor

talit

yR

icha

rdC

[5]

Sep

ticsh

ock

and/

orA

RD

S67

6R

ando

miz

edco

ntro

lled

tria

lN

odi

ffere

nce

inm

orta

lity

ES

CA

PE

[6]

Hea

rtfa

ilure

433

Ran

dom

ized

cont

rolle

dtr

ial

No

diffe

renc

ein

mor

talit

y;in

-hos

pita

ladv

erse

even

tsw

ere

mor

eco

mm

onam

ong

patie

nts

inth

ePA

Cgr

oup

PAC

-Man

[7]

Inte

nsiv

eca

reun

itpa

tient

s10

41R

ando

miz

edco

ntro

lled

tria

lN

odi

ffere

nce

inm

orta

lity

Flu

idan

dC

athe

ter

Trea

tmen

tTria

l(F

AC

TT

)[8

]

Acu

telu

ngin

jury

1000

Ran

dom

ized

cont

rolle

dtr

ialw

ithex

plic

itm

anag

emen

tpro

toco

lN

odi

ffere

nce

inm

orta

lity,

ICU

days

ortim

eon

vent

ilato

r


valve and the mitral valve remain closed as long as the pressure in the ventriclesexceeds the pressure in the atria.

Blood flow in the heart occurs due to pressure gradients. As ventricularpressure decreases during isovolumetric ventricular relaxation (Figure 2.2a),it reaches a point where atrial pressure is higher than ventricular pressureand the mitral and tricuspid valves open due to hydrostatic pressure. Theventricles now rapidly fill (rapid ventricular filling; Figure 2.2b). This rapidfilling characterizes the first phase of diastole. During the second phase ofdiastole, the pressure in the atria and the ventricles are the same. During thistime only a small amount of blood normally flows into the ventricles (slowventricular filling). This is primarily blood returning to the atria from the greatveins or pulmonic veins, which then passes into the ventricle. Near the endof diastole, the atria contract (atrial systole) which generates an increase inpressure in the atrium and enhances blood flow into the ventricles (Figure 2.2c).

Systole begins when electrical activation of the ventricles (QRS complex)leads to mechanical activation of the ventricles. Very shortly after onset ofventricular contraction the pressure in the ventricle increases to the point whereit exceeds atrial pressure and the mitral and tricuspid valves close (all fourvalves are now closed). As the ventricles contract (isovolumetric contraction;Figure 2.2d), right and left ventricular pressures increase until RV pressureexceeds PA pressure and LV pressure exceeds aortic pressure. At these points,the pulmonic and aortic valves open and ejection occurs (ventricular ejection;Figure 2.2e). At the end of systole, isovolumetric ventricular relaxation beginsand ventricular pressures fall.

A point to remember is that as heart rate increases, myocardial oxygen con-sumption increases while supply decreases. This is because time in diastoledecreases and time in systole increases as heart rate increases while most coro-nary artery filling occurs during diastole but oxygen consumption of the heartis maximal during systole. About 2/3 of blood flow in the left coronary arteryand 1/2 of blood flow in the right coronary artery occurs during diastole.During systole intramural arteries and capillaries are compressed by the con-tracting heart muscle, which limits blood flow.

Vascular accessA PA catheter can be inserted via any vein that has uninterrupted access to theRA. The most commonly used insertion sites include subclavian veins, internaljugular veins, and femoral veins. The site is generally chosen based on patientpreference and operator experience. All things being equal, the right internaljugular vein provides the shortest and straightest path to the heart. Compli-cations associated with venous access include bleeding, hematoma, arterialpuncture, infection, pneumothorax, and hemothorax (Table 2.2). Advantagesand disadvantages of various access sites are listed in Table 2.3.

Right heart catheterization and placement of a PA catheterA PA catheter can be positioned using either fluoroscopy or pressure moni-toring. Fluoroscopy is preferred for femoral artery insertion and/or if there is


Ventricular ejection

Isovolumetric relaxation

Ventricular filling

Atrial systole

Isovolumetric contraction

(a)

(b)

(c)

(d)

(e)

Figure 2.2 Schematic of blood flow in the heart. Phases represented include isovolumetricventricular relaxation (a), rapid ventricular filling (b), atrial contraction (c), isovolumetric contraction(d), and ventricular ejection (e).


Table 2.2 Complications of PA catheters

Complications ComplicationsComplications associated with PA associated with Problems causingassociated with catheter insertion or indwelling PA inaccurate datavascular access removal catheters interpretation

Hematoma Arrhythmias Damage to tricuspid or Improper calibrationpulmonic valve Improper leveling of transducer

Bleeding Damage to tricuspid Infection Air in tubingor pulmonic valve Interpretation without using ECG

Arterial Infection Thromboembolic Effects of respiration not notedpuncture

Pneumothorax Pulmonary infarction Pulmonary infarction Inaccurate computationRight bundle branch constant used to determine

block cardiac outputHemothorax Pulmonary artery Pulmonary artery Use of improper pressure scale

rupture rupture

Table 2.3 Advantages and disadvantages of venous access sites

Site Advantages Disadvantages

Subclavian vein Patient mobility; sterility Risk of pneumothorax(especially in patients withCOPD or on positivepressure ventilation);noncompressible site

Internal jugular vein Compressible site; patientmobility; easily accessible

Inadvertent puncture ofcarotid; catheter can bedislodged from pulmonarycapillary wedge positionwith neck movement; risk ofpneumothorax

Femoral vein Compressible site; easilyaccessible

Patient must remain in bed;infection risk

significant hardware in the right heart (e.g. biventricular pacer). If continuouspressure monitoring is used, transduce the distal tip. Insert the PA catheter20 cm and inflate the balloon with air. Central venous pressure (CVP) is ini-tially obtained. As the catheter is being “floated” into the heart it first entersthe RA and then it goes into the RV. Atrial waveforms consist of an A wave,a V wave, a C wave, and X and Y descents (Figure 2.3a). The A wave in theRA waveform coincides with atrial contraction and can be found under the PRinterval on the ECG whereas the peak of the V wave in the RA waveform canbe found near or at the end of the T wave. A further discussion of interpretationof atrial waveforms can be found in Chapter 5.

Entry of the catheter into the RV can be identified by the appearance ofventricular systole on the pressure tracing (Figure 2.3b). There will be a rapid


1mV1mV

1mV1mV

1mV1mV

1mV1mV

1mV1mV

1mV1mV

IIIV5

(a)

(b)

(c)

(d)

25

r r r r r

a a

r r r r

20

15

10

5

01 2 3 4 5 6 7 8 9

SERIES: IVMET MEDICALS

SERIES: IVWITH MEDICALS

SERIES: IVWITH MEDICALS

0.19 mmHgmm

0.38 mmHgmm

0.38 mmHgmm

25mm/s >[-15-]

25mm/s >[-17-]

25mm/s >[-23-]

IIIV5

IIIV5

IIIV5

r

r

ss

ss

s s s

s

d

aa

aa a a a a

vv

vv

vv

v

v

r r r r r r r

dd

d d d d

r r r r r r r

r

ss s s s s s s s

e

dd d

d RV

PA

PWP

RA

d d d d d

r r r r r r r

50

0

50

50

25

01 2 3 4 5 6 7 8 9

25

01 2 3 4 5 6 7 8 9

25

e e e e e e e e

1mV1mV

1mV1mV

1mV1mV

1mV1mV

1mV1mV

1mV1mV

Figure 2.3 Pressure tracings encountered during the insertion of a PA catheter.


upstroke reflective of ventricular contraction and a rapid decline reflective ofventricular relaxation. Diastole will be similar to the atrial tracing. A further dis-cussion of interpretation of ventricular waveforms can be found in Chapter 3.Normal RV systolic pressure is approximately 25 mm Hg, with diastolic pres-sures ranging from 0 to 8 mm Hg. The only time you should see an RV tracingis when the PA catheter is being inserted. A RV pressure tracing from the distalport of the catheter appearing on the monitor at any other time, is an indicationthat the catheter is no longer correctly placed. With the tip of the catheter in theRV, the risk of ventricular tachycardia is markedly increased. Note that somecontinuous cardiac output catheters use a proximal port in the RV to monitorfor correct positioning of the thermistor – the warning here is that if the pres-sure from the distal port of the catheter changes from PA to RV that impliesthat the catheter tip is now residing in the RV.

As the catheter is advanced, it will leave the RV and enter the PA (Figure 2.3c).Remember that RV and PA systolic pressures are the same (both should coin-cide with the T wave on the ECG). Thus passage into the PA will be evidentfrom diastolic pressures. If a RV waveform is still present approximately 20 cmafter the initial RV pattern appears, the catheter may be coiling in the RV.

PCWP can be obtained by inflating the balloon on the catheter and advanc-ing the catheter until it “wedges” (Figure 2.3d). When the balloon is “wedged,”blood flow is stopped in that portion of the PA. Pressure equalizes in a non-flowing segment and since there are no valves between the pulmonary arteriesand pulmonary veins or between the pulmonary veins and the LA, pressureat the tip of the balloon is reflective of LA pressure. The A and V waves ofthe PCWP tracing will be delayed relative to RA tracing and ECG with the Awave starting in or at the end of the QRS complex, and the V wave startinglater in the T–P interval. This is because LA pressure is reflected back to thePA catheter via the pulmonary veins and through a static column of blood.PCWP is an important measurement since it is an estimate of LV end diastolicpressure and thus LV preload.

Tempe et al. [9] measured the distance to various chambers in 300 adultpatients who had PA catheters placed via the right internal jugular vein. Theright ventricle was reached at 25 ± 3 cm, the PA at 36 ± 4 cm, and wedgeposition at 43 ± 6 cm. The length of catheter required was directly related toheight of the patient and greater in patients with valvular heart disease.

In general, indwelling PA catheters rest in the PA (Figure 2.4), continuouslygiving a read out of PA systolic and PA diastolic pressure. Mean RA pressurecan be obtained if the CVP port is transduced.

Always deflate the balloon before withdrawing the PA catheter.

Ensuring that accurate data is obtained from a PA catheterRemember the old saying—”Trash in equals trash out.” The integrity of the dataobtained from a PA catheter is dependent upon the accuracy of the system. Afew simple rules to start out with:� The PA catheter needs to be “zeroed” at the level of the RA.


Figure 2.4 Chest X-ray showing placement of a PA catheter. The catheter was inserted via theright internal jugular vein. It passes through the superior vena cava (black arrow), RA, and RV. Thetip rests in the right PA (white arrow).

� Minimize the length of pressure tubing, the number of stopcocks, and thenumber of connections. Avoid narrow tubing as the smaller the diameter ofthe tubing, the more pressure damping that occurs.� Eliminate any air bubbles in the pressure line.� The proper pressure scale needs to be used� Pressure tracings need to be interpreted in conjunction with an ECG tracing.� The effects of respiration have to be identified.It is essential that the PA catheter be properly “zeroed” at the level of theRA, which in the supine patient is assumed to be in the fourth intercostalspace at the midaxillary line. Proper “zeroing” removes the effect of gravityon pressures. To locate the level of the RA, two imaginary lines are drawn(Figure 2.5) and the place where they intersect is called the phlebostatic axis.To locate the phlebostatic axis; locate the fourth intercostal space on the edge ofthe sternum, draw an imaginary line along the fourth intercostal space laterally,along the chest wall, and draw a second line from the axilla downward, midwaybetween the anterior and posterior chest wall (i.e. mid-axillary line).

For every centimeter that the transducer is not correctly leveled, pressureswill change by 0.75 mm Hg. If the transducer is too high, pressures will be


Phlebostatic axis

Figure 2.5 Phlebostatic axis.

falsely low. Alternatively, if the transducer is too low pressures will be falselyhigh.

Pressure is transmitted from the catheter to the transducer via tubing filledwith fluid, usually saline. Small changes in pressure in the PA are transmittedthrough the tubing and cause deflection of the transducer membrane, whichconverts these changes into electrical signals. Anything, which hampers trans-mission of these small pressure changes to the membrane, will distort pres-sure measurements and cause pressure “damping.” Thus, it is important touse semirigid, noncompliant tubing, to minimize the length of tubing and tomake sure that there is no air in the pressure tubing. Other causes of pressuredamping include long or compliant tubing, multiple stopcocks, transducermalfunction, loose connections, and partial occlusion of the port (e.g. withthrombus).

It is important that the proper scale be used when interpreting the pressuretracings. For example, when measuring LV pressure a scale of 0–200 mm Hgis useful. However, if this scale is used when examining RA pressure tracingsmuch of the information is lost that would be apparent if a scale of 0–25 mm Hgwas used.

It is important that pressure tracings be interpreted in conjunction with anECG recording. The most accurate method of obtaining hemodynamic data isto use a dual channel strip recorder, which will allow analysis of the waveformand ECG together.

When interpreting atrial pressure waveforms, emphasis can be placed onmean pressures or pressures obtained at end-expiration. In many intensive careunits, all waveforms are read at end-expiration because this is when intrapleu-ral pressures are negligible which means that the waveforms are an accuratereflection of cardiac pressures.

In a spontaneously breathing patient, intrathoracic and thus intracar-diac pressures will decline during inspiration and rise during expiration(Figure 2.6). The diaphragm pulls downward creating a negative pleural pres-sure during inspiration while the diaphragm relaxes and the elastic recoil ofthe lungs, chest wall, and abdominal structures compress the lung duringexpiration.

In patients who are mechanically ventilated, end-expiration is still used torecord pressures. The difference is that pressures increase with inspiration anddecrease with expiration. If the patient is not breathing over the ventilator, then


s s

s

d

a

v

a

v

av

av

aa a

a

a

a

a a

a

a a

v

v

a a

aa

a

v

v

v

v

v

v v

dd d d

d dd d d d d

d dd d d d d d d

s

ss s

s

s

s s s

s

s

ss

s

s

ss

Aorta

PCWP

RA

r r r r r r r r r r r r r r r rIGain 100IIGain 100V5Gain 100

50

1mV

1mV

1mV

1mV

1mV

1mV

25

aa

a

a

a

a

a

a

a

aa

aa

a

a

a

v

vv

v

v

v

v

v v v

v

01 2 3 4 5 6 7 8 9

25mm/s >[-11-]

Figure 2.6 Effect of respiratory variation on RA and PCW pressures.


interpretation is straightforward. If however the patient is ventilated and alsobreathing spontaneously, then waveform interpretation becomes more difficultas inspiration can be associated with decreases in pressure (spontaneous res-piration) and increases in pressure (ventilator). The spontaneous breath withpressure support will result in an initial decrease in pressure which can be usedto locate end expiration (just prior to the inspiratory effort marked by the dip inpressure). In patients receiving intermittent mandatory ventilation, identify-ing end-expiration can be a challenge. There are newer ventilators which showgraphic respiratory waveforms that provide a means for accurately identifyingend expiration.

In the intensive care unit, manual interpretation of waveforms using aprinted copy is necessary to ensure that pressures are interpreted at end-expiration unless a monitor is available that displays the ECG, respiratoryrecording and PA pressure waveforms on a single screen.

Cardiac outputA brief overview of cardiac output is presented here. For a more detaileddiscussion, please see Chapter 6.

There are two primary methods used to calculate cardiac output when uti-lizing a PA catheter, thermodilution and assumed Fick. The thermodilutionmethod requires that a substance that is cooler than blood (usually saline)be injected through the CVP port of the PA catheter. This injection should besmooth and take less than 4 seconds. The temperature in the PA is measuredand the change over time, as the cooler injectate passes through the PA, is usedto calculate cardiac output. The computer in the cardiac output machine usesthe following formula:

CO = CC × (Tb − Ti)area under the curve

where CC is the computation constant, Tb the blood temperature, and Ti theinjectate temperature.

A computation constant needs to be entered to provide the cardiac outputcomputer with information on the amount and temperature of the injectate.Each PA catheter comes with an insert describing computation constants to beused under various conditions.

If you enter a computation constant that calls for 10 cc and you inject 9 cc, themeasured cardiac output will be incorrect. Most of the time room temperaturesaline is used as the injectate. If a different temperature injectate is used (e.g.iced saline), the computation constant must be changed.

The other method used to estimate cardiac output was first described byAdolph Fick in 1870. He postulated that the total uptake or release of a sub-stance by an organ is the product of the blood flow through that organ andthe arteriovenous difference (A − VO2) of the substance. To measure oxygenconsumption one requires a Water’s Hood for the nonventilated patient or a


metabolic cart for the ventilated patient. This is cumbersome and rarely usedin daily practice. Instead, oxygen consumption is “assumed” (methods usedto arrive at oxygen consumption are described in detail in Chapter 6).

The A − VO2 gradient in the body is determined by comparing arterialoxygen saturation (measured directly or obtained via a pulse oximeter) with“mixed venous” oxygen saturation. Blood is considered “mixed venous” whenit has circulated through the right side of the heart, to the PA. This mixingincludes blood from the inferior and superior vena cava and the heart itself.

The formula used to estimate cardiac output using the assumed Fickapproach is

O2 consumption = CO (A − VO2 difference)

or

CO = O2 consumptionA − VO2 difference

CO = 130 × BSA1.36 × Hgb × 10 × (SaO2 − SvO2)

where BSA = body surface area; 130 mL/min = standard oxygen consumption;1.36 = a gram of hemoglobin (Hgb) holds 1.36 mm of oxygen; Hgb = reportedas grams per deciliter; 10 = to convert deciliters to liters you need to multi-ply the Hgb by this; SaO2 = either from an ABG or from the pulse oximeter;SvO2 = either from an oximetric PA catheter or by during oxygen saturationon a blood sample drawn from the PA port.

An example:

If BSA = 17; Hg = 12, SaO2 = 0.98, and SvO2 = 0.55

CO = 130 × 1.71.36 × 12 × 10 × (0.98 − 0.55)

= 22170.1

= 3.48 L/ min

Cardiac index (CI) enables comparison of cardiac output between patientsof different size. It is calculated by dividing cardiac output by BSA. Normalcardiac index is 2.5–4 L/min.

Calculating systemic vascular resistance and pulmonaryvascular resistanceVascular resistance is an indicator of ventricular afterload. Systemic vascularresistance (SVR) can be calculated using the following formula:

SVR = MAP − CVPCO

× 80

Normal SVR is 800–1200 dynes/sec/cm5.


An example:If MAP = 93 mm Hg, CVP = 3 mm Hg, and CO = 5 L/min then

SVR = (93 − 3)/5 = 18 × 80 = 1440 dynes/sec/cm5

Next, we will figure out pulmonary vascular resistance (PVR), which is anestimate of RV afterload. Normal PVR is in the range of 40–150 dynes/sec/cm5.PVR can be calculated using the following formula:

PVR = Mean PAP − PCWPCO

× 80

Mean pulmonary artery pressure (PAP) is calculated similar to mean arterialpressure.

Mean PAP = (PA systolic pressure + (2 × PA diastolic pressure))/3

An example:

If PAS = 30 mm Hg, PAD = 15 mm Hg, PWP = 10 mm Hg, and CO = 5 L/min

PVR = 20 − 105

× 80 = 2 × 80 = 160 dynes/sec/cm5

Occasionally, PVR will be reported in Wood units (named after Paul Wood, aBritish cardiologist). It is the same formula as above, however the conversionfactor of 80 is not used. To calculate PVR in Wood units, do the same calculationas above but do not multiply by 80 (i.e. PVR = 2 Wood units).

PVR is an important parameter in patients being evaluated for a heart trans-plant. If PVR is elevated, the right heart in the transplanted heart will notbe able to generate enough pressure to maintain cardiac output. In general,survival in transplanted patients is better if PVR < 3 Wood units.

SvO2 monitoringA full discussion of mixed venous oxygen saturation monitoring (SvO2) isbeyond this chapter; however, we will cover some of the basic concepts ofSvO2 monitoring.

The body maintains a balance between the oxygen delivered and the oxy-gen consumed. The supply side consists of cardiac output, oxygenation, andhemoglobin. The demand side is basal energy consumption plus anythingthat increases oxygen use, for example, pain, shivering, infection, agitation,or any number of disease processes. Oxygen use is decreased by various fac-tors including anesthesia, hypothyroidism, paralyzing agents, sleep, and painmedication. SvO2 is a function of oxygen delivered–oxygen consumed. Forexample, assume that 1000 mL/O2/min of oxygenated blood is delivered tothe tissues. Next, assume that the tissues use or consume 250 mL/O2/min. Thiswould leave 750 mL/O2/min to return to the lungs through the venous system.


A total of 1000 − 250 = 750 mL or 75% of the oxygenated blood is returned tothe lungs. This 75% is the venous oxygen reserve.

If a healthy individual’s oxygen consumption increases, the body willincrease cardiac output and respiratory rate to ensure that demand is met andblood is kept fully oxygenated. In a critically ill patient, the situation is differ-ent and some patients are incapable of meeting even resting oxygen demands.When supply fails to meet demand despite maximal extraction of oxygen fromthe blood, anaerobic metabolism and lactic acidosis may occur.

SvO2 can be measured continuously if the patient has an oximetric PAcatheter in place. The oximetric catheter has fiberoptic technology that enablesmeasurement of oxygen saturation of hemoglobin within the PA. Alternatively,if the patient does not have a SvO2 catheter in place, oxygen saturation in PAblood can be measured via an oximeter by obtaining intermittent blood sam-ples from the PA via the catheter. There are several uses of continuous oximetryincluding measuring tissue oxygen consumption, obtaining real time estima-tion of cardiac output and monitoring responses to medications.

Important points about SvO2 monitoring1. Mixed oxygen saturation monitoring and arterial saturation monitoringmay allow identification of an imbalance of oxygen supply and demand.2. Cardiac output is the largest determinant of oxygen supply.3. Normal SvO2 is 60–80%.4. Oximetric catheters allow continuous monitoring of SvO2.5. If the patient does not have an oximetric catheter in place, intermittentPA blood samples can be used to determine SvO2. The disadvantage of thismethod is that you only know what the SvO2 is at that specific time.6. SvO2 may be high in sepsis because of impaired oxygen extraction. Inthis case, tissues are generally hypoxic despite high SvO2. Some practitionersare using ScvO2- Central venous oxygen saturation monitoring in the patientwith sepsis. The concept is the same, however; the blood sample is taken fromdistal port of a triple lumen that has been placed in the right subclavian vein,therefore; this saturation measurement does not include venous return fromthe heart.

Complications of pulmonary artery catheterizationThere are several risks associated with PA catheter insertion that are inde-pendent of vascular access (Table 2.2). These include arrhythmias, transientright bundle branch block (RBBB), infection, pulmonary infarction, PA rupture,myocardial perforation, damage to the pulmonic or tricuspid valves, throm-boembolic complications, air embolism, and knotting of the catheter.

Arrhythmias primarily occur during passage through the RV and areincreased in patients with acute ischemia, hypocalcemia, hypokalemia, hypo-magnesemia, hypoxemia, acidosis, shock, or digitalis toxicity. Atrial arrhyth-mias can also be precipitated by PA catheter placement.

RBBB can occur during passage of the catheter through the tricuspid valve.The right bundle is a fairly discrete and superficial structure in the septal wall


and thus subject to trauma during the catheter advancement. RBBB is usuallytransient and does not cause any problems unless the patient has a preexistingleft bundle branch block (LBBB) in which case complete heart block develops.In patients with LBBB, it is prudent to have a transcutaneous pacemaker readilyavailable. In one series of 82 patients with LBBB, two episodes of complete heartblock occurred. Both of these episodes occurred in patients with recent onsetLBBB and both occurred one day after the catheter had been inserted [10].

PA catheters have been associated with an increase in the risk of pulmonaryembolus. In a randomized trial of PA catheters in 1994 high-risk surgicalpatients, there was a higher rate of pulmonary embolism in the catheter groupthan in the standard-care group (8 events versus 0 events, P = 0.004) [4]. Oneclue to the appearance of thrombus on the tip of a PA catheter can be pressuredamping.

PA catheters have also been associated with valve damage and endocarditis.A study of autopsies from 55 patients who had undergone PA catheterizationfound that 53% had right-sided endocardial lesions. These included 22% withsubendocardial hemorrhage, 20% with thrombus, 4% with hemorrhage andthrombus, and 7% with infective endocarditis. The pulmonic valve was themost common site of lesions (56%) followed by tricuspid valve (15%), RA(15%), right ventricle (10%), and PA (5%) [11].

There are two complications that can occur with wedging of the PA catheter:PA ischemia and/or infarction and PA rupture. As the PA catheter wedges itblocks blood flow in a branch of the PA and therefore can cause ischemia and/orinfarction if the balloon is left inflated too long. Alternatively, the catheter canmigrate into a small branch and occlude flow without balloon inflation (called“spontaneous wedge”). During the first few hours that the PA catheter is inplace it warms up and can migrate further into the distal lung vasculature.

The most feared complication of PA catheters is PA rupture. The presentationcan be dramatic with sudden onset of hemoptysis followed by cardiovascularcollapse. It generally occurs when the balloon is inflated in a small nondis-tensible branch but can also be caused by the tip of the catheter perforatingthe artery. Risk factors for PA rupture include pulmonary hypertension, age,female gender, anticoagulation and frequent wedging. This complication isfatal approximately 50% of the time.

To avoid complications from “wedging” the PA catheter, inflate the balloonslowly with continuous waveform monitoring. If a PCWP tracing is obtained,stop inflating the balloon. If the PA catheter wedges with 0.5 cc of air or less,then it probably needs to be repositioned. A couple of other important points:try to never wedge longer than 2 respiratory cycles, never inflate the balloonwith more than 1.5 cc of air, continuously monitor waveforms to ensure thatthe PA catheter does not spontaneously wedge and never flush the catheterin a wedged position. When a PA catheter overwedges the waveform usuallygoes straight up or down. If you see this waveform you should immediatelylet the air out of the balloon and obtain a chest X-ray to determine the locationof the tip of the catheter.


If PA diastolic pressure and PCWP are closely related (within 6 mm Hg),there is no need to repeatedly obtain wedge pressures. PA diastolic pressurecan be used to make clinical decisions.

One of the most common sources of errors when using PA catheters is inac-curate interpretation of data. Multiple studies have demonstrated inconsistent,incomplete, and improper data interpretation. It is important to obtain a com-plete understanding of how best to analyze the data obtained at the time of PAcatheter placement in order to optimally use the PA catheter.

Pulmonary artery catheter education project

For more information on the proper setup and use of PA catheters, see the Pul-monary Artery Catheter Education Project at http://www.pacep.org/. Thisweb-based educational effort is a collaboration of the American Associationof Critical Care Nurses, the American Association of Nurse Anesthetists, theAmerican College of Chest Physicians, the American Society of Anesthesiolo-gists, the American Thoracic Society, the National Heart Lung and Blood Insti-tute, the Society of Cardiovascular Anesthesiologists, and Society of CriticalCare Medicine.

Case studies

1. You are working in surgical ICU caring for a 75 year old male who was in amotor vehicle accident 36 hours ago. Despite adequate fluid resuscitation witha CVP reading of 10 mm Hg, he remains hypotensive, acidotic and has a basebalance of -3. Vital signs: HR 120 bpm; BP 90/40 mm Hg with MAP 56 mm Hg,urine output = 15 ml/hour. Ventilator settings are as follows: SIMV/PRVC;FiO2 = 70%, TV 450 ml, Rate 14, PS 10 cm H2O & PEEP of 5 cm H2O. He is notbreathing over the ventilator.

The decision is made to place a PA catheter to evaluate his cardiac status.His PA catheter numbers are: CVP = 10 mm Hg, PA pressures = 40/20 mm Hg,PCWP = 16 mm Hg, CO/CI = 3.5/1.8 L/min.

a. What is his stroke volume (SV)?b. What is his SVR?c. Based on the PA numbers would you change your therapy?d. On his pressure waveforms how will you identify end expiration?

2. You are working in the medical ICU when a 69 year old female is admittedwith the diagnosis of severe dyspnea. There isn’t any documented history ofheart disease. Vital signs: HR130 bpm; BP 80/30 mm Hg with MAP 46 mm Hg,urine output = 10 ml/hour, Temp 35◦C orally, lactate 4 mg/dl. Ventilator set-tings = SIMV/PRVC; FiO2 = 100%, TV 400 ml, Rate 22, PS 10 cm H2O, PEEP10 cm H2O. Initially she had a CVP of 2 mm Hg and a ScvO2 of 65%. Despitean increase in CVP to 10 mm Hg after fluid resuscitation and the addition oftwo vasopressors, her blood pressure does not improve. Her chest x-ray showsdiffuse pulmonary edema. To answer the question of whether the pulmonary


edema is cardiogenic or non-cardiogenic, a PA catheter is inserted and the fol-lowing information obtained: CVP = 10 mm Hg, PA pressure = 25/15 mm Hg,PCWP = 12 mm Hg, SvO2 = 60%, and CO = 8 L/min with CI = 3.5 L/min

a. Based on the information available, is the primary etiology of the pul-monary edema cardiogenic or noncardiogenic?

3. You are working in the CCU when a 62 year old female is admitted withthe diagnosis of cardiogenic shock. The patient became extremely fatigued48 hours ago but only recently came to the ER when she became very shortof breath. The ECG shows Q waves in leads I, aVL, and V1-V6, consistentwith a large antero-lateral MI. Her vital signs are as follows: HR 120 bpm, BP86/40 mm Hg with MAP 55 mm Hg, RR 32 times per minute. Pulse Oximeteryis reading 85% on room air. Temp = 36.5 Celsius.

A PA catheter is inserted to assist with her treatment and the following infor-mation is obtained: CVP = 15 mm Hg, PA pressure = 45/30 mm Hg, PCWP =26 mm Hg, and CO = 3 L/min with CI = 1.7 L/min

a. What is her SV?b. What is her SVR?c. Based on the PA numbers would you change your therapy?d. Would you use beta blockers to decrease the heart rate at this time?

4. You are working in the cardiac catheterization laboratory and a 33-year-oldmale is referred for evaluation for potential orthotropic heart transplant. Hewas well until 18 months ago when he developed severe LV dysfunction ofunclear etiology but thought to be a sequela of a viral syndrome. He has had asteady downward course during the past 3 months during which his symptomshave progressed to Class III (i.e. he cannot walk across a room without havingto stop to catch his breath). Ventriculography confirms severe LV dysfunction.Coronary angiography is normal. A right heart catheterization is performedand the following information obtained: RA pressure = 20 mm Hg, PA pres-sure = 78/41 (mean of 58 mm Hg), PCWP = 38 mm Hg, and cardiac output =4.6 L/min.

a. What is the PVR?b. What would you do next?

Nitroprusside infusion was begun and titrated until systemic systolicblood pressure was approximately 80 mm Hg. The following informationwas obtained from the PA catheter: PA pressure=65/30 (mean of 46 mm Hg),PCWP = 24 mm Hg, and cardiac output = 7.9 L/min.c. What is the PVR now?d. Is the patient a candidate for heart transplantation?

Case study answers

1. a. SV = CO/HR = 3500 ml/min/120 bpm = 29 ml.The patient’s stroke volume is 29 mL.


b. SVR = ((MAP − CVP) × 80)/CO = ((56 − 10) × 80)/3.5 = 1051 dynes/sec/cm5.c. Preload is reflected by CVP =10 mm Hg and PCWP = 16 mm Hg withan afterload (SVR) of 1051 dynes/sec/cm5. Cardiac output is low despitereasonable preload suggesting LV dysfunction (possibly due to a cardiaccontusion suffered at the time of the accident). An echocardiogram would beuseful. Giving IV fluids to increase PCWP might increase cardiac output (atthe risk of worsening oxygenation) and/or an inotrope such as Dobutaminemight be used to increase contractility.d. The patient is receiving positive pressure ventilation and thus inspirationis characterized by an increase in pressure on RA and PCWP waveforms.End expiration will be just prior to the ventilator stimulated increase. If thepatient has any spontaneous respirations, remember to ignore the initialinspiratory dip in pressure – end expiration will be just prior to this briefdrop in pressure.

2. a. A PCWP less than 18 mm Hg favors pulmonary edema of noncardiacorigin rather than of cardiogenic origin. According to Schwarz and Albert(Chest 2004;125:1530–5), “acute lung injury (ALI) and ARDS (ALI/ARDS)are defined as follows: (1) an illness having an acute onset, (2) an arterialoxygen tension/inspired oxygen fraction �200 mm Hg (or �300 mm Hg forALI), (3) the presence of bilateral infiltrates on frontal chest radiographs,and (4) a pulmonary artery occlusion pressure �18 mm Hg if measured, orno clinical evidence of left atrial hypertension when not measured.”

3. a. SV = CO/HR = 3000 ml/min/120 bpm = 25 ml.b. SVR = ((MAP − CVP × 80)/CO = ((55 − 15) × 80)/3 = 1067 dynes/sec/cm5.c. Her preload is high (CVP = 15 mm Hg and PCWP = 26 mm Hg), her after-load is within the normal range and her cardiac output is low. These num-bers confirm the diagnosis of cardiogenic shock. Treatments might includepharmacologic diuresis to decrease preload and thus improve LV functionaccording to the Frank-Starling curve. An IABP could be inserted to augmentcoronary artery perfusion, increase MAP and decrease afterload. Some prac-titioners might also use dobutamine to improve LV contractility.d. No as the heart rate is necessary to maintain blood pressure. The heartrate has increased in order to enhance cardiac output in the setting of adecreased stroke volume.

4. a. PVR = (Mean PAP − PCWP)/CO = (58 − 38)/4.6 = 4.4 Wood units.b. In severe LV dysfunction, total PVR includes both a ’fixed’ componentand a ’dynamic’ component which is dependent upon LA pressure. Follow-ing a heart transplant, LA pressure is less and the dynamic component ofPVR is greatly reduced or eliminated. The fixed component of PVR persistshowever and the new RV must be able to maintain cardiac output againstthis resistance. In determining which patients are at risk for acute RV failurefollowing heart transplantation, it is essential to quantify the fixed com-ponent of PVR. This can be done by measuring PA pressures, PCWP and


cardiac output during infusion of a vasodilator (e.g. prostacyclin, adenosineor nitroprusside).c. The PVR is now (46 − 24)/7.9 = 2.8 Wood units.d. This level of PVR would not exclude the patient from receiving a hearttransplant in most centers.

References

1 Swan HJ, Ganz W, Forrester J, Marcus H, Diamond G, Chonette D. Catheterization ofthe heart in man with use of a flow-directed balloon-tipped catheter. N Engl J Med 1970;283:447–451.

2 Shah MR, Hasselblad V, Stevenson LW, et al. Impact of the pulmonary artery catheter incritically ill patients: meta-analysis of randomized clinical trials. JAMA 2005; 294:1664–1670.

3 Connors AF, Jr, Speroff T, Dawson NV, et al. for SUPPORT Investigators. The effectivenessof right heart catheterization in the initial care of critically ill patients. JAMA 1996; 276:889–897.

4 Sandham JD, Hull RD, Brant RF, et al. A randomized, controlled trial of the use ofpulmonary-artery catheters in high-risk surgical patients. N Engl J Med 2003; 348:5–14.

5 Richard C, Warszawski J, Anguel N, et al. Early use of the pulmonary artery catheter andoutcomes in patients with shock and acute respiratory distress syndrome: a randomizedcontrolled trial. JAMA 2003; 290:2713–2720.

6 Binanay C, Califf RM, Hasselblad V, et al. Evaluation study of congestive heart failure andpulmonary artery catheterization effectiveness: the ESCAPE trial. JAMA 2005; 294:1625–1633.

7 The National Heart, Lung, and Blood Institute Acute Respiratory Distress Syndrome(ARDS) clinical trials network: pulmonary-artery versus central venous catheter to guidetreatment of acute lung injury. N Engl J Med 2006; 354:2213–2224.

8 Harvey S, Harrison DA, Singer M, et al.: Assessment of the clinical effectiveness of pul-monary artery catheters in management of patients in intensive care (PAC-Man): a ran-domised controlled trial. Lancet 2005; 366:472–477.

9 Tempe DK, Gandhi A, Datt V, et al. Length of insertion for pulmonary artery cathetersto locate different cardiac chambers in patients undergoing cardiac surgery. Br J Anaesth2006; 97(2):147–149.

10 Morris D, Mulvihill D, Lew WY. Risk of developing complete heart block during bedsidepulmonary artery catheterization in patients with left bundle-branch block. Arch InternMed 1987; 147:2005–2010.

11 Rowley KM, Clubb KS, Smith GJ, Cabin HS. Right-sided infective endocarditis as a con-sequence of flow-directed pulmonary-artery catheterization. A clinicopathological studyof 55 autopsied patients. N Engl J Med 1984; 311:1152–1156.

CHAPTER 3

Normal hemodynamics

Alison Keenon, Eron D. Crouch, James E. Faber,George A. Stouffer

Introduction

The mechanical events of the cardiac cycle as they occur in series are as fol-lows: right atrium (RA) contracts, left atrium (LA) contracts, left ventricle (LV)begins contraction, mitral valve closes, right ventricle (RV) begins contrac-tion, tricuspid valve closes, pulmonary valve opens, aortic valve opens, aor-tic valve closes, pulmonary valve closes, tricuspid valve opens, mitral valveopens. The events of the cardiac cycle superimposed upon cardiac pressuretracings and an ECG tracing, often called the Wiggers diagram after Dr CarlWiggers who first arranged data in this useful manner, is helpful in illus-trating the relationship and timing of events in the cardiac cycle to hemo-dynamic phenomena (a stylized version of a Wiggers diagram is shown inFigure 3.1).

The cardiac cycle is often divided into systole (ventricular contraction) anddiastole (ventricular filling) (Figure 3.2). Systole can be further divided intofour stages: isovolumetric contraction, rapid ejection, reduced ejection, andisovolumetric relaxation (Figure 3.3). The isovolumic phase of LV contractionis that time in which ventricular pressure is increasing but not sufficient toopen the aortic valve (Figure 3.1). This phase ends when the aortic valve opensand blood is ejected (ejection phase). On an LV pressure tracing, the onset ofventricular contraction coincides with the R wave on the ECG and is indicatedby the earliest rise in pressure after atrial contraction. During ventricular sys-tole, the atrial pressure initially declines. This is thought to be due to stretchof the atria as the base of the heart descends. Following this brief period, atrialpressure progressively increases during the rest of ventricular systole as bloodflow into the atria increases.

Diastole is the period during the cardiac cycle between aortic valve closureand mitral valve closure. A common mistake is to think of diastole as lessimportant than systole, or more inaccurately as merely the absence of systole.Our appreciation of diastole has increased over the last several years as wehave realized the importance of diastolic dysfunction in patients with conges-tive heart failure. Diastole is commonly divided into three components: rapidfilling phase, slow filling phase, and atrial systole (Figure 3.3). A fourth phaseof diastole, isovolumic relaxation, is sometimes utilized. This is the period

37


120

0

20

40

60

80

100

Pre

ssur

e(m

m H

g)

Atr

ial s

ysto

le

Iso

volu

met

ricco

ntra

ctio

n

Rap

id e

ject

ion

Red

uced

eje

ctio

n

Iso

volu

met

ricre

laxa

tion

Mitral valve

closes

Aortic valve opens

Aortic valve

closes

Ven

tricu

lar

volu

me

(mL) 140

100

60

Ven

ous

puls

e(m

m H

g)

0

3

Rap

id v

ent

ricul

ar

fillin

g

Red

uced

vent

ricul

arfil

ling

ECG P

Q S

T

R

Mitral valve opens

ac

Heart sounds

1 234

Left ventricle

Left atrium

Aorta

Pre

ssur

e(m

m H

g)

0

2030

10

Tricuspid valve closes

Pulmonary valve opens

Pulmonary valve closes

Tricuspid valve opens

Right ventricleRightatrium

Pulmonary artery

LVESV

LVEDV

0.1 sec

a cv

v

120

0

20

40

60

80

100

Atr

ial s

ysto

le

Iso

volu

met

ricco

ntra

ctio

n

Rap

id e

ject

ion

Red

uced

eje

ctio

n

Iso

volu

met

ricre

laxa

tion

Mitral valve

closes

Aortic valve opens

Aortic valve

closes

140

100

60

0

3

Rap

id v

ent

ricul

ar

fillin

g

Red

uced

vent

ricul

arfil

ling

ECG P

Q S

T

R

Mitral valve opens

ac

Heart sounds

1 234

Left ventricle

Left atrium

Aorta

0

2030

10

Tricuspid valve closes

Pulmonary valve opens

Pulmonary valve closes

Tricuspid valve opens

Right ventricleRightatrium

Pulmonary artery

LVESV

LVEDV

0.1 sec

a cv

v

Figure 3.1 Wiggers diagram.

Chapter 3 Normal hemodynamics 39

Great cardiac vein

Mitral valve

Posterior vein

Pulmonary artery

The heart in diastole

The heart in systole

Aortic valve

Middle cardiac vein

Posterior interventricularartery

Tricuspid valve

Pulmonary valve

Figure 3.2 Cardiac structures during systole and diastole.

between the closing of the semilunar valve and the opening of the AV valve.It is very brief and dependent upon the relaxation properties of the ventri-cle. Analysis of this phase is useful experimentally but has limited applicationclinically.

The rapid filling period occurs immediately upon opening of the AV valve.In normal individuals, this phase accounts for the majority of ventricular fill-ing. This phase is characterized by the Y descent on atrial pressure tracings andby a rapid fall (because the ventricle is still relaxing) and then equally rapid


Ventricularsystole

Ventriculardiastole

Atrial systole

Isovolumetriccontraction

Rapidejection

Reducedejection

Isovolumetricrelaxation

Rapidfilling

Reducedfilling

Ventricularsystole

Ventriculardiastole

Atrial systole

Isovolumetriccontraction

Rapidejection

Reducedejection

Isovolumetricrelaxation

Rapidfilling

Reducedfilling

Figure 3.3 Mechanical events of the cardiac cycle.

rise in ventricular pressure. As the rate of ventricular filling declines, the slowfilling phase is entered following by diastasis (a period when ventricular fillingceases). During the slow filling period in a normal heart, the atrial and ventric-ular pressures are similar and continued filling of the ventricles is dependentupon blood returning from the peripheral circulation (RV) or the lungs (LV).The slow filling period lasts longer than the rapid filling period in the normalheart. In patients with noncompliant ventricles, true diastasis may not occur.Rather, in the setting of elevated atrial pressure, passive blood flow from theatrium to the ventricle continues throughout diastole with the rate of flow andincrease in ventricular pressure dependent upon the compliance of the diseasedventricle.

Atrial systole is the last phase of diastole. It occurs soon after the beginningof the P wave on the electrocardiogram and is the cause and thus occurs con-current with the A wave on the atrial pressure tracing. In the normal heart atrest, atrial systole contributes relatively little blood to the already filled ven-tricle. In the diseased heart, or in the exercising normal heart, the contributionof atrial systole assumes a larger importance. In clinical practice, it is commonto see patients with hypertrophied ventricles who do well until they developatrial fibrillation. The more rapid rate and the loss of atrial contraction can thenprecipitate congestive heart failure.


Table 3.1 Normal hemodynamic values

Pressure Normal value Temporal relations ofmeasurement (mm Hg) ECG to mechanical events

Right atrium (RA) Onset of P wave to RA contraction is ∼60-80 msMean pressure 2–8

Right ventricle (RV) Onset of Q wave to RV contraction is ∼65 msPeak-systolic pressure 17–32End-diastolic pressure 2–8

Pulmonary artery Onset of Q wave to RV ejection is ∼80 msMean pressure 9–19Peak-systolic pressure 17–32End-diastolic pressure 4–13

Left atrium (PCWP) Onset of P wave to LA contraction is ∼85 ms.Mean pressure 2–12 Onset of P wave on ECG to A wave on

PCWP tracing will vary and maybe >200 msLeft ventricle (LV) Onset of Q wave to LV contraction is ∼52 ms

Peak-systolic pressure 90–140End-diastolic pressure 5–12

Aorta Onset of Q wave to LV ejection is ∼115 msMean pressure 70–105

Cardiac chambers

Right atriumNormal RA pressure is 2–8 mm Hg (Table 3.1) and is determined by centralvenous pressure (CVP), RA compliance, tricuspid valve function, and RVcompliance (Figure 3.4). CVP, and thus RA pressure, is primarily influenced byblood volume, body position, reflex changes in venous tone, skeletal musclecontraction, and the respiratory cycle.

A close examination of the RA pressure waveform in a patient in normalsinus rhythm reveals two major positive deflections, the A and V waves, andtwo negative deflections, the X and Y descents (Figure 3.5). A third minorpositive deflection, the C wave, may be seen in some settings between the Aand V waves. The A wave results from RA contraction and occurs immedi-ately following the P wave on a surface EKG, with the peak of the A wavefollowing the peak of the P wave by approximately 60-80 milliseconds. Fol-lowing RA contraction, the pressure declines, and this is visualized as the Xdescent. The X descent is caused by atrial relaxation as well as downwardmotion of the atrioventricular junction during early ventricular systole. TheC wave, when seen, will interrupt the X descent, and is caused by tricuspidvalve closure. The C wave marks the onset of RV systole. It will be seen as aminor positive deflection immediately following the QRS complex, and willfollow the A wave by a duration equals to the PR interval. After the peak ofthe C wave, the X descent continues (and is then called X prime; x′) as atrialrelaxation occurs and the pressure declines. The V wave is seen following the


Right atrium

Right ventricle

Superior vena cava

Pulmonary veins

Tricuspid valve

Septal band

Orifice of coronary sinus

Aorta

Pulmonary artery

Pulmonary veins

Inferior vena cava

Figure 3.4 Schematic of RA and RV.

nadir of the X descent and represents venous filling of the right atrium whilethe tricuspid valve is closed. The peak of the V wave occurs at the end of ven-tricular systole, just prior to tricuspid valve opening. V waves generally occursimultaneously with the T wave on a surface ECG. The Y descent follows the Vwave and denotes the onset of RV diastole. This descent reflects the rapid fallof RA pressure in conjunction with the exit of blood from the RA into the RV


Figure 3.5 Right atrial pressure waveform.

when the tricuspid valve opens (see Chapter 5 for a further discussion of atrialwaveforms).

Right ventricleNormal RV pressures are 2–8 mm Hg at the end of diastole and 17–32 mm Hgduring systole. The pressure in the RV rises rapidly during systole, and whenit exceeds the pressure in RA, the tricuspid valve closes. Pressure rises fur-ther as contraction continues in the closed chamber, and when the pres-sure exceeds that in the pulmonary artery, the pulmonary valve will open.The waveform reflecting RV systole has a rapid upstroke and is rounded,and will occur with or immediately following the QRS complex (Figure 3.6).The peak of this waveform is the RV systolic pressure. As systole ends andventricular relaxation occurs, pressure in the RV will rapidly fall to base-line, marking the onset of ventricular diastole. The pulmonary valve willclose when the pressure in the RV falls below pressure in the pulmonaryartery. When the pressure falls below that in the RA, the tricuspid valve willopen.

RV diastole can be divided into three phases (Figure 3.7). The first phasecomes with the opening of the tricuspid valve, and accounts for early fillingof the ventricle, approximately 60–75% of total filling in the normal heart. Thesecond phase accounts for 15–25% of filling, and is the slow filling period. Insome patients, this phase will include a period of diastasis in which ventricularfilling slows or ceases. Finally, the third phase occurs with RA systole, and


Normal sinus rhythm

1mV

1mV1mV

1mV1mV

1mV

0.38 mm Hg/mm

IGain:200

IIGain:200

V5Gain:100

50

25s

d 1 2d d d

e

s s

025mm/s >[-16-]

e e

Figure 3.6 Ventricular pressure tracing.

accounts for 10–25% of total filling. At this point, the tricuspid valve is open,creating essentially one right-sided chamber, so it follows that this waveformis simultaneous with and identical in morphology and amplitude to the Awave on the RA pressure tracing. The pressure at the end of this waveformrepresents RV end-diastolic pressure.

Left atrial pressure (pulmonary capillary wedge pressure)Normal LA pressure ranges from 2 to 12 mm Hg, and is typically measuredindirectly, via a balloon-tipped catheter placed in a distal branch of the


4

d

SFRF

Diastasis

A waveEDP

d

Figure 3.7 Ventricular diastole—the ventricular diastolic pressure waveform can be divided intothree phases: the early or rapid filling (RF) phase occurs with the opening of the tricuspid valve.The slow filling (SF) phase follows and extends until the onset of right atrial systole (A wave).The nadir in pressure following the A wave is the right ventricular end-diastolic pressure (EDP).

pulmonary artery. Inflation of the balloon obstructs flow and allows pressureto equilibrate along the column of blood, thus giving an estimate of LA pres-sure. This pressure is called the pulmonary capillary wedge pressure, or simplywedge pressure. LA mechanical events are transmitted retrograde through thiscolumn of blood in a wedge pressure tracing, thus the deflections on the wedgepressure waveform are often dampened and delayed relative to events in theLA (and on the ECG). The LA (or PCWP) pressure wave, like an RA pressurewave, consists of positive A, C, and V waves, and negative X and Y descents.

Left ventricleNormal LV pressure is between 90 and 140 mm Hg during systole and 5–12 mm Hg during diastole. While the pressures in the LV are much higher thanin the RV, the pressure waveform components are similar. LV systole leads to arapid increase in pressure. The mitral valve closes when the pressure exceedsthat in the LA (Figure 3.8). The pressure rises with continued contraction, andwhen it exceeds the pressure in the aorta, the aortic valve opens, and ventricu-lar ejection begins. LV pressure then continues to rise through the rapid ejectionphase of systole, and peaks during the T wave of a surface ECG. Pressure in theLV then begins to decline through the reduced ejection phase of systole. Whenthe pressure in the LV falls below that in the aorta, the aortic valve closes and


I

II

V5

1mV

1mV

1mV

1mV

1mV

1mV

50

25

025mm/s >[-19-]

r

e

d

d d

d

e

ee

a

v v

v v

a

PCWP

LV

a a

r r r

0.38mm Hg/mm 1 2 3 4

Figure 3.8 Simultaneous LV and PCWP tracings. The pressure tracing was taken from a57-year-old male with severe ischemic cardiomyopathy. Note the large V wave and prominent Awave in the PCWP tracing.

LV pressure continues to decline due to isovolumetric ventricular relaxation.When the pressure drops below that of the left atrium, the mitral valve opensand the left atrium empties into the LV. The mitral valve opening marks theonset of ventricular diastole. LV diastole, like RV diastole, consists of threephases.

Pulmonary arteryNormal pulmonary artery pressures range from 17 to 32 mm Hg during systoleand from 4 to 13 mm Hg during diastole. The pulmonary artery waveformreflects the pressure seen during systole as blood is ejected from the RV intothe pulmonary artery. Because of the pulmonic valve being open at this point,this waveform will be similar to the right ventricular systolic waveform. Thepressure in the pulmonary artery and the RV declines as systole comes to anend. When the pressure in the RV falls below that in the pulmonary artery,the pulmonic valve closes, marked by a dicrotic notch on the downslope ofthe pulmonary artery pressure waveform. The dicrotic notch notes the end ofventricular ejection. The decline in pressure in the pulmonary artery continuesgradually after closure of the pulmonic valve as blood flows to the pulmonaryarteries and veins toward the left atrium. Note that the peak of the pulmonaryartery waveform occurs within the T wave on a surface ECG.


AortaThe shape of the pressure waveform generated in the aorta is similar to that gen-erated in the pulmonary artery, with the pressures being significantly higherin the aorta. The upstroke of the aortic pressure waveform denotes the onset ofleft ventricular ejection, and occurs once the pressure in the left ventricle hasexceeded the pressure in the aorta and the aortic valve has opened (Figure 3.1).In the absence of aortic valve or perivalvular pathology, the aortic systolicpressure will be equal to the left ventricular systolic pressure. Aortic and leftventricular pressures decline with reduced ejection and contraction, and whenthe ventricular pressure drops below that of the aorta, the aortic valve closes.This is sometimes represented in the pressure waveform as a dicrotic notch inthe downslope of the aortic pressure tracing.

The definition of hypertension has changed over time as the deleteriouseffects of even minor increases in blood pressure have been realized. The classi-fication of normal blood pressure, prehypertension, and hypertension outlinedin the Seventh Report of the Joint National Committee on Prevention, Detec-tion, Evaluation, and Treatment of High Blood Pressure (JNC 7) is describedin Chapter 4.

1mV

1mV

1mV

1mV

1mV

1mV

Normal sinus rhythm

I

II

V5 50

25

025 mm/s >[-20-]

rr r r

ss

e e e e

d

e

d

e

d

ed

d

1 2 3 4

d d d

e

s s

0.38mm Hg/mm

LV

RV

Figure 3.9 Abnormal EDP. Simultaneous RV and LV pressures from the 57-year-old male withsevere ischemic cardiomyopathy depicted in Figure 3.8. Note that both LVEDP and RVEDP areelevated demonstrating that both RV and LV are poorly compliant.


100

LV(a) (b)

(c)

(d)

(f) (g)

(e)

d d d

d d

e e e

d

NIBP: / < >

r r r r

1 2

e

e

e

76

e

d dd

I

II

V5

I

II

V5

I

II

V5

I

II

V5

50

25

50

25

ss

e e e e e edd d d d dd

ss s

ss

01 2 3 4 5 6 7

0.38 mmHg/mm

0.38 mmHg/mm25mm/s >

[-14-]

0mm/s >

r r r r r r

r

s

200

100

0e e e e e e

d

r r

s s

s s

s

s s

s

r r r r r r r

d d d d d

s s s s s

r r r r r

50

25

25 mm/s>[-27-]

01 2 3 4 5 6 7

e

dd

d

d d

d d

ee e

ee e

e

Figure 3.10 Examples of different ventricular diastolic waveforms. LV tracings taken from apatient with aortic insufficiency ((a); note the rapid rise in diastolic pressure), constrictivepericarditis ((b); note the dip and plateau configuration), hypertrophic cardiomyopathy ((c); notethe elevated end diastolic pressure and accentuated A wave), restrictive cardiomyopathy ((d); notethe dip and plateau configuration), ischemic cardiomyopathy ((e); note the elevated end diastolicpressure and accentuated A wave); and pericardial tamponade before (f) and after (g)pericardiocentesis.


Left ventricular function

Myocardial contractile dysfunction is an important cause of morbidity andmortality and LV systolic function is an important determinant of survivalin patients with cardiomyopathy and/or coronary artery disease. Despite itstremendous importance, clinical measurement of LV function is limited by theinability to directly assess myocyte performance. Instead, various surrogatessuch as ejection fraction are used. In interpreting these tests, it is importantto remember that LV function is a dynamic process that varies not only withcontractility but also with preload and afterload.

Measurement of left ventricular preloadPreload for the ventricles is defined as amount of passive tension or stretchexerted on the ventricular walls (i.e. intraventricular pressure) just prior tothe initiation of systole. This load determines end-diastolic sarcomere lengthand thus the force of contraction. The Frank–Starling law states that the pas-sive length to which the myocardial cells are stretched at the end of diastoledetermines the active tension they develop when stimulated to contract. TheFrank–Starling law is an intrinsic property of myocytes and is not dependentupon extrinsic factors such as the autonomic nervous system or circulatinghormones. The general principle is that increased preload causes increasedforce of contraction, which increases stroke volume and thus cardiac output.

200

100LV

pressure(mm Hg)

00 ESV EDV100

a

SVd bEDPVR

ESPVR

Aorticvalve

opening

Aorticvalve

closing c

LV volume (mL)

200

Figure 3.11 Schematic of pressure volume loop in a normal heart. Following mitral valve opening,ventricular filling occurs with only a small increase in pressure despite a large increase in volume(a). The first segment of systole is isovolumic contraction (b). When the aortic valve opens,ejection begins and LV volume falls as LV pressure continues to rise (c). After closure of the aorticvalve, isovolumic relaxation occurs (d). This point marks the end-systolic pressure–volume pointon the curve. [ESPVR = end-systole pressure-volume relationship and EDPVR = end-diastolepressure-volume relationship].


Effect of increased preload on LV pressure–volume loop

Effect of increased afterload on LV pressure–volume loop

200

(a)

(b)

150

LV pressure(mm Hg)

LV pressure(mm Hg)

LV volume(mls)

100

50

0

200

150

100

50

00 100

Ees Ea

LVEDV

200

0 100LV volume (mls)

Loop 2 has an increased preload (increased LVEDV) as compared to loop 1Note: Loop 2 has a larger stroke volume than loop 1.

The afterload & contractility have remained constant. The afterload lines for the 2 loops are parallel sothey have the same afterload. Both end-systolic points are on the same contractility line so the 2 loopshave the same contractility.(See discussion of contractility & afterload lines on p44)

LVEDV for loop 2

200

End-systolicpressure–volume line

21

Endsystolicpressure–volume line

2

1

(c) Effect of increased contractility on LV pressure–volume loop

LV pressure(mm Hg)

LV volume(mls)

Note the increased stroke volume for loop 2 (which has the increased contractility).• The increased slope of the end-systolic pressure-volume line is an index of the increased contractility.• The end-systolic points of both loops lie on the same ‘afterload line’ so there afterload is the same for the 2 loops.• The LVEDV is the same for the loops so the pre-load is the same.

200

150

100

50

00 100

Ees Ea200

2 1

Figure 3.12 Effect of changes in preload, afterload, and contractility on pressure–volumerelationship.


100(mm Hg)

50 100 (mL)

BA

C

D

LV volume

LV pressure

(a) (b)

(c) (d)

100(mm Hg)

100 200 (mL)

BA

CD

LV volume

LV pressure

100(mm Hg)

100 200 (mL)

BA

C

D

LV volume

LV pressure

100(mm Hg)

50 100 (mL)

LV volume

LV pressure

Figure 3.13 Schematics of pressure volume loops in various disease states. In aortic stenosis((a); dotted line is normal ventricle), note the marked increase in afterload, minimal increase inpreload and increased contractility. Chronic aortic insufficiency ((b); dotted line, normal ventricle;dashed line, acute AI; full line, chronic AI) is characterized by marked increase in preload with asmaller increase in afterload and decreased contractility. Acute aortic insufficiency manifests asincreased preload with minimal change in afterload or contractility. Chronic mitral regurgitation (c)is associated with marked increase in preload, decreased afterload and decreased contractility.Mitral stenosis (d) is primarily characterized by decreased preload.

In a given patient, elevated filling pressures do not differentiate primarydiastolic dysfunction from primary systolic dysfunction but do give an indi-cation of preload required to obtain a specific cardiac output.

Left ventricular preload as measured by LV end-diastolic pressure (LVEDP)is a common tool used clinically in assessment of LV performance. The LVEDPimmediately precedes isometric ventricular contraction (Figure 3.6). This point,also known as the “Z” point, is located on the downslope of the LV “A” waveat the crossing over of LA and LV pressures and is coincident with the R waveon the surface ECG. Since the mitral valve is open during atrial contraction, theA wave is represented in the LV pressure tracing coincident with the A wavein the atrial tracing. Identification of the true LVEDP can at times be difficult,especially when using fluid-filled catheters.

LVEDP is influenced by ventricular compliance and intravascular volumestatus. LVEDP is normally <12 mm Hg but may be elevated when the LVexperiences volume (e.g. mitral regurgitation or aortic regurgitation) or pres-sure (e.g. hypertension or aortic stenosis) overload. Impairment of myocardial


stro

ke w

ork

(car

diac

out

put)

Filling pressure (mm Hg)

Failure

Normal

Figure 3.14 Ventricular function curve. Schematic ofthe relationship between filling pressure and cardiacoutput (or stroke volume) in a normal and in a failingleft ventricle.

contractility also alters the diastolic pressure–volume relationship and shiftsthe end-diastolic pressure point upward (Figure 3.9).

Inspection of the waveform in diastole provides information aboutLV diastolic function that can supplement LVEDP. Myocardial relaxationabnormalities are suggested by the continuing decline of pressure duringearly diastole with the pressure nadir occurring midway through the dias-tolic period. Alternatively, a stiff, poorly compliant ventricle is suggested byan abnormally tall A wave resulting in a marked elevation of LVEDP althoughthe diastolic pressure prior to atrial contraction can be normal (Figure 3.10).

Pressure–volume loopsA useful method of displaying the relationship between ventricular volumeand pressure during the cardiac cycle is the pressure–volume loop (Figure 3.11).Stroke volume (change in volume) and stroke work (stroke volume × MAPor alternatively the area inscribed by the P–V loop) can be calculated usingPV loops. Different pressure–volume loops are obtained following changes inpreload, afterload, or contractility (Figure 3.12), and in different disease states(Figure 3.13).

Ventricular performance can be depicted by plotting LVEDP versus strokevolume or cardiac output (Figure 3.14). This curvilinear relationship is com-monly called the LV function curve. Ventricular function curves are shiftedupward by inotropes and downward by interventions impairing inotropicactivity. Afterload may also significantly influence the elevation or declineof the ventricular function curve.

Indices of contractilityA completely satisfactory clinical index of contractility that is independentof preload and afterload has not been defined. The maximal rate of myocytefiber shortening in the isolated heart correlates well and is little affected bypreload or afterload but obviously of no use clinically. In the intact heart,contractility is best measured by the pressure volume point as the aortic valvecloses. Another index of contractility is dP/dtmax which is the derivative ofthe maximal rate of ventricular pressure rise during the isovolumetric period.


s

d d

a a

a a

a

aa

aa

aa

a

a

a

aa

v v

a

Monitor Length: 10 sec.

v

v

v

v

v

v

v

v

v

aa

v

d dd d

d d d d dd d d d d d d d d

d

s

s s

ss

s

s

s

s s s

s

ss

s

s

s

s

ssAorta

PCWP

1mV

1mV

1mV

1mV

1mV

1mV

IGain:100

IIGain:100

V5Gain:100

50

25

aa

a

a

a

a

a

a

aa

a

v

a

a

v

vv

v

v

v

v v

a

v

a

01 2 3 4 5 6 7 8 9

25mm/s >[-11-]

r r r r r r r r r r r r r r r r r

Normal sinus Rythmn

RA

Figure 3.15 Respiratory variation. Aortic, PCWP, and RA pressure in a patient with largerespiratory variation.


Figure 3.16 Effect of breath hold on RA and PCW pressures (all pressures recorded on a25 mm Hg scale).

A complete discussion of indices of contraction is beyond the scope of thischapter.

Respiratory variation

The act of breathing influences hemodynamic measurements (Figure 3.15).Even in healthy patients intravascular pressure in the thoracic aorta and venacava (and thus preload and afterload) may be significantly altered by normalrespiration. During normal, spontaneous respiration, intrathoracic pressuremay drop from −3 to −4 mm Hg at end-expiration to −7 to −8 mm Hg duringend-inspiration. With regards to cardiac chambers, this decreases transmural


pressure in the normally compliant left atrium, resulting in an underestimationof wedge pressure. Conversely, during mechanical ventilation, intrathoracicpressure may increase to >10 mm Hg during end-inspiration. PCWP measure-ment is also particularly vulnerable to elevations in intrathoracic pressure,because elevated intrathoracic pressure blocks retrograde transmission of leftatrial pressure events to the catheter tip resulting in a significant underesti-mation of wedge pressure. In both cases, since end-expiration more closelyresembles atmospheric pressure, intracardiac pressures should be recorded atend-expiration (Figure 3.16).

CHAPTER 4

Arterial pressure

George A. Stouffer

Arterial pressure

Arterial blood pressure is one of the most fundamental measurements in hemo-dynamics. Because of the need to quantify pressure levels, emphasis is placedon mean, systolic, or diastolic values (see Table 4.1). These values, while tremen-dously useful, do not provide complete information to characterize the com-posite pressure wave either within the aorta or as it undergoes significantchanges during propagation within the arterial system.

The aortic pressure wave is not just determined by left ventricular mechanicsbut rather results from hydraulic interactions between blood ejected from thecontracting left ventricle and the systemic arterial system. The central aorticpressure wave is composed of a forward-traveling wave generated by left ven-tricular ejection and a backward-traveling wave reflected from the periphery.Ejection of blood into the aorta generates a pressure wave that is propagatedto other arteries throughout the body at a given velocity (termed pulse wavevelocity, PWV). At any discontinuity of the vascular wall, but mainly at thearteriolar branching points, the wave is reflected and comes back toward theheart at the same PWV. One way to think about reflected waves is to note thatthey would be absent if the aorta were an open tube rather than part of a closedsystem.

Pressure wave reflection in the arterial system serves two beneficial pur-poses. When normally timed, the reflected wave returns to the central aorta indiastole and therefore enhances diastolic perfusion pressure in the coronarycirculation. Partial wave reflection also returns a portion of the pulsatile energycontent of the waveform to the central aorta where it is dissipated by viscousdamping. Thus, wave reflection limits transmission of pulsatile energy into theperiphery where it might otherwise damage the microcirculation.

The velocity at which the outgoing and reflected waves travel is dependenton the properties (especially elasticity) of the arteries along which they prop-agate. PWV increases with stiffness and is defined by the Moens–Kortewegequation, PWV = (Eh/2�R), where E is Young’s modulus of the arterial wall(a measurement of elasticity), h is wall thickness, R is arterial radius at the endof diastole, and � is blood density. An important (although simplified) con-cept is that the longitudinal velocity of pressure waves traveling in distensibletubes is slowed by the extent that the vessel expands with each pulsation. In

57


Table 4.1 Definition of terms used to describe blood pressure

Term Definition

Systolic pressure Maximum pressure (peak of arterial pressure wave)Diastolic pressure Minimum pressure (trough of arterial pressure wave)Pulse pressure Difference between systolic and diastolic pressureMean arterial pressure Average pressure during the cardiac cycle

the normal arterial system, there is a steep gradient of increasing arterial stiff-ness moving outward from the heart. In a young adult, PWV is only 4–6 m/sin the highly compliant proximal aorta and increases to 8–10 m/s in the stifferperipheral muscular arteries.

Disturbed reflection, e.g. with increased PWV, can cause the reflected wavesto occur earlier, affecting the central arteries during systole and not duringdiastole. Thus as aortic elasticity declines, transmission velocity of both for-ward and reflected waves increase, which causes the reflected wave to arrive atthe central aorta earlier in the cardiac cycle and thus augment pressure in latesystole. This in turn contributes to an increase in systolic blood pressure andpulse pressure and a decrease in diastolic blood pressure [1]. These changesincrease left ventricular afterload and decrease coronary perfusion pressure.

The concept of PWV was originally described early in the twentieth centurybut recent advances in noninvasive technologies have greatly increased theinterest in using PWV as a surrogate for vascular disease. There are manydifferent ways to measure PWV but the concept is similar. The time delaybetween arterial pulse wave arrival at a proximal artery (e.g. carotid) and amore distal artery (e.g. the femoral) is measured noninvasively. The distancetraveled by the pulse wave is estimated (unless aortic imaging is availableto enable direct measurement) and PWV is then calculated as distance/time(m/s). Measurement of PWV has been shown to correlate with aortic stiffnessand more importantly, to be predictive of clinical events [2, 3]. Not surprisingly,PWV is increased in individuals with hypertension, diabetes mellitus, tobaccouse, atherosclerosis, and end-stage renal disease.

A simplified relationship to remember is stiffer arteries → increased PWV→ earlier arrival of reflected waves → augmentation of systolic rather thandiastolic pressure → increased pulse pressure.

Aortic pressure

The definition of hypertension has changed over time as the deleterious effectsof even minor increases in blood pressure have been realized. The classificationof normal blood pressure, prehypertension, and hypertension outlined in theSeventh Report of the Joint National Committee on Prevention, Detection,Evaluation, and Treatment of High Blood Pressure (JNC 7) is described inTable 4.2 [4].

Chapter 4 Arterial pressure 59

Table 4.2 Classification of hypertension by the Seventh Report of the Joint National Committeeon Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7)

SBP (mm Hg) DBP (mm Hg)

Normal <120 and <80Prehypertension 120–139 or 80–89Stage 1 hypertension 140–159 or 90–99Stage 2 hypertension ≥160 or ≥100

Aortic pressure is the primary determinant of the afterload against which theleft ventricle (LV) must pump blood. Given consistent preload and contractility,an increase in afterload will reduce cardiac output. The primary determinantsof afterload are mean arterial pressure, aortic compliance (distensibility), andaortic valve resistance (the normal aortic valve presents minimal resistanceto flow). Thus, hypertension, atherosclerosis, and aortic stenosis all representconditions of increased afterload.

Mean arterial pressure

The mean arterial pressure (MAP) is the average pressure during the cardiaccycle. Calculation of the true MAP requires integration of the arterial pressureover time. In clinical medicine, a useful approximation of MAP when the heartrate is 60 bpm can be obtained using the following formula:

MAP = DBP + 1/3 (pulse pressure)

where DBP is the diastolic blood pressure. This formula is only applicable atlower heart rates because the relative amount of time spent in systole increasesat higher heart rates.

The major determinant of systolic pressure is stroke volume. Lesser influ-ences include diastolic pressure and aortic compliance. The major determinantof diastolic pressure is systemic vascular resistance (in turn, primarily deter-mined by arteriolar resistance). Lesser determinants include systolic pressure,aortic compliance, and heart rate. The normal aorta is distensible and diametercan increase up to 15% during LV systole. During diastole, this stored potentialenergy is released as the aorta recoils thus helping to maintain diastolic bloodpressure (sometimes called the diastolic pump). With age, compliance of theaorta decreases as elastin and collagen change in both amount and properties.Systolic pressure, more so than diastolic pressure, tends to increase with agein individuals over 50 years old and this can be at least partially explained bychanges in aortic stiffness. Diastolic pressure, largely determined by periph-eral arterial resistance, increases until middle age and then tends to fall. Incontrast, systolic pressure and pulse pressure, influenced more by the stiffness


of large arteries, as well as peripheral pulse wave reflection and the pattern ofleft ventricular ejection, increase continuously with age.

Pressure waveform

A pressure waveform generated in the aorta has a characteristic waveform(Figure 4.1). When pressure in the LV exceeds aortic pressure, the aortic valveopens and blood flows rapidly from the LV into the aorta. The steep upstroke(or anacrotic limb) coincides with opening of the aortic valve and reflects thestroke volume ejected by the LV into the aorta (Figure 4.2). The rounded partat the top of the waveform (or anacrotic shoulder) reflects continued flow fromthe LV to aorta but at a reduced rate. The downslope of the pressure tracing (ordicrotic limb) is divided by the dicrotic notch (or incisura), which representsclosure of the aortic valve. The location of the dicrotic notch varies according tothe timing of aortic closure in the cardiac cycle and will be delayed in patientswith hypovolemia (Figure 4.3).

Aortic and LV pressures decline with reduced ejection and contraction, andwhen the ventricular pressure drops below that of the aorta, the aortic valvecloses. This is represented in the pressure waveform as a dicrotic notch in thedownslope of the aortic pressure tracing. This marks the end of LV ejection.

Figure 4.1 Schematic of aortic pressure.


ventricular Left systole

Atr

ial s

ysto

le

Isov

olum

etric

cont

ract

ion

Rap

id e

ject

ion

Red

uced

eje

ctio

n

Isov

olum

etric

rela

xatio

n

Rap

id v

entr

icul

arfil

ling

Red

uced

vent

ricul

arfil

ling

Left ventricular diastole(through atrial systole)

ECG

120

100

80

Pre

ssur

e(m

m H

g)

60

40

20

0

Heart 41

sounds As nT

LVEDV

AC V

LVESV

APVp

32

140

100

Ven

tric

ular

volu

me

(mL)

Pre

ssur

e(m

m H

g)

Ven

ous

puls

e(m

m H

g)

60

3

0

30 Pulmonaryvalve opens

Pulmonaryvalve closes

Tricuspid valveopens

Tricuspid valvecloses

Pulmonaryartery

Right ventricle

20

10

0.1 sec

0

P T

R

Q S

Aorticvalveopens

Aorticvalvecloses

Aorta

Mitralvalvecloses

Mitralvalveopens Left atrium

Left ventricle

A C V

Rightatrium

Figure 4.2 Wiggers diagram.

Diastolic pressure declines gradually as blood flows from the aorta into theperipheral vessels.

The contour of the aortic pressure tracing can provide clues to various diseasestates (Table 4.3). For example, aortic pressure that rises rapidly, dips and then


200

aVR

I

II

150

after fluid bolus

100

50

0

A0A0

Figure 4.3 Aortic pressure in a hypovolemic patient. Aortic pressure tracings taken in a patientbefore and after an IV fluid bolus. The aortic waveform has a characteristic shape when leftventricular filling pressures are low. Aortic pressures are low, pulse pressure decreases, thedicrotic limb is steeper, and the dicrotic notch is delayed. The diastolic phase of the aorticpressure tracing gives an indication of peripheral runoff and may also be abnormal. In severedehydration, there is a sudden descent in aortic pressure and then an almost flat diastolic phase.

rises again (pulsus bisferiens) in hypertrophic obstructive cardiomyopathy orrises slowly (pulsus parvus et tardus) in aortic stenosis.

The augmentation index (AIx) is the proportion of central PP that resultsfrom arterial wave reflection and is a commonly used measure of arterial stiff-ness (Figure 4.4). AIx is a function of the timing of the arrival of the reflectedwave at the proximal aorta (determined mostly by large artery PWV) and themagnitude of the reflected waves. The magnitude of reflected waves is deter-mined by the diameter and elasticity of small arteries and arterioles and thusinfluenced by vasoactive drugs. AIx is positively correlated with age and bloodpressure, and inversely correlated with height and heart rate.

Effects of respiration on aortic pressure

Changes in thoracic pressure with respiration will influence blood pressure.During normal, spontaneous respiration, intrathoracic pressure decreases dur-ing inspiration, which in turn, causes a decrease in pericardial and right atrialpressures. This results in augmented systemic venous return to right-sidedchambers and decreased venous return to left-sided chambers. It is thus normalto have a slight decrease in systolic blood pressure (approximately 5 mm Hg)with inspiration (Figure 4.5). Pulsus paradoxus is an exaggerated decrease


Table 4.3 Disease states associated with various pulse characteristics

Label Characteristics Disease states

Bounding Large pulse pressure Aortic insufficiency, hyperkineticstates (e.g. fever,thyrotoxicosis, anemia), AVfistula

Pulsus parvus et tardus Slow and weak pulse Aortic stenosisPulsus bisferiens Two systolic peaks prior

to dicrotic notchHypertrophic obstructive

cardiomyopathy and aorticinsufficiency

Pulsus alternans Alternating strong andweak pulses

Severe LV dysfunction

Pulsus paradoxus Excessive decrease(>10 mm Hg) insystolic pressure withinspiration

Cardiac tamponade; lessfrequently it can be observedwith acute aortic regurgitation(i.e. acute aortic dissection),elevated LVEDP, atrial septaldefect, pulmonaryhypertension or rightventricular hypertrophy

Dicrotic pulse Two systolic peaks withone occurring afterdicrotic notch

Low cardiac output + lowperipheral resistance; rare inpatients >45 years old

s

d d

Systolicpressure

Inflectionpressure

Distolic pressure

Figure 4.4 Augmentation index—the augmentationindex (AIx) is given by AIx = ± Ps−Pi

Ps−Pdwith Ps , Pd,

and Pi indicating systolic, diastolic, and inflectionpressure, respectively.

in systolic pressure with inspiration and is variously defined as a drop of>12 mm Hg, a drop of≥10 mm Hg, or a drop of≥9% during normal inspiration.

Not surprisingly, positive pressure ventilation causes an inversion in thenormal relationship between respiration and blood pressure. Blood pressurewill increase during inspiration (as thoracic pressure increases).


1mV

1mV1mV

1mV1mV

1mV

1mV

1mV

1mV

1mV

1mV

1mV

25 mm/s > 1.50 mm Hg/mm

I

II

V5

I

II

V5

IGain:200

IIGain:200

V5Gain:400

200

100

200

100

0

200

100

0

0

r

s s s s s ss s s s

d

r

s

d d

s

d

1 2 3 4

dd d

d dd

s ss s

s

r

ss

d d dd d d d

dd d d

s ss s

ss

s ss s

s

r r r r r r r r r r r r

d d d d d d d d

r r r r r r r rr(a)

(b)

(c)

Figure 4.5 Effect of respiration on aortic pressure. The tracing in (a) is from a healthy 43-year-oldfemale. Note the slight decrease (<10 mm Hg) in systolic pressure with respiration. The tracing in(b) is from a patient with pleural effusions. Note that the systolic pressure declines by18–22 mm Hg with respiration. The tracing in (c) is taken from a patient in cardiac tamponade.Systolic pressure decreases by approximately 30 mm Hg with inspiration. Note also that thewaveform is very abnormal with a steep dicrotic limb, narrow ejection phase, and flat diastolicphase. These findings are consistent with decreased stroke volume and vasoconstriction.


r

s s

s

s

s s

s s

s s

s s

s s

s

s

e e e e edd d d d d d d

dddd

dddd

r r r r r r

I

II

V5

200

200Aorta

RFA

025mm/s >11 12 13 14 15 16 17

Figure 4.6 Peripheral amplification. Simultaneous aortic and right femoral artery pressures in a49-year-old male with aortic insufficiency. Note that the systolic pressure is higher and the rapidincrease phase narrower in the RFA.

Peripheral amplification

Peripheral amplification is said to occur when systolic blood pressure is higherin peripheral arteries (e.g. femoral or brachial) than in the aorta (mean pressureswill be the same). It is primarily observed in young individuals and is causedby the reflected pressure waves returning to the aorta during diastole, makingpulse pressure higher in peripheral than in central arteries (Figure 4.6).

Noninvasive measurement of blood pressure

Intra-aortic pressures are measured directly in the catheterization laboratory.Most clinical decisions are, however, based upon noninvasive measurementof blood pressure. This introduces two potential problems: inaccuracies asso-ciated with noninvasive measurement of arterial pressure and inaccuraciesassociated with the use of the brachial artery, rather than the aorta, as the siteof measurement. Despite these limitations, there is a large amount of infor-mation on more than 1,000,000 patients [5] demonstrating the usefulness ofnoninvasively determined brachial blood pressure in clinical decision making.

Determination of blood pressure by sphygmomanometer utilizes soundsthat are thought to originate from a combination of turbulent flow and arterialwall oscillations. As the blood pressure cuff is deflated from a supraphysiologicpressure, flow through the brachial artery will begin once the systolic pressureis reached. Turbulent flow (and arterial oscillation) ceases once cuff pressurefalls below diastolic pressure.

In 1905, at a conference at the Imperial Medical Academy, Dr NicolaiKorotkoff announced a new way to measure blood pressure. Since that time,the sounds heard by a stethoscope placed over the brachial artery during bloodpressure cuff deflation have been called Korotkoff sounds and can be dividedinto five phases. Phase 1 occurs when the cuff pressure equals the systolicpressure and is characterized by a sharp tapping sound. As cuff pressure is


lowered, phase 2 occurs which is characterized by softer and longer sounds.Phase 3 is defined by a resumption of crisp tapping sounds, similar to thoseheard in phase 1. Phase 4 begins when there is an abrupt muffling of soundas turbulent flow decreases. Finally, phase 5 is when turbulent flow ceasesand thus no sound is heard. There is agreement that the onset of phase 1corresponds to systolic and that the disappearance of sounds (phase 5) cor-responds to diastolic pressure. Although some investigators have advocatedusing phase 4 to define diastole, general practice now is to use phase 4 onlyin situations in which sounds are audible even after complete deflation of thecuff such as in pregnancy, arteriovenous fistulas, and aortic insufficiency. Noclinical significance has been attached to phases 2 and 3. It is important tonote that using Korotkoff sounds to measure blood pressure tends to underes-timate intra-arterial systolic pressure and overestimate intra-arterial diastolicpressure [6].

References

1 Safar ME, Levy BI, Struijker-Boudier H. Current perspectives on arterial stiffness and pulsepressure in hypertension and cardiovascular diseases. Circulation 2003;107:2864–2869.

2 Blacher J, Asmar R, Djane S, London GM, Safar ME. Aortic pulse wave velocity as a markerof cardiovascular risk in hypertensive patients. Hypertension 1999;33:1111–1117.

3 Laurent S, Boutouyrie P, Asmar R, et al. Aortic stiffness is an independent predictor of all-cause and cardiovascular mortality in hypertensive patients. Hypertension 2001;37:1236–1241.

4 Chobanian AV, Bakris GL, Black HR, et al. The Seventh Report of the Joint National Com-mittee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: theJNC 7 report. JAMA 2003;289:2560–2572.

5 Lewington S, Clarke R, Qizilbash N, Peto R, Collins R. Age-specific relevance of usual bloodpressure to vascular mortality: a meta-analysis of individual data for one million adults in61 prospective studies. Lancet 2002;360:1903–1913.

6 Pickering TG, Hall JE, Appel LJ, et al. Recommendations for blood pressure measurementin humans and experimental animals. Part 1: blood pressure measurement in humans: astatement for professionals from the Subcommittee of Professional and Public Educationof the American Heart Association Council on High Blood Pressure Research. Circulation2005;111:697–716.

CHAPTER 5

The atrial waveform

David P. McLaughlin, George A. Stouffer

Introduction

A fundamental understanding of the atrial waveform is extremely importantfor anyone applying hemodynamics to patient care. Data regarding volumestatus, valvular pathology, and ventricular compliance are contained withinthe tracings when examined carefully. The right atrial waveform is typicallymeasured directly with a fluid-filled pressure transducer placed in the rightatrium (RA). The left atrial waveform in contrast can be measured directly bytranseptal atrial puncture or more commonly indirectly via a balloon-tippedpulmonary artery catheter. In most cases, the pulmonary capillary wedge pres-sure (PCWP) accurately reflects mean left atrial (LA) pressure; however, thepressure wave deflections are delayed and damped on a PCWP tracing com-pared to LA pressures [1]. Similarly, the timing of the atrial waves and descentsin reference to the ECG will occur later in the PCWP tracing than the RA trac-ing, because of the delay in transmission of pressures from the left atriumthrough the pulmonary vein, capillaries, and pulmonary artery to the tip of thecatheter.

The components of the atrial wave

In patients who are in sinus rhythm the atrial waveform is composed of twopositive and two negative deflections (Figure 5.1). The A wave is caused byatrial contraction and usually occurs 60–80 milliseconds after the onset of theP wave on the ECG. The A wave is delayed to about 200 milliseconds afterthe onset of the P wave when measured using PCWP due to the time delay intransmission of the reflected wave. The A wave is absent in atrial fibrillation(because of the lack of atrial contraction) and is exaggerated in patients with anoncompliant ventricle or in the presence of tricuspid or mitral stenosis.

The decay of the A wave is the X descent, which is due to the decrease inatrial pressure as a consequence of atrial relaxation (Figure 5.1). In the RAthe X descent is often interrupted by a small positive deflection, the C waveresults from closure of the tricuspid valve during isovolumetric contraction ofthe ventricle. The C wave is more commonly seen in RA compared to PCWPtracings and is prominent in patients with prolonged PR intervals. AdditionalX descent after the C wave is referred to as the X′ descent.

67


A VY

X′X

mv

mv

V5 mv

1mv

1mv

1mv

I

II

50

25

025mm/s>[18]

r r r r r r r r r r r r

A

VV

V

V V

V

VV

V V

V

A AA

AA

A AA

AA

A

0.38mmHg/mm 1 2 3 4 5 6 7 98

(a)

(b)

(c)

Figure 5.1 Atrial pressure waveforms. Schematic (a) and actual tracing (b) of RA pressureshowing A and V waves and X and Y descents. Panel (c) is a PCWP tracing showing V > A.(The schematic in (a) is courtesy of Jonathan Mark, Duke University and was originally publishedin Mark JB: Atlas of cardiovascular monitoring. New York, Churchill Livingstone, 1998.)

Chapter 5 The atrial waveform 69

The second positive deflection in the atrial pressure tracing is the V wavewhich represents rapid atrial filling during ventricular systole while the atri-oventricular valve is closed. The peak of the V wave usually corresponds tothe T wave on the ECG. The decay of the V wave is referred to as the Y descent(Figure 5.1). The Y descent corresponds to rapid ventricular filling in earlydiastole.

The waveforms are dependent upon heart rate. Tachycardia reduces theduration of diastole and can abbreviate the Y descent thus causing the V andA waves to merge. Bradycardia causes the waves to become more distinct andcan elicit an H wave, a mid- to late-diastolic plateau that follows the Y descentand precedes the A wave.

The mean LA pressure is higher than the mean RA pressure in the normalheart. The A wave is usually higher than the V wave in the RA; in the LAeither the V wave is greater than the A wave or the A and V waves are nearlyequal. Also in the normal heart, mean atrial pressure falls with inspiration.The opposite effect, in which the mean RA pressure does not decrease withinspiration, is seen in constrictive pericarditis and occasionally other conditions(referred to as Kussmaul’s sign).

It is important to remember that atrial pressure tracings represent pressure(and provide some indirect information on atrial volume) but not flow. Venousreturn to the RA is dependent upon the pressure gradient between the venoussystem and the RA. Consequently, venous return to the RA is maximal duringperiod of low pressure (X and Y descents).

Abnormalities in atrial pressures

Elevations in RA pressure are often a clue to cardiovascular pathology.Increased mean RA pressure is seen in valvular heart disease, constrictivepericarditis, restrictive cardiomyopathy, cardiac tamponade, fluid overload(e.g. renal failure), pulmonary hypertension, left to right shunts, pulmonaryembolus, RV infarction, LV systolic or diastolic failure, and congenital heartdisease. It is important to note that mean RA pressures may appear normal inpatients with heart disease who are dehydrated.

The A wave is accentuated in patients with noncompliant ventricles andtherefore can be increased in patients with ventricular ischemia, infarction, orventricular hypertrophy from pressure or volume overload. A prominent RA Awave can occur in tricuspid regurgitation, pulmonic regurgitation, or stenosisas well as pulmonary embolus or cor pulmonale from any cause (Figure 5.2aand Tables 5.1 and 5.2). Intermittent cannon A waves can be seen in completeheart block or ventricular tachycardia as the atria are contracting intermittentlyagainst a closed AV valve (Figure 5.2b). In patients with atrial flutter or atrialtachycardia, A waves may be present at an accelerated rate. Small A waves arerare in RA tracings but common in LA tracings (Figure 5.2c).

Abnormalities in the X descent are seen predominantly with severe TR orMR when the X descent and C wave can be interrupted by a prominent V wave.


1mV1mV

A

25

20

15

10

5

0

A A

A

A A A

A

V V V V

V

PW25

0

Q

V

50

(a)

(b)

Figure 5.2 Examples of prominent A waves (a) and V waves (b).

The V wave reflects atrial pressures during ventricular systole and thus canbe useful as a clue to the presence of AV valve regurgitation. In the RA aprominent V wave can be seen in patients with tricuspid regurgitation andthe RA pressure tracing can take on a ventricularized character when TR issevere (Figure 5.3). Pitts et al. reported that neither prominent right atrial Vwaves nor elevated mean right atrial pressures reliably predicted the presenceof moderate or severe tricuspid regurgitation (as determined by echocardio-graphy) but that the absence of right atrial V waves and elevated mean rightatrial pressures were relatively specific for the excluding significant tricuspidregurgitation [2]. For the purposes of their study, a “prominent” right-sidedV wave was defined as: (1) V wave >15 mm Hg, (2) a difference between Vwave and mean RA pressure >5 mm Hg, or (3) a ratio of V wave to mean RApressure >1.5.


Table 5.1 Conditions associated with abnormal right atrial pressure tracings

Prominent Blunted

A wave A–V dissociation (e.g. complete heart block) Atrial arrhythmia(or ventricular tachycardia)

Decreased ventricular complianceTricuspid regurgitation or stenosisPulmonary stenosis or regurgitation

X descent Tamponade Atrial arrhythmiaRV ischemia (normal RA) RA ischemiaAtrial septal defect Tricuspid regurgitation

V wave Tricuspid regurgitation TamponadeRV failure Constrictive pericarditis

HypovolemiaY descent Constrictive pericarditis Tamponade

Restrictive cardiomyopathy RV ischemiaTricuspid regurgitation Tricuspid stenosis

Commonly, however, the left-sided V wave receives the most attention andthat is what we will focus on in this section (examples are shown in Figure 5.4).The PA systolic pressure waveform and left atrial V wave are generated atthe same time (assuming the ventricles contract at the same time), however,the V wave in the PCWP tracing will be seen later because of the delay intransmission of left atrial pressure to the catheter tip. Moore et al., in theirstudy of 13 patients with large V waves, found that the onset of the V wavewas approximately 110 milliseconds after the onset of the pulmonary arterialupstroke [3].

The V wave in the PCWP (or LA) tracing is dependent on the pressure–volume relationship of the left atrium and left atrial compliance as well as thevolume of blood entering the atrium. A large V wave indicates a rapid rise in LApressure during ventricular systole and can occur in setting of a noncompliantleft atrium and normal flow (e.g. atrial ischemia) or a normally compliant leftatrium with large flow (e.g. acute mitral regurgitation). Conversely, if the leftatrium has a large capacitance than a prominent V wave may be absent evenin the presence of severe mitral regurgitation.

In the left atrium or pulmonary capillary wedge pressure tracing, a promi-nent V wave suggests mitral regurgitation but is neither sensitive nor specific.Some investigators have advocated using a scale where a V wave that is twicethe mean wedge pressure is suggestive of severe MR and a V wave maximumof three times mean wedge pressure is diagnostic of severe MR but this has notbeen widely validated (Table 5.3). Though not typical, a prominent V wave canoccasionally be observed in patients with mitral stenosis if the LA is relativelynon-compliant with limited ability to accept expanded filling volumes.

In patients with severe MR and a large V wave, mean PCWP can overestimateLV end-diastolic pressure. In a study by Haskell and French [4], mean PCWPpressure overestimated LVEDP by approximately 30% (in this study, a large V


Tab

le5.

2R

ight

atria

lpre

ssur

etr

acin

gfin

ding

sin

vario

usco

nditi

ons

Vw

ave

Yde

scen

tA

wav

eX

desc

ent

RA

mea

npr

essu

reM

isc

Nor

mal

Atr

ialfi

lling

follo

win

gR

apid

vent

ricul

arA

tria

lcon

trac

tion

0–5

A>

Vve

ntric

ular

cont

ract

ion

fillin

gph

ase

LA:V

>A

norm

alde

lay

inpr

essu

retr

ansm

issi

onfr

omLA

topu

lmca

p:14

0–20

0m

sC

onst

rictio

nP

rom

inen

tP

rese

rved

Ele

vate

dtr

acin

gsca

nha

veM

orW

form

Kus

smau

l’ssi

gnTa

mpo

nade

Sm

all/a

bsen

tP

rese

rved

Ele

vate

dE

ffusi

veco

nstr

ictiv

eP

rom

inen

tP

ersi

sten

tel

evat

ion

afte

rpe

ricar

dioc

ente

-si

s

RA

pres

sure

rem

ains

elev

ated

desp

iteco

mpl

ete

drai

nage

(per

icar

dial

pres

sure

=0)

Res

tric

tion

Pro

min

ent

Pre

serv

edE

leva

ted

RA

trac

ings

can

have

Mor

Wfo

rmR

Vin

farc

tIm

paire

dE

leva

ted

RA

inta

ct:i

ncre

ased

Aw

ave

and

stee

pX

desc

ent

RA

infa

rct:

decr

ease

dA

wav

ean

dim

paire

dX

and

Yde

scen

ts


I

II50

25

0

I

II

V 5

V 5

I

II

V 5

I

II

V 5

50

25

0

r

r r r r r r r r r

r r r r r r r r r

a a

a a a a a a a a a a

a a a a a a a av

v

a av v v v v v v v v

vv v v v v v v v

vv v v v v v v

1 2 3 4

1mV

1mV

1mV

1mV

1mV

1mV

1mV

1mV

1mV

1mV

1mV

mV

200

100

025mm/s >[-1-]

75mm/s >[-1-]

r r r r r r r r r r r

a a a a a a

1.50mm Hg/mn

0.38mm Hg/mn

1

1 2 3

2 3 4 5 6 7 8 9SERIES IV- WITH BIOMEDICAL

Normal Sinus Monitor Length: 10sec

r r r r r r r

vv

vv

v v

aa

50

25

0SERIES IV- WITH BIOMEDICAL

(a)

(b)

(c)

Figure 5.3 Pressure tracings from a patient with severe tricuspid regurgitation. Panels (a) and(b) show RA pressures on 50 and 200 scale, respectively. Note the “ventricularization” of the atrialpressure. Panel (c) shows simultaneous RA and RV pressures.


1mV

1mV1mV

1mV

1mV

1mV

r r r r r r r

a aa a a

aa a

v v

v

v v

v

v v

1 2 3 4 5 6 7 8

50

25

0

e e

d d d dd

d

e e

ee

a aa a

aa

vv v v

v

v

Figure 5.4 Examples of prominent V waves.


Table 5.3 Important points about V waves

The V wave on PCWP tracing corresponds to: (1) T wave on ECG and(2) descent of pressure tracing on LV recording

V wave >2x mean PCWP is suggestive of severe MRV wave >3x mean PCWP is diagnostic of severe MRAbsence of significant V waves does not exclude significant MRWhen large V waves are present, trough of Y descent is better predictor of LVEDP

than is mean PCWP [4]V waves are affected by arterial systolic blood pressure (i.e. ventricular afterload)

wave was defined as being greater than the A wave by more than 10 mm Hg).A better estimate of LVEDP was obtained by using the trough of the Xdescent.

The Y descent or decay in the V wave occurs during the opening of theAV valve and represents rapid emptying of the atria into the ventricle (whichresults in a drop in pressure). Anything that prevents or enhances early diastolicemptying affects the character of the Y descent. Tricuspid or mitral stenosiscauses an attenuation or flattening of the terminal Y descent (Figure 5.5). Note

I

II

V5

100

50

075 mm/s >

[-8-]0.75 Hg/mm SERIES IV-WITH MEDICAL

s ss

sr r r r

LV

PCWP

d

d d

d

ee

e e

a

a

a

vv

v

1mV

1mV

1mV

1mV

1mV

1mV

Figure 5.5 Example of simultaneous PCWP and left ventricular pressure tracings in a patient withmitral stenosis and mitral regurgitation. Note the persistent pressure gradient between thechambers throughout diastole. Note also that the PCW pressure is delayed relative to LV pressureand accurate determination of diastolic pressure gradient and mitral valve area requires shiftingthe PCWP tracing to the left. The V wave on the PCWP tracing should coincide with thedownstroke (isovolumetric relaxation) of the LV tracing.


r

v aa a a a a

avv v v v v

v

r r r r r r r

Figure 5.6 RA tracing in a patient with constrictive pericarditis. Constrictive pericarditis orrestrictive cardiomyopathy are two hemodynamic states that cause significant abnormalities of theatrial wave with prominent Y descents. There are typically high atrial pressures and prominentearly diastolic filling waves.

also the prominent V wave. Venous pressures are elevated and Y descents areprominent in various conditions including constrictive pericarditis (Figure 5.6)and right ventricular infarction (Figure 5.7). Prominent Y descents can also bepresent in cardiac tamponade (during inspiration; Figure 5.8) and become moreapparent following pericardiocentesis. The Y descent can vary with positionof the patient in the rare instance of large atrial myxoma.

Physical exam

Physical examination of jugular venous pulse (JVP) provides valuable infor-mation to clinicians as the JVP is a manometer of pressure in the right atrium.To determine JVP, sit the patient at 45◦ and turn his or her head slightly away


r rr r

r rr r

r r r rr

a aa a

a a a a a a av v v v vv

vv v v

Figure 5.7 RA tracing in a patient with right ventricular myocardial infarction. RV infarction canoccasionally cause a pattern of pseudoconstriction with a prominent X and Y descent. This occursdue to acute distention of the volume overloaded RV into a contained pericardial space.

from you. Use the internal jugular vein (not external jugular vein) medialto the clavicular head of sternocleidomastoid and measure JVP in centime-ter above the sternal notch (vertical not diagonal distance). The venous pulsecan usually be analyzed more readily using the right internal jugular veinas the right innominate and jugular veins extend in an almost straight linefrom the superior vena cava. Sir Thomas Lewis in 1930 proposed a sim-ple bedside method for measuring central venous pressure. He found thatthe catheter-measured CVP was equal to 5 cm plus the vertical distance ofthe JVP from the sternal angle and that this measurement was independentof the individual’s position (e.g. supine, semiupright, or upright). Remem-ber that CVP in cm of blood must be divided by 1.34 to get pressure inmm Hg. Also note that the distance between the sternal angle and the mid-portion of the RA, which Lewis estimated at 5 cm, varies considerably betweenindividuals and is affected by age and anterior – posterior diameter of thechest [5].

JVP can be used to assess right heart filling pressures and the waveformcan provide information suggestive of specific diagnoses. Raised JVP withnormal waveform is suggestive of right heart failure and/or fluid overload.The lack of pulsations in a patient with elevated JVP would be suggestive ofSVC obstruction. Various abnormalities of the A wave, V wave, X descent, or



I

II

V5

50

25

aa

aa a a

a

a

aav v v

v v vv v v

0m/s >

1 2 3 4 5 60.38 mm H/mmg

Figure 5.8 RA tracing in a patient with pericardial tamponade. Tamponade results in pandiastolicrestriction of ventricular filling as pericardial pressure exceeds RA pressure throughout diastole.This results in blunting of both the X and Y descents, which often return following successfulpericardial drainage.

Y descent suggest diseases outlined in Tables 5.2 and 5.3. An increase in JVPwith inspiration (Kussmaul’s sign) is suggestive of constrictive pericarditis.

Important points� The atrial waveform is composed of the positive atrial (A) and ventricular(V) waves and the X and Y descents.� The A wave is prominent with abnormalities of RV and LV compliance (e.g.hypertrophy, ischemia).� The V wave is accentuated in cases of atrioventricular valve regurgitation(RA V wave in TR, LA V wave in MR).� The X and Y descents are diminished or absent in pericardial tamponade.� The X and Y descents are accentuated in constriction and restrictive myocar-dial disease.


Historical anecdote

Karel Wenckebach described the arrhythmia that bares his name (also known asMobitz type 1 AV block) in a paper entitled “On the analysis of irregular pulses”that was published in 1899 before the advent of clinical electrocardiography. Hedescribed the arrhythmia based on a careful examination of the arterial pulseand the A wave in the jugular venous waveform in a “forty-year old lady . . .

with a pulse that was small, soft and intermittent after 3 to 6 contractions” whohad consulted him (more details can be found in Ann Intern Med 1999;130:58–63or Tex Heart Inst J 1999;26:8–11).

References

1 Batson GA, Chandrasekhar KP, Payas Y, Rickards DF. Comparison of pulmonary wedgepressure measured by the flow directed Swan–Ganz catheter with left atrial pressure. BrHeart J 1971;33:616.

2 Pitts WR, Lange RA, Cigarroa JE, Hillis LD. Predictive value of prominent right atrial Vwaves in assessing the presence and severity of tricuspid regurgitation. Am J Cardiol 1999;83:617–618, A10.

3 Moore RA, Neary MJ, Gallagher JD, Clark DL. Determination of the pulmonary capillarywedge position in patients with giant left atrial V waves. J Cardiothorac Anesth 1987;1:108–113.

4 Haskell RJ, French WJ. Accuracy of left atrial and pulmonary artery wedge pressure inpure mitral regurgitation in predicting left ventricular end-diastolic pressure. Am J Cardiol1988;61:136–141.

5 Seth R, Magner P, Matzinger F, van Walraven C. How far is the sternal angle from themid-right atrium? J Gen Intern Med 2002;17:852–856.

CHAPTER 6

Cardiac output

Frederick M. Costello, George A. Stouffer

Introduction

A key component of evaluating the hemodynamic status of a patient is a mea-surement of cardiac output. Cardiac output is the amount of blood moved perunit time from the venous system (i.e. the vena cava) to the arterial system (i.e.aorta). It is a dynamic process and tightly regulated so that blood flow throughthe heart equals perfusion needs of the body.

Cardiac output is primarily regulated by preload, afterload, heart rate, andmyocardial contractility. Many factors have rapid effects on cardiac outputincluding metabolic demands, posture, volume status, adrenergic state, andrespiratory rate (e.g. cardiac output increases with inhalation and decreaseswith exhalation or the Valsalva maneuver). Cardiac output decreases with age[1] and increases with exercise with maximal observed increases in cardiacoutput with exercise being sixfold in trained athletes (note that oxygen extrac-tion also increases with exercise and thus oxygen delivery to tissues increases12- to 18-fold). Cardiac index is the cardiac output divided by the body sur-face area. A cardiac index below <1 L/min/m2 is generally incompatiblewith life.

Cardiac output is a measurement of the forward flow of blood in the vascularsystem and is equal to the heart rate × stroke volume (in the absence of valvularregurgitation). Measurement of the heart rate is easily attainable but an accu-rate measurement of stroke volume is more challenging (see Table 6.1). Becauseof the difficulty in measuring left ventricular stroke volume, several methodshave been developed that calculate blood flow through the right heart. In thecardiac catheterization laboratory or in the intensive care unit, cardiac outputis primarily measured using either the Fick method or the indicator dilutionmethod (e.g. thermodilution). Each of these techniques has advantages andpitfalls that are discussed in detail below (see Table 6.2).

Fick method

In 1870, Adolph Fick described the first method to estimate cardiac output inhumans. He postulated that oxygen uptake in the lungs is entirely transferredto the blood and therefore that cardiac output can be calculated knowing oxy-gen consumption of the body and the difference in oxygen content between

81


Table 6.1 Important formulas involving cardiac output.

SVR (in dynes s/cm5) = (mean arterial pressure − right atrial pressure) × 80cardiac output

where SVR = systemic vascular resistance. Normal SVR = 700–1600 dynes s/cm5

PVR (in dynes s/cm5) = (mean PA pressure − PCWP) × 80cardiac output

where PA = pulmonary artery, PCWP = pulmonary capillary wedge pressure, and PVR = pulmonaryvascular resistance. Normal PVR = 20–130 dynes s/cm5

CO = SV × HR

where CO = cardiac output, SV = stroke volume, and HR = heart rate

Table 6.2 Comparison of Fick and thermodilution methods of determining cardiac output.

Fick Thermodilution

Underlying principle Fick principle(Conservation ofmass)—the uptake orrelease of a substanceby an organ is theproduct of blood flow tothe organ × difference inthe concentration of thesubstance betweenblood entering and bloodleaving the organ

Conservation of energy - that is, thatthere is no loss of cold injectatebetween the site of injection anddetection. Other underlyingassumptions are that mixing of theindicator and blood is completeand that the temperature changeelicited by injection of saline canbe discriminated accurately fromthe fluctuations in baselinetemperature in the pulmonaryartery.

Sources of error Determination of oxygensaturations

Warming of injectate during transitthrough catheter

Assumption of steady state Irregular heart ratesAssumed oxygen

consumption valuesTricuspid regurgitationTemperature of blood in PA varies

with respiratory and cardiac cyclesAdvantages More accurate than TD in

low output states,tricuspid regurgitation,and irregular heart rates

More accurate than Fick in highoutput states

Variation underideal conditions

10% when measuringoxygen consumption

5–20%

arterial and mixed venous blood. In modern practice, oxygen consumptionis rarely directly measured and is usually “assumed.” Direct measurement ofan individual’s oxygen consumption is cumbersome and time consuming andrequires a tight fitting gas exchange system. An alternative approach, althoughless accurate, is to estimate the oxygen consumption based on the patient’sweight (this method to determine cardiac output is called the “assumed Fick”).

Cardiac output can be calculated using the Fick method knowing the oxygensaturation of arterial and mixed venous blood, the hemoglobin concentration,

Chapter 6 Cardiac output 83

and the estimated oxygen consumption (discussed in more detail) based onthe equation below:

Cardiac output

= oxygen consumption(arterial − venous)O2 × hemoglobin concentration × 1.36 × 10

(the constant 1.36 is expressed in mL O2/g hemoglobin).The assumed Fick method is used frequently but it is important to realize

its limitations. The initial evaluation of the assumed Fick method was per-formed in young healthy adults. Subsequently, Krovetz and Goldbloom [2],LaFarge and Miettinen [3], and Bergstra et al. [4] derived empiric formulasto estimate VO2, while others advocate taking oxygen consumption to be125 mL/min/m2 in adults with or without a correction for age. There is, how-ever, large variability in oxygen consumption and thus use of assumed valuescan introduce errors. In a study of 108 patients (mean age 49 yr) undergoingcardiac catheterization, Dehmer et al. found the mean oxygen consumptionto be 126 mL/min/m2 but with a large standard deviation of 26 mL/min/m2

[5]. Oxygen consumption was not affected by age or sex but varied withlevel of sedation. In a study of 80 patients, Kendrick et al. compared assumed(calculated using five different estimation methods) and directly measuredvalues of oxygen consumption and found significant discrepancies, with overhalf the estimated values differing by more than 10% from directly measuredamount and 35% differing by more than 25% [6]. Similarly, Wolf et al. directlymeasured oxygen consumption in 57 nonsedated patients (mean age 52 yr)undergoing evaluation in a metabolic laboratory and found the mean tobe 126.5 mL/min/m2 but with large variation (71–176 mL/min/m2) [7].Comparison with assumed values calculated using the LaFarge and Miettinenequation showed that there were large, unpredictable errors when assumedvalues were used.

The value ascribed to the assumption of oxygen consumption is generallythe largest error in using the assumed Fick method; however, other sourcesof error exist including measurement of oxygen saturation (see the chapter onIntracardiac Shunts for a more complete discussion of potential errors) andthe assumption of steady state. In the best settings, the error associated withmeasuring cardiac output using the assumed Fick method will be 10–15%.Under less stringent conditions this variation can rise significantly.

In low output states, the Fick method is the most accurate measurementavailable. This is because the differences in oxygen saturations between arterialand mixed venous blood are large, which minimizes errors introduced in themeasurement of oxygen saturation. It is also the most accurate method whenthe patient’s heart rate or rhythm is irregular such as in atrial fibrillation orventricular bigeminy.

Supplemental oxygen use during measurement of the cardiac output shouldbe avoided. If oxygen is required, maintaining a steady state is imperative toobtaining the most accurate measurement.


Thermodilution method

The indicator dilution method was first used in 1897. In its simplest form,it involves injection (either via bolus or continuous infusion) of an indicatorsubstance into the circulation and then measurement of this substance at apoint in the circulation downstream of the injection. The amount of indica-tor that is measured, per unit time after injection, will be a function of car-diac output. The substance must be nontoxic, capable of mixing completelywith blood, not taken up from the blood stream and capable of being accu-rately measured. Numerous indicators have been used over the years includ-ing indocyanine green but currently saline is the substance most commonlyused.

The thermodilution method involves injecting saline into the right atriumand then measuring the temperature of blood in the pulmonary artery. The typ-ical technique involves placement of a right heart catheter with multiple portsinto the proximal pulmonary artery. Ten milliliters of cold saline is injectedinto the right atrium and temperature changes are measured in the pulmonaryartery. Accurate measurement requires that a “mixing chamber” be presentbetween the injection site and the measurement site. Thus, the proximal portshould lie in the right atrium and the distal catheter should be beyond thepulmonary valve.

Injection of saline into the right atrium will cause a transient decrease inpulmonary artery temperature and a thermodilution curve is generated byplotting the temperature of the pulmonary artery versus time. The curve gen-erally has a smooth upslope and a more gentle decline (Figure 6.1). The areaunder the thermodilution curve is inversely related to the cardiac output andcan be used to calculate the cardiac output. The calculation of the area underthe curve and calculation of cardiac output are performed by computer but anestimate of the validity of the measurement can be obtained by checking thewaveform of the thermodilution curve.

Significant tricuspid regurgitation is generally considered a contraindicationto the use of thermodilution for measurement of cardiac output. An early studyby Konishi et al. [8] did not find a significant difference in cardiac outputs mea-sured using thermodilution or Fick in the presence of tricuspid regurgitation.Later studies, however, found that a high degree of tricuspid regurgitationwas associated with underestimation of cardiac output by the thermodilutiontechnique. Balik et al. in their study of 27 patients undergoing cardiac surgeryreported a significant increase in the variation of cardiac output measurementswith increasing degrees of tricuspid regurgitation [9]. Cigarroa et al. in theirstudy of 30 patients (mean age 50 yr) found excellent agreement between ther-mal dilution cardiac output and Fick or indocyanine green dye cardiac outputin patients without tricuspid regurgitation [10], but results of thermodilutionwere consistently lower (by approximately 20%) than those obtained by theFick or indocyanine green dye methods in the 17 patients with tricuspid regur-gitation (see Table 6.2).


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23SERIES IV:WITH MEDICAL

SERIES IV:WITH MEDICAL

37.00

36.60

36.20

24

CO = 6.7 L/min

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

37.00

36.60

36.20

24 25

CO = 4.2 L /min

Figure 6.1 Thermodilution curves. Pulmonary artery temperature versus time is plotted inpatients with thermodilution cardiac outputs (CO) of 6.7 L/min and 4.2 L/min.

Of note is that mechanical ventilation can have an effect on cardiac out-put measurement with the thermodilution technique by causing or worseningtricuspid regurgitation. In a small study, Artucio et al. found that the use of pos-itive end-expiratory pressure (PEEP) caused tricuspid regurgitation to developor worsen in six out of seven patients studied [11]. Subsequently, Balik et al.[9] found that thermodilution underestimated cardiac output in 27 ventilatedpatients (using transesophageal echocardiographic cardiac output as the goldstandard) and that inaccuracies increased proportional to the amount of tri-cuspid regurgitation.

For the sake of simplicity, room temperature saline is often used but this canlead to less accurate measurements. Berthelsen et al. reported a significant risein variation and a loss of precision with the use of 20◦C saline versus the useof 0◦C saline [12]. They also noted, however, that if room temperature salinewas injected through a right heart catheter that had both proximal and distalthermistors, the variation was within an acceptable range. If ice-cold saline isused, careful attention must be paid to not allowing the saline to warm prior toinjection. Iced injectate in a plastic syringe warms by 1◦C for every 28 secondsat room temperature. The rate of warming increases if the injectate is held in agloved hand.


Potential sources of error for thermodilution include “catheter warming”and temperature variation in the pulmonary artery. Catheter warming refersto the increase in temperature of the injectate as it passes through the catheter.Commercially available systems include an empiric constant that corrects forthis warming. Variations in the temperature of blood in the pulmonary arteryare generally minimal except in cases of deep inspiration or positive pressureventilation. In some cases, the variation in the temperature of the blood isenough to affect cardiac output determination. An additional limitation tothermodilution is the inaccuracy associated with determining accurate cardiacoutput with irregular heart rates or rhythms.

Thermodilution is most accurate when used in high output states as theerrors introduced by injectate warming are proportionally smaller. As cardiacoutput declines, the accuracy of thermodilution cardiac output declines (incontrast to the Fick method in which accuracy increases with decreasing car-diac output). van Grondelle et al. compared 57 cardiac output measurements bythe thermodilution and Fick methods in 26 patients and found that thermodi-lution values were higher in all 16 cases in which Fick outputs were less than3.5 L/min. In 10 cases where Fick values were less than or equal to 2.5 L/min,thermodilution and Fick measurements differed by an average of 35% [13].They postulated that thermodilution is less accurate at low cardiac outputsbecause of warming of the injectate during conditions of low flow. Under idealconditions, thermodilution can determine cardiac output within an error rangeof 5–20%.

Doppler echocardiographic measurement of cardiac output

Doppler echocardiography can also be used to determine cardiac output. Theunderlying principle is that the volume of flow can be determined by multi-plying the velocity of flow by the area through which the blood is flowing.Thus, cardiac output can be calculated as the product of the heart rate, aortictime velocity integral, and the area of the left ventricular outflow tract (LVOT).This equation assumes that the area is constant and unchanging and the LVOTis frequently used as it is the least dynamic. Because blood flow is pulsatile,a simple velocity cannot be used. Instead, a time velocity integral must bederived from Doppler measurements of flow.

This technique is very useful and can provide an accurate measurement ofcardiac output but as with all techniques, it too has pitfalls. The measurementof the time velocity integral using Doppler requires diligence. One must mea-sure with pulse wave Doppler at the precise location where the LVOT diameterwill be measured. Additionally, the pulse wave Doppler must be parallel to theflow. Any variance greater than 20◦ off of the direction of flow will significantlychange the calculated result. Another source of potential area is the measure-ment of the diameter of the LVOT. The area of the outflow tract is calculatedby squaring the diameter and multiplying by 0.785. Even very small errors in


the measurement of the LVOT will result in large variations in the area becausethe diameter is squared in this calculation.

References

1 Brandfonbrener M, Landowne M, ShocK NW. Changes in cardiac output with age.Circulation 1955;12:557–566.

2 Krovetz LJ, Goldbloom S. Normal standards for cardiovascular data. I. Examination ofthe validity of cardiac index. Johns Hopkins Med J 1972;130:174–186.

3 LaFarge CG, Miettinen OS. The estimation of oxygen consumption. Cardiovasc Res1970;4:23–30.

4 Bergstra A, van Dijk RB, Hillege HL, Lie KI, Mook GA. Assumed oxygen consumptionbased on calculation from dye dilution cardiac output: an improved formula. Eur Heart J1995;16:698–703.

5 Dehmer GJ, Firth BG, Hillis LD. Oxygen consumption in adult patients during cardiaccatheterization. Clin Cardiol 1982;5:436–440.

6 Kendrick AH, West J, Papouchado M, Rozkovec A. Direct Fick cardiac output: are assumedvalues of oxygen consumption acceptable? Eur Heart J 1988;9:337–342.

7 Wolf A, Pollman MJ, Trindade PT, Fowler MB, Alderman EL. Use of assumed versusmeasured oxygen consumption for the determination of cardiac output using the Fickprinciple. Cathet Cardiovasc Diagn 1998;43:372–380.

8 Konishi T, Nakamura Y, Morii I, Himura Y, Kumada T, Kawai C. Comparison of thermod-ilution and Fick methods for measurement of cardiac output in tricuspid regurgitation.Am J Cardiol 1992;70:538–539.

9 Balik M, Pachl J, Hendl J. Effect of the degree of tricuspid regurgitation on cardiac outputmeasurements by thermodilution. Intensive Care Med 2002;28:1117–1121.

10 Cigarroa RG, Lange RA, Williams RH, Bedotto JB, Hillis LD. Underestimation of cardiacoutput by thermodilution in patients with tricuspid regurgitation. Am J Med 1989;86:417–420.

11 Artucio H, Hurtado J, Zimet L, de Paula J, Beron M. PEEP-induced tricuspid regurgitation.Intensive Care Med 1997;23:836–840.

12 Berthelsen PG, Eldrup N, Nilsson LB, Rasmussen JP. Thermodilution cardiac output. Coldvs room temperature injectate and the importance of measuring the injectate temperaturein the right atrium. Acta Anaesthesiol Scand 2002;46:1103–1110.

13 van Grondelle A, Ditchey RV, Groves BM, Wagner WW, Jr, Reeves JT. Thermodilutionmethod overestimates low cardiac output in humans. Am J Physiol 1983;245:H690–H692.

CHAPTER 7

Detection, localization, andquantification of intracardiac shunts

Frederick M. Costello, George A. Stouffer

Introduction

In the normal circulation, blood passes through the right heart, into the pul-monary circulation, then into the left heart, and finally into the systemic circu-lation in a continuous, unidirectional manner. In certain conditions, however,oxygenated blood is shunted from the left heart directly to the right heart as inthe case of atrial septal defects (ASD; Figure 7.1) or ventricular septal defects(VSD; Figure 7.2) or from the aorta to the pulmonary artery (patent ductusarteriosus, PDA). More rarely, unoxygenated blood can be shunted from theright heart to the left heart (e.g. Eisenmenger’s syndrome). Intracardiac shuntscan be either congenital (Figures 7.1 and 7.2) or acquired (e.g. VSD as a com-plication of myocardial infarction).

Intracardiac shunting of blood results when there is an opening betweenthe right and left heart chambers and a pressure difference between the con-nected chambers. Because pressures (both systolic and diastolic) on the leftside of the heart are generally higher than on the right side, most shunts arepredominantly left-to-right although right-to-left and bidirectional shunts areseen (predominantly in Eisenmenger’s syndrome). Of note is that arterial tovenous shunts can exist outside of the heart (e.g. intrapulmonary, intrahepatic,AV malformation) and effect oxygen saturations. We will not address theseshunts in this chapter but rather concentrate on intracardiac shunts.

Detection of an intracardiac shunt

The presence of a shunt can be determined either invasively or noninvasively(e.g. radionuclide studies, MRI, or Doppler echocardiography). In current prac-tice, many shunts are suspected based on physical exam or ECG, diagnosed atechocardiography and then quantified using right heart catheterization.

There are several methods by which an intracardiac shunt can be detected,localized, and quantified in the catheterization laboratory. The indicator dilu-tion method is of historical interest but rarely used outside of research studies.It involves injecting a substance, such as indocyanine green, into a right-sideheart chamber and then monitoring its appearance in the systemic circulation.

89


SVC

Ostiumprimum

Sinusvenous

Ostiumsecundum

Sinusvenosus

IVC

Coronarysinus

RV

RA

Figure 7.1 Schematic showing the location of various types of atrial septal defects.

Dye curve measurements are very accurate but are slow and require additionalspecialized equipment that is rarely available in modern cath labs. Contrastangiography, in which contrast dye is injected into the higher pressure cham-ber of a suspected shunt (e.g. left ventricle) is occasionally used to identify aVSD or PDA. It has limited ability to diagnose ASDs and is a poor tool forquantifying a shunt.

Oximetry, or measurement of the oxygen saturations in various locations inthe venous system and the right heart (“oxygen saturation run”) is the mostfrequently used invasive technique due to its simplicity and reliance on readilyavailable equipment.

Oxygen saturation runThe oxygen saturation run is performed as a catheter is passed through thevenous system, right heart, and pulmonary circulation. The samples need tobe acquired with the patient breathing room air or a gas mixture containing nomore than a maximum of 30% oxygen [1]. Saturation data may be inaccuratein patients breathing more than 30% oxygen, as a significant amount of oxygenmay be present in dissolved form in the pulmonary venous sample. Dissolvedoxygen is not factored into calculations when saturations are used and thuspulmonary flow will be overestimated and the amount of shunt exaggerated.

The oxygen saturation run typically begins with femoral vein access andis complete when all venous samples have been obtained. Samples can beobtained as the right heart catheter is advanced although many clinicianschoose to place the catheter in the pulmonary artery first and then obtainsamples as the catheter is withdrawn. This latter technique can improve thespeed of obtaining samples, which can become an issue if there is difficultyadvancing the catheter into the pulmonary artery.

Chapter 7 Intracardiac shunting of blood 91

Figure 7.2 Ventriculogram in an LAO projection showing efflux of dye from the left ventricle intothe right ventricle via a VSD. LV, left ventricle; RV, right ventricle; IVS, intraventricular septum;VSD, ventricular septal defect.

Blood samples should be taken from the locations listed in Table 7.1 and oxy-gen saturations determined. Inferior vena cava (IVC) saturation varies depend-ing on where the sample is obtained, and the sampling site should be at the levelof the diaphragm to ensure that hepatic venous blood is taken into account. Anarterial blood sample should be collected at the same time the venous samplesare being obtained.

Multiple samples may have to be taken in various chambers to ensure accu-racy; however, it is important to note that variability in oxygen saturation (inabsence of shunt) decreases as blood flows through the heart. Oxygen contentvaries by approximately 2 mL O2/100 mL blood in the right atrium (RA), 1 mLO2/100 mL blood in the right ventricle (RV), and 0.5 mL O2/100 mL bloodin the pulmonary artery (PA) [2]. These values were determined by direct


Table 7.1 Sites from which oxygen saturation should bemeasured when quantifying intracardiac shunts

Superior vena cava just above the junction with the right atriumInferior vena cava just below the diaphragmLow right atriumMid-right atriumHigh right atriumRight ventricleLeft, right and/or main pulmonary arteryPulmonary capillary wedge (or left atrial)Left ventricle and/or arterial

measurement of oxygen content in the days prior to the development of oxime-try. Oxygen content can be approximated by the following formula:

Oxygen content ( mL O2/100 mL blood)

= hemoglobin (g/dL) × 1.36 ( mL O2/g of Hg) × percent saturation

In a patient with a hemoglobin of 15 mg/dL, a difference in saturation of 5%is approximately equal to a difference in oxygen content of 1 mL O2/100 mLblood.

The goal of the oxygen saturation run is to measure differences in oxygensaturation in various chambers of the heart. There are several practical factorsthat must be kept in mind to ensure accurate results:� The oxygen saturation method assumes that the body is in a steady stateduring the collection of samples. To ensure the results are as accurate as possi-ble, the samples must be collected as close in time as good technique permits.For this reason, many physicians perform a retrograde oxygen saturation runin which they begin with the catheter in the PA.� Blood samples should not be withdrawn into the syringe too rapidly. Mattaet al. found that rapid aspiration increased oxygen saturations [3].� Complete mixing of blood is assumed in all chambers.

Limitations of using oximetry to detect and quantifyintracardiac shuntsUsing the saturation “step-up” method to detect and quantify intracardiacshunts has some inherent limitations that should be noted. First, because of theinherent variability in determining oxygen saturations, the oximetry methodof quantifying shunts loses accuracy when determining small shunts or inthe presence of high cardiac output (which decreases AVO2 difference). Sec-ond, the magnitude of the step-up varies with the oxygen-carrying capacity ofblood and the cardiac output. Saturation step-ups are increased if hemoglobinconcentration is low or cardiac output is low. As Shepherd et al. [4] note, a 5%step-up occurs with a shunt flow of 1300 mL/min when systemic blood flow


is 5 L/min but occurs with a shunt flow of only 285 mL/min if systemic bloodflow is 2.5 L/min or requires a shunt flow of 3400 mL/min when systemicblood flow is 7.5 L/min.

Third, the relationship between the magnitude of step-up and the shunt flowis nonlinear and with increasing left-to-right shunting, a given change in shuntflow produces less of a change in the saturation step-up. In contrast, quantifica-tion of shunt size using Qp/Qs is not affected by hemoglobin concentration andthe relationship between Qp/Qs and shunt flow is linear. Qp/Qs is sensitiveto cardiac output and some investigators advocate using exercise in patientswith low cardiac output to improve accuracy.

Diagnosis of intracardiac shunts at right heart catheterizationPatients are often referred for cardiac catheterization to confirm and quantifya shunt that has been detected by physical exam and/or imaging. Occasion-ally, however, a patient will have a shunt that is unsuspected prior to rightheart catheterization. These patients can be identified by routinely measuringoxygen saturation in the PA and the right atrium (RA). Hillis et al. found thatthe difference (mean ± SD) between RA and PA saturations in 980 patientswithout intracardiac shunts was 2.3% ± 1.7%. In this same population, thedifference (mean ± SD) between superior vena cava (SVC) and RA saturationswas 3.9% ± 2.4%. Using threshold values of 8.7% and 5.7% (mean + 2 SD),respectively, between SVC and RA and RA and PA as “cut-offs” to identifypatients with intracardiac shunts had excellent sensitivity and specificity [5].Similarly, differences in oxygen saturation of 8% between SVC and PA and 5%between RV and PA can be used to detect intracardiac shunts.

Quantifying a left-to-right shuntQuantifying a left-to-right shunt is generally done in two ways. The ratio of pul-monary blood flow versus the systemic blood flow can be calculated, termedthe QP /QS, where QP is the pulmonary blood flow and QS is the systemicflow. Alternatively, the actual flow of the shunt can be calculated. This is thedifference between the pulmonary blood flow and the systemic blood flow(these two are equal in a normal heart).

The pulmonary (QP ) and systemic (QS) blood flow are calculated usingthe Fick method of estimating cardiac output (Table 7.2). The only differencebetween the pulmonary flow equation and the systemic flow equation is thearterial and venous saturations used. The QP and QS equations are shownbelow as well.

Cardiac output

= oxygen consumption(arterial O2 − venous O2)×hemoglobin concentration × 1.36 × 10

QP = oxygen consumption(PVO2 − PAO2) × hemoglobin concentration × 1.36 × 10


Table 7.2 Important formulas in quantifying shunts

QP = Oxygen consumption (mL/min)Oxygen content in pulmonary veins − oxygen content in pulmonary artery (mL/L)

QS = Oxygen consumption (mL/min)Oxygen content in systemic artery − oxygen content in mixed venous sample

Simplified formula for calculating QP/QS in patients with left-to-right shunts.

QP

QS= (arterial sat − mixed venous sat)

(pulm vein sat − pulm artery sat)

� Pulmonary vein samples are rarely obtained. Instead, pulmonary capillary wedge samples orleft atrial samples can be used. Alternatively, arterial saturation (in the absence of right-to-leftshunt) can be substituted or an assumed value of 98% may be used� Left ventricular may be substituted for arterial saturation, provided that there is no right-to-leftshunt� For PDA, use RV sample to determined mixed venous O2� For VSD, use RA sample to determined mixed venous O2� For ASD, calculate using mixed venous O2 = 3(SVC) + IVC/4 (see Table 7.3 for a more com-plete discussion of calculating mixed venous O2 from SVC and IVC samples)

For right-to-left and bidirectional shunts:

Calculate QP, QS, and QEP (the amount of oxygenated blood delivered to the body) using thefollowing equations:

Constant to use in all calculations = oxygen consumption (either measured or estimated inmL O2/min)/hemoglobin (g/dL) × 1.36 (mL O2/g)

where PVO2 is the pulmonary venous oxygen saturation and PAO2 is thepulmonary artery oxygen saturation

QS = oxygen consumption(SAO2 − MVO2) × hemoglobin concentration × 1.36 × 10

where SAO2 is the systemic arterial oxygen saturation and MVO2 is the mixedvenous oxygen saturation

These equations can be simplified to calculate QP /QS. The simplified equa-tion is the difference of the systemic arterial oxygen saturation minus the mixedvenous oxygen saturation divided by the pulmonary venous oxygen concen-tration minus the pulmonary arterial oxygen concentration. This equation isshown below.

QP

QS= (SAO2 − MVO2)

(PVO2 − PAO2)

Since pulmonary veins are rarely entered during a cardiac catheterization, apulmonary catheter wedge sample or left atrial sample (if the left atrium isentered via an ASD) can be used in its place. Alternatively, arterial saturation


(in the absence of right to left shunt) can be substituted or an assumed valueof 98% may be used.

There are four factors that will help in using this simplified equation. Thefirst is that the difference in the arterial–venous oxygen saturations is sys-temic over pulmonary, despite the fact that the ratio is QP /QS. This is a bitcounter intuitive but a quick reference to the flow equations above will revealthe reason for this (i.e. cardiac output is inversely related to the differencein saturations). The second factor is that the difference between the arterialand venous oxygen saturations in the QP equation is PVO2 – PAO2. This isbecause the difference is between oxygenated blood and unoxygenated blood.The anatomical pulmonary arteries carry unoxygenated blood and the pul-monary veins carry oxygenated blood, thus (oxygenated – unoxygenated) isequal to (PVO2 – PAO2). The third factor is the mixed venous oxygen satura-tion (MVO2) in the QS equation. Different values should be used depending onwhere the shunt is located. The MVO2 should be from the chamber precedingthe shunt, thus if there is a VSD the average of samples from the right atriumshould be used. If there is an ASD, the MVO2 is calculated from the superiorvena cava (SVC) and the inferior vena cava (IVC). While theoretically coro-nary sinus blood flow contributes to mixed venous blood along with bloodfrom IVC and SVC, the contribution is so small as to be safely ignored. Theequation most frequently used is (3 × SVC + IVC)/4. This is the most commonbut certainly not the only calculation proposed as an appropriate estimate ofthe MVO2. At least six different calculations have been suggested as ways toestimate MVO2 (see Table 7.3). Lastly, this equation can be used to calculatethe ratio of pulmonary blood flow to the systemic blood flow but should notbe confused as a measure of the actual shunt blood flow. Shunt flow can bedetermined by calculating QP and QS; the shunt flow is the difference betweenthe pulmonary and systemic flow, QP − QS. This can be useful in assessing thesignificance of the shunt.

Table 7.3 Determination of mixed venous oxygen saturation

In the absence of an intracardiac shunt, the pulmonary artery provides a site of mixed venousblood. If there is a VSD, RA blood can be assumed to be mixed venous. In the presence of anASD, mixed venous saturation needs to be estimated. In these cases, the formula most commonlyused to determine mixed venous oxygen saturation is MV = (3 × SVC + IVC)/4 (where MV = mixedvenous, SVC = superior vena cava, and IVC = inferior vena cava). Formulas that have beenadvocated for estimating mixed venous oxygen saturation include:

First author Year Formula

Flamm 1969 MV = (3 × SVC + IVC)/4Iskandrian 1976 MV = (2 × SVC + IVC)/3Swann 1954 MV = (SVC + 2 × IVC)/3Barratt-Boyds 1957 MV = (SVC + IVC)/2Miller 1974 MV = IVCGoldman 1968 MV = SVC


Shunt managementOnce a shunt has been detected, management varies depending on the severityand patient symptoms. QP /QS can be a very useful tool in making decisionsabout the need for repair of a shunt. The following numbers are generalizationsabout left-to-right shunts:� A QP /QS of 1–1.5—observation is generally recommended.� A QP /QS ratio of 1.5–2.0—significant enough that closure (either surgicallyor percutaneously) should be considered if the risk of the procedure is low.� A QP /QS ratio of greater than 2—closure (either surgically or percuta-neously) should be undertaken unless there are specific contraindications.These recommendations are not absolutes and the entire clinical scenarioshould be considered along with these findings. Shunt flows can change overtime. For example, left-to-right shunts via an ASD can increase with age as leftatrial pressures rise in response to decreased left ventricular compliance. Alter-natively, left-to-right shunts via an ASD can decrease in response to pulmonaryhypertension.

Right-to-left shuntingRight-to-left shunting is unusual except in the case of Eisenmenger’s syn-drome. In right-to-left shunting, the effective pulmonary flow is reduced bythe amount of the shunt (flow through the pulmonary valve + flow throughthe shunt = flow through the aortic valve). The amount of blood shunted asa percentage of cardiac output can be calculated using the QP /QS equation.In this case, the value will be less than 1. Other calculations that are useful inquantifying right-to-left shunts are shown in Table 7.2.

Sample caseThe patient is a 50-year-old woman with hypertension and a known ASD. TheASD was detected 10 years ago and cardiac catheterization demonstrated aQP /QS ratio of 1.4:1. The ASD was not closed and the patient was lost to follow-up. She now returns with complaints of exertional dyspnea and paroxysmalatrial fibrillation and is referred for repeat cardiac catheterization to quantifythe shunt. The following oxygen saturations were obtained:SVC 69%IVC 64%RV 82%PA 84%FA 98%Using the equations discussed above, we can determine the QP /QS ratio forthis patient. The equations and calculations are shown below. Because theshunt is at the level of the atrium, the MVO2 is calculated from the SVC andIVC saturations.

MVO2 = [(3 × SVC) + (1 × IVC)]/4

MVO2 = [(3 × 69) + (1 × 64)]/4


MVO2 = (207 + 64)/4

MVO2 = 271/4

MVO2 = 67.75

With the MVO2 calculated the other variables are obtained directly from thecollected oxygen saturations.

QP

QS= (SAO2 − MVO2)

(PVO2 − PAO2)QP

QS= (98 − 68)

(98 − 84)QP

QS= (30)

(14)QP

QS= 2.2

This case illustrates an ASD with a significant left-to-right shunt developingover the course of many years. The magnitude of left-to-right shunting is deter-mined primarily by the relative compliance of the two ventricles. Because ofthis individual’s hypertension and the aging process, her left ventricular com-pliance decreased over time resulting in increased left to right shunting.

References

1 Wilkinson JL. Haemodynamic calculations in the catheter laboratory. Heart 2001;85:113–120.2 Dexter L, Haynes FW, Burwell CS, Eppinger EC, Seibel RE, Evans JM. Studies of congenital

heart disease. II. The pressure and oxygen content of blood in the right auricle, right ventri-cle, and pulmonary artery in control patients, with observations on the oxygen saturationand source of pulmonary capillary blood. J Clin Invest 1947;26:554–560.

3 Matta BF, Lam AM. The rate of blood withdrawal affects the accuracy of jugular venousbulb. Oxygen saturation measurements. Anesthesiology 1997;86:806–808.

4 Shepherd AP, Steinke JM, McMahan CA. Effect of oximetry error on the diagnostic valueof the Qp/Qs ratio. Int J Cardiol 1997;61:247–259.

5 Hillis LD, Firth BG, Winniford MD. Variability of right-sided cardiac oxygen saturations inadults with and without left-to-right intracardiac shunting. Am J Cardiol 1986;58:129–132.

PART I I

Valvular heart disease

CHAPTER 8

Aortic stenosis


Introduction

The basic function of the aortic valve is to separate the aorta from the left ven-tricle cavity during diastole. At peak left ventricular (LV) ejection the aorticvalve opens briskly to an average area of 4 cm2. Although rarely LV outflowtract obstruction can be supravalvular (e.g. Noonan’s syndrome) or subvalvu-lar (e.g. hypertrophic obstructive cardiomyopathy) the overwhelming majorityof cases are due to valvular stenosis. Pathologic thickening and fusion of theaortic valve leaflets results in restricted leaflet mobility and a decrease in theeffective aortic valve area leading to aortic valve stenosis (in this chapter, wewill use the abbreviation AS to refer to aortic valvular stenosis).

The etiology of aortic valve disease in developed countries has changeddramatically in the last few decades. With the aging of the population andconcurrent decrease in rheumatic fever, degenerative calcific AS is now by farthe most common cause of AS. Many recent studies have suggested that AS ishistologically similar to atherosclerosis with lipid infiltration and inflammationand there is suggestive evidence that drugs that delay progression of coronarydisease may also be effective in AS. Congenital bicuspid aortic valve is one ofthe most common congenital cardiac defects and is currently the second mostcommon underlying lesion in AS.

As aortic disease progresses and the resistance to blood flow increases, apressure gradient develops from the LV to aorta in order to maintain strokevolume (Figure 8.1). This leads to elevations in left ventricular systolic pressureand this “pressure overload” of the LV leads in turn to compensatory concen-tric LV hypertrophy. This adaptation allows the LV to generate the necessarypressure to maintain cardiac output but can lead to abnormalities in diastolicLV function, coronary perfusion, and eventually LV systolic dysfunction.

Progressive increase in LV/aorta pressure gradient and cardiac(mal)adaptation explain the stages of hemodynamic findings that patientsgo through as AS progresses. In mild AS, intracardiac pressures and cardiacoutput will appear normal. As the valve becomes more stenotic, the patientmay have normal hemodynamic findings at rest but may be unable to increasecardiac output during exercise. Progressive narrowing of the valve leads todecreased stroke volume and cardiac output even at rest. In moderate tosevere AS, patients may develop elevated filling pressures to compensate for

101

102 Part 2 Valvular heart disease

120

100

80

60

40

20

0

0 1 2 3 4 5

AV

pre

ssur

e gr

adie

nt(m

m H

g)

Aortic valve area(cm2)

Figure 8.1 AV pressure gradient versus aortic valve area in a hypothetical patient with cardiacoutput = 5 L/min, HR = 80 bpm, and SEP = 0.33 seconds.

the increase in LV end-diastolic pressure. In a minority of patients LV systolicfailure also occurs, leading to further elevation in intracardiac pressures. Itis important to remember that the pressure gradient across the aortic valveincreases exponentially (not linearly) with decreasing aortic valve area. Thus,in patients with severe AS, small changes in aortic valve area can lead to largechanges in hemodynamics (Figure 8.1).

Adaptive mechanisms of the LV and circulation allow most patients toremain asymptomatic until advanced narrowing of the aortic valve orificeoccurs. The prognosis of patients with asymptomatic severe AS is very goodwith a sudden death rate of less than 0.5% per year. The onset of symptoms,however, heralds a marked change in the natural history of this lesion withpoor outcomes unless the patient undergoes aortic valve replacement. Theclassic clinical presentation of severe AS is typically an insidious onset of anyof the triad of effort angina, dyspnea, or exertional syncope. The average sur-vival in untreated patients with severe AS is classically thought to be 2, 3, or 5years after the onset of heart failure, syncope, or chest pain, respectively. Impor-tantly, many patients report being asymptomatic, however careful questioningreveals that they have gradually decreased their level of activity and would, infact, be symptomatic at their previous level of exertion. Extensive questioningoften involving a family member is imperative. Occasionally, careful exercisetesting can be useful in the nominally asymptomatic patient.

Physical exam

The physical examination of patients with AS can often be helpful in predictingwhich patients have hemodynamically severe disease. A low volume late onset

Chapter 8 Aortic stenosis 103

Table 8.1 Physical examination of severe AS

• Parvus et tardus pulse (low amplitude, delayed upstroke)• Sustained LV impulse• Late-peaking systolic murmur (though murmur can be absent)• Diminished or absent aortic component of the second heart sound• Parodoxical splitting of the second heart sound• Prominent fourth heart sound• Gallavardin’s phenomenon (apical murmur radiation in the elderly)

pulse referred to as parvus et tardus is often present in severe aortic stenosis.This finding can be appreciated on palpation of the carotid upstroke and radialartery. It may not be manifest in the elderly owing to a noncompliant vascu-lature. Other physical findings include a systolic murmur that is crescendo–decrescendo in intensity. The duration will vary with the severity of disease butthe murmur always begins after S1 and ends prior to S2. The timing of the loud-est portion of the murmur correlates with the severity of AS with late-peakingmurmurs being more common in critical AS. The murmur is generally heardbest in the right second intercostal space and can radiate to the carotid arteries.A diminished aortic component of the second heart sound is not uncommonand additionally the second heart sound can also split paradoxically in severeAS (delayed aortic component with narrowing of the splitting with inspira-tion). Elderly patients with heavily calcified valves may developed systolicmurmurs that radiate prominently to the apex (Gallavardin’s phenomenon).Physical exam findings that are suggestive of severe AS include diminishedcarotid pulses, late-peaking systolic murmur, and absent aortic component ofthe second heart sound; however, it is important to realize that even in expe-rienced hands the physical exam can be unreliable in terms of quantifying ASseverity (see Table 8.1).

Echocardiographic hemodynamics

Two-dimensional echocardiographic evaluation of aortic valve disease hasgreatly facilitated the management of these patients. Two-dimensional echoprovides important information about valve morphology and left ventricularfunction as well as important prognostic ancillary findings such as degree ofleft ventricular hypertrophy, presence of mitral regurgitation, and pulmonaryartery pressures. Doppler evaluation enables the noninvasive measurementof aortic valve gradient and an estimation of valve area. The AV gradient isestimated by measuring blood flow velocity across the aortic valve and thenusing the Bernoulli equation to determine pressure gradient.

Bernoulli equation

Pressure gradient = 4 × velocity2


Aortic stenosisCatheterization vs. Doppler

200 mm Hg

100 mm Hg

0 mm Hg

Peak instantaeousgradient (56 mm Hg)

Mean gradient(38 mm Hg)

Peak to peak(30 mm Hg)

2 m/s

4 m/s Doopler transaortic velosity(m/s)

Figure 8.2 Comparison of invasive and echocardiographically derived hemodynamics in ahypothetical patient with AS. (Courtesy of Jatin Joshi.)

It is common for different AV gradients to be reported by the echocardio-graphic laboratory and the cardiac catheterization laboratory in the samepatient (Figure 8.2). This has led to the common misconception that AV gradi-ents vary depending on whether echocardiography or left heart catheterizationis used and that there is poor correlation between hemodynamics found in thecatheterization laboratory and in the echocardiography laboratory. This mis-conception is based on comparing apples to oranges. In the catheterizationlaboratory, gradients are commonly expressed as “peak to peak” whereas theechocardiographic lab reports peak instantaneous gradients (Figure 8.2). Thepeak Doppler-derived gradient by echo correlates well with the peak instanta-neous transvalvular gradient at cardiac catheterization. Likewise the Doppler-derived mean gradient obtained from the time velocity integral correlateswell with cardiac catheterization mean gradients. The peak to peak invasivecath lab gradient is a nonphysiologic measure and does not have any echocorrelate.

Invasive hemodynamics

Hemodynamic findings associated with AS include fixed outflow tract obstruc-tion and diastolic dysfunction. In contrast to a dynamic outflow tract obstruc-tion (e.g. hypertrophic obstructive cardiomyopathy), LV/FA gradient willdecrease with preload reduction and pulse pressure will increase after a PVC(negative Brockenbrough sign). Over time, the left ventricle in patients with


100 mm Hg

0 mm Hg SEP

LV

Ao

Paper sheet: 100 mm/s

Figure 8.3 Simultaneous LV–aortic pressure tracings with systolic ejection period (SEP) andsystolic transvalvular gradient (shaded area) indicated.

AS hypertrophies in response to the pressure overload. Increased wall thick-ness causes the left ventricle to become stiffer and less compliant and thushigher end-diastolic pressures are required to maintain left ventricular fill-ing. This, in turn, leads to higher left atrial pressures and elevated pulmonarycapillary wedge pressures. PA pressures are normal or only mildly elevatedearly in the disease but increase as LV failure (both systolic and diastolic)worsens.

In the catheterization laboratory, the calculation of aortic valve area is tradi-tionally performed using the Gorlin formula. This formula was initially derivedin the early 1950s in patients with mitral stenosis. Though the Gorlin for-mula is seemingly straightforward, there are a number of potential pitfalls andmeticulous detail is necessary to obtain an accurate estimation of aortic valvearea.

The Gorlin formula requires measurement of cardiac output, heart rate, sys-tolic ejection period, and mean transvalvular gradient. The transvalvular gra-dient requires simultaneous LV and aortic pressure measurement (Figure 8.3).The cardiac output can be obtained by either thermodilution or via the Fickmethod. The systolic ejection period extends from when the aortic valve opens(i.e. intraventricular pressure rises above aortic pressure) and ends with aorticvalve closure.

Gorlin formula for estimating aortic valve area

AVA(cm2) = CO/(SEP) (HR)

44.3 × √mean pressure gradient

CO = cardiac output (mL/min); HR = heart rate; SEP = systolic ejection period(s/beat); 44.3 = constant.


200 mm Hg

100 mm Hg

0 mm Hg

FA

Ao

Peripheral amplification

Figure 8.4 Simultaneous measurement of aortic and femoral artery pressure demonstratingperipheral amplification.

Clinical pearl

Aortic valve area can be estimated in patients with severe AS by dividing thecardiac output by the square root of the peak-to-peak systolic pressure gradientmeasured in the cath lab. This is the so called Hakki formula.

AVA(cm2) = CO(L/min)

(√

peak LV systolic pressure – peak aortic systolic pressure)

Hakki AH, Iskandrian AS, Bemis CE, et al. A simplified valve formula for thecalculation of stenotic cardiac valve areas. 1981;63:150–1055 [1].

Angel J, Soler-Soler J, Anivarro I, Domingo E. Hemodynamic evaluation ofstenotic cardiac valves: II. Modification of the simplified formula for mitraland aortic valve area calculation. Cathet Cardiovasc Diagn 1985;11:127 [2].

Common pitfalls

There are many common pitfalls in the estimation of aortic valve area in thecatheterization laboratory.

� Peripheral amplification� Aortoiliac stenosis� Aortic regurgitation� Low gradient AS� Poor fidelity LV tracing

If femoral artery pressure is used as a surrogate for central aortic pressure,obtaining the true LV–aortic gradient can be challenging in patients withperipheral amplification or in patients with peripheral vascular disease.


Peripheral amplification refers to the increase in peak systolic pressure andpulse pressure in peripheral arteries as compared to the central aorta (Figure8.4; see also Chapter 4). It is important to account for peripheral amplificationbecause determination of AV gradient by comparing femoral artery and LVpressure will underestimate transvalvular gradient compared to a comparisonof LV and aortic pressure. In contrast, the presence of iliac artery diseaseleading to decreased femoral artery pressures will result in overestimation ofthe severity of AV disease if comparison is made between femoral artery andLV pressure.

Both these potential pitfalls can be remedied by using either a pressure wireor a double lumen pigtail thus allowing simultaneous measurement of leftventricular and proximal aortic valve pressure. Another approach is to use along sheath that enables comparison of LV and distal aortic pressure. Finally,if femoral artery pressure is used in patients with AS, it is important to simul-taneously measure femoral artery and aortic pressure.

Careful attention to detail and examination of the left ventricular waveformfidelity is imperative. It is not uncommon to find LV–aortic tracings in which thesideholes of the pigtail catheter are in the aorta; this will lead to underestima-tion of the true valve gradient and overestimation of the valve area. Anotherpotential misinterpretation is shown below. In this case, a pigtail catheter is“pulled-back” from the LV to the aorta. Comparison of systolic pressures indi-cates a 40 mm Hg difference between the last beat in the LV and the first in theaorta (Figure 8.5). More careful examination, however, reveals that the heartrate has slowed. Within five beats, the aortic systolic pressure was the same asthat observed in the LV.

Another potential mistake is illustrated in Figure 8.6. There appears to bea 40–50 mm Hg gradient between LV and femoral artery and upon catheterpullback the gradient resolves. Closer inspection, however, reveals that thefemoral artery pressure increased following pullback and now correspondedwith LV systolic pressure. In patients with severe AS, aortic pressure canincrease following removal of a catheter from the left ventricle (Carrabello’ssign; discussed below), however, it is never this dramatic and aortic systolicpressure will always be less than LV systolic pressure. The cause of the pressure

r

s

IGain 100

II

V5Gain 50

Gain 50

200

100

0d d

1 2 3 4 5 6 7 8 9d dd d

dd d

e e e

e e e

r r r r

s s ss s s

s ss s

r r r r r r r r r r

Figure 8.5 Pullback of pigtail catheter from LV to aorta.


IGain:50

IIGain:50

V5Gain:50

200

100

025 mm/s >

[-1-]

r

s s ss

s

r r r r r r r r

1.50mmHg/mm SERIES IV- WITH MEDICAL1 2 3 4 5 6 7 8 9

1mV

1mV

1mV

1mV

1mV

1mV

Figure 8.6 Pullback of pigtail catheter from LV to aorta while simultaneously measuring pigtailand femoral artery pressure.

gradient in this patient was an obstruction in the femoral artery sheath thatwas dislodged with manipulation of the catheter.

The challenge of low-gradient AS

In patients with low cardiac output and a small pressure gradient across theaortic valve, calculation of aortic valve area with the Gorlin formula can beproblematic. The fundamental question in these patients is whether true severeAS with secondary left ventricular systolic dysfunction coexists or alternativelywhether the low gradient is due to a low cardiac output from the cardiomy-opathy and the valve stenosis is minimal. Differentiating these two entitiesis critical as the former may respond favorably to aortic valve replacementwhereas the latter does not.

Increasing cardiac output pharmacologically (typically with dobutamine ornitroprusside) either in the catheterization lab or during 2DE can be helpful inthis regard (Figure 8.7). Patients with true severe AS who will likely improvewith aortic valve replacement manifest contractile reserve (25% increase instroke volume) and/or show an increase in their transvalvular gradient andcardiac output with no change or a decrease in the calculated aortic valvearea. In contrast the patients with pseudo AS will not have contractile reserve,will not significantly increase their LV–aortic pressure gradient, and will oftenhave an increase in calculated aortic valve area with dobutamine. These latterpatients typically have myocardial disease as their primary lesion and “inci-dental” AS. They do not respond favorably to aortic valve replacement.


True aortic stenosis

Base

100

LV

Ao

0Cardiac Output: 10 L/min

Mean Gradient: 25 mm Hg

Aortic Valve Area: 0.8 cm

Cardiac Output: 4 L/min



Cardiac Output: 2.5 L/min



Cardiac Output: 3.4 L/min



Dobutamine(20 mcg)

Base Dobutamine(20 mcg)

Pseudo aortic stenosis

Figure 8.7 Effects of dobutamine infusion in patients with and without valvular aortic stenosis.

The challenge of estimating aortic valve area in patientswith AS and significant AR

Another major challenge is accurate calculation of aortic valve area in patientswith concurrent aortic regurgitation (AR). The problem lies in the numeratorof the Gorlin equation, namely, the accurate determination of flow across theaortic valve. Commonly used ways to estimate cardiac output such as Fick andthermodilution measure effective and right-sided cardiac output, respectively,and thus will underestimate the flow across the aortic valve (flow across theaortic valve = cardiac output + regurgitant volume) and thus overestimate theseverity of AS.

Case study

A 70-year-old man with 3 months of progressive chest pain presents with lossof consciousness while ascending a hill. Examination reveals a blood pres-sure of 136/80 and a delayed carotid upstroke. The second heart sound issplit paradoxically. There is a late-peaking systolic ejection type murmur heardbest at the right upper sternal border radiating to the carotids. A poor quality2D echo demonstrates a calcified trileaflet aortic valve with decreased leafletmobility, normal left ventricular function, and severe LV hypertrophy. Doppler


evaluation demonstrated an aortic velocity of 4 m/s corresponding to a peakinstantaneous gradient of 64 mm Hg and a mean gradient of 50 mm Hg.Invasive hemodynamics are shown in Figure 8.8. Note the large meantransvalvular gradient of 110 mm Hg and slow upstroke of the aortic tracing.A cardiac output of 4.2 L/min, with a systolic ejection period of 0.33 s/beat ata heart rate of 70 beats/min yields the following in the Gorlin equation:

AVA = (4200 mL/ min)/(70 beats/ min)(0.33 s/beat)

44.3√

110AVA = 0.4 cm2

This tracing also demonstrates a very interesting hemodynamic finding. Notethe rise in aortic pressure on pullback from LV to aorta (arrow). Carabelloet al. demonstrated that this finding correlated with critical AS. A rise in aorticpressure on LV pullback of 10 mm Hg or greater is highly suggestive of criticalaortic stenosis.

This patient went on to aortic valve replacement and recovered fully.

200 mm Hg

0 mm Hg

Figure 8.8 LV to aortic pullback of a pigtail catheter with simultaneous measurement of femoralartery pressure. (Courtesy of Cardiovillage.com.)

Carabello’s sign

In patients with AS, Carrabello’s sign is defined as a rise in peak aortic sys-tolic pressure by greater than 5 mm Hg when a catheter is removed from theleft ventricle (Figure 8.8). This phenomenon occurs in the setting of a highlystenosed valve because the additional occlusive effect of the catheter across thevalve is enough to further decrease aortic pressure. Carrabello and colleaguesfound this sign in 15 of 20 patients with AVA < 0.6 cm2 and in none of 22patients with AVA > 0.7 cm2. (Am J Cardiol 1979;44:424)


References

1 Hakki AH, Iskandrian AS, Bemis CE, et al. A simplified valve formula for the calculationof stenotic cardiac valve areas. 1981;63:150–1055.

2 Angel J, Soler-Soler J, Anivarro I, Domingo E. Hemodynamic evaluation of stenotic cardiacvalves: II. Modification of the simplified formula for mitral and aortic valve area calculation.Cathet Cardiovasc Diagn 1985;11:127.

CHAPTER 9

Mitral stenosis

Robert V. Kelly, Chadwick Huggins

Introduction

The cross-sectional area of the mitral valve is 4–6 cm2 in healthy adults. Mitralstenosis (MS) occurs when this area decreases, resulting in increased resistanceto flow from the left atrium to left ventricle and increased left atrial (LA) pres-sure. The increased LA pressure is transmitted to the pulmonary circulation,resulting in elevated right heart pressures (Figure 9.1).

Almost all cases of MS are rheumatic in origin. Isolated MS occurs in40% of all patients presenting with rheumatic heart disease and womenwith MS outnumber men by approximately 2:1. Rheumatic mitral valvechanges including thickening of the mitral valve leaflets, calcification andcommissural fusion occur over the course of decades. Other rare causes ofMS include systemic lupus erythematosis, rheumatoid arthritis, carcinoid,mucopolysaccharidosis, mitral annular calcification, and congenital valvedeformity. Conditions that cause increased LA pressure can mimic MS. Exam-ples include LA myxoma, pulmonary vein obstruction, and cor triatriatum(a thin membrane across the left atrium which obstructs pulmonary venousinflow).

MS is a progressive disorder with symptoms worsening as mitral valve areadecreases. Symptoms do not usually appear until mitral valve area <2 cm2.Initially, symptoms occur only with exertion, emotional stress, or pregnancyand consist primarily of dyspnea. Symptoms can progress over time and insevere cases eventually include dyspnea at rest, orthopnea, and paroxysmalnocturnal dyspnea. More rarely, there may be angina, palpitations, recumbentcough, and/or hemoptysis.

Patients with MS generally pass through four separate hemodynamic stages(Figure 9.2):� Normal hemodynamics� Normal filling pressures at rest but increased filling pressures and normalcardiac output during exercise� Increased filling pressures at rest and inability to increase cardiac outputwith exercise� Markedly increased filling pressures at restIn addition to increasing resistance to flow, MS also increases transvalvu-lar blood velocity and the ratio of turbulent to laminar flow across the

113


Figure 9.1 Hemodynamic changes of mitral stenosis. [RA = right atrium, RV = right ventricle,PA = pulmonary artery, PV = pulmonary veins, LA = left atrium, LV = left ventricle].

Figure 9.2 Hemodynamic of mitral stenosis effects of different degrees/(Calculations performedassuming HR × DFP = 32.).

mitral valve. For example, left ventricular (LV) inflow velocity acrossthe normal valve is approximately 1 m/s but can reach 3 m/s in severeMS. The effects of accelerated velocity of blood are not completelyunderstood.

Chapter 9 Mitral stenosis 115

Table 9.1 Findings at cardiac catheterization.

• Increased LA pressure• Increased right heart pressures with exercise and/or at rest• Prominent A wave on RA tracing• Decreased slope of Y descent on RA tracing• Gradient between LA and LV persists throughout diastole

Cardiac hemodynamics in patients with MS

The classic hemodynamic findings associated with MS include elevated LApressure and a gradient between the LA and LV that persists throughout dias-tole (Table 9.1). There are several other findings that are discussed below inrelationship to the cardiac chamber involved.

Left atriuma. LA pressure is increased and there is a gradient between left atrium andleft ventricle during diastole (Figure 9.3).b. Chronic LA hypertension causes dilatation over time in patients with MS.LA dilation is accelerated in patients with atrial fibrillation.c. LV filling is slowed accounting for decreased slope of Y descent.d. Atrial contraction component of ventricular filling becomes moreimportant.� Accounts for prominent A wave on RA and pulmonary capillary wedgepressure (PCWP) tracing (Figure 9.4).� Atrial arrhythmias are poorly tolerated.� Somewhat paradoxically, the contribution of active atrial contraction totransvalvular flow decreases as the severity of MS increases. Meisner et al.in a study of 30 patients with MS found that atrial contraction accounted for29 ± 5% of filling volume in mild MS and only 9 ± 5% in severe MS. Computermodeling suggested that increased heart rate, rather than loss of atrial contrac-tion, was the primary mechanism responsible for decompensation of patientswith severe MS who develop atrial fibrillation [1].

Pulmonary arteryIn mild MS, pulmonary artery pressures may be normal or slightly elevatedat rest but increase with exercise. In severe MS, pulmonary artery pressurewill be elevated at rest. Pulmonary hypertension in severe MS is determinedboth by elevated LA pressures and increased pulmonary vascular resistancedue to pulmonary arteriolar constriction and obliterative changes in the pul-monary vascular bed. It has been speculated that the increased resistance inthe pulmonary veins protects the patient from pulmonary edema.

Evidence for a reversible component of pulmonary vascular resistance wasshown in a study of 18 women with MS and pulmonary hypertension. Inhalednitric oxide reduced pulmonary artery systolic pressure by approximately


PCWP

LV

575

L.Ventricle

L.Atrium

120

100

80

60

40

20

mm

Hg30

4

70/4

70/30

120/875%

75%

75%

95%

97%

3095%

3097%

95%

Mitralstenosis

Gradient

(a)

(b)

12070

Figure 9.3 Transvalvular pressure gradient in a patient with MS. Schematic (a) and actualrecordings (b) of simultaneous LV and PCWP pressures in a patient with MS. Note that in (b) thePCWP is delayed relative to LV and should be phase shifted prior to performing any calculations.

15% and pulmonary vascular resistance by 30% without having any effect onLV end-diastolic pressure, LA pressure, cardiac output, or systemic vascularresistance [2]. Following mitral valve surgery or percutaneous mitral valvu-loplasty, pulmonary artery pressures decrease immediately due to decreasedLA pressure. Over time, pulmonary artery pressures continue to decrease aspulmonary vascular resistance decreases. This was demonstrated in a study of21 patients with severe MS and pulmonary hypertension undergoing mitralballoon valvuloplasty. Mean PCWP decreased from 27 ± 5 (preprocedure)to 15 ± 4 mm Hg (immediately postprocedure) and then remained constantat a mean follow-up of 1 year. Mean mitral valve gradient decreased from18 ± 4 to 6 ± 2 mm Hg and then remained constant. Pulmonary artery sys-tolic pressure decreased from 65 ± 13 to 50 ± 13 mm Hg to 38 ± 9 mm Hg.


AIR REST Moniter Length 10 sec

r rr

r r r r r r r

a

aa

aa

a a

aa

a a

v

vv

v v

v v

v v

2 3 4 5 6 7 8 9

Figure 9.4 RA tracing from a patient with severe MS. Note the prominent A wave.

Pulmonary vascular resistance decreased from 461 ± 149 to 401 ± 227 to 212± 99 (dyn s/cm5) [3].

Left ventriclea. Filling is impaired and determined by degree of MS.b. LV filling is slowed accounting for delayed rise in LV diastolic pressure.c. LV systolic function is generally normal.

The LV ejection fraction (LVEF) is normal in most patients with MS.d. LV diastolic function maybe abnormal and directly influences LA pressure.

In most patients with MS, the LV end-diastolic pressure (LVEDP) is normal.It is reduced in about 15% of cases. In the presence of heart failure or mitralregurgitation, the LVEDP can be elevated. Since flow across the mitral valvein MS is directly proportional to transvalvular pressure gradient, as LVEDPincreases LA pressure must also increase to maintain the transvalvular pressure


gradient. Also, reduced LV compliance from displacement of the interventricu-lar septum secondary to early filling of the right ventricle or interstitial fibrosisfrom rheumatic fever may hinder LV filling during the second half of diastole.e. MS can unmask the ability of the LV to develop negative pressures in earlydiastole. The presence and magnitude of negative pressures is variable andmay require advanced recording techniques to quantify.

Right ventricleRight ventricular systolic pressures rise in line with increases in pulmonaryartery systolic pressures.

Cardiac outputa. Rise in stroke volume is limited as mitral valve disease progresses and thuscardiac output will rise only if filling pressures rise.b. The length of diastolic filling period is important in determiningtransvalvular flow. Tachycardia impairs LV filling, further increasing LA pres-sure.

Quantification of severity MS

The severity of MS can be assessed by catheterization. The mitral valve gradientmay be directly measured by comparing pressures in the LV and left atriumor PCWP (Figure 9.5). Simultaneous pressure measurements are recorded andthe gradients measured. For best results, the fastest paper recording speed

(AIR REST) Simultaneous gradient calculation25 mm/sec Time: 09:28:50 Wave Y:13 Line: 20 sec Normal staus 100mm Hg/mm

I

II

V5

Date: 10.11.2000

rr r r r r r

50

0

a

119/ -6,25 42/ 55(42)LV PVLV offset 36 msec

v

a a

d

d

77

Figure 9.5 Transvalvular gradient in a patient with MS. Simultaneous recording of LV and PCWPpressure tracings. The darkened portion represents transvalvular gradient.


(generally 100 mm/s) and a 0–50 mm Hg scale is used. The gradient across themitral valve should be measured using an average of five cardiac cycles inpatients with normal sinus rhythm and 10 cardiac cycles in patients with atrialfibrillation.

The severity of obstruction in MS can be quantified using mitral valve area,mitral valve resistance, or mitral valve gradient. Mitral valve area remainsthe standard used in most laboratories because it includes the three majorhemodynamic variables: transvalvular pressure gradient, cardiac output, anddiastolic filling period. By convention, a mitral valve area <1 cm2 is consideredsevere MS. A valve area of 1–1.5 cm2 is moderate stenosis and a valve area>1.5 cm2 is mild stenosis.

Mitral valve resistance is a useful research tool although its clinical use lagsbehind mitral valve area. It is a measurement of the opposition to blood flowof the stenotic mitral valve and is computed by dividing pressure gradient bytransvalvular flow. Mitral valve resistance has been advocated as being moreaccurate in patients with low cardiac output and low pressure gradients.

The degree of MS can also be defined by the gradient across the mitral valve.By convention, a mean gradient >10 mm Hg at rest is severe MS. A mean gradi-ent of 5–10 mm Hg at rest is moderate stenosis and a mean gradient <5 mm Hgat rest is mild MS. Transvalvular gradient will increase in direct proportionto flow across the valve and thus conditions that increase cardiac output (e.g.anemia, hyperthyroidism, anxiety, exercise) will result in increased gradients.

Some patients may report significant symptoms but have only a mild tomoderate resting mitral valve gradient. In these patients, measuring hemo-dynamic response to exercise can be very helpful. With exercise, blood flowacross the mitral valve increases and this can dramatically change the gra-dient. If the mitral valve pressure gradient increases with exercise and symp-toms develop, the patient’s symptoms can be attributed to mitral valve disease(Figure 9.6).

Calculating mitral valve area

The Gorlin formula is the standard for mitral valve area calculation (see textbox). It was derived by the Gorlins (father and son) based on Torricelli’s lawfor flow across a round orifice [4]. An empiric constant was added (initially0.7 and later changed to 0.85) to account for the difference between calculatedand actual mitral valve areas.

Gorlin formula for calculating mitral valve area

MVA = (CO/DFP × HR)/(44.3 × 0.85 × square root of mean gradient)MVA = mitral valve area (cm2)CO = cardiac output (cm3/min)HR = heart rate (bpm)Mean gradient = average diastolic gradient across mitral valve (mm Hg)


DFP = diastolic filling period (s/beat) [measured between initiation of dias-tole (PCWP/LV crossover) and end of diastole (peak of R wave on ECG)]44.3 × 0.85 = 37.7 is an empiric constant.

I

II

V5

I

II

V5

100

100

50

0

50

0

r r

aa

v

v

r

Rest

Exercise

(a)

(b)

Figure 9.6 Exercise hemodynamics in a patient with MS. Transvalvular pressure gradient(simultaneous recording of LV and PCWP pressure tracings) at rest (a) and with exercise (b).


Potential sources of error in quantifying the severity of MSby using the Gorlin formula in the cardiac catheterizationlaboratory

1. Use of PCWP: To avoid a transseptal puncture, PCWP is commonly used inplace of LA pressure. In this scenario, it is essential to ensure that an accuratewedge tracing is obtained. An oxygen saturation >95% helps ensure that thetracing is not a damped pulmonary artery tracing. Potential sources of errorwhen using PCWP tracing include:

� Using damped PA tracing instead of PCWP.� Inability to obtain PCWP in patients with severe pulmonary hypertension.� Delay in transmittal of LA pressure to proximal PA leading to alignmentmismatch in PCWP/LV pressure tracing (in general, the pulmonary capillarywedge pressure is 40–120 ms delayed compared to LA pressure).� PCWP does not approximate LA pressure in patients with venoocclusivedisease or cor triatriatum.Lange et al. compared PCWPs and LA pressures in 10 patients with MS.

PCWP was measured using a Goodale–Lubin catheter and confirmed byoximetry. The mean and phasic LA pressure and PCWP were similar. Useof the PCWP with adjustment for time delay resulted in similar values fortransvalvular pressure gradient and mitral valve area compared to use of LApressure. In contrast, when PCWP was used without adjustment for time delay,the transvalvular pressure gradient and mitral valve area were significantlydifferent from the values obtained with use of LA pressure [5].2. Measurement of cardiac output: See Chapter 6 for a discussion of the limi-tations of measuring cardiac output.3. Atrial fibrillation: The Gorlin formula was developed based on sinusrhythm. It is commonly used in patients with atrial fibrillation, but since meantransvalvular gradient will vary with differing diastolic filling periods, it isimportant to average at least 10 consecutive heart beats.4. Poor calibration of equipment.5. Mitral regurgitation—this will increase flow across the mitral valve and canlead to an overestimation of the severity of MS if cardiac output is used in theGorlin equation.

Physical examination in MS

On physical exam, patients with severe MS may have “mitral facies” as aresult of low cardiac output and systemic vasoconstriction. The amplitude ofthe arterial pulse will be small. Jugular venous pulse will be elevated and havea prominent A wave. In atrial fibrillation, this is absent and a CV wave predom-inates. Palpation of the apex may reveal a presystolic wave and a diastolic thrillmay be felt in the left lateral position. An RV heave in the left parasternal areaand a loud P2 in left second intercostal space suggest pulmonary hypertension.

On auscultation, there is classically a loud S1, an opening snap, and a diastolicrumble. The first heart sound may be lessened with a more calcified mitral


valve. In addition, there may be a tricuspid regurgitation murmur, pulmonaryregurgitation murmur (Graham Steel murmur), and an S4. An S3 may occur ifmitral regurgitation or aortic regurgitation are present. In MS, the low-pitchedrumbling diastolic murmur is best heard at the apex with the bell of stethoscopeand patient in left lateral position. It may radiate if loud. The intensity andduration of the diastolic rumble increases as the gradient across the mitral valveincreases. The murmur persists as long as the mitral valve gradient is more than3 mm Hg. If a diastolic rumble is not heard, perform maneuvers to increasethe patient’s heart rate. The increased heart rate decreases the diastolic fillingperiod and therefore causes the LA pressure to increase. Despite maneuvers,some patients with severe MS will not have an audible diastolic rumble dueto body habitus.

The duration between A2 and the opening snap is indicative of the severityof MS. As the LA pressure increases, the A2–OS interval decreases. In mild MS,the A2–OS interval is >100–110 milliseconds. In severe MS, the A2–OS intervalis <60–70 milliseconds.

In acute mitral valvulitis due to acute rheumatic fever, the Carey Coombsmurmur is heard. This is a soft early diastolic murmur that varies from day today. It is also higher pitched than the classic MS murmur.

Echocardiography

Echocardiography is essential for diagnosing MS and can be quite helpful inassessing the severity of obstruction. In the parasternal long-axis view, themitral valve has a “hockey-stick” deformity. In the short-axis view, the mitralvalve area can be measured by planimetry. The mitral valve area may also becalculated using the pressure half-time method or continuity equation. Thepressure half-time method may not be accurate in the setting of abnormal LAor left ventricular compliance. In these situations, the continuity equation ispreferred. The mean mitral valve gradient may also be determined by Dopplerexamination. An echocardiographic grading system, based on 2D imagingof the valve and the subvalvular apparatus, is often used in decision mak-ing for mitral valve surgery versus percutaneous balloon mitral valvuloplasty(PBMV). There are four categories, including leaflet thickening, leaflet calcifi-cation, leaflet mobility, and subvalvular fusion. A score of 1–4 is assigned toeach category with 1 being least involvement and 4 being severe involvement.In general, a score ≤8 suggests a pliable, noncalcified valve in which PBMVmay be technically possible. A full discussion of echocardiographic assessmentof the mitral valve is beyond the scope of this book.

Hemodynamics of mitral valve surgery and percutaneousballoon mitral valvuloplasty (PBMV)

Two treatment options exist for MS—surgery and PBMV. Among the sur-gical options, there is closed commisurotomy, open commissurotomy, and


mitral valve replacement. Closed commissurotomy was developed prior tocardiopulmonary bypass. It involves a lateral thoracotomy and a splitting ofthe commissures with either a finger or dilator. The success is limited by theamount of calcification present. An open commissurotomy utilizes cardiopul-monary bypass. The surgeon has direct visualization of the valve and is ableto incise the commissures and perform subvalvular repair if needed. Mitralvalve replacement is a third option. This option is often necessary in patientswith heavily calcified valves.

Both PBMV and mitral valve surgery significantly improve mitral valve area,decrease transmitral pressure gradient, reduce pulmonary artery pressure, andincrease cardiac output. In a multicenter registry of 290 patients undergoingPBMV, mitral valve area increased from 1.0 to 1.7 cm2, mean LA pressuredecreased from 24 to 19 mm Hg, transvalvular pressure gradient decreasedfrom 13.4 to 6.1 mm Hg, mean pulmonary artery pressure decreased from 34to 29 mm Hg, and cardiac output increased from 4.1 to 4.4 L/min immediatelypostprocedure [6].

References

1 Meisner JS, Keren G, Pajaro OE, et al. Atrial contribution to ventricular filling in mitralstenosis. Circulation 1991;84:1469–1480.

2 Mahoney PD, Loh E, Blitz LR, Herrmann HC. Hemodynamic effects of inhaled nitric oxidein women with mitral stenosis and pulmonary hypertension. Am J Cardiol 2001;87:188–192.

3 Fawzy ME, Mimish L, Sivanandam V, et al. Immediate and long-term effect of mitral balloonvalvotomy on severe pulmonary hypertension in patients with mitral stenosis. Am Heart J1996;131:89–93.

4 Gorlin R, Gorlin SG. Hydraulic formula for calculation of the area of the stenotic mitralvalve, other cardiac valves, and central circulatory shunts. I. Am Heart J 1951;41:1–29.

5 Lange RA, Moore DM, Jr, Cigarroa RG, Hillis LD. Use of pulmonary capillary wedgepressure to assess severity of mitral stenosis: is true left atrial pressure needed in thiscondition? J Am Coll Cardiol 1989;13:825–831.

6 Feldman T. Hemodynamic results, clinical outcome, and complications of Inoue balloonmitral valvotomy. Cathet Cardiovasc Diagn 1994;(suppl 2):2–7.

CHAPTER 10

Aortic regurgitation

George A. Stouffer

Introduction

The aortic valve separates the left ventricle from the aorta. In normal hearts, itis pliable, opens widely, and presents minimal resistance to flow. The normalaortic valve is a trileaflet structure that is composed of three equal-sized bowl-shaped tissues that are referred to as cusps (left coronary cusp, right coronarycusp, and noncoronary cusp). The aortic valve serves an essential hemody-namic function in isolating the left ventricle from the arterial circulation andcan fail in two ways. The valve can fail to open properly thus inhibiting ejectionof blood from the left ventricle (aortic stenosis). Alternatively, the valve canfail by becoming incompetent thus enabling backflow into the ventricle duringdiastole (aortic regurgitation).

The hemodynamic changes associated with aortic regurgitation differdepending on the time–course of the valve dysfunction. If AR develops rapidly(i.e. acute or subacute AR), the LV is unable to handle the pressure and volumeoverload causing a rapid increase in LV pressures during diastole, markedlyelevated pressures at end diastole, and premature closure of the mitral valve.Systemic diastolic pressures may be low but generally there is a minimalincrease in pulse pressure; in very severe cases of acute AR, cardiac outputmay fall leading to hypotension.

In chronic AR, stroke volume increases to maintain effective forward flow.This leads to dilation of the LV, leading in some patients to the develop-ment of a massively dilated left ventricle termed cor bovinum (the largestleft ventricles are seen in patients with chronic AR). The body’s adapta-tion to chronic AR results, at least in part, in the classic physical exami-nation findings of a widened pulse pressure and low aortic diastolic pres-sure. If and when the regurgitant flow overwhelms the adaptive mechanisms,this leads to uncompensated chronic AR with signs and symptoms of heartfailure.

Before the development of aortic valve replacement, severe AR had an omi-nous prognosis. The availability of surgical treatment has greatly reduced themortality of this disease. The important question now in treating patients withthis disease is the timing of aortic valve replacement. An understanding of thehemodynamic changes associated with chronic AR can help in determiningcardiac adaption to the regurgitant flow, and also help determine the severity

125


of AR in patients with moderate AR by angiography or echo but decreasedejection fraction.

Chronic aortic regurgitation

PathophysiologyHemodynamic changes associated with chronic AR result from aorta to leftventricle blood flow during diastole, compensatory responses of the left ven-tricle to increased diastolic filling and adaptations by the cardiovascular sys-tem to maintain systemic blood flow. In mild AR, increased ejection frac-tion and/or heart rate lead to an increase in cardiac output across the aorticvalve thus maintaining systemic blood flow. As the severity of AR increases,the left ventricle begins to remodel and dilate. This remodeling allows themaintenance of relatively normal filling pressures as left ventricular volumesincrease. Left ventricular dilation progresses as the severity of AR worsensand patients with severe chronic AR, if untreated, often develop the largestend-diastolic volumes associated with any heart disease (cor bovinum). LVEDPincreases, however, if progression of valve incompetence exceeds the rate ofleft ventricular remodeling (acute AR), if limits of left ventricular remodel-ing are reached or if systemic vascular resistance increases in such a man-ner as to increase regurgitant flow above the ability of the left ventricle toadapt.

Aortic pressuresThe two most common blood pressure manifestations associated with chronicAR are a wide systemic pulse pressure and a low aortic diastolic pressure. Thewide pulse pressure results from an increased stroke volume (that is neededto deliver enough blood to the aorta to maintain systemic blood flow despitelarge regurgitant fractions) and decreased peripheral vascular resistance. Thelow diastolic pressure results from a relatively low resistance to aortic diastolicflow: both backward flow into the LV and forward flow into the periphery. Asthe competence of the aortic valve decreases, resistance to flow from aorta toLV decreases, regurgitation increases, and diastolic blood pressure drops. Awidened and elevated systemic arterial pressure without a dichrotic notch issometimes observed.

There are other, more subtle, changes in arterial pressure that are associ-ated with AR. Peripheral arterial pressures are generally higher than aor-tic pressures because of amplification due to summation of pressure wavereflections. In AR, because of the accelerated velocity of ventricular ejec-tion, peripheral amplification is more profound than normal. This phe-nomenon may also lead to the appearance of a bisferiens systolic arterialpressure waveform in which the first peak is due to ventricular empty-ing and the second peak to reflections propagated back from the arterialcirculation.

Chapter 10 Aortic regurgitation 127

Left ventricular pressuresThe LV hemodynamic findings associated with chronic AR depend upon theextent of ventricular remodeling. In mild to moderate AR, filling from theaorta enables the left ventricle to increase stroke volume by increasing LV end-diastolic volume (LVEDV) (i.e. Starling’s law). As the AR becomes more severe,the LV dilates in order to handle increased volumes with minimal changes inventricular diastolic pressures. Thus patients with longstanding, severe ARmay have very large hearts with normal diastolic pressures. In some patientswith AR and dilated left ventricles, myocardial function begins to decline forreasons that are poorly understood. The development of LV dysfunction leadsto increased LV diastolic pressures and eventually to development of pul-monary congestion and symptoms of dyspnea.

Hemodynamic findings in aortic regurgitation

Aorta� Wide pulse pressure� Low diastolic pressure� Increased peripheral amplification

Left ventricle� Increased LV end-diastolic pressure (LVEDP) (in severe AR, LVEDP, andaortic diastolic pressure will be equal)� Early closure of the mitral valve during diastole

Hemodynamic changes detected by physical exam

There are several physical findings associated with chronic AR. In the absenceof left ventricular failure, patients generally have a wide pulse pressure (thediastolic pressure is usually less than one half of the systolic pressure). Thiswide pulse pressure produces a variety of physical findings that are describedin the text box. A low diastolic pressure is observed with Korotkoff soundsoccasionally persisting to zero. On auscultation, a systolic murmur (usuallydue to increased and/or turbulent flow across the aortic valve) may be present.A2 may be soft or absent and a blowing, high-pitched diastolic murmur may beheard along the sternal border. The diastolic murmur usually increases duringmaneuvers that increase peripheral resistance (e.g. handgrip, squatting). Ifthe AR is heard best along the left sternal border, the etiology may be eithervalvular or from aortic root pathology while AR heard best along the rightsternal border is typically due to aortic dilation (so-called Harvey’s sign). Theduration and intensity of the murmur correlate poorly with severity of AR.

A mid- to late-diastolic apical rumble that resembles the murmur of mitralstenosis may be heard in patients with AR (so called “Austin Flint murmur”).This murmur was initially thought to reflect rapid antegrade flow across astructurally normal mitral valve that is partially closed in response to rapid


increases in LV diastolic pressures. More recent studies have suggested thatthe murmur is present in patients in which the AR jet is directed at the anteriormitral valve leaflet. Shuddering of the MV causes vibrations and shock wavesthat distort the AR jet leading to the murmur. Mitral stenosis may coexist inpatients with AR (particularly in patients with rheumatic disease); the presenceof a loud S1 and/or an opening snap are clues that the mitral valve is abnormal.Exercise, amyl nitrite inhalation, or any maneuver that decreases peripheralresistance intensifies the murmur of mitral stenosis while diminishing the mur-mur of aortic regurgitation and the Austin Flint murmur. Handgrip and othermaneuvers that increase peripheral resistance will have the opposite effect.

In patients with chronic AR, cardiac palpation generally reveals that theapical impulse is hyperdynamic, diffuse, and displaced laterally and inferiorly.In patients with left ventricular systolic dysfunction, the precordial impulsemay be less prominent and there may be findings of left-sided heart failure(e.g. pulmonary rales, S3).

Physical exam findings of chronic aortic regurgitation

� Wide pulse pressure� Low diastolic pressure� Blowing, high-pitched diastolic murmur ± a systolic murmur� A2 may be soft or absent� Austin Flint murmur—a mid- to late-diastolic apical rumble that resemblesthe murmur of mitral stenosis� Apical impulse is hyperdynamic, diffuse, and displaced laterally andinferiorly.

Classic findings described in patients with severe chronic AR

� de Musset’s sign—head bobbing with systolic pulse� Muller’s sign—uvula bobbing with systolic pulse� Corrigan’s pulse—booming pulse with quick collapse� Quincke’s pulse—capillary pulsations seen in digits� Traube’s sign—pistol shot sounds heard over femoral artery during systole� Duroziez’s sign—systolic murmur heard over femoral artery with proximalcompression and diastolic murmur heard with distal compression� Hill’s sign—popliteal systolic pressure exceeds brachial pressure by30 mm Hg

Hemodynamic changes detected by echocardiography

Echocardiography plays an important role in the evaluation of the patientwith AR. Echocardiography (transthoracic or transesophageal) enables


visualization of the valve and identification of congenital abnormalities,rheumatic changes, vegetations, thickening, and/or calcification. Echocardio-graphy also provides information about the size of the LV and aortic root andLV systolic function. Severity of AR may be estimated based on measurementof velocity of regurgitant flow between aorta and left ventricle during diastoleby continuous wave Doppler. The rate of decay of the velocity of flow betweenaorta and left ventricle during diastole (i.e. pressure half-time) gives an estimateof the gradient between the aorta and left ventricle. A rapid decay indicatesequalization of left ventricular and aortic pressures and suggests severe AR.The amount of regurgitant flow can be estimated by several techniques but adiscussion of the relative advantages and disadvantages of these methods isoutside the scope of this chapter. Other parameters that can indirectly indicatethe hemodynamic severity of AR include: 1) measurement of the velocity oftricuspid regurgitation which enable an estimate of pulmonary artery pres-sures, which in turn, may be related to elevated left atrial pressures, and 2)M-mode echocardiography of the mitral valve which can demonstrate ‘flutter’of the valve from AR and/or premature closing of the mitral valve duringdiastole.

Cardiac output tends to increase in patients with AR as they exercise becausethe increase in heart rate shortens diastole and peripheral resistance decreasesas blood flow to skeletal muscle increases. As the patient exercises more vig-orously, increases in cardiac output are dependent upon increased ejectionfraction. Thus, some investigators advocate the use of exercise echocardiogra-phy in patients with moderate AR. The inability to augment ejection fractionis an indication for earlier surgery.

Acute aortic regurgitation

Acute AR generally results from endocarditis, aortic dissection, trauma or fail-ure of a prosthetic valve. The clinical presentation is usually dramatic withhypotension and pulmonary edema. LV diastolic pressures increase markedlyas the LV is suddenly exposed to regurgitant flow from the aorta. In acute AR,the LV does not have time to dilate in response to the regurgitant flow and thusmarkedly increased diastolic pressures are found.

In acute AR, the “classic” hemodynamic changes associated with chronicAR, e.g. wide pulse pressure or low diastolic pressure, are absent. There are,however, several characteristic hemodynamic changes that can be observedin these patients. Left ventricular and aortic pressures may equalize duringdiastole. The rate of increase (slope) of the LV pressure during diastole is ele-vated due to rapid filling from the aorta. This is due to the large volume ofblood entering a normal, relatively noncompliant chamber. Rapid filling fromthe aorta can lead to premature closure of the mitral valve. In a heart with acompetent aortic valve, LV pressure does not exceed left atrial pressure duringdiastole and the mitral valve remains open until the left ventricle begins tocontract. In acute AR, however, filling of the ventricle from the aorta during


diastole leads to a rapid increase in LV pressure. If LV pressure exceeds leftatrial pressure, the mitral valve will close before the end of diastole. Prematureclosure of the mitral valve protects the left atrium and pulmonary circulationfrom elevated diastolic pressures but reduces flow of blood from the left atriumto the LV. Decreased systemic flow leads to activation of compensatory mech-anisms which cause left atrial pressure to rise and can result in elevations ofpulmonary capillary wedge pressure and development of pulmonary edema.

Pharmacological treatment of AR

The preferred treatment of patients with severe AR is valve replacement butthere are patients with less severe disease who will be treated medically. Phar-macological treatment is based on the principle that the amount of regurgitation is afunction of the valve incompetence, length of diastole, and peripheral arterial resis-tance. Thus in patients with AR, drugs that decrease peripheral resistance (e.g.nifedipine, angiotensin converting enzyme inhibitors) are beneficial whereasdrugs that lengthen diastole (e.g. �-blockers or rate slowing calcium channelblockers) are deleterious.

Hemodynamic tracings of a patient with severe AR(Figure 10.1)

100

LV(a)

Figure 10.1 Aortic and left ventricular tracings alone (a), with femoral artery pressure (b) and withPCPW (c) in a patient with severe chronic AR (all tracings are on a 200 mm Hg scale). (Continued)


LV/FA

(b)

100LV

866

PCWP

(c)

Figure 10.1 (Continued)

Case study

A 65-year-old male with chronic severe aortic regurgitation was referred forcardiac catheterization prior to valve replacement. He had carried the diagnosisof severe AR for several years and was being treated with enalapril. Annual


echocardiograms confirmed the presence of AR but showed no chamber dila-tion. He had been asymptomatic for years but recently developed dyspnea onexertion.

The hemodynamic tracings are shown in Figure 10.2.Note that the pulse pressure is 60 mm Hg with an aortic diastolic pressure

of 62 mm Hg. A clear dichrotic notch is present on the aortic tracing. Whilethe LV diastolic pressure rises more rapidly than normal, the LVEDP is only15 mm Hg.

The hemodynamics are consistent with mild, not severe, AR. This was con-firmed by aortic angiography that revealed 2+ AR. In the absence of ele-vated filling pressures and with normal LV size, there is no evidence that thispatient will improve with valve replacement. Looked at a different way, valvereplacement will probably not result in a favorable change in hemodynamics(in AR this usually involves a decrease in LVEDP and an increase in aorticdiastolic pressure).

A severe lesion was found in the right coronary artery that was treated byplacement of an intracoronary stent. Following the procedure, the patient’ssymptoms resolved.

IGain 50

Gain 50

Gain 50200

100

0

II

V5

IGain 50

Gain 50

Gain 50

50

25

050 max

II

V5

IGain 50

Gain 50

Gain 50200

100

0

II

V5

Figure 10.2 Left ventricular tracing on a 50 mm Hg scale showing the rise in pressure duringdiastole.


Angiographic classification of severity of aortic regurgitation (basedon the amount of contrast dye entering the left ventricle during anaortogram)

� 1+ some dye enters ventricle but clears with every systole� 2+ the ventricle becomes completely opacified after several heart beatsand remains opacified throughout the cardiac cycle� 3+ after several heart beats—the ventricle becomes as dark as the aorta� 4+ after several heart beats—the ventricle becomes darker than the aorta

CHAPTER 11

Mitral regurgitation

Robert V. Kelly, Mauricio G. Cohen

Introduction

Sudden onset of pulmonary edema is a hallmark of acute mitral regurgi-tation (MR) whereas in the setting of gradual progression of MR, patientsmay remain asymptomatic for years. The natural history of chronic MR isvariable and depends on the regurgitant volume, left ventricular (LV) func-tion, and the underlying cause of MR. Compensatory mechanisms enable thepatient to adapt to severe MR (Figure 11.1). Symptom onset may be insidiousand occur initially only with exertion. The development of symptoms at restcan be an ominous finding especially if coupled with decrease in LV systolicfunction.

Pathology

Mitral regurgitation can be caused by structural abnormalities of the mitralleaflets, papillary muscles, chordae tendineae, or mitral annulus. A partial listof causes of mitral regurgitation includes myxomatous changes, congenitalabnormalities, chordal rupture, papillary muscle dysfunction or rupture sec-ondary to myocardial infarction, endocarditis, trauma, and rheumatic degen-eration. In patients with MR due to myocardial infarction, the posterior leafletis most likely to be incompetent as the posterior papillary muscle is suppliedsolely by the posterior descending artery whereas the anterolateral papillarymuscle has a dual blood supply including diagonal branches of the left anteriordescending coronary artery and often marginal branches from the circumflexartery.

Mitral regurgitation can occur in the setting of a normal mitral valve andapparatus when there is pathology of the mitral annulus. In normal adults,the annulus is about 10 cm in circumference. It is a soft and flexible structure,which contributes to valve closure by transmitting valvular constriction, dueto the contraction of the surrounding LV muscle. Dilation of the left ventriclecan cause dilatation of the mitral valve annulus and mal-apposition of themitral leaflets. Mitral annular calcification (MAC) can also cause regurgitationin severe cases. MAC is associated with hypertension, diabetes, aortic stenosis,Marfan’s syndrome, Hurler’s syndrome and chronic renal failure. In severecalcification, a rigid ring of calcium encircles the mitral orifice and calcific spurs

135


No MR Severe MR

Figure 11.1 Mitral insufficiency. Still frames from left ventriculograms showing no mitralregurgitation on the left and severe mitral regurgitation on the right. In both cases, dye has beeninjected into the left ventricle. Note that the left atrium is opacified on the right but not on the leftindicative of mitral regurgitation. The arrow represents blood flow. With a competent mitral valve,all of the blood ejected by the left ventricle goes into the aorta. In patients with mitral regurgitation,a portion of left ventricular stroke volume goes into the left atrium. LV, left ventricle; LA, left atrium;and Ao, aorta.

may project into the adjacent myocardium. Severe mitral annular calcificationmay also immobilize the mitral leaflets resulting in MR.

Acute MR

Acute MR is a rare but potentially life-threatening condition that is generallyassociated with abrupt onset of dyspnea, heart failure, and shock. Characteris-tic hemodynamic changes include increased LV preload, increased total strokevolume, compromised forward stroke volume, and decreased LV end-systolicdiameter. Causes of acute MR include ruptured papillary muscles (e.g. duringacute MI) or chordae (e.g. from myxomatous disease), myocardial ischemialeading to papillary muscle dysfunction, bacterial endocarditis, and trauma.Acute MR is characterized hemodynamically by regurgitation in the absenceof any compensatory dilation of the left atrium or ventricle. The lack of timefor compensatory mechanisms to develop results in an abrupt rise in left atrial(LA) pressure which is generally accompanied by large V waves on pulmonarycapillary wedge pressure (PCWP) tracing, appearance of large V waves in thePA tracing (so-called Camelback PA tracing in which V waves are reflectedthrough a compliant pulmonary vasculature; see Figure 11.4) and a rapid Ydescent as the distended LA quickly empties.

Hemodynamic concepts in patients with chronic MR

1. The amount of mitral regurgitation is labile and dependent upon:a. Size of regurgitant orifice—the primary determinant of the size of theregurgitant orifice is the underlying pathology. Note, however, that the

Chaper 11 Mitral regurgitation 137

cross-sectional area of the annulus may be influenced by LV size. LV dila-tion will result in an increase in the annulus size and the regurgitant orificewhereas a decrease in LV size (e.g. by diuretics, inotropes, or vasodilators)will cause a reduction in the regurgitant orifice.b. LA compliance—the left atrium dilates over time in patients with chronicMR so that large regurgitant volumes can be accommodated with minimalincreases in pressure.c. Mean pressure difference between LV and LAd. Duration of systolee. Afterload—systolic blood flow from the left ventricle goes in twodirections—into a high pressure, high capacitance system (the aorta) anda low pressure, low capacitance system (the left atrium). Thus, the amountof regurgitant flow is influenced by the ratio of resistance to flow across theaortic valve to resistance to flow across the mitral valve. Increases in aorticafterload (e.g. blood pressure, aortic stenosis, etc.) will increase mitral regur-gitation. Because of blood flow into the left atrium the effective afterload seenby the LV is reduced regardless of the aortic afterload.

2. In the LV pressure–volume loop there is no true isovolumetric contractionphase because regurgitant flow across the mitral valve occurs before the aorticvalve opens.3. Left ventricular end diastolic volume (LVEDV) increases to maintain strokevolume and compensate for the regurgitant volume. Increased diastolic vol-umes result in increased contractility (via Starling’s law). In MR, there isincreased stroke volume and stroke work, although effective forward strokevolume may be normal or reduced.4. The compliance of the left atrium and pulmonary veins is an importantdeterminant of the severity of MR symptoms. In patients with a normal-sizedLA and severe MR, marked elevation in LA pressure occurs, with a promi-nent V wave on the PCWP tracing and significant pulmonary congestion. Inpatients with acute MR, the left atrium initially operates on the steep portion ofthe Frank–Starling curve, with a marked rise in pressure for a small increase involume. Over time, the LA dilates and is able to accommodate rapid filling. Thisusually occurs in a 6–12 month period. Increased LA compliance is a featureof longstanding MR (i.e. increased LA size with minimal increases in LA pres-sure). In this situation, longstanding MR shifts the LA Starling curve to right,minimizing the increases in LA pressures in response to large volume increases.5. LV loading conditions in mitral regurgitation are favorable for preservedejection fraction because LV preload is increased while LV afterload is normalor reduced. In patients with chronic MR, LV systolic contractility can becomeprogressively impaired with clinical indexes of LV function (e.g. ejection frac-tion and fractional shortening) remaining normal. This is why a fall in ejec-tion fraction in patients with severe MR is such an ominous finding and whyejection fraction may worsen after mitral valve surgery (as opposed to aorticregurgitation in which ejection fraction may improve after valve replacementbecause afterload is decreased).


6. Symptoms commonly occur initially with exercise in patients with MR.There is controversy about whether exertional dyspnea in patients with chronicMR is due to increases in pulmonary artery pressure or due to inability toincrease cardiac output. Hasuda et al. found that exertional dyspnea did notcorrelate with pulmonary artery pressures but did correlate with the rise inpulmonary artery pressures per unit of cardiac output in their group of 20patients with MR or AR [1].

Compensatory mechanisms in chronic MR

The hemodynamic changes of chronic MR can be predicted by understandingthe compensatory changes that occur. The heart adapts to chronic MR primarilyby LA and LV dilation. The LV adapts to substantial regurgitant volumes byincreasing LVEDV to maintain adequate forward cardiac output. According toLaplace’s law (wall tension is related to radius × intraventricular pressure), theincreased LVEDV increases wall tension to normal or supranormal levels. Inchronic MR, LVEDV and LV mass increase, usually in proportion to the degreeof LV dilatation. The degree of hypertrophy correlates with the amount ofchamber dilation so that the ratio of LV mass to end-diastolic volume remainsin the normal range (in contrast to the situation in patients with LV pressureoverload). At the same time there is greater volume at a given pressure resultingin a shift in the pressure–volume relationship (Figure 11.2). In most patients,LV compensation is maintained for years but eventually LV failure occurs. Thisresults in an increase in preload and LV end-systolic volume (LVESV), and adecrease in LVEF and stroke volume.

LVPressure

LVPressure

LV Volume

Normal

Normal

Acute MR

Severe chronic MR

LV Volume

(a)

(b)

Figure 11.2 Pressure–volume loops in acute MR (a) and chronic MR (b).


LVESV is an important preoperative prognostic marker, especially in termsof mortality, post-op heart failure, and post-op LV systolic function. LVESVindex is an important marker of LV function in MR patients and helps withmitral valve surgery decisions. End-systolic diameter (ESD) on echo is also auseful prognostic indicator. An ESD >45 mm is generally used as an indicationfor surgery although if mitral valve repair can be accomplished a lower cut-offvalue has been advocated.

The transition from a compensated state to a decompensated state in chronicMR is not completely understood. LV systolic function begins to decline,LVEDV increases, and filling pressures rise. Thus, chronic decompensated MRresults in both systolic and diastolic dysfunction.

Cardiac catheterization and MR hemodynamics

The primary findings at catheterization in chronic MR include:� Increased right ventricular and atrial pressures.� In severe MR, effective cardiac output is usually depressed while stroke vol-ume (i.e. the combination of forward and regurgitant flow) is usually increased.Functional capacity during exercise depends primarily on cardiac output andnot the regurgitant volume.� V waves may or may not be present in the PCWP tracing (Figure 11.3). Theyoccur during ventricular systole and coincide with the T wave on the ECGand descent of pressure (isovolumetric relaxation) on the LV tracing. The Vwave represents the rise in LA pressure during ventricular contraction and theheight of the V wave is determined by the volume of blood entering the LA, LAcompliance, and LA size. Because the size of the V wave is influenced by factorsother than regurgitant flow, large V waves are neither sensitive nor specific forMR and the absence of significant V wave does not rule out significant MR (fora more detailed discussion of V waves see Chapter 5).� There may be a “Camelback” appearance to the PA waveform. This tracingis characterized by a bifid appearance in which the systolic peak is followedby a second peak, which is a reflected V wave (Figure 11.4).� In MR, the A wave is generally not affected. In cases of pure MR, the Ydescent in the pulmonary capillary wedge pressure is rapid (pseudoconstric-tion pattern) as the distended left atrium empties rapidly during early diastole.However, in patients with mixed mitral valve disease, the Y descent is grad-ual. A brief early diastolic pressure gradient between LA and LV may occur inpatients with isolated severe MR as a result of the rapid blood flow across anormal size mitral orifice early in diastole.

Physical examination

The pulse is sharp in severe chronic MR and the pulse volume is usually nor-mal. The apical impulse is brisk and hyperdynamic. It is often displaced to theleft with a prominent LV filling thrill. On auscultation, the first heart sound


Figure 11.3 PCWP tracing in a patient with severe MR showing a V wave (arrow).

is diminished in severe MR. A wide splitting of S2 is common. It results fromthe shortening of the LV ejection and an earlier aortic valve closure as a conse-quence of reduced resistance to LV outflow. If severe pulmonary hypertensiondevelops, P2 is louder than A2.

There is a pansystolic murmur. In severe MR, it commences immediatelyafter a soft S1. It may extend beyond A2 because of the persisting pressuredifference between the LV and LA after aortic closure. The murmur is a blowinghigh-pitched murmur and is loudest at the apex. It often radiates to the axilla,but it may also radiate to the sternum or aortic valve area if the posteriorleaflet is involved, mimicking aortic stenosis. The murmur of MR is usuallyaccentuated by isometric exercise. In papillary dysfunction, the MR murmurmay occur later in systole with preservation of a normal S1 because the initialclosure of the mitral valve is unaffected. A third heart sound but not a fourthheart sound is typical of chronic MR.

In acute severe MR, the patient is almost always symptomatic and oftenin heart failure. A systolic murmur will be present although it may not beholosystolic and may disappear in mid or late systole if left atrial pressure iselevated. The murmur of acute MR is generally lower in pitch and softer thanthe murmur of chronic MR. The S3 and S4 are common but there may not bea hyperdynamic apical impulse if the ventricle is normal in size.


50

25

0

1 2 3

1mV

1mV

1mV

1mV

1mV

1mV

r

s

s

d

s

r r

d

d

2.31 mm Hg/mm

Figure 11.4 Pulmonary artery tracing in a patient with severe MR (the same patient as inFigure 11.2) showing bifid appearance characteristic of a reflected V wave (arrow).


Echocardiography

The mitral valve, valve apparatus, chordae, and papillary muscles can be visu-alized on echocardiography. The LA and LV size can be measured. Severityof MR can be defined by several different echo parameters. The intensity ofthe Doppler signal, the ratio of the regurgitant flow (RF) to the forward flow(FF), the regurgitant jet area, the ratio of regurgitant jet to LA area, and effec-tive regurgitant orifice (ERO) all correlate with the severity of MR. Reversal ofDoppler flow in pulmonary veins during systole can also be assessed and isan important indicator of severe MR.

Important points� The amount of MR is labile and dependent upon

� Size of regurgitant orifice� LA compliance� Mean pressure difference between LV and LA� Duration of systole� LV afterload (i.e. blood pressure)

� Acute MR is generally associated with abrupt onset of heart failure, normalsize LV, hyperdynamic LV, and softer, shorter murmur than chronic MR� Chronic MR is associated with reduced LV afterload. This is why a fall inejection fraction is ominous and why ejection fraction does not improve aftervalve replacement� LVEDV increases to maintain stroke volume� Pronounced V waves on PCWP tracing are suggestive of MR but are neithersensitive nor specific

Hemodynamics of mitral regurgitation

Acute� Elevated PCWP and PA� Prominent V wave� Pseudoconstriction (rapid dilation of left atria, RV, and right atria can sim-ulate hemodynamic findings of constrictive pericarditis)� Hyperdynamic LV function� LV is normal in size� May have hypotension and shock

Chronic compensated� Normal to mild right heart pressure elevation� Less prominent V wave� Mild to moderate LV dilation� Normal ejection fraction


Chronic decompensated� Elevated PCWP, PA, and right heart pressures� Marked LV dilation� Decreased ejection fraction

Pseudomitral stenosis in a patient with severe mitral regurgitation

The pressure gradient across a stenotic valve is a function of resistance and flow(P = R × F). In patients with mild MS and severe MR, the diastolic gradientacross the mitral valve can simulate severe MS because of the excessive flowacross the valve. Remember, in MR, flow across the mitral valve is equal tocardiac output + regurgitant volume. Thus in severe MR, flow across the mitralvalve may exceed 2 × cardiac output.

Grading severity of mitral regurgitation on LV angiography

1+ Contrast enters LA but clears with each beat and never fills the entire LA2+ Faint filling of entire LA but the opacification of the LA remains less thanthat of the LV3+ Complete opacification of LA with opacification of the LA equal to that ofthe LV4+ Complete and dense filling of the LA with the initial beat and opacificationof the LA becoming darker than LV; evidence of contrast in the pulmonary veins

Reference

1 Hasuda T, Okano Y, Yoshioka T, Nakanishi N, Shimizu M. Pulmonary pressure–flow rela-tion as a determinant factor of exercise capacity and symptoms in patients with regurgitantvalvular heart disease. Int J Cardiol 2005;99:403–407.

CHAPTER 12

The tricuspid valve

David A. Tate

Introduction

The tricuspid valve separates the right atrium (RA) from the right ventricle(RV). Its three leaflets are less well defined than those of the other cardiacvalves. The leaflets are also thinner and more translucent, a design which is ade-quate to the relatively lower pressures and hemodynamic stresses to which theright-sided valves are subjected under normal circumstances. Indeed, althoughstructural tricuspid valvular abnormalities do occur, hemodynamic perturba-tions involving the tricuspid valve are far more commonly seen in patients withmorphologically normal valves subjected to abnormal hemodynamic stresses.

Tricuspid regurgitation

PathophysiologyTricuspid regurgitation due to primary structural valvular abnormalities israre, but will occasionally be seen due to rheumatic heart disease, myxomatousdisease (prolapse), infective or marantic endocarditis, carcinoid heart disease,anorectic drugs, trauma, Marfan’s syndrome, or Ebstein’s anomaly. In addition,the tricuspid valve can be damaged during placement of pacemaker leads orimplantable cardioverter-defibrillators.

In contrast, secondary functional tricuspid regurgitation is very commonlyencountered as a consequence of any condition associated with increased pul-monary arterial pressures such as left ventricular failure, mitral regurgitation,mitral stenosis, primary pulmonary disease, and primary pulmonary hyper-tension. Dilatation of the tricuspid annulus may also be seen in right ventricularinfarction or due to dilated cardiomyopathy.

The backward flow of blood from the RV to the RA is often clinically subtledue to the relatively compliant RA and the more conspicuous manifestationsproduced by the underlying primary disease process. Nevertheless, as theregurgitation becomes more severe, the signs and symptoms of elevated sys-temic venous pressure become evident, and, in severe disease, cardiac outputis diminished.

145


Hemodynamic changes detected by physical examSymptoms and signs are often due to associated left-sided heart disease orpulmonary disease but, if there is significant tricuspid regurgitation, signs ofthis are generally evident on the physical exam. As with any cause of rightheart failure, there are likely to be congestive findings on the physical examsuch as pedal edema, ascites, and hepatic enlargement.

Inspection of the jugular veins will generally demonstrate both disten-tion correlating with elevated RA pressure, and a distinct and prominent CVwave reflecting systolic regurgitant flow into the RA. The typical murmur isholosystolic and located at the left sternal edge. Augmentation of the murmurwith inspiration helps to distinguish tricuspid from mitral regurgitation.

Hemodynamic changes detected by echocardiographyDoppler echocardiography is invaluable in the evaluation of tricuspid regurgi-tation. The echocardiographic portion of the study evaluates the structure of thevalvular apparatus, but, as noted above, this is normal in the overwhelmingmajority of cases. RA and RV size, however, give important clues as to theduration of the volume and pressure overload.

The Doppler component of the study yields specific hemodynamic informa-tion. Color flow and pulse-wave examination reveals the presence, direction,and magnitude of the regurgitant jet. Detection of systolic flow reversal inthe inferior vena cava and hepatic veins is generally indicative of severetricuspid regurgitation. Finally, continuous wave Doppler and the modifiedBernoulli equation can be used to estimate the RV and pulmonary arterysystolic pressures. In tricuspid regurgitation, the gradient between the RVand the RA during systole equals four times the square of the velocity. Thisgradient is then added to the estimated right atrial pressure (the jugularvenous pressure) to estimate RV systolic pressure. In the absence of pulmonicstenosis, this also equals pulmonary systolic pressure. It is important torecognize that this calculation estimates the severity of the pulmonaryhypertension, not the volumetric severity of the tricuspid regurgitationitself.

Hemodynamic changes evident at catheterizationIn the catheterization laboratory, the findings associated with tricuspidregurgitation are most evident in the RA pressure tracing (Figure 12.1). Thenormal RA waveform consists of an A wave associated with atrial contraction,a C wave associated with ventricular contraction, and a V wave associatedwith rising atrial pressure just prior to opening of the tricuspid valve. Withvolumetrically important tricuspid regurgitation, there is a large systolic wavein the right atrial tracing reflecting retrograde ejection of blood and consequenttransmission of pressure from the RV to the RA. This waveform is variouslytermed an S wave, or a CV wave as it occurs during systole and subsumes the

Chapter 12 The tricuspid valve 147

r r r r r r r r r

I

II

V5

I

II

V5

50

25

025mm/s >

25mm/s >

RA

RV

0.38mm Hg/mm1 2 3 4 5 6 7 8 9 SEREAS IV:WITH MEDICAL

r1mV

1mV

1mV1mV

1mV

1mV

1mV

1mV

1mV

1mV

1mV1mV

r r r r r r r r r r

s

s

s

s

ss

s s

s

dd

d d

d

d d

d

50

25

0 1 2 3 4 5 6 7 8 9

(a)

(b)

Figure 12.1 RA and RV pressure tracings from a patient with severe tricuspid regurgitation.


Figure 12.2 Simultaneous RA and RV pressure in a patient with severe tricuspid regurgitation.

normally independent C and V waves. The magnitude of the systolic wave isdetermined both by the severity of the regurgitation, and the compliance of theRA.

With severe tricuspid regurgitation, the systolic wave in the RA becomes soprominent that the tracing resembles the right ventricular tracing (Figure 12.1).Indeed, in very severe tricuspid regurgitation the RA and RV pressure tracingsare virtually identical, reflecting the fact that in the absence of a competenttricuspid valve, the RA and RV become, functionally, a single chamber. Whilethis may be most evident with the simultaneous placement of two fluid-filledcatheters in the RA and RV (Figure 12.2), this is procedurally cumbersomeand the catheter across the tricuspid valve introduces the possibility of artefac-tual tricuspid regurgitation. Careful performance and observation of RV to RA“pullback” pressures, paired with Doppler and echocardiographic analysis,render simultaneous RA and RV pressure determination generally unneces-sary. Similarly, while right ventricular angiography can help assess the severityof tricuspid regurgitation, the catheter across the tricuspid valve and the ectopyassociated with right ventricular contrast injection often introduce artefactualerror. With current echocardiographic and Doppler techniques, angiographygenerally is of little additional utility.

Catheterization data do not allow definitive distinction between structuraland functional tricuspid regurgitation. However, severe tricuspid regurgita-tion associated with relatively low RV and PA systolic pressures (less than 40mm Hg) are more likely to be at least partially due to organic valvular disease.


On the other hand, tricuspid regurgitation associated with very high RV sys-tolic pressures is much more likely to be functional.

TreatmentIn the case of functional tricuspid regurgitation, the mainstay of therapy is treat-ment of the condition causing pulmonary hypertension. This may be medicaltherapy as, for example, in the setting of left ventricular failure, or procedu-ral as, for example, in the setting of mitral stenosis. Diuretics may be usefulfor refractory fluid retention. With structural valve disease, tricuspid valverepair or replacement is appropriate for patients refractory to medical therapyor sometimes at the time of surgery for coexistent mitral valve disease. Oftena prosthetic ring is used for annuloplasty. If valve replacement is necessary,bioprostheses are favored because the tricuspid valve may be relatively proneto thrombosis.

Tricuspid stenosis

PathophysiologyAcquired tricuspid stenosis is uncommon. Most cases are due to rheumaticheart disease, carcinoid heart disease, or prosthetic valvular stenosis of a valveinitially placed for tricuspid regurgitation. When rheumatic tricuspid stenosisis present, it is generally associated with mitral stenosis, which accounts formost of the presenting signs and symptoms. The signs and symptoms oftricuspid stenosis may be mimicked by tumors (myxoma or metastasis) orvegetations that obstruct RV inflow.

Hemodynamic changes detected by physical examThe signs and symptoms of tricuspid stenosis are primarily due to increasedsystemic venous pressure. Peripheral edema, ascites, hepatic enlargement, andright upper quadrant discomfort may develop with chronic tricuspid steno-sis or regurgitation. Decreased cardiac output may cause pronounced fatigue.Jugular venous pressure is increased, and, if the patient is in sinus rhythm,there is a prominent A wave due to impaired RV filling during atrial sys-tole. However, due to the increased right atrial pressure, these patients areoften in atrial fibrillation. Though clinically subtle, the finding of a blunted orabsent Y descent supports the diagnosis. The murmur of tricuspid stenosis isa low-pitched diastolic murmur at the lower left sternal edge. However, this isoften obscured by or difficult to differentiate from the usually associated mitralstenosis murmur. Accentuation of the murmur during inspiration may help toidentify a component of tricuspid stenosis, even if there is concurrent mitralstenosis.

Hemodynamic changes detected by echocardiographyEchocardiography typically reveals thickened tricuspid leaflets, decreasedmobility, scarred chordae, and sometimes doming of the leaflets if they remain


pliable. Carcinoid heart disease is associated with a distinctive morphologyof a thickened tricuspid valve that is narrowed and fixed in the open position.The pandiastolic gradient of tricuspid stenosis is reflected in both the slow Eto A slope of the M-mode echocardiogram and in the slowly falling velocityof the turbulent flow on Doppler assessment. Doppler evaluation also allowsestimation of the diastolic pressure gradient by the modified Bernoulliequation.

Hemodynamic changes evident at catheterizationWith modern echocardiographic and Doppler techniques, cardiac catheteriza-tion is generally not necessary for the diagnosis of tricuspid stenosis. Whenthe diagnosis is in doubt, however, careful invasive determination of thetransvalvular gradient is necessary. In most cases, a careful RV to RA pull-back will reveal the gradient. However, the gradients are generally small, inthe range of 4–8 mm Hg. If the cardiac output is low, tricuspid gradients are par-ticularly likely to be low and may not be adequately evaluated with a catheterpullback. In this case, separate simultaneous catheters should be placed in theRA and RV. Clinically, significant tricuspid stenosis is usually associated witha valve area of 1.5 cm2 or less.

TreatmentTreatment of tricuspid stenosis includes diuretics and occasionally nitrates torelieve venous congestion. Refractory patients can undergo tricuspid valvereplacement, but in most cases the concomitant mitral valve disease primarilydetermines the indication and timing of surgery. A surgical approach mayalso be indicated for debulking of obstructive tumors or myxoma. The earlyexperience with percutaneous balloon valvuloplasty for tricuspid stenosis isencouraging.

Case study of a patient with tricuspid stenosis

A 49-year-old male is referred for catheterization due to atypical chest pain,mild pedal edema, and an echocardiogram suggesting recurrent tricuspidstenosis with a mean gradient of 6 mm Hg (Figure 12.3a). Ten years previ-ously, he had undergone tricuspid valve replacement with a Hancock porcinebioprosthetic valve for tricuspid stenosis due to carcinoid tumor. Coronaryangiography revealed no hemodynamically significant stenoses. To evaluatethe gradient across the tricuspid valve, separate catheters were placed in theRA and RV.


(a)

(b) IGain:100IIGain:50

V5Gain:50

1mV

1mV

1mV1mV

1mV1mV25

20

15

10

5

01 2 3 4 5

6

7 8 925mm/s >

r

s

v

e

e

e ee e

e

e e

ee

d

d

d d

dd

d d

d

v vv

v

v v RA

RV

rs

rs

rs r r rs

ss

r r r r rss s

s

(c) 25 mm/secDate:02.10.2003Time:13:14:17Wave # 15 len:20 sec

(AIR REST)

(AIR REST)

Simultaneous gradient calculation

rs

20

15

10

5e

e

dd

d d

d d

d

e

v v v

ee e

ee

0RV 28 1,8 RA13/14 (13)

ss

srr r r r r

Figure 12.3 Doppler echocardiography and invasive hemodynamics in a patient with severetricuspid stenosis. Doppler recording across the tricuspid valve (a) and simultaneous RA and RVpressures (b) were recorded in a patient with severe tricuspid stenosis. Panel (c) highlights thetransvalvular gradient during diastole.

Hemodynamic tracings showed a pandiastolic gradient between the RA andRV (Figure 12.3). The mean RA pressure was 13 mm Hg, and the RV end-diastolic pressure is 8 mm Hg. The mean gradient was 7.3 mm Hg, and thecalculated valve area was 1 cm2.

Because of the relatively mild symptoms and the morbidity of a repeat oper-ation, the patient was managed conservatively with diuretic therapy.

PART I I I

Cardiomyopathies

CHAPTER 13

Hypertrophic cardiomyopathy

Jayadeep S. Varanasi, George A. Stouffer

Introduction

Hypertrophic cardiomyopathy (HCM) is the name given to a heterogeneousfamily of disorders characterized by genetic defects involving myocyte sar-comeric proteins. The most common cause is a mutation in the gene thatencodes beta-myosin heavy chain but defects in other genes encoding for sar-comeric proteins including troponin T, troponin I, myosin light chains, alphatropomyosin, and myosin-binding protein C have been implicated in causingHCM. Inheritance is usually autosomal dominant with variable penetrance.Classically, HCM was defined by excessive septal hypertrophy (usually thewidth of the septum on echocardiography exceeding 1.5 cm) with or withoutthe presence of outflow tract obstruction and even today most studies of HCMuse echocardiographic, rather than genetic, criteria to determine eligibility.

Various microscopic features of HCM distinguish it from other diseases thatcause left ventricular hypertrophy (LVH). The principal aspect of HCM ismyocyte disarray in which cells lose their normal parallel arrangement. Indi-vidual myocytes often show variability in size and may form circular patternsaround areas of fibrous tissue. Myocyte disarray can be seen in other illnesses;however, it tends to affect less of the left ventricle when caused by hyperten-sion or aortic stenosis. In HCM, intramyocardial arteries are often obliteratedcreating small areas of fibrosis within the ventricular wall, especially withinthe interventricular septum.

On gross examination in patients with HCM, there is increased myocar-dial mass with a normal or small left ventricular chamber size. The diseaseusually affects the left ventricle more than the right ventricle with the inter-ventricular septum and anterolateral free wall being the most common seg-ments involved. The hypertrophy is often asymmetric with differing wallthickness noted in contiguous segments. Left atrial enlargement is commonbecause of both impaired ventricular relaxation as well as mitral regurgita-tion. Along with fibrosis in the ventricular wall, a fibrotic plaque is sometimesseen on the interventricular septum and is thought to be the result of repet-itive contact between the anterior mitral leaflet and the septum. Five to tenpercent of patients develop “burned out” HCM in which there is LV wallthinning, chamber enlargement, and systolic dysfunction resembling dilatedcardiomyopathy.

155

156 Part 3 Cardiomyopathies

There are several anatomic variants of HCM that have been described. Themost well-known phenotype from a hemodynamic standpoint is hypertrophicobstructive cardiomyopathy (HOCM; also known as idiopathic hypertrophicsubaortic stenosis). It was formally described as a distinct clinical entity in1958, and consists of a narrowed left ventricular outflow tract (LVOT) witha dynamic pressure gradient (Figure 13.1). This anatomic pattern occurs in25–50% of cases of HCM and will be the type of HCM that we concentrateon in this chapter. Other anatomic forms of HCM that have been describedinclude apical HCM [1], concentric hypertrophy, localized hypertrophy (e.g.posterior portion of septum, posterobasal free wall of the LV, or at the midven-tricular level), and HCM of the elderly characterized by severe concentric LVH,a small LV cavity, and hypertension [2]. The vast majority of intraventricularpressure gradients are localized to the LVOT, however, mid-cavity obstructionhas also been reported. Note that in the rest of this chapter that HCM will referto all types of hypertrophic cardiomyopathy while HOCM will refer only tohypertrophic cardiomyopathy in which there is a pressure gradient in the leftventricle at rest or with provocation.

The most feared complication of HCM is sudden cardiac death. HCM is oneof the most common causes of sudden death in individuals between the agesof 12 and 35 and is also a common cause of death in young athletes. Not allpatients with HCM are symptomatic but in those that are, dyspnea, angina,and palpitations are common.

Physical exam

Signs of LVH, such as an S4, and laterally displaced precordial impulse may bepresent. Most patients with intraventricular gradients have a double or tripleapical impulse. Atrial contraction may be noted as a presystolic apical impulse,as well as a prominent A wave in the jugular venous pulsation. The carotidpulse may display a “spike and dome” configuration. In patients with an LVOTobstruction, there is a harsh midsystolic murmur that commences well afterthe first heart sound and which becomes louder with maneuvers that decreaseLV size (e.g. standing from squatting, Valsalva, or dehydration). The murmuroften decreases with handgrip exercise. A murmur of mitral regurgitation maybe present as well.

Hemodynamics

An understanding of the hemodynamic manifestations of HOCM can be facili-tated by considering causative factors separately while realizing that in a givenpatient there will be significant interactions between these factors as well aspotential contributions from other hemodynamic factors and/or arrhythmias,which will not be considered here. The four factors that will be discussedinclude diastolic dysfunction, LVOT obstruction, mitral regurgitation, andabnormal coronary flow reserve (Table 13.1).

Chapter 13 Hypertrophic cardiomyopathy 157

Reduced outflow

Hypertrophiedbasal septum

I

II

V5

25 mm/s>

Femoral artery

LV Apex

400

200

0

Mitral regurgitation

Hypertrophic obstructive cardiomyopathy

Figure 13.1 Eccentric hypertrophy and LV outflow tract gradient in HOCM. HOCM ischaracterized by asymmetric septal hypertrophy (a) which causes an obstruction to LV outflow anda pressure drop in the LV outflow tract (b).

Diastolic dysfunction is present, in some degree, in almost all patients withHCM. Diastolic dysfunction progresses as the ventricle hypertrophies andbecomes less compliant. Higher pressures are required to fill the ventricle lead-ing to increased left atrial, pulmonary artery (PA), right ventricular (RV), andright atrial pressures. Progressive diastolic dysfunction results in clinical man-ifestations of congestive heart failure in some patients with HOCM.


Table 13.1 Hemodynamic findings in HOCM.

Spike and dome configuration ofarterial pulse

On aortic pressure tracing, there is a rapid riseduring systole which is followed by a mild dropin pressure and then a secondary peak

Systolic intraventricular pressuregradient

Simultaneous intraventricular and aortic pressuretracings will show a difference in maximumsystolic pressure at rest and/or with provacation

Diastolic dysfunction LV end-diastolic pressure will be elevated, the rapidphase of LV filling will be prolonged and atrialcontribution to LV filling will be accentuated

Brockenbrough sign Aortic pulse pressure fails to widen during apost-extrasystolic beat

Outflow tract obstruction in patients with HOCM is “dynamic,” whichmeans the resistance to flow changes depending on filling pressures, after-load (e.g. aortic pressure) and force of contractility. LVOT obstruction changesin severity depending on the force of systolic contraction and the dimensionsof the LV (in contrast to aortic stenosis which, since it is valvular and not mus-cular, is a fixed obstruction). Thus, patients with HOCM may not have anyobstruction while at rest but may develop significant pressure gradients dur-ing any activity which increases the force of cardiac contraction (e.g. exercise,emotional stress) or at times when their LV is smaller because of incomplete fill-ing (e.g. dehydration). This obstruction can cause pressure gradients of greaterthan 100 mm Hg and is thought to be one etiology of exercise-induced syncopein these patients (another possible etiology is arrhythmias). LVOT gradientscan be provoked using methods that decrease filling pressures and stroke vol-ume (e.g. Valsalva maneuver or nitroglycerin) or methods that increase theforce of contraction (post-PVC or isoproterenol infusion). Dobutamine shouldbe avoided as it can cause subaortic pressure gradients in normal hearts dueto catecholamine stimulant effects.

Mitral regurgitation is thought to be caused by systolic anterior motion(SAM) of the mitral valve during midsystole which interferes with normalvalve closure. Mitral regurgitation is usually mild but can become significant,especially in association with large subaortic pressure gradients. The magni-tude of pressure gradients measured at cardiac catheterization correlate wellwith Doppler velocities and with the duration of SAM. Prolonged mitral valve-septal coaptation (>30% of systole) is invariably associated with high LVOTgradients and mild to severe mitral regurgitation. Although the role of SAMin producing the gradient is controversial there is a close relationship betweenthe degree of SAM and the size of the LVOT gradient in patients with subaorticobstruction (as opposed to mid-cavity obstruction) [3, 4].

Coronary flow reserve, the ratio of the maximal to the resting coronary bloodflow and thus an index of the ability to increase coronary flow under physio-logic stress, is abnormal in most patients with HOCM. A study of 20 patientswith HCM found higher resting coronary blood flow and lower coronary


resistance in comparison to 28 controls. Most patients with HCM achievedmaximum coronary vasodilation and flow at modest increases in heart rate.Higher heart rates resulted in severe myocardial ischemia, elevation in LV end-diastolic pressure (LVEDP), and a decline in coronary flow [5]. The same groupof investigators, in a subsequent study of 50 patients with HCM, found thatpatients with LVOT obstruction at rest had higher coronary flow and regionalmyocardial oxygen consumption at rest and with pacing compared to patientswith HCM but without obstruction. In patients with obstruction, transmuralcoronary flow reserve was exhausted at a heart rate of 130 beats/min. Inter-estingly, in patients without obstruction, myocardial ischemia occurred at alower coronary flow than in patients with obstruction [6]. A more recent studycomparing coronary blood flow in eight patients with symptomatic HCM toeight matched controls found that patients with HCM had higher resting coro-nary blood flow, lower coronary resistance, and lower coronary flow reserve.Patients with HCM also had abnormal phasic coronary flow characteristics.These results are consistent with the reduction of coronary flow reserve inpatients with HCM being caused by near maximal vasodilation of the micro-circulation in the basal state [7].

Findings at cardiac catheterization

Cardiac catheterization is usually not needed to confirm the diagnosis butoccasionally patients with HCM will undergo catheterization for evaluation ofchest pain or congestive heart failure.

Left atrium or pulmonary capillary wedge pressurePulmonary capillary wedge pressure (PCWP) is generally elevated in patientswith HCM, even more so when there is significant mitral regurgitation. TheA wave may be accentuated due to a stiff ventricle and the V wave can beincreased, either from mitral regurgitation or reduced LA compliance. RVsystolic and PA pressures are generally only mildly elevated, and cardiac out-put is preserved, until LV failure occurs.

As an example of filling pressures in patients with HCM, a study of 20patients with HCM (9 with significant outflow tract gradients) compared to 28controls found higher LVEDP (16 ± 6 versus 11 ± 3 mm Hg) and mean PCWP(13 ± 5 versus 7 ± 3 mm Hg) [5].

LV pressureLVEDP is generally elevated but there are no pathognomonic findings forHCM in the LV pressure tracing. The rapid filling phase of LV may be sig-nificantly prolonged and a prominent atrial contraction wave may be present(Figure 13.2). The upstroke of the LV pressure tracing may have a notch whichcorresponds with the anterior leaflet of the mitral valve coming in contact withthe septum because of SAM.


Figure 13.2 LV tracings in a patient with HOCM. In panel (a), pressures are obtained at baseline.Note the elevated LVEDP, prominent A wave in LV tracing and lack of an outflow tract gradient.Panel (b) was obtained after inhalation of amyl nitrate. LVEDP is increased and an A wave isprominent. Aortic pressure is decreased compared to basal conditions and a small outflow tractgradient is present.


Aortic pressureA variable finding in HOCM is that the dynamic nature of the outflow tractobstruction causes a “spike and dome” configuration of the aortic or peripheralpulse (Figure 13.3). This is known as a bisferiens pulse, a name derived fromLatin: Bis (= two) + Feriere (= to beat). The bisferiens pulse is characterizedby an initial rapid rise in aortic pressure (spike), followed by a slight dropin pressure (dip) and then a secondary peak (dome) and is most prominent incentral aortic pressure but can also be transmitted to the carotids. It is enhancedby maneuvers that increase intraventricular pressure gradients (e.g. followinga PVC or by the Valsalva maneuver) and in patients with obstruction onlyduring provocation, a normal appearing pressure tracing will often be replacedby the spike and dome during provocative maneuvers. Bisferiens pulse shouldbe not confused with the dicrotic pulse, a pulse with an exaggerated dicroticwave. A bisferiens pulse is most commonly associated with HOCM but canalso be seen in severe aortic regurgitation.

Outflow tract gradientA gradient can be demonstrated by comparing pressure in the LV apex (usingan endhole catheter) to aortic pressure or by a slow pullback of the catheterthrough the LV (Figure 13.4). In the cardiac catheterization laboratory, dynamicoutflow tract obstruction can be provoked by a variety of maneuvers including

Figure 13.3 Aortic pressure in a patient with HOCM demonstrating “spike and dome”configuration.


1mV

1mV

1mV

1mV

1mV

1mV

r r r r r r r r r r r r r

s

d

dd d d

d d ddd

e ee

ss

s s ssss

s

Figure 13.4 LV to aortic pullback using a dual lumen end-hole catheter during amyl nitrateadministration in a patient with HOCM. Aortic pressure is measured continuously through thesecond lumen.

methods that decrease preload and/or afterload (e.g. Valsalva maneuver,administration of amyl nitrate; Figure 13.5) or methods that increase the forceof contraction (e.g. isoproterenol infusion or PVC; Figure 13.6). LV outflowobstruction can be reduced or eliminated by maneuvers that increase chambersize (e.g. hydration).

RV intraventricular gradient is common in infants and children with HOCMbut rarely seen in adults. Maron [8] reported that RV obstruction was observedin 60% of infants with HOCM and that RV outflow tract obstruction was fre-quently greater than LV outflow tract obstruction.

Echocardiography

Since the initial development of ultrasonic evaluation of the heart, echocar-diography has been the gold standard for diagnosing HOCM. The cardinalechocardiographic features of HOCM are LVH, an intraventricular pressuregradient and SAM of the mitral valve. There is considerable variability in the



s

ed

r

s

s

e1 2

Normal sinus

3 4 5 6 7 8 9

e e ee

e e

ee e

d d dd d d d d

d d d dd d d

dd

s

s s

s s

ss

ss

s s

s s

s

s

s

r r r r r r r r r r

d

d d

d d

d d

d d d d

dd ddd

de

1 2 3 4 5 6 7 8 9e

e e e e e ee

d

ss ss ss

sss

ss s s

s ss

r r

s

s s ss

s

ssssss

e

1 2 3 4 5 6 7 8 9

e ee e

e

e

d1.50mm Hg/mm

d

d

d

dd

d

d d d

d

r r r r r

(a)

(b)

(c)

Figure 13.5 Demonstration of a provocable gradient with inhalation of amyl nitrate. There is noleft ventricular outflow tract gradient under basal conditions (a). Following inhalation of amylnitrate, aortic pressure falls, and a gradient is apparent (b). Over time, aortic pressure continues tofall and the gradient increases (c). All tracings are recorded on a 200 mm Hg scale.


Monitor Length: 10 sec

1mV1mV

1mV1mV

1mV1mV


s ss

s s s s

d dddd

ddd

200

100

01 2 3 4 5 6 7 8 91.50mm Hg/mm

Figure 13.6 Demonstration of a provokable gradient with PVCs during administration of amylnitrate.

degree and pattern of hypertrophy and in some patients there is variation inthe extent of LVH from region to region. The finding of a thickened septumthat is at least 1.3–1.5 times the thickness of the posterior wall when measuredin diastole just prior to atrial systole has been a commonly used criterion forthe diagnosis of asymmetrical septal hypertrophy.

An intraventricular pressure gradient as measured by continuous waveDoppler interrogation in the aortic outflow tract is present in some patients.Provocative maneuvers (e.g. inhalation of amyl nitrate, Valsalva maneuver) canbe performed in the echocardiography laboratory. Some investigators have crit-icized these maneuvers as being nonphysiologic and advocated using echocar-diography in conjunction with treadmill or bicycle exercise testing to determineprovocable gradients.

The American College of Cardiology/European Society of Cardiology Clin-ical Expert Consensus [9] has recommended classifying patients with HCMinto one of three groups based on the representative peak instantaneous gra-dient as assessed with continuous wave Doppler: (1) resting obstruction—gradient under basal (resting) conditions equal to or greater than 30 mm Hg(2.7 m/s by Doppler), (2) provocable obstruction—gradient less than 30 mmHg under basal conditions and equal to or greater than 30 mm Hg with provo-cation, and (3) non obstruction—gradient less than 30 mm Hg at rest and withprovocation.


1mV1mV

1mV1mV

1mV1mV1mV

s

ss

s

s

s

s

s

d

dddd

d1 2 3 4 5 6

d

e

ee

e

d

dd

s

sIGain:50

IIGain:50

V5Gain:50

200

100

01.50mm Hg/mm

25 mm/s >[-17-]

...157/66 (100)

200

150

100

50

0

(a)

(b)

Figure 13.7 Brockenbrough–Braunwald–Morrow sign. Pulse pressure falls in the first normal beatpost-PVC (arrows) in HOCM (a) but not in aortic stenosis (b).


Other common echocardiographic findings in HCM include normal systolicfunction, diastolic dysfunction with reduced LV compliance, and a mitral valveE/A ratio less than 1.0 (usually <0.8) and an enlarged left atrium.

Treatment

Initial treatment of HOCM usually includes beta-blockers and/or verapamilor diltiazem. In rare cases unresponsive to first line medical therapy, disopy-ramide can be used. If medical treatment is unsuccessful, invasive therapiessuch as alcohol septal ablation, surgical myomectomy, or DDD pacing can beutilized to relieve the obstruction. Defibrillators are useful for patients at highrisk of sudden death.

Invasive treatments are reserved for patients who have severe refractorysymptoms (Class III or IV) despite medical treatment and intraventricular gra-dients of more than 50 mm Hg. The classic Morrow myectomy is performedthrough an aortotomy and involves resection of a relatively small amount oftissue from the proximal septum. Surgical resection of the hypertrophied sep-tum has an initial success rate of 90% in decreasing symptoms and LV outflowobstruction. In a study of 64 patients who underwent transaortic myectomy,there was a sustained improvement in symptoms and a decrease in LVOTresting gradient from 73.2 ± 14.8 mm Hg to 13.6 ± 2.7 mm Hg at an averagefollow-up of 4.6 years (4 mo to 12 yr) [10]. Another study of 20 patients whounderwent either myectomy or mitral valve replacement as treatment of anoutflow tract gradient, found a sustained reduction in gradient, a decrease inbasal coronary blood flow and a decrease in coronary blood flow and myocar-dial oxygen consumption during rapid atrial pacing [11].

Alcoholic septal ablation has been developed as a treatment for LV outflowobstruction due to septal hypertrophy. In this procedure, alcohol is injectedinto a septal branch of the left anterior descending artery causing localizedinfarction of the septum. Alcohol septal ablation appears to elicit a sustainedreduction in outflow tract gradient. In a study of 50 patients, septal abla-tion (mean creatine kinase value 413 ± 193 U/L) reduced LVOT gradients,from 80 ± 33 to 18 ± 17 mm Hg after 4–6 months and to 17 ± 15 mm Hg after12–18 months [12].

Another potential treatment for patients who remain symptomatic despitemaximal medical therapy is placement of a dual chamber pacemaker. RV pac-ing causes the interventricular septum to move paradoxically. In theory, thisincreases LVOT dimensions and thus reduces LVOT blood velocities, whichresults in less SAM of the mitral valve and less mitral regurgitation. While ini-tially proposed in 1975, interest in AV sequential pacing for HCM was revivedby a study [13] of DDD devices in 84 patients with obstructive HCM and severesymptoms refractory to medical therapy. Symptoms were abolished in 33.3%of the patients, reduced in 56% and unchanged in 8%. At a mean follow-up of2.3 ± 0.8 years only four patients had to undergo a surgical procedure. Alsoin this period of time, there was significant improvement in NYHA class, inexercise capacity and in resting LVOT gradient.


Subsequent studies by other groups produced less enthusiastic results anddual-chamber pacing has fallen out of favor as a treatment for HOCM. In amulticenter, randomized, double-blind, crossover study of permanent DDDpacing in 48 patients with drug-refractory HOCM, Maron et al. found an aver-age reduction of outflow gradient of 40% but no change in exercise capacity,peak oxygen consumption, or LV wall thickness in the overall group [14].

Case study

A 75-year-old male with a history of atrial fibrillation reported increasing dys-pnea on exertion and lower extremity edema. A nuclear stress test was sugges-tive of ischemia and he was referred for cardiac catheterization. Simultaneousmeasurement of LV apex and femoral artery pressures revealed no gradientunder basal conditions. Initiation of a PVC elicited a Brockenbrough sign(Figure 13.8a) and Valsalva maneuver precipitated a marked, transient increasein LVOT gradient (Figures 13.8b,c). These findings are consistent with HOCM.

IGain:50IIGain:50V5Gain:50



200

100

0 1.50mmHg/mm

1.50mmHg/mm

25mm/s >[-42-]

25mm/s >[-40-]

r1 mV1 mV

1 mV1 mV

1 mV1 mV

1 mV1 mV

1 mV1 mV

1 mV1 mV

r r r r r rv

v

r

v

a

1 2 3 4 5 6

aa a

a a

(a)

(b) (c)

200

200

100

0

100

v v vv v

v

v

ss

s s ss

d

d d1 2 3 4

d d

d d d d

dd

d

ss

ss

s

ss

e e e e e e

dddddd

a a a a a

a a a

0

r r r r r r r r

Figure 13.8 Illustrative case—simultaneous LV apex and femoral artery pressures demonstratinga Brockenbrough sign (a) and precipitation of a intraventricular gradient during Valsalva maneuver(onset in (b) and offset in (c)).


References

1 Louie EK, Maron BJ. Apical hypertrophic cardiomyopathy: clinical and two-dimensionalechocardiographic assessment. Ann Intern Med 1987;106:663–670.

2 Topol EJ, Traill TA, Fortuin NJ. Hypertensive hypertrophic cardiomyopathy of the elderly.N Engl J Med 1985;312:277–283.

3 Maron BJ, Bonow RO, Cannon IRO, Leon MB, Epstein SE. Hypertrophic cardiomyopa-thy: interrelations of clinical manifestations, pathophysiology, and therapy. N Engl J Med1987;316:780–789; 844–852.

4 Panza JA, Petrone RK, Fananapazir L, Maron BJ. Utility of continuous wave Dopplerechocardiography in the noninvasive assessment of left ventricular outflow tract pressuregradient in patients with hypertrophic cardiomyopathy. J Am Coll Cardiol 1992;19:91–99.

5 Cannon RO, III, Rosing DR, Maron BJ, et al. Myocardial ischemia in patients with hyper-trophic cardiomyopathy: contribution of inadequate vasodilator reserve and elevated leftventricular filling pressures. Circulation 1985;71:234–243.

6 Cannon RO, III, Schenke WH, Maron BJ, et al. Differences in coronary flow and myocardialmetabolism at rest and during pacing between patients with obstructive and patients withnonobstructive hypertrophic cardiomyopathy. J Am Coll Cardiol 1987;10:53–62.

7 Yang EH, Yeo TC, Higano ST, Nishimura RA, Lerman A. Coronary hemodynamics inpatients with symptomatic hypertrophic cardiomyopathy. Am J Cardiol 2004;94:685–687.

8 Maron BJ. Hypertrophic cardiomyopathy. Curr Probl Cardiol 1993;18:637–704.9 Maron BJ, McKenna WJ, Danielson GK, et al. American College of Cardiology/European

Society of Cardiology clinical expert consensus document on hypertrophic cardiomyopa-thy. A report of the American College of Cardiology Foundation Task Force on ClinicalExpert Consensus Documents and the European Society of Cardiology Committee forPractice Guidelines. J Am Coll Cardiol 2003;42:1687–1713.

10 Minami K, Woltersdorf H, Kleikamp G, Bothig D, Koertke H, Koerfer R. Long-term resultsafter myectomy in 64 patients with hypertrophic obstructive cardiomyopathy (HOCM).Morphological and hemodynamic aspects. J Cardiovasc Surg (Torino) 2000;41:801–806.

11 Cannon RO, III, McIntosh CL, Schenke WH, Maron BJ, Bonow RO, Epstein SE. Effect of sur-gical reduction of left ventricular outflow obstruction on hemodynamics, coronary flow,and myocardial metabolism in hypertrophic cardiomyopathy. Circulation 1989;79:766–775.

12 Boekstegers P, Steinbigler P, Molnar A, et al. Pressure-guided nonsurgical myocardialreduction induced by small septal infarctions in hypertrophic obstructive cardiomyopa-thy. J Am Coll Cardiol 2001;38:846–853.

13 Fananapazir L, Epstein ND, Curiel RV, Panza JA, Tripodi D, McAreavey D. Long-termresults of dual-chamber (DDD) pacing in obstructive hypertrophic cardiomyopathy. Cir-culation 1994;90:2731–2742.

14 Maron BJ, Nishimura RA, McKenna WJ, Rakowski H, Josephson ME, Kieval RS. Assess-ment of permanent dual-chamber pacing as a treatment for drug-refractory symptomaticpatients with obstructive hypertrophic cardiomyopathy. A randomized, double-blind,crossover study (M-PATHY). Circulation 1999;99:2927–2933.

CHAPTER 14

Heart failure

Steven Filby, Patricia P. Chang

Introduction

Heart failure (HF) is a clinical syndrome resulting from the failure of the heart tofill properly or to pump blood effectively and meet the demands of the body. Itcan be classified by type of dysfunction (systolic versus diastolic), by anatomyand physiology (dilated, hypertrophic, restrictive cardiomyopathies), and byetiology (Table 14.1).

The physical examination, chest X-ray, and echocardiogram are essential inthe diagnosis of heart failure and are often used to guide medical managementin chronic heart failure patients admitted with decompensation. However, itis difficult to accurately estimate the true hemodynamic status of the chronicHF patient using these tools alone. Physical examination findings of HF can beabsent in as many as 40% of patients with elevated pulmonary capillary wedgepressures (PCWPs) [1, 2]. Additionally, the chest X-ray often does not showevidence of pulmonary congestion in advanced HF patients due to increasedpulmonary lymphatic drainage [3]. With this in mind, there are many situationsin which right heart catheterization with a pulmonary artery catheter is helpfulin the management of hospitalized HF patients (Table 14.2).

Still, right heart catheterization has never been shown to improve mortalityor major clinical outcomes in patients admitted with HF [4–11]. In the Evalu-ation Study of Congestive Heart Failure and Pulmonary Artery Catheteriza-tion Effectiveness (ESCAPE) trial, the use of pulmonary artery catheterization(PAC) in decompensated HF patients did not affect mortality or length of hos-pitalization but was associated with higher adverse event rates than thosepatients who did not receive PAC-guided therapy [11]. Nevertheless, rightheart or PAC can be very useful in guiding the management of severely illpatients when clinically indicated.

The measurements presented in this chapter describe advanced HF patientswith poor systolic function. The hemodynamic findings of other HF states(e.g. restrictive cardiomyopathy, constrictive pericarditis, or hypertrophic car-diomyopathy) will be discussed elsewhere in this book.

Intracardiac pressures

The pressures of the right heart are typically mildly to moderately elevatedin chronic HF (Figure 14.1). Patients with chronic HF usually have higher

169


Table 14.1 Common etiologies of heart failure.

Coronary artery diseaseHypertensionInfectious (Chagas’ disease, coxsackie, echovirus, HIV, Lyme disease)Toxins (ethanol, adriamycin)Metabolic (hypothyroidism)Connective tissue disease (systemic lupus erythematosis)Infiltrative disease (amyloidosis, sarcoidosis, hemachromatosis)Hypertrophic cardiomyopathyPrimary valvular diseaseTachycardia-induced cardiomyopathyCongenital heart diseaseInherited (Neuromuscular disorders [Duchenne’s muscular dystrophy],

X-linked, Mitochondrial [MELAS syndrome], Storage disease)High-output states (arteriovenous fistula, hyperthyroidism, beriberi)Stress-induced cardiomyopathyMiscellaneous (peripartum, giant cell myocarditis, eosinophilic myocarditis/hypersensitivity)Idiopathic

Table 14.2 Indications for right heart catheterization in heart failure.

1 Patients with coexisting disease or injury associated with hemodynamic instability2 Patients requiring therapy for coexisting disease that requires close monitoring

(e.g. perioperative management)3 Patients requiring parenteral therapy (inotropes or vasodilators)4 Patients hospitalized with refractory HF5 Patients with pulmonary hypertension6 Patients with valvular disease, intracardiac shunts, tamponade, or pericardial constriction

baseline intracardiac pressures than those with acute HF because chronic HFpatients typically require higher filling pressures to maintain optimal cardiacoutput (i.e. Frank–Starling mechanism; Figure 14.2). A PCWP of 17–20 mmHg is often needed to optimize cardiac output in these patients; however,very high filling pressures may result in decreased stroke volume and cardiacoutput. Each HF patient has different optimal filling pressures and findingthe ideal filling pressure can require serial measurement of PCWP and cardiacoutput.

Typically, the mean PCWP is greater than the mean right atrial pressure inchronic HF. However, tricuspid regurgitation can raise the mean right atrialpressure above the mean wedge pressure. In patients with predominantly rightHF, the mean right atrial pressure can dominate the right atrial waveform witha prominent V wave and steep Y descent.

The PCWP waveform in HF patients is often characterized by a prominent Vwave (Figure 14.1). This increased V wave may be due to a noncompliant leftventricle and/or atrium or concomitant mitral regurgitation (as a consequenceof left ventricular dilatation). The mean pulmonary capillary wedge and pul-monary artery waveforms may be falsely elevated as the increased V waves are

Chapter 14 Heart failure 171

50

a va

a a

av

v

v v

Figure 14.1 Pulmonary capillary wedge pressure tracing. This PCWP tracing was taken from apatient with dilated cardiomyopathy and severe mitral regurgitation. The mean PCWP is elevatedat approximately 25 mm Hg and prominent V waves are also noted (arrow).

Left ventricular end-diastolic pressure (LVEDP)

Normal

Mild LV dysfunction

Severe LV dysfunction

Cardiac output

Figure 14.2 Cardiac output as a function of LVEDP. This figure shows idealized Frank–Starlingcurves for patients with varying degrees of left ventricular dysfunction. In the normal heart, anincrease in preload results in a steep increase in cardiac output. With left ventricular dysfunction,the curves are displaced down and rightward such that higher filling pressures are needed tomaintain the same amount of cardiac output.

transmitted to the pulmonary vessels. The Y descent of the PCWP waveformmay be steep reflecting the rapid emptying of a distended left atrium.

HF patients typically have mildly to moderately elevated pulmonary arterypressures, and the pulmonary artery diastolic pressure usually correlatesclosely with the mean PCWP. With increasing pulmonary hypertension andincreasing pulmonary vascular resistance (PVR), the pulmonary artery dias-tolic pressure may exceed the PCWP. If the pulmonary artery diastolic pressure


exceeds the PCWP by more than 5 mm Hg, the diagnosis of pulmonaryembolism should be considered.

Cardiac output and cardiac index

Cardiac output and cardiac index in advanced HF are low. In new acute HF,a cardiac index < 2.2 L/min/m2 characterizes cardiogenic shock. However,patients with chronic HF adapt to a low cardiac index by increasing the oxy-gen tissue extraction resulting in a decrease in the mixed venous oxygen sat-uration. As such, chronic HF patients may have cardiac indices as low as1.0–1.5 L/min/min2 without demonstrating clinical signs of shock. As dis-cussed in the chapter on cardiac output, the Fick method for determinationof cardiac output is preferred in patients with low output HF. HF patientsoften have tricuspid regurgitation and low cardiac output states that renderthe thermodilution method inaccurate.

Mixed venous oxygen saturation

Normal mixed venous oxygen saturation (SvO2) is approximately 75%. TheSvO2 is a marker of adequacy of tissue perfusion and decreases when oxygendelivery falls or tissue oxygen demand increases. The SvO2 is directly relatedto cardiac output, hemoglobin, and arterial saturation and inversely related tothe metabolic rate. The SvO2 in HF patients is low, often less than 60% and maybe as low as 30% in severely decompensated patients. Measurement of SvO2 isuseful to monitor progression of disease in HF patients and has been proposedas an alternative to more conventional hemodynamic parameters (i.e. cardiacoutput) to estimate hypoperfusion.

Left ventricular end-diastolic pressure

The left ventricular end-diastolic pressure (LVEDP) is often used as a measureof left ventricular preload. This is generally acceptable in HF patients who typ-ically have elevated LVEDP reflecting an increased volume status. However,the relationship between left ventricular pressure and volume is curvilinear(Figure 14.3). For example, in a poorly compliant, stiff left ventricle, a highLVEDP may actually reflect a relatively small left ventricular end-diastolicvolume (LVEDV). It is important to monitor changes in LVEDP and the effectson cardiac output as the dilated ventricle will require higher preload than anormal ventricle for optimal cardiac output.

Aortic pressure and pulsus alternans

In contrast to acute HF, aortic pressure in chronic HF may be normal or evenelevated. Changes in aortic pulse pressure correlate with changes in cardiacindex. Pulsus alternans may also be seen on the aortic pressure tracing (see


Left ventricular end-diastolic volume (LVEDV)

XY

LVEDP

Figure 14.3 Relationship between left ventricular end-diastolic volume (LVEDV) and leftventricular end-diastolic pressure (LVEDP). With increasing left ventricular volume, there is asteep increase in pressure. The shift from curve Y to X reflects a decrease in LV complianceillustrating that a noncompliant ventricle is associated with a higher LVEDP.

Figure 14.4), noted by the beat-to-beat variation in the peak systolic pressureamplitude. While the precise mechanism of pulsus alternans is not known, itis a sign of severe left ventricular dysfunction. Proposed mechanisms includechanges in preload and contractility leading to alterations in stroke volume[12, 13]. Postectopic potentiation and increased preload may play a role in theinitiation of pulsus alternans. Pulsus alternans is more common during tachy-cardia and occurs more frequently with inotropic therapy. It should be notedthat the absence of pulsus alternans does not exclude severe left ventriculardysfunction. In one study, only 19.1% of patients with moderate to severe leftventricular dysfunction (mean EF of 35%) had pulsus alternans at rest [13].Pulsus alternans as a result of severe right HF may also be seen on pulmonaryartery pressure tracings.

Important points: hemodynamics in HF� Invasive hemodynamic monitoring can be useful in the assessment andmanagement of HF patients.� Right and left heart filling pressures are typically elevated.� The Fick method is preferred for assessment of cardiac output.� Mixed venous saturation (SvO2) is a useful surrogate measurement for theassessment of cardiac output.� Pulsus alternans is a sign of a failing left ventricle.


BP1

r r r r r r r r r r

s

s s

s s

s

s s s

ddd

dddd

(a)

(b)

Figure 14.4 Aortic pressure tracing. The pressure tracing in panel (a) represents pulsusalternans. The beat-to-beat variation in aortic pressure amplitude is evident. The pressure tracingin panel (b) represents “pseudo-pulsus alternans”. In this example, the beat-to-beat variation inaortic pressure amplitude is due to the patient’s underlying arrhythmia: atrial flutter.

Case study (see Figure 14.5)

A 29-year-old female with dyspnea on exertion underwent echocardiographythat showed severe dilation of the right atrium and right ventricle (A). Onright heart catheterization, right atrial pressure was markedly elevated (meanpressure of 30 mm Hg) with an exaggerated Y descent (B). Right ventricularand pulmonary artery pressures were elevated (C). Intermittently, there werebeat-to-beat alterations in the PA pressures in the absence of respiratory orcycle length variations, which is consistent with pulsus alternans (D).

Pulsus alternans is a rare finding in the pulmonary circulation and maybe associated with severe right ventricular failure, severe reactive airway


disease, left ventricular diastolic dysfunction, or pulmonary embolus. Theexact mechanical disturbance accounting for the beat-to-beat alteration in ven-tricular pressure is poorly understood, but several theories exist includingalterations in right ventricular end-diastolic size with subsequent alteration instroke output, inherent beat-to-beat differences in right ventricular contractil-ity, or deranged intracellular calcium kinetics.

(a)

(b) (c)

(d)

Figure 14.5 Pulmonary artery pulsus alternans. An apical 4 chamber still frame from anechocardiogram showed severe dilation of the right atrium and right ventricle (a). Right atrialpressure was markedly elevated (mean pressure of 30 mm Hg) with an exaggerated Y descent(b). Right ventricular and pulmonary artery pressures were elevated (c). Intermittently, there werebeat-to-beat alterations in the PA pressures in the absence of respiratory or cycle lengthvariations, which is consistent with pulsus alternans (d).


Pulsus alternans

Pulsus alternans is characterized by alternating strong and weak pulses andis due to beat-to-beat variation in systolic pressure (Figure 14.4a). It was origi-nally described in 1872. The precise mechanism underlying pulsus alternans isan age old controversy—Is pulsus alternans caused by hemodynamic factorsor intrinsic defect in myocytes? The hemodynamic hypothesis holds that alter-ations in filling and afterload cause changes in myocardial contractility basedon Frank–Starling mechanism. Alternatively, the myocytes hypothesis statesthat metabolic deficiencies and abnormal calcium handling cause a group ofmyocytes to contract only every other beat.

Pulsus alternans was originally described as alternating pulses and is gen-erally sought in aortic pressure tracings. Be aware of pseudo-pulsus alternansthat can occur due to atrial arrhythmias (Figure 14.4b). Note however, thatpulsus alternans can also be observed in left ventricular, right ventricular, orpulmonary artery pressure tracings (Figure 14.5).

References

1 Stevenson LW, Perloff JK. The limited reliability of physical signs for estimating hemody-namics in chronic heart failure. JAMA 1989;261:884–888.

2 Butman SM, Ewy Ga, Standen JR, Kern KB, Hahn E. Bedside cardiovascular examinationin patients with severe chronic heart failure: importance of rest or inducible jugular venousdistention. J Am Coll Cardiol 1993;22:968–974.

3 Chakko S, Woska D, Martinez H, et al. Clinical, radiographic, and hemodynamic corre-lations in chronic congestive heart failure: conflicting results may lead to inappropriatecare. Am J Med 1991;90:353–359.

4 Matthay MA, Chatterjee K. Bedside catheterization of the pulmonary artery: risks com-pared with benefits. Ann Intern Med 1988;109(10):826–834.

5 Fein AM, Goldberg SK, Walkenstein MD, Dershaw B, Braitman L, Lippman ML. Is pul-monary artery catheterization necessary for the diagnosis of pulmonary edema? Am RevResp Dis 1984;129(6):1006–1009.

6 Harvey S, Harrison DA, Singer M, et al. Assessment of clinical effectiveness of pulmonaryartery catheters in the management of patients in intensive care units (PAC-Man): a ran-domized controlled trial. Lancet 2005;366:472–477.

7 Bernard GR, Sopko G, Cerra F, et al. Pulmonary artery catheterization and clinical out-comes: National Heart, Lung, and Blood Institute and Food and Drug AdministrationWorkshop Report. Consensus Statement. JAMA 2000;283(19):2568–2572.

8 Connors AF, Jr, McCaffree DR, Gray BA. Evaluation of right-heart catheterization in thecritically ill patient without acute myocardial infarction. N Engl J Med 1983;308(5):263–267.

9 Connors AF, Jr, Speroff T, Dawson NV, et al., for SUPPORT Investigators. The effec-tiveness of right heart catheterization in the initial care of critically ill patients. JAMA1996;276(11):889–897.

10 Eisenberg P, Jaffe AS, Schuster DP. Clinical evaluation compared to pulmonary arterycatheterization in the hemodynamic assessment of critically ill patients. Crit Care Med1984;12(7):549–553.


11 Binanay C, Califf RM, Hasselblad V, et al. Evaluation study of congestive heartfailure and pulmonary artery catheterization effectiveness: the ESCAPE trial. JAMA2005;294(13):1625–1633.

12 Gleason WL, Braunwald E. Studies on Starling’s law of the heart. VI. Relationshipsbetween left ventricular end diastolic volume and stroke volume in man with observationson the mechanism of pulsus alternans. Circulation 1962;25:841–848.

13 Kodama M, Kato K, Hirono S, et al. Linkage between mechanical and electrical alternansin patients with chronic heart failure. J Cardiovasc Electrophysiol 2004;15(3):295–299.

CHAPTER 15

Restrictive cardiomyopathy


Introduction

Restrictive cardiomyopathy is a disease of the myocardium characterized byimpaired relaxation and diastolic dysfunction of either or both ventricles. Thedisease is progressive and as the ventricles become less distensible, diastolicfilling and cardiac output are impaired and filling pressures increase. A variablereduction in systolic function may also be present as the disease progresses.Initial symptoms are usually low output type complaints (e.g. exertional dysp-nea or exercise intolerance) because of inability to augment stroke volume. Asright atrial pressures increases to 10–18 mm Hg, symptoms of systemic venouscongestion may predominate. Further increases in right atrial (RA) and pul-monary capillary wedge (PCW) pressures will be accompanied by symptomsof orthopnea and paroxysmal noctural dyspnea. The clinical presentation ofrestrictive cardiomyopathy can be similar to constrictive pericarditis and dif-ferentiation of these processes can be difficult. Additionally, some disease pro-cesses (e.g. amyloid) can involve both myocardium and pericardium leadingto a mixed hemodynamic profile.

Causes of restrictive cardiomyopathy include amyloidosis, homochromo-cytosis, metabolic storage diseases, hypereosinophilic syndrome, metastaticmalignancies, sarcoid, carcinoid, idiopathic, endomyocardial fibrosis, andmediastinal radiation.

Hemodynamic principles

The hemodynamic findings of restrictive cardiomyopathy and constrictivepericarditis are similar. Chapter 16 on constrictive pericarditis contains a moredetailed description of these findings.1. Diastolic filling is impaired and determined by degree of restriction.

� Pressures during diastole are elevated and equal in all cardiac chambers.Right and left ventricular diastolic pressures are elevated but generally end-diastolic pressures differ by more than 5 mm Hg.� The difference between the left ventricular end-diastolic pressure(LVEDP) and right ventricular end-diastolic pressure (RVEDP) is accen-tuated by exercise.

179


� Stroke volume is decreased. Cardiac output will fall in the absence oftachycardia.� Severe pulmonary hypertension is more common than in constrictivepericarditis. RV and pulmonary systolic pressures may exceed 50 mm Hg.� Pulsus paradoxus is not found in restrictive cardiomyopathy because ofa stiff, noncompliant septum.

2. Almost all ventricular filling occurs in early diastole.� Exaggerated Y descent in atrial tracings.� Usual respiratory variation of atrial pressures is reduced, but the Ydescent may become deeper during inspiration.� RA tracing may have classic M or W configuration.� Dip and plateau configuration in RV and LV tracings.

Differentiating restrictive cardiomyopathy fromconstrictive pericarditis

The hemodynamic effects of restrictive cardiomyopathy are similar to thoseof constrictive pericarditis and it is generally very difficult to make a correctdiagnosis based on hemodynamics alone. Hemodynamic criteria that favorrestrictive cardiomyopathy include:� Equalization of left and right ventricular filling pressures but with a dif-ference of more than 5 mm Hg between LVEDP and RVEDP. Remember thatthe studies from which the 5 mm Hg cutoff was determined were generallyperformed with high-fidelity manometric catheters.� RVEDP does not exceed one-third of the level of the right ventricular systolicpressure.� RV systolic pressure may be greater than 50 mm Hg.� Concordance—this is defined as an inspiratory decrease in both RV and LVsystolic pressures. This is in contrast to discordance, which is an increase in RVsystolic pressure simultaneous with a decrease in LV systolic pressure (seen inconstrictive pericarditis). The difference in ventricular interdependence is dueto the effects of respiration on ventricular filling. It is imperative to record RVand LV simultaneously on 200 mm Hg scale on fast sweep speed and to assessseveral respiratory cycles.

A review of 82 cases of constrictive pericarditis and 37 cases of restrictivecardiomyopathy found that the predictive accuracy of a difference betweenRVEDP and LVEDP >5 mm Hg was 85%, the predictive accuracy of RV sys-tolic pressure >50 mm Hg was 70%, and the predictive accuracy of a ratio ofRVEDP to RV systolic pressure of less than 0.33 was 76%. If all three criteriawere concordant, the probability of having correctly classified the patient wasgreater than 90%. However, one-fourth of patients could not be classified byhemodynamic criteria [1].

A more detailed description of hemodynamic findings and differentiat-ing restrictive cardiomyopathy from constrictive pericarditis can be found inChapter 16.

Chapter 15 Restrictive cardiomyopathy 181

a

r(a)

(b)

(c)

(d)

(e)

r r r

aa

v v v

5 6 7 8 9SERIES IV:WITH MEDICAL

IGain:100

IIGain:50

V5Gain:50

IGain:100

IIGain:50

V5Gain:50

IGain:100

IIGain:50

V5Gain:50

IGain:50

IIGain:50

V5Gain:50

100

50

01 2 3 4 5 6

ed d d d d d

e ee e ev v v

25 mm/s >[-27-]

1mV1mV

1mV

1mV

1mV

1mV

1mV1mV1mV1mV

1mV1mV

r r r r r r

ss s s

s s

0.75mm Hg/mm

50

25

01 2 3 4 5 6 7

25mm/s >

e

d dd ddd d d d

dd

e e ee

ee e e

s s ss

s

50

25

50

25

0

0

a v v v v ve

ee

e ee

aa a

a

dd d

d

r r r r r r

r

s

ee

ddd

d dd dd

dde

e e ee

eee

s

s

s

s

r r r r

Normal sinus Monitor Len 19/10 sec

Normal sinus Monitor Len 19/10 sec

1mV

1mV

1mV

1mV

1mV

1mV

Figure 15.1 Pressure tracings from a patient with cardiac sarcoidosis. See case study for details.

Echocardiography

The ventricles are usually normal in size. LV hypertrophy is frequently present(of note is that decreased R wave amplitude on ECG coupled with LV hyper-trophy on echo is frequently a clue to restrictive cardiomyopathies). Mitral-inflow velocity in restrictive cardiomyopathy is typically one of a prominentE wave, decreased A wave, an increased ratio of early diastolic filling to atrialfilling (>2), decreased deceleration time (<150 ms), and a decreased isovolu-mic relaxation time (<70 ms). Cardiac valves may be involved in restrictivecardiomyopathy leading to regurgitation.


Respiratory variation in mitral valve inflow is less in restrictive cardiomy-opathy compared to constrictive pericarditis. An echocardiographic study of30 patients (19 with constrictive pericarditis and 11 with restrictive cardiomy-opathy) found that significant respiratory variation in the ventricular inflowpeak velocity ≥10% predicted constrictive pericarditis with a sensitivity andspecificity of 84 and 91%, respectively [2]. Remember that mitral valve inflowvelocity is dependent upon preload and that marked elevation in LA pressuremay mask respiratory variation [3]. Reducing preload (e.g. by head-up tilting)may accentuate respiratory variation in inflow velocity.

Case study

A 49-year-old female with pulmonary sarcoidosis reported increasing dysp-nea associated with chest tightness. Echocardiogram showed normal LV cham-ber size, wall thickness, and function. Initial hemodynamic evaluation in thecardiac catheterization laboratory showed a mean RA pressure of 8 mm Hg.The waveform had a “W” configuration with accentuation of the Y descentduring inspiration (Figure 15.1a). RA and LV pressures during diastole weresimilar (Figure 15.1b). The RV pressure tracing showed a “dip and plateau”configuration and RV diastolic pressures approximated LV diastolic pressures(Figure 15.1c). Following rapid infusion of 500 mL of saline, RA, RVEDP, andLVEDP increased markedly. RA, RV, and LV pressures were similar duringdiastole but some separation of RA and LV at the end of diastole (Figure 15.1d)and RVEDP from LVEDP (Figure 15.1e) was evident. These findings are con-sistent with restrictive cardiomyopathy from sarcoid.

References

1 Vaitkus PT, Kussmaul WG. Constrictive pericarditis versus restrictive cardiomyopathy: areappraisal and update of diagnostic criteria. Am Heart J 1991;122:1431–1441.

2 Rajagopalan N, Garcia MJ, Rodriguez L, et al. Comparison of new Doppler echocardio-graphic methods to differentiate constrictive pericardial heart disease and restrictive car-diomyopathy. Am J Cardiol 2001;87:86–94.

3 Ha JW, Oh JK, Ling LH, Nishimura RA, Seward JB, Tajik AJ. Annulus paradoxus: trans-mitral flow velocity to mitral annular velocity ratio is inversely proportional to pul-monary capillary wedge pressure in patients with constrictive pericarditis. Circulation2001;104:976–978.

PART IV

Pericardial disease

CHAPTER 16

Constrictive pericarditis


Introduction

The pericardium is a two-layered sac that encircles the heart. The visceral peri-cardium is a mesothelial monolayer that is adherent to the epicardium andwhich is reflected back on itself at the level of the great vessels. The parietal layeris a tough, fibrous outer layer. In the potential space that exists between thesetwo layers there is normally 5–50 mL of serous fluid. The normal pericardiumserves three primary functions: fixing the heart within the mediastinum, lim-iting the spread of adjacent infections, and limiting acute cardiac distentionduring sudden increases in intracardiac volumes.

Constrictive pericarditis is a condition characterized by a dense, fibrousthickening of the pericardium that adheres to and encases the myocardiumresulting in impaired diastolic ventricular filling. The general paradigm is thatconstrictive pericarditis occurs over a period of years either due to an acuteinsult (e.g. viral infection) that elicits a chronic fibrosing reaction or a chronicinsult which stimulates a persistent reaction (e.g. renal failure). The true inci-dence of constrictive pericarditis is unknown. In the past, the most commonetiology was tuberculosis but currently idiopathic pericardial constriction isthe most frequent offender.

Clinically, constrictive pericarditis is generally a chronic disease with symp-tom progression over a period of years. The clinical presentation is that ofright-sided heart failure and may resemble restrictive cardiomyopathy, cirrho-sis, or cor pulmonale among other conditions. Occasionally, patients will go foryears without the correct diagnosis being made. Recently, the advent of newerdiagnostic technologies and a change in the predominant etiologies of con-striction has led to increasing recognition of subacute presentations occurringover a period of months [1].

Constriction occurring after acute pericarditis of any cause is rare but a his-tory of prior symptoms suggestive of acute pericarditis should raise one’ssuspicion. Constrictive physiology, often transient, is not uncommon in theweeks following cardiac surgery or acute pericarditis. This often resolves andthe possible transient nature must be carefully considered when planning treat-ment. In patients with permanent constrictive pericarditis, the presentation isusually years after the initial insult. Constriction following radiation therapy

185

186 Part 4 Pericardial disease

Table 16.1 Hemodynamic findings in constrictive pericarditis.

• Almost all ventricular filling occurs in early to mid-diastole• Elevation of atrial pressures• Equalization of early and mid-diastolic pressures (e.g. RV and LV or RA and LV)• Increase in RA pressure during inspiration (Kussmaul’s sign)• Exaggerated X and Y descents• Intrathoracic pressure is not transmitted to the cardiac chambers• Ventricular systolic discordance—defined as inspiratory augmentation of RV systolic pressure

simultaneous with a decrease in LV systolic pressure. Occurs due to the effects ofrespiration on ventricular filling

for malignancy can be a difficult diagnostic dilemma as it can coexist withradiation-induced restrictive myocardial disease.

Hemodynamics of constrictive pericarditis

Elevation of and equalization of diastolic pressures is the hallmark of con-striction (Table 16.1). Right atrial (RA), left atrial (LA), right ventricular (RV),and left ventricular (LV) diastolic pressures are elevated and nearly identi-cal. In contrast to the normal heart where pressures in the cardiac cham-bers are independent (i.e. unrelated) during diastole, in constrictive peri-carditis the stiff pericardium limits expansion of the cardiac chambers. Thechambers can fill beyond a certain limited point only by compressing otherchambers and thus the diastolic pressures equalize. The ventricular waveformhas a characteristic square root sign due to the rapid rise in early diastolicpressure prior to reaching the constraining effects of the rigid pericardium(Figure 16.1).

I

II

V5

50

25

0

r r sss

rr

d

1 2 3 4d d

d dd

ddd

e eeee

eee

Figure 16.1 Ventricular pressure tracings in constrictive pericarditis. Simultaneous LV and RVtracings in a patient with constrictive pericarditis. Note the “dip and plateau” configuration and thenear equalization during mid- and late-diastole. Operative findings in this patient included a dense,adherent pericardium. Central venous pressure dropped by approximately 15 mm Hg in theoperating room with removal of the pericardium from the anterior RV and the RA.

Chapter 16 Constrictive pericarditis 187

Hemodynamic principles1. Diastolic filling is impaired and determined by degree of constriction.

� Pressures during diastole are elevated and equal in all cardiac chambers.Right and left ventricular end-diastolic pressures (RVEDP and LVEDP) areelevated and within 5 mm Hg of each other.� Stroke volume is decreased. Cardiac output will fall in the absence oftachycardia.� Hypovolemia may mask constriction (especially in patients who areoverdiuresed). Fluid challenge may help in diagnosis.� Severe pulmonary hypertension is rare. Usually RV and pulmonaryartery (PA) systolic pressures are <50 mm Hg.

2. Almost all ventricular filling occurs in early diastole (Figure 16.2).� Exaggerated Y descent.� RA tracing may have classic M or W configuration (Figure 16.3).� Dip and plateau configuration in RV and LV tracings.

3. In severe constrictive pericarditis, intrathoracic pressure is not communi-cated to intrapericardial space (Figure 16.4).

Figure 16.2 Comparison of intracardiac pressures in the normal heart and in constrictivepericarditis.


40

(a)

(b)

0

40

Mean RA pressure

0

RA-mean

RAScale-40

No respiratory variationScale-40

Figure 16.3 RA tracing from a patient with constrictive pericarditis. Note the “W” configuration inpanel (a) and the lack of respiratory variation in panel (b).

� Central venous pressure (CVP) and RA pressure do not decrease, andmay actually increase, with inspiration. Classically, this is described asKussmaul’s sign, which is defined as a reversal of the normal pattern ofdecreasing jugular venous pressures during inspiration [2]. Venous flow toRA does not accelerate with inspiration.� Pulsus paradoxus is minimal or absent in rigid constrictive pericarditisbecause ventricular filling is not effected by intrathoracic pressure. In some


Hemodynamic changes of constrictive pericarditis during inspiration

RA LA

Pulmonary volume

± Intrapericardial pressure

Blood flow to left heart

Blood flow to right heart

Intrapleural pressure

RA pressureRV pressureLA pressureLV pressure

Inspiration

50

25

Transmural aortic root pressure

Stroke volume

Mild septal shift

Myocardium normal

Thickened constrictive pericardium

LV

LV

RV

RV

Figure 16.4 Effect of respiration on blood flow in mild constrictive pericarditis.

cases, the pericardium retains some elasticity as it thickens and pulsusparadoxus may be present.� Interdependence of ventricular filling—on inspiration, intrathoracicpressure and pulmonary venous pressure decreases but LA pressure doesnot. A reduced pulmonary vein to LA pressure gradient results in decreasedflow into the LA and LV. Decreased LV filling during diastole allows forincreased RV filling, which leads to an increase in flow across the tricus-pid valve. On expiration, increased LV filling occurs at the expense of RVfilling and the opposite effect on mitral valve and tricuspid valve flowoccurs.

4. Constrictive physiology can be mimicked by restrictive cardiomyopathyand acute volume overload of the heart.

The differentiation of constrictive pericarditis from restrictive cardiomyopa-thy based solely on hemodynamics is difficult and will be discussed in moredetail later in this chapter. Pseudoconstriction occurs when LV or RV vol-ume overload occurs rapidly such that the pericardium does not have timeto expand to accommodate the increased size. It is defined as constrictiveappearing physiology in the absence of pericardial pathology and as such is adiagnosis of exclusion. Listed below are the most common causes of pseudo-constrictive physiology. Bradycardia is listed but does not cause compressive


physiology per se. It is mentioned here only in that the ventricular wave-form can manifest a square root appearance in patients with heart rates inthe 40s.� Large RV infarcts� Acute RV volume overload (e.g. pulmonary embolus)� Acute severe mitral regurgitation� Acute severe tricuspid regurgitation� Acute severe aortic regurgitation� Severe bradycardia� Aneurysms

Physical exam

The most obvious findings are related to elevated right heart pressures andinclude elevated CVP, hepatomegaly, ascites, and peripheral edema. The gen-eral physical examination in patients with constrictive pericarditis is not unlikethose with right heart failure of any cause. There are, however, a few fairly spe-cific physical findings that are important to recognize.

Careful examination of CVP will invariably reveal elevation. Normally,inspiration leads to a fall in CVP as the negative thoracic pressure generatedin the thorax is transmitted to the right-sided cardiac chambers. Kussmauldescribed a paradoxical rise in CVP with inspiration in patients with peri-cardial constriction that is rare in restrictive myocardial disease. Occasionally,prominent X and Y descents can be seen on examination of the neck veins.Friedreich’s sign, the sudden collapse of previously distended neck veins dur-ing early diastole, is found in patients with constrictive pericarditis. A high-pitched early diastolic sound or pericardial knock occurring before a typicalS3 is often heard.

Pericardial imaging techniques

Plain chest X-ray has some value in chronic pericardial constriction as cal-cification of the pericardium can be seen in up to 25% of cases. Pericardialthickening can be seen on echocardiography but must be distinguished fromgain artifact or harmonics. Pericardial effusion is occasionally present and typ-ically small. CT scanning and cardiac MRI are now the gold standard of cardiacimaging modalities used in diagnosing constriction. A pericardial thickness of>3 mm is suggestive of pericardial constriction but it is important to note thatpericardial thickening on imaging is absent in up to 20% of cases. Similarly,not all patients with thickened pericardium will have constrictive pericarditis,however, a thickness of >6 mm adds considerable specificity to the diagnosis.


1. Diastolic pressures are elevated. Equalization of diastolic pressures can beobserved in the normal heart when filling pressures are low and paradoxically


constrictive pericarditis can be masked with hypovolemia. Thus, it is essen-tial that IV fluids must be given in sufficient quantifies to increase diastolicpressures before the diagnosis of constrictive pericarditis can be considered.2. Prominent Y descent on RA tracing. X descent is usually preserved.3. Right and left ventricular diastolic pressures are elevated and similar. Ina study of 15 patients with surgically proven constrictive pericarditis, LVEDPminus RVEDP (determined using high-fidelity manometric catheters) at endexpiration was 4 ± 4 mm Hg [3]. While comparison of LVEDP and RVEDP isthe time-honored way to evaluate for constrictive pericarditis, a comparison ofmean pressures in the RA and the LA (or PCWP) may be more useful as suchrecordings are less subject to artifacts.4. LV diastolic pressures and RA pressures are similar.5. Characteristic dip and plateau configuration to LV and RV pressuretracings.6. RA pressure does not decrease with inspiration (i.e. Kussmaul’s sign). TheY descent may become more prominent with inspiration in patients with someremaining elasticity in their pericardium.7. Ventricular systolic discordance is thought by some to be the most specifichemodynamic sign of constrictive pericarditis. This is defined as inspiratoryaugmentation of RV systolic pressure simultaneous with a decrease in LV sys-tolic pressure and occurs due to the effects of respiration on ventricular filling(Figure 16.5). Inspiration augments right heart filling at the expense of left heartfilling. This is reflected as a drop in LV systolic pressure on particular cycleswhen the RV is rising. It is imperative to record RV and LV simultaneouslyon 200 mm Hg scale on fast sweep speed and to assess several respiratorycycles.8. The RVEDP is usually greater than 1/3 the RV systolic pressure. In a study of15 patients with surgically proven constrictive pericarditis, the ratio of RVEDPto RVSP was 0.57 ± 0.14 [3].9. Pulmonary artery diastolic pressure (and occasionally PCWP) can be lessthan LVEDP during inspiration. The pulmonary artery, but not the LA, is sub-ject to intrathoracic pressures in patients with constrictive pericarditis and thuswill fall with inspiration.

Sensitivity and specificity of various hemodynamic findingsin constrictive pericarditis

A study from the Mayo Clinic of 15 patients with constrictive pericarditis and 7patients with restrictive cardiomyopathy provides some useful information [3].Note however that this study was performed with high-fidelity manometriccatheters. The small, soft fluid-filled catheters that are commonly used incardiac catheterization laboratories can distort waveforms to a significantextent.� Sensitivity and specificity of an LVEDP–RVEDP ≤5 mm Hg for the diagnosisof constrictive pericarditis were 60% and 71%, respectively.


Discordance

Concordance

Figure 16.5 Examples of concordance and discordance. (Courtesy of Cardiovillage.com.)


� Sensitivity and specificity of an RVEDP/RV systolic pressure >1/3 for thediagnosis of constrictive pericarditis were 93% and 57%, respectively.� Ventricular concordance/discordance was 100% sensitive and specific.� Respiratory change in RA pressure <3 mm Hg was 93% sensitive and 48%specific.

Another study found that Kussmaul’s sign was neither sensitive nor spe-cific for constrictive pericarditis. In a study of 135 patients, Kussmaul’s signwas present in only 21% of patients with surgically proven constrictive peri-carditis [1]. Kussmaul’s sign may also occur in right-sided heart failure, rightventricular infarction, and tricuspid stenosis.

Findings on echocardiography

The two-dimensional and Doppler echocardiographic features of constric-tion include a thickened pericardium, abnormal ventricular septal motion,flattening of the left ventricular posterior wall during diastole, respiratoryvariation in ventricular size, dilated inferior vena cava (IVC), impaired dias-tolic filling, and dissociation of intracardiac and intrathoracic pressures [4, 5].In constrictive pericarditis, there is a decrease in the mitral inflow veloc-ity of greater than 25% with respiration. Also with respiration, variation inventricular filling can sometimes be seen. During inspiration, a decrease inLV filling makes more room for RV filling as the interventricular septummoves to the left and hepatic diastolic flow velocities increase during inspi-ration. During expiration, left ventricular filling increases, which decreasesright heart filling and therefore hepatic diastolic forward flow velocity isdecreased. In constriction, diastolic forward flow is usually greater than sys-tolic forward flow. Additionally, hepatic diastolic flow reversal is increased,as the inflow across the tricuspid valve is interrupted by the pericardiumand movement of the septum toward the right ventricle with expiration.Plethora of the IVC is common and failure of the IVC to decrease in diam-eter by 50% during the respiratory cycle is the echo equivalent of Kussmaul’ssign. M-mode imaging can demonstrate an early diastolic notch of the inter-ventricular septum corresponding to rapid cessation of filling, the pericar-dial knock and the onset of the plateau phase of the ventricular diastolicwaveform.

Differentiation of constrictive pericarditis andrestrictive cardiomyopathy

Constrictive pericarditis and restrictive cardiomyopathy are very differentdiseases sharing a similar hemodynamic profile. Both have as the primaryhemodynamic abnormality altered mid- to late-diastolic filling of the ventriclesleading to a syndrome of congestive heart failure. Often the heart failureis insidious in onset and predominantly right-sided. These syndromes may


Table 16.2 Hemodynamic findings in constrictive pericarditis and restrictive cardiomyopathy.

Constrictive pericarditis Restrictive cardiomyopathy

LV systolic function Usually normal May be reducedPA systolic pressure Usually <50 mm Hg May be >50 mm HgRV/LV systolic pressure Discordant ConcordantRVEDP/LVEDP separation <5 mm Hg >5 mm HgRVEDP/RV systolic pressure >1/3 <1/3Kussmaul’s sign Present Absent

mimic many other disease entities and it is common for both conditions to goundiagnosed for years.

In the case of constrictive pericarditis the impediment to filling is causedby the thickened unyielding pericardium. In restrictive cardiomyopathy theabnormality is a result of a poorly compliant myocardium that limits theability of the ventricles to expand and accept the filling volume of the atria.Rarely there is overlap and both entities can coexist (e.g. radiation-inducedmyopericardial disease). The possibility of constriction or restriction shouldbe entertained in any patient presenting with heart failure and normal sys-tolic function (although systolic function will decline as restrictive cardiomy-opathy progresses) particularly when other causes of this entity are notpresent.

Though the differentiation of these two entities can be quite challenging tothe clinician, a thorough understanding of both the similarities and differencesis imperative to arriving at the correct diagnosis. A listing of hemodynamic fac-tors that are helpful in differentiating constrictive pericarditis from restrictivecardiomyopathy are listed in Table 16.2, although experience teaches that it israre to arrive at a firm diagnosis of either condition based on hemodynamicsalone.

Case study

A 54-year-old male with end-stage renal failure was admitted to the hospitalwith signs and symptoms of right heart failure. He had been admitted 10 timesduring the previous 2-year period. Past medical history was remarkable for anepisode of uremic pericarditis 8 years previously requiring pericardiocentesis.The pericardium was calcified on fluoroscopy (Figure 16.6a). Hemodynam-ics revealed elevated RA pressure with prominent Y descent and minimalvariation with respiration (b). Simultaneous recording of RV and LV pressures(c) showed end-diastolic equalization and systolic discordance. LV and RApressures were equal during diastole (d). These findings are all consistent withconstrictive pericarditis.


(a) (b)

(c)

(d)

Figure 16.6 Case of constrictive pericarditis. Note the calcification of the pericardium onfluoroscopy of the heart (a, LAO projection). RA pressure is elevated with prominent Y descentand minimal variation with respiration (b). Simultaneous recording of RV and LV pressures (c)showed diastolic equalization and systolic discordance. LV and RA pressures were equal duringdiastole (d). These findings are all consistent with constrictive pericarditis.


References

1 Ling LH, Oh JK, Schaff HV, et al. Constrictive pericarditis in the modern era: evolving clinicalspectrum and impact on outcome after pericardiectomy. Circulation 1999;100:1380–1386.

2 Bilchick KC, Wise RA. Paradoxical physical findings described by Kussmaul: pulsus para-doxus and Kussmaul’s sign. Lancet 2002;359:1940–1942.

3 Hurrell DG, Nishimura RA, Higano ST, et al. Value of dynamic respiratory changes in leftand right ventricular pressures for the diagnosis of constrictive pericarditis. Circulation1996;93:2007–2013.

4 Hatle LK, Appleton CP, Popp RL. Differentiation of constrictive pericarditis and restrictivecardiomyopathy by Doppler echocardiography. Circulation 1989;79:357–370.

5 Oh JK, Hatle LK, Seward JB, et al. Diagnostic role of Doppler echocardiography in constric-tive pericarditis. J Am Coll Cardiol 1994;23:154–162.

CHAPTER 17

Cardiac tamponade

Siva B. Mohan, George A. Stouffer

Introduction

The pericardium is a fibroserous sac consisting of two parts: (1) a strong exter-nal layer composed of tough fibrous tissue, called the fibrous pericardium,and (2) an internal double-layered sac (visceral and parietal pericardium) com-posed of a transparent membrane called the serous pericardium. The parietalpericardium is fused to the internal surface of the fibrous pericardium. Thevisceral pericardium is reflected onto the heart where it forms the epicardium,the external layer of the heart wall. The potential space between the parietaland visceral layers of the serous pericardium is called the pericardial cavity. Itnormally contains a thin film of serous fluid that enables the heart to move andbeat in a frictionless environment. The pericardium stabilizes the heart withinthe thoracic cavity by virtue of its ligamentous attachments and also preventsextreme dilatation in the setting of a sudden rise of intracardiac volume.

The normal pericardial space contains 15–40 mL of pericardial fluid. A myr-iad of conditions produce pathological accumulation of fluid in the pericardialcavity. Fluid that accumulates in the pericardial space can be serous, serosan-guineous, hemorrhagic, chylous, or a combination of the above. Once a criticalthreshold of pericardial fluid is reached, the pericardial pressure rises in a non-linear fashion resulting in hemodynamic compromise or cardiac tamponade(Figure 17.1).

Hemodynamic pathophysiology

The primary hemodynamic pathophysiologic process in the development oftamponade is increased pericardial pressure that impairs diastolic filling [1–6].Normal pericardial pressure is zero; any increase can have hemodynamic con-sequences. Another way of looking at tamponade is that the effusion com-petes with the cardiac chambers to occupy the limited volume within andsurrounded by the pericardium. Resultant elevations in intracardiac diastolicpressures impair systemic and pulmonary venous return leading to venouscongestion and reduced cardiac output (Figure 17.2). As pericardial pres-sures continue to increase, a variety of compensatory mechanisms are elicitedincluding sympathetic nervous system activation, which leads to tachycardia,increased ejection fraction, peripheral vasoconstriction, and sodium and fluidretention.

197


Inspiratoryfall in

systolicpressure

Cardiacoutput

PCWPRV

PericardialPhase 1 Phase 3Phase 2

40 mm Hg

20 mm Hg

0 mm Hg

3 L/min

20 mm Hg

0 mm Hg

6 L/min

Figure 17.1 Schematic of the phases of cardiac tamponade. (Courtesy of Cardiovillage.com.)

The pericardium is a relatively noncompliant structure and thus the rela-tionship between pericardial pressure and volume is nonlinear. Pericardialpressure rises exponentially as pericardial volume increases after a certainthreshold is reached. Factors that influence the development of hemodynamiccompromise in the setting of a pericardial effusion include the rate at whichthe fluid accumulates, the amount of pericardial fluid, the compliance of thepericardium, the size of the pericardial space, and ventricular compliance.The pericardium expands only minimally in the acute setting but can greatlyincrease in size over prolonged periods of time. Thus, patients with a slowlydeveloping effusion may be asymptomatic even with a large volume effusion.In contrast, if the rate of accumulation is rapid, even a small volume of effusionmay trigger hemodynamic compromise (e.g. in the setting of cardiac trauma).The inherent compliance of the pericardium also plays a role in determining thehemodynamic significance of an effusion. Several disease processes may ren-der the pericardium more stiff or noncompliant allowing for a smaller effusionto cause cardiac tamponade. Left and right ventricular hypertrophy rendersthe heart most resistant to elevated pericardial pressures.

The hemodynamic effects of increased pericardial pressure occur on a spec-trum and cardiac tamponade can be divided into three phases [5, 7]. As thepericardial fluid accumulates, the pericardial pressure rises (nearing or equal-ing right atrial pressure). At this point, clinical signs of tamponade may beabsent as the right heart (RV stroke volume) but not the left heart is compro-mised (phase 1). As pericardial fluid continues to accumulate and pericardialpressure rises, pulmonary capillary wedge pressure begins to rise and exter-nal compression of both left and right ventricles is present (phase 2). End-diastolic pressures throughout the cardiac chambers are elevated and within5 mm Hg of each other (RA–RVEDP–PAD–PCWP–LVEDP). At this stage, clas-sic signs and symptoms of cardiac tamponade are usually seen. While strokevolume is diminished, compensatory tachycardia attempts to maintain cardiac

Chapter 17 Cardiac tamponade 199

RA LA

RA pressureRV pressureLA pressureLV pressure

Inspiration

50

25

Pulmonary volume

Intrapericardial pressure

±

Blood flow to left heart

Blood flow to right heart

Intrapleuralpressure

Transmuralaortic rootpressure

Strokevolume

Marked septal shift

Fluid-filled pericardial space

LV LVRV

RV

Neck veinsdistended

Heart soundsdistant

Venous pressureelevated

Decreased arterialand pulse pressuresoften exist but notpathognomonic

Pulsusparadoxusis frequentlypresent

Systolic arterial pressure

Ven

ous

pres

sure

(m

m H

g)

Art

eria

l pre

ssur

e (m

m H

g)

Venous pressure

In cardiac tamponade venous pressure risesprogressively and linearly; arterial pressure

may be normal or elevated and isdiagnostically unreliable

18

16

14

12

10

8

6

140

120

100

80

60

40

20

160

Cardiac tamponade

Hemodynamic changes of cardiac tamponade during inspiration

Figure 17.2 Cardiac tamponade.


Table 17.1 Findings on catheterization.

Elevated diastolic pressures and equalization of end-diastolic pressuresX descent is preserved but Y descent is small or absent on RA pressure tracingPericardial pressure is elevatedRV pressure does not fall below pericardial pressure in early diastole

(i.e. there is no dip and plateau configuration)Pulsus paradoxus

output. Eventually, if pericardial pressures continue to rise and approximatesRA and RV diastolic pressures, cardiac tamponade may progress to shock withimpaired tissue perfusion (phase 3).

Hemodynamic findings (Table 17.1)1. Elevated pericardial pressure (Figure 17.2).

2. Elevated and near equalization of end-diastolic pressures (right atrial,right ventricular diastolic, pulmonary arterial diastolic, and pulmonarycapillary wedge pressure). Blood flow through the heart is determinedby pressure gradients. Since cardiac filling pressure is intracardiac minuspericardial pressure, intracardiac pressures must increase as pericardialpressure increases to maintain a pressure gradient.

3. Elevated right atrial pressures with a prominent X descent and blunted Ydescent. In the setting of a hemodynamically significant pericardial effu-sion, diastolic filling is impaired and affected by pericardial pressure. Aunique feature of cardiac tamponade is the continuous compression of theheart throughout the cardiac cycle. This limits the fall in ventricular pres-sure during early diastole and thus causes a decrease in the normal atriumto ventricular pressure gradient in early diastole. This, in turn, restricts theamount of early diastolic filling and blunts the Y descent seen in the atrialpressure tracings (Figure 17.3).

In patients with tamponade, a relatively larger amount of atrial fillingoccurs following ventricular contraction since as blood is ejected to thegreat vessels during ventricular systole, ventricular volume transientlyfalls enabling atrial expansion and filling. Right atrial pressure drops imme-diately following ventricular contraction as atrial volume increases; thislarge swing in right atrial pressure gradient in the setting of cardiac tam-ponade elicits an exaggerated right atrial X descent. Conversely, the Ydescent corresponds to ventricular filling and the total cardiac volume isunchanged. The X descent may be exaggerated during inspiration as adecrease in pericardial pressure leads to augmentation of blood flow to theright ventricle during inspiration (Figure 17.2).

Following removal of pericardial fluid with a resulting reduction inpericardial pressure, the Y descent in the RA tracing will become moreapparent (Figure 17.4).


40

RA

Figure 17.3 RA pressure tracing in a patient with tamponade. Note the increased RA pressures,preserved respiratory variation, and blunted Y descent.

4. RV pressure tracing lacks the “dip and plateau” of constrictive pericardi-tis. Unlike constrictive pericarditis, cardiac tamponade is not character-ized by rapid ventricular filling in early diastole nor does RV pressure fallbelow pericardial pressure in early diastole. Therefore, there is no “dip andplateau” configuration in the RV tracing. Right ventricular hypertrophy isone exception to the rule. Because of the thickness of the right ventricularfree wall, early passive diastolic filling may still occur and, therein, the Ydescent is preserved.

5. Pulsus paradoxus—Pericardial pressure tracks intrathoracic pressure.During inspiration, pericardial and right atrial pressures decrease becauseof negative intrathoracic pressure. This results in augmented systemicvenous return to right-sided chambers and a marked increase in the rightventricular volume. This results in an increase in RV filling during inspi-ration which, in the setting of cardiac tamponade, compromises LV filling,the fundamental principle of pulsus paradoxus. Note that it is normalto have a drop in systolic blood pressure with inspiration and that pul-sus paradoxus is defined by the magnitude of the decrease (an “exagger-ated” drop). Pulsus paradoxus is variously defined as a drop of >12 mmHg, a drop of ≥10 mm Hg, or a drop of ≥9% during normal inspiration(Figure 17.5). Note that pulsus paradoxus may not be present in patientswith tamponade and� acute aortic regurgitation (i.e. acute aortic dissection)� elevated LVEDP


aa a a

dd

d

dd

d

d d

aa a aa a

a aa a a a as

s s s

s

RA

s

s s ss s sv v

vv v v v v v v v v v

vv

v

1 mV1 mV

1 mV1 mV

1 mV1 mV

r r r r r r r r r r r r r r r r

Initial pressure

r

d d

s s

d1 2 3 4 5 6 7 8 9 SEREEA IV-WITH MEDICAL0.38mm Hgmm25 mm/s >

[-1-]

aaaa

aaaaaa

aaaa

aa vvvvvv

vvvvv

vvvvv

25

0

1mV

1mV

1mV1mV

1mV

1mV50

P 400 mL of pericardial fluid removed

1 2 3

Pericardialpressure

4 5 6 7 8 925 mm/s >[-1-]

(a)

(b)

25

0

50

I

II

V5

I

II

V5

Normal sinus Monitor Length 10 sec

Figure 17.4 Simultaneous right atrial and pericardial pressure tracings in a patient withtamponade.

� atrial septal defect� pulmonary hypertension� right ventricular hypertrophy

6. Respiratory variation in atrial pressures—Negative thoracic pressure istransmitted to the fluid-filled pericardial space, unlike the situation in con-strictive pericarditis, and influences atrial pressures.


d

dd d d d

aa

a

a

d

0

100

200

0 1 2 3 4 5 6 7SERIES IV: WITHY MEDICAL8

Figure 17.5 Aortic pressure tracing in a patient with tamponade showing pulsus paradoxus.(Courtesy of Cardiovillage.com.)

7. Inspiratory traction—described by Boltwood et al. this is the diastolic equal-ization of pulmonary capillary and right atrial pressures predominantlyduring inspiration [8]. This is due to inspiratory traction of the taut peri-cardium by the diaphragm.

8. Stroke volume is decreased. Cardiac output will fall in absence of tachy-cardia.

Several clinical scenarios can mask the clinical diagnosis of tamponade. Inthe setting of hypovolemia and suspected tamponade, the typical hemody-namic findings may be masked. Low-pressure effusions equilibrate only withright-sided diastolic pressures and at first only during inspiration. A fluidbolus may elicit classic findings. Severe left ventricular dysfunction may alsomask tamponade. In this setting LV diastolic pressures are higher than RVdiastolic pressures and the diagnosis of cardiac tamponade should be depen-dent upon demonstrating elevated pericardial pressures equal to RA pressures.Regional cardiac tamponade can occur when some, but not all, cardiac cham-bers are compressed by loculated effusions. Typical hemodynamic abnormal-ities maybe found only in the compressed chambers or zones. In some cases,loculation may also produce classic tamponade, presumably by “stretching”the pericardium.

Physical exam findings

The physical exam may vary greatly in cardiac tamponade. Sinus tachycardiaand hypotension with narrow pulse pressure are hallmarks. The clinician must


remain aware that blood pressure may remain normal or even elevated untilcardiovascular collapse is imminent.

Pulsus paradoxus, as described above, is an “exaggerated” drop in sys-tolic blood pressure during normal inspiration (not forced inspiration). Thisis a common finding in moderate to severe tamponade. To measure pul-sus paradoxus, sphygmomanometer should be inflated to suprasystolic pres-sure, deflating until Korotkoff sounds initially heard in expiration only.Then, the cuff should be deflated until Korotkoff sounds heard through-out respiratory cycle. The difference of these two pressures is the “para-dox” (normal pulsus paradox is <10–12 mm Hg). This finding is neitherspecific nor sensitive and may be provoked by various conditions includ-ing constrictive pericarditis, severe obstructive pulmonary disease, restric-tive cardiomyopathy, pulmonary embolus, obesity, and right ventricularinfarction.

Pericardial friction rub may be heard in the setting of pericarditis. Jugularvenous distension is usually present with a preserved X descent. In tamponade,venous waves in the neck are not outward pulsations but rather X descentsare apparent as a collapse from a high standing pressure level. The Y descentis typically not seen. Kussmaul’s sign, which was originally described as anincrease in the jugular venous pressure with inspiration but is now commonlyused to describe the lack of a decrease in JVP with inspiration, is not classicallyseen in tamponade without a component of constrictive disease (i.e. effusive–constrictive disease) [9]. Beck’s Triad, described in 1935, is a cluster of findings incardiac tamponade characterized by hypotension, muffled heart sounds, andjugular venous distension.

Hemodynamics of cardiac tamponade as measuredwith echocardiography

The echocardiogram is a useful tool in the diagnosis of cardiac tamponade;however, it is important to note that cardiac tamponade is a clinical diagnosis.The echocardiogram is a quick, noninvasive test, which may aid the clinicianto delineate the size, morphology, hemodynamic impact, and location of theeffusion. Important findings on echocardiography include:� The presence of an effusion is the only sensitive sign in tamponade. Theabsolute size of the effusion is not as useful in predicting tamponade with thecaveat that larger effusions confer greater risk of tamponade.� Early diastolic collapse of the right ventricular free wall.� Late diastolic collapse of the right atrial free wall.� Small chamber sizes.� Inferior vena cava plethora with attenuated respiratory variation.� Significant changes in inflow across both mitral and tricuspid valves duringinspiration.� Abnormal septal motion. On inspiration the ventricular septum movestoward the left heart whereas the septum moves toward the right heart on


Respirometer

Note the respiratory variationin mitral valve flow

Figure 17.6 Doppler recording of LV inflow with a superimposed respirometer in a patient withtamponade. Note the decrease in mitral valve flow during inspiration.

expiration. The filling of the different ventricles is dependent on the positionin the respiratory cycle; an echocardiographic demonstration of pulsus para-doxus.

The major echocardiographic findings indicative of hemodynamic compro-mise are right atrial collapse, right ventricular diastolic collapse, inferior venacava plethora, and exaggerated respiratory variation in mitral valve flow. Rightatrial collapse occurring in late diastole or early systole is a relatively early signof increased pericardial pressure. Inferior vena cava plethora, which appearsto correspond to the increased right atrial pressure, is commonly observedin patients with tamponade. Right ventricular diastolic collapse occurring inearly diastole is believed to occur only with higher pericardial pressures, how-ever, the lack of RV diastolic collapse does not indicate a nonhemodynamicallysignificant effusion.

Doppler transmitral flow velocity paradoxus, a reciprocal respiratory vari-ation in transvalvular right- and left-sided flow velocities, has been thoughtto be a sensitive sign of tamponade. Normally, there is no substantial varia-tion in early diastolic filling velocities throughout the respiratory cycle. Withcardiac tamponade, there are exaggerated increases in right-sided flow veloci-ties and exaggerated decreases in left-sided flow velocities during inspiration(Figure 17.6). A substantial decrease in Doppler transmitral flow velocity withinspiration (>25%) may serve as an indicator of flow velocity paradoxus andtamponade physiology.

Tremendous efforts are expended in determining whether a patient is “intamponade” and, thus, qualifies for emergent therapy. It is important to notethat increased pericardial pressure is a clinical spectrum rather than an all-or-none phenomenon. The important questions are “what is the approximate


pericardial pressure?” and “what effect is the pericardial fluid having on car-diac output, arterial blood pressure, and intracardiac pressures?” The determi-nation of the need for pericardiocentesis must weigh the risk of the procedureversus these questions (and other important considerations including the rateof accumulation and the potential diagnostic value of obtaining pericardialfluid).

(a)

(b)

Figure 17.7 Echocardiographic findings from the patient described in case report. Panel (a) is astill 2D frame from the parasternal long axis showing a 1.1-cm pericardial effusion. Panel (b) is aDoppler image with the probe at the level of the mitral valve.


Prepericardiocentesis Postpericardiocentesis

I

II

V5

I

II

V5

I

II

V5

I

II

V5

I

II

V5

50

25

0

50

25

0

50

25

0

50

25

0

1 mV

1 mV

1 mV

1 mV

1 mV

1 mV

1 mV

1 mV

1 mV

1 mV

1 mV

1 mV

1 mV

1 mV

1 mV

1 mV

25/ mm/s>0.38.mm Hg mm

1 2 3 4 5 6

50

25

0

[-21-]

25 mm/s>[-14-] 0.38.mm Hg mm1 2 3 4 5 6

25 mm/s>[-27-]

0.38.mm Hg mm1 2 3 4 5 6

25 mm/s>[-28-] 0.38.mm Hg mm

1 2 3 4 5 6

25 mm/s>[-15-]

0.38.mm Hg mm1 2 3 4 5 6

r r r r r r r r r r r r r r r r

rrrrrrrr r r r r r r r

r r r r r r r r

a

a

a a a a

aa

aa

aaa

v v

vv v

v

v v

vv v

v

s ss

s

s

sss

e e e e e e ee e

e e

ee

e ed dd d d d d d d

d

dd

d

d d

dd

aa

aa

aa

av

v vv

s

s s

s

s s

ss

vv v v v v vv

a a a a a a a a

(a)

RA

RV

(b)

(c)

RA

Pericardium

Figure 17.8 Pressure tracings from the patient described in the case report. RA (a) and RV (b)tracings before and after pericardiocentesis. Simultaneous RA and pericardial pressures after theremoval of several hundred milliliters of pericardial fluid (c). Note that the scale is not accurate in(c) because the patient is no longer completely supine but rather the thorax is partially inclined inorder to facilitate pericardiocentesis.


Case study

A 42-year-old female with metastatic lung cancer presented with fatigue anddyspnea on minimal exertion. She was found to have a large pericardial effu-sion with tamponade physiology on echocardiography (Figure 17.7). On exam,she had muffled heart sounds, JVP at 16 cm, and a blood pressure of 92/68 mmHg with a heart rate of 112 bpm. Her pulsus paradoxus was 15–18 mm Hg.She was referred for right heart catheterization and pericardiocentesis. Theinitial RA waveform showed exaggerated X descent with blunted Y descent(Figure 17.8a). RV waveform showed elevated RV diastolic pressures withminimal change from early diastole to late diastole (Figure 17.8b). Pericardialpressure was elevated and equalized with RA pressure during inspiration(Figure 17.8c). Pericardiocentesis was performed and 600 mL of fluid wasremoved. The final pericardial pressure was 0 mm Hg. Following pericardio-centesis, Y descents are apparent on RA tracing and evidence of early diastolicfilling is also apparent on RV tracing.

References

1 Kern MJ, Aguirre FV. Interpretation of cardiac pathophysiology from pressure wave-form analysis: pericardial compressive hemodynamics. Part I. Cathet Cardiovasc Diagn1992;25:336–342.

2 Kern MJ, Aguirre FV. Interpretation of cardiac pathophysiology from pressure wave-form analysis: pericardial compressive hemodynamics. Part III. Cathet Cardiovasc Diagn1992;26:152–158.

3 Kern MJ, Aguirre FV. Interpretation of cardiac pathophysiology from pressure wave-form analysis: pericardial compressive hemodynamics. Part II. Cathet Cardiovasc Diagn1992;26:34–40.

4 Spodick DH. Acute cardiac tamponade. N Engl J Med 2003;349:684–690.5 Reddy PS, Curtiss EI, Uretsky BF. Spectrum of hemodynamic changes in cardiac tampon-

ade. Am J Cardiol 1990;66:1487–1491.6 Reddy PS, Curtiss EI, O’Toole JD, Shaver JA. Cardiac tamponade: hemodynamic observa-

tions in man. Circulation 1978;58:265–272.7 Spodick DH. Pathophysiology of cardiac tamponade. Chest 1998;113:1372–1378.8 Boltwood C, Rieders D, Gregory KW. Inspiratory tracking sign in pericardial disease. Cir-

culation 1984;70(suppl II):103.9 Bilchick KC, Wise RA. Paradoxical physical findings described by Kussmaul: pulsus para-

doxus and Kussmaul’s sign. Lancet 2002;359:1940–1942.

CHAPTER 18

Effusive--constrictive pericarditis

Eric M. Crespo, Sidney C. Smith

Introduction

Effusive–constrictive pericarditis is an uncommon syndrome characterized byconstriction of the heart by the visceral pericardium in the presence of a tensepericardial effusion. It is diagnosed when elevated right atrial pressures persistdespite reduction of intrapericardial pressures to normal levels by pericardio-centesis. Although first observed in the 1960s [1], it was not well described untilthe publication of a 13-patient case-series by Hancock in 1971 [2]. Since that timethere has been a paucity of medical literature on the topic, and it was not untilthe recent publication of a case-series by Sagrista-Sauleda that more completeinformation about etiology, incidence, and prognosis of effusive–constrictivepericarditis became known [3].

In their series, the authors prospectively evaluated 1184 patients present-ing to their institution with pericarditis of any type over a 15-year period.Based on their data the authors estimate that the prevalence of effusive–constrictive pericarditis is approximately 1.3% of all patients with pericarditisand 6.8% among patients with clinical tamponade. Additionally, they foundthat although effusive–constrictive pericarditis can occur with all types of peri-carditis, it is relatively more frequent with radiation-related pericardial diseaseand less-often associated with postsurgical pericarditis. The etiology and inci-dence, however, will likely vary between institutions and between differentparts of the world based on the most common causes of pericarditis in eachlocation. The majority of patients diagnosed with effusive–constrictive peri-carditis will progress to chronic constriction, but a proportion of those withidiopathic disease may have only temporary cardiac constriction and then goon to eventual full resolution [3, 4].

Patients tend to have a subacute presentation and are often initially believedto have cardiac tamponade. The true diagnosis can only be made after pericar-diocentesis reveals the presence of underlying constrictive physiology. Thus,continuous monitoring of intracardiac filling pressures during pericardiocen-tesis is required to arrive at the correct diagnosis.

209


Hemodynamics of effusive--constrictive pericarditis

The hallmark of effusive–constrictive pericarditis is the persistence of elevatedright atrial pressure after intrapericardial pressure has been reduced to normallevels by pericardiocentesis. As in both constrictive pericarditis and cardiactamponade the expansion of the cardiac chambers during diastole is limitedand there is elevation and equalization of diastolic pressures in the atria andventricles. Prior to drainage of the effusion the hemodynamic findings mostresemble cardiac tamponade with a preserved X descent and an absent or atten-uated Y descent on the right atrial pressure tracing. Once the effusion has beendrained, constrictive physiology predominates with return and exaggerationof the Y descent leading to a classic M- or W-shaped configuration of the rightatrial pressure waveform. The ventricular pressure tracings demonstrate thesquare-root sign due to rapid ventricular filling in early diastole. For detaileddescriptions of the hemodynamic principles and findings of constrictive peri-carditis and cardiac tamponade refer to Chapters 16 and 17.

Physical exam

The findings on initial physical exam are similar to a patient presenting withcardiac tamponade. Patients will have elevated jugular venous pressure (JVP),and careful examination of the neck veins may reveal a prominent X descentand an absent Y descent. Kussmaul’s sign, an increase in central venous pres-sure with inspiration, may be present and manifest as either increased orunchanged (i.e. failure to decrease) JVP during inspiration. Other sequelae ofelevated right heart pressures may be noted including hepatomegaly, ascites,and peripheral edema. Arterial pulsus paradoxus (inspiratory decrease in sys-tolic pressure by ≥10 mm Hg) may also be noted. Additionally, a pericardialfriction rub or pericardial knock may sometimes be auscultated. It should bekept in mind, however, that the sensitivity and specificity of different physicalexam findings is not well defined.

Pericardial imaging techniques

A discussion of pericardial imaging can be found in Chapter 16. It is possi-ble that pericardial calcification may be less common among patients witheffusive–constrictive pericarditis than it is among those with strictly constric-tive disease. In the series by Sagrista-Sauleda and colleagues, none of thepatients with effusive–constrictive pericarditis was noted to have pericardialcalcification on radiographic examination [3].

Findings on echocardiography

As with the hemodynamic findings during catheterization, the echocardio-graphic findings will fall somewhere on a spectrum between the findings of

Chapter 18 Effusive–constrictive pericarditis 211

cardiac tamponade or constriction depending on whether the effusion has beendrained and intrapericardial pressure has been normalized. Refer to Chapters16 and 17 for further discussion of the echocardiographic findings in constric-tive pericarditis and cardiac tamponade.

Case study

A 53-year-old previously healthy man presented to the hospital with acomplaint of progressive dyspnea on exertion, mild pretibial edema, andpleuritic central chest pain. His symptoms had been progressing over thecourse of 2 weeks, and had only become prominent over the last few days. Oneweek prior to the onset of symptoms he had suffered a fall of approximately8 feet in which he landed directly on his left chest and shoulder. Chest X-rayrevealed cardiomegaly but no pericardial calcification or pulmonary edema.Echocardiogram revealed a large (>2 cm) circumferential pericardial effusionwith mild early diastolic collapse of the right ventricle and exaggeratedrespiratory variation in mitral inflow velocity. Right heart catheterizationrevealed the following:

Before After pericardiocentesispericardiocentesis (230 mL of serosanguinous(mm Hg) fluid) (mm Hg)

Right atrium 17 14Right ventricular 16 16

end-diastolic pressurePulmonary capillary 17 14

wedge pressurePericardial pressure 13 0

The persistent elevation in right atrial pressure despite normalizationof intrapericardial pressure is consistent with the diagnosis of effusive–constrictive pericarditis. In this case the etiology was felt to be his recent traumawith bleeding into the pericardium leading to pericardial irritation and inflam-mation. Despite continued elevation of cardiac filling pressures the patient’ssymptoms completely resolved. The patient was prescribed high-dose ibupro-fen and scheduled for outpatient follow-up.

References

1 Spodick DH, Kumar S. Subacute constrictive pericarditis with cardiac tamponade. Dis Chest1968;54(1):62–66.

2 Hancock EW. Subacute effusive–constrictive pericarditis. Circulation 1971;43(2):183–192.


3 Sagrista-Sauleda J, Angel J, Sanchez A, Permanyer-Miralda G, Soler-Soler J. Effusive–constrictive pericarditis. N Engl J Med 2004;350(5):469–475.

4 Sagrista-Sauleda J, Permanyer-Miralda G, Candell-Riera J, Angel J, Soler-Soler J. Transientcardiac constriction: an unrecognized pattern of evolution in effusive acute idiopathic peri-carditis. Am J Cardiol 1987;59(9):961–966.

PART V

Miscellaneous

CHAPTER 19

Right ventricular myocardial infarction

Robert V. Kelly, Mauricio G. Cohen

Introduction

Right ventricular (RV) dysfunction occurs to some degree in approximately 30–50% of inferior wall and 10% of anterior wall infarcts. Severe RV dysfunctionleading to the classic hemodynamic changes occurs in approximately 10% ofinferior myocardial infarctions and is associated with higher rates of mortality,cardiogenic shock, sustained ventricular arrhythmias, and advanced AV block.Cardiogenic shock patients with RV involvement have a significantly higherinhospital mortality compared with those without RV involvement. WhileRV dysfunction during acute arterial occlusion has classically been labeledas RV infarction, profound RV ischemia may be more accurate as patients withRV infarctions who survive their initial hospitalization tend to do well in thelong term as RV function generally recovers.

The effects of ischemia on the right ventricle

The RV is pyramidal shaped with a triangular base and a thin crescentic freewall. It is designed as a volume pump ejecting into the low resistance pul-monary circulation. The RV is less susceptible to ischemia than the left ventri-cle (LV) for two reasons: oxygen demand is less and the blood supply is moreredundant. Oxygen demand is less due to the smaller muscle mass of the RV(approximately 15% of muscle mass of LV), lower preload and lower afterload.Less oxygen extraction at rest means that the RV has greater extraction reserveduring stress.

The RV receives its blood supply from several sources. The septum is per-fused by the left anterior descending (LAD) artery and posterior descendingartery, the lateral wall by marginal branches of the right coronary artery (RCA)and the anterior wall by the conus branch of the RCA and the moderator branchartery of the LAD. The RV is also thought to receive oxygen directly from theventricular cavity by diffusion.

Coronary perfusion to the RV occurs approximately equally during systoleand diastole both under normal conditions and with collateral perfusion dur-ing RCA occlusion. RV perfusion during systole is greater than LV perfusionbecause of lower systolic intramyocardial pressure and less diastolic intracav-itary pressure.

215

216 Part 5 Miscellaneous

The vast majority of RV infarcts occur with occlusion of the RCA although theLAD and circumflex can also cause RV infarcts. There is a large interindividualvariation in the number and origin of RV branches from the RCA (Figure 19.1)but a direct correlation exists between the proximity of RCA occlusion and theextent of RV infarction.

Clinical presentation, ECG changes, and echocardiographicfindings in RV infarction

The diagnosis of RV myocardial infarction is made using the clinical presen-tation in combination with ECG, echocardiographic, and/or hemodynamiccriteria. The classic clinical triad described with RV infarction is hypotension,clear lung fields, and elevated jugular venous pressure in the setting of aninferior wall myocardial infarction.

The ECG, including both normal precordial leads and right-sided leads, isvery useful in the diagnosis of RV myocardial infarction. The sensitivity of STelevation in V4R, V3R, and V1 is 93%, 69%, and 28%, respectively, while thespecificity is 95%, 97%, and 92%.

Echocardiographic findings that support the diagnosis of RV infarctioninclude RV dilation, RV wall motion abnormalities, paradoxical motion of theinterventricular septum, tricuspid regurgitation, decreased descent of the RVbase, and plethora of the IVC.

Hemodynamics of RV infarction

In proximal RCA occlusion there is compromised perfusion to the RV freewall, with RV dyskinesis, and depressed RV function, which is reflected inthe RV waveform as a sluggish upstroke, depressed broadened systolic peak,delayed relaxation phase, and diminished RV stroke work (Figure 19.2). RVsystolic dysfunction reduces transpulmonary delivery of LV preload leadingto decreased cardiac output despite intact LV contractility. RV dysfunction iscompensated for by augmented RA contraction and by left ventricular septalcontraction that bulges in piston-like fashion into the RV during systole andgenerates force against a stiff RV free wall.

In acute RV ischemia, diastolic biventricular function is key. Depressed RVcontractility results in RV dilation and ischemia impairs RV relaxation. The RVbecomes stiff and dilated in early diastole resulting in increased impedance toinitial RV inflow. As filling progresses during diastole the noncompliant RVascends a steep pressure volume curve leading to the pattern of rapid diastolicpressure elevation (Figure 19.2). Acute RV dilation and elevated RV diastolicpressure cause the interventricular septum to bulge into the LV during diastoleleading to impaired LV compliance and potentially limiting LV filling.

Abrupt RV dilatation within a noncompliant pericardium leads to elevatedintrapericardial pressure. As both ventricles fill during diastole and competefor space within the pericardium the effects of pericardial constraint cause a

Chapter 19 Right ventricular myocardial infarction 217

RV marginal(lateral wall)

Posterior descendingcoronary artery

(septum)

Atrial branches

RV marginals

(b)

(a)

Figure 19.1 Right ventricular blood supply. Panel (a) is an AP cranial projection of RCAdemonstrating RA branches, a large RV marginal branch, and the posterior descending coronaryartery. Panel (b) is from a different patient and demonstrates several RV marginal branchesoriginating from the mid-portion of the RCA. From this angiogram, it is easy to see that the moreproximal the occlusion in the RCA, the greater the extent of RV ischemia.


3 4 5 6 7 8 9

e

e e

e e

e e

ee

d

d

d

d

d

d

d

d

d

d

s

s

ss

s

s

s

rr

r r rr

r r r

Figure 19.2 RV pressure tracing in a patient with RV infarction.

pattern of equalized diastolic pressures and the RV “dip and plateau” pat-tern suggestive of constrictive pericarditis. Right atrial (RA) waveforms in RVinfarction are characterized by diastolic impedance imposed by a stiff dilatedRV. Reduced inflow velocity across the tricuspid valve gives rise to a blunted Ydescent (more prominent in expiration) rather than a rapid Y descent as seen inconstrictive pericarditis. In the RA, the blunted Y descent signifies pandiastolicRV dysfunction (Figure 19.3).

The ischemic RV causes preload and afterload of the right atrium to increaseresulting in enhanced RA contractility reflected in a “W” pattern in RA wave-form characterized by a rapid upstroke and increased peak A wave amplitude,sharp X descent reflecting atrial relaxation, and blunted Y descent of pandias-tolic RV dysfunction. Early studies reported that a prominent Y descent wasfrequently found in RV infarction; however, later studies that timed RA events


10.38mmHg/mm 2 3 4 5 6 7 8 6SERIES IV:WITH MEDICAL

v v v v v vv v

v

v vv va a

a a

aa a

aa a

a a a

r

1mV

1mV

1mV1mV

1mV

1mV

r r r r r r r r r rr r

Figure 19.3 RA pressure tracing in a patient with RV infarction. Note the “W” configuration.

to RV events (rather than to the ECG) showed that the X descent was predomi-nant [1]. Patients with intact RA function demonstrate prominent A waves andX descents and diminished Y descents. An exception will be in patients withtricuspid regurgitation in which V waves and Y descent will be prominent.

More proximal RCA occlusions compromising atrial blood supply resultsin atrial as well as ventricular ischemia. Atrial dysfunction results in furtherelevation in mean RA pressure, an “M” waveform pattern in RA tracing char-acterized by depressed A wave and X descent and blunted Y descent. In acuteischemia there is RA dysfunction, decreased RV filling, with resultant decre-ments in LV preload and cardiac output.


RV infarction is characterized by increased RA and RV diastolic pressures,low cardiac output, and systolic hypotension. LV dysfunction (e.g. inferiormyocardial infarction) increases RV afterload, which can further reduce cardiacoutput.1. RA waveform:

� Pressure is elevated� Y descent is impaired� Failure of RA pressure to decrease with inspiration (Kussmaul’s sign) hasbeen shown in small studies to be sensitive and specific for RV infarction


798

Figure 19.4 RA pressure tracing in a patient with RA and RV infarction. Note the “M” configuration.

� Comparison of X and Y descents approximates degree of RA impairment.i. RV infarction with intact RA function—RA pressure tracing demon-strates increased A wave, steep X descent, and impaired Y descent (“W”pattern; Figure 19.3).ii. Combined RA and RV infarction—RA pressure tracing demon-strates decreased A wave and impaired X and Y descents (“M” pattern;Figure 19.4).

2. RV waveform:� Elevated diastolic pressures� Rapid rise in diastolic pressures� Slow upstroke in systole� Diminished peak pressure� Delayed relaxation� In severe, abrupt RV dilation, RV and LV diastolic pressures are elevatedand equalized and demonstrate “dip and plateau” configuration (similarto constrictive pericarditis). Y descent is blunted unlike the accentuated Ydescent seen in constrictive pericarditis.

3. Aortic and pulmonary artery waveform:� Relative hypotension� Pulse pressure is reduced� Pulsus paradoxus maybe present

Diagnosis of RV infarction with hemodynamics

Various criteria have been advanced to aid in the diagnosis of RV infarction1. Both a mean RA pressure of 10 mm Hg or more and an RA/pulmonaryartery wedge pressure ratio of 0.8 or more. Volume loading may be usedto unmask these findings. An abnormal RV radionuclide angiogram was


observed in 12 of 13 patients with these hemodynamic measurements in astudy of 53 patients with acute inferior myocardial infarction [2].2. RA pressure greater than 10 mm Hg and within 1–5 mm Hg of thepulmonary–capillary wedge pressure had a sensitivity of 73% and a speci-ficity of 100% in identifying RV infarction in 60 patients with acute myocardialinfarction (22 with RV involvement) [3].3. A severe noncompliant pattern, defined as a Y descent deeper than the Xdescent in RA pressure tracing, was present in 55% patients with, and 3% ofpatients without, RV infarction (sensitivity of 54.5% and specificity of 97.4%)in 60 patients with acute myocardial infarction [3].4. RA > pulmonary capillary wedge pressure (PCWP). The sensitivity andspecificity of this criteria was 45% and 100% respectively in a study of 60patients with acute myocardial infarction [3].

Management

Management of RV infarction can be summarized as follows: early recognition+ reperfusion + volume expansion ± dobutamine. The cornerstone of treat-ment of myocardial infarction is reperfusion, either via balloon angioplasty orthrombolytic therapy. In this section we will concentrate on the hemodynamicmanagement of patients with RV infarction. Aggressive fluid resuscitation isvital to restore preload and raise RV filling pressure in order to maximize car-diac output. This can require several liters of saline. Once pulmonary wedgepressure increases above 15 mm Hg, further fluid expansion is unlikely toimprove hemodynamics as further dilation of the RV compromises LV filling.

Once RV preload is optimized, dobutamine may improve RV function andcardiac output. In one small study of patients with RV infarction, volumeloading increased RA and PCWP pressure but did not improve cardiac index.Addition of dobutamine, but not nitroprusside, increased cardiac index (byan average of 35%) and RV ejection fraction [4]. Other therapeutic optionsto support blood pressure include intraaortic balloon pumping and cardiacpacing to maintain AV synchrony in the setting of advanced AV block.

Hemodynamic principles of RV infarction

� RV systolic and diastole function is impaired.� In the setting of a large RV infarction, the RV becomes merely a conduit (anda stiff one at that) from the systemic veins to the pulmonary circulation.� In the setting of a large RV infarction, the difference between RA and LApressure becomes the driving force between the systemic veins and the LV. RAtransport is important in maintaining LV preload and cardiac output.� Dilation of RV can impair LV function.� Abrupt dilation of RV can give rise to a “pseudo” constrictive pericarditisappearance to hemodynamics including a dip and plateau configuration in RVtracing and a positive Kussmaul’s sign.


� Cardiac output is reduced.� Systemic hypotension and reduced pulse pressure are common.

References

1 Goldstein JA, Barzilai B, Rosamond TL, Eisenberg PR, Jaffe AS. Determinants of hemody-namic compromise with severe right ventricular infarction. Circulation 1990;82:359–368.

2 Dell’Italia LJ, Starling MR, Crawford MH, Boros BL, Chaudhuri TK, O’Rourke RA. Rightventricular infarction: identification by hemodynamic measurements before and after vol-ume loading and correlation with noninvasive techniques. J Am Coll Cardiol 1984;4:931–939.

3 Lopez-Sendon J, Coma-Canella I, Gamallo C. Sensitivity and specificity of hemodynamiccriteria in the diagnosis of acute right ventricular infarction. Circulation 1981;64:515–525.

4 Dell’Italia LJ, Starling MR, Blumhardt R, Lasher JC, O’Rourke RA. Comparative effectsof volume loading, dobutamine, and nitroprusside in patients with predominant rightventricular infarction. Circulation 1985;72:1327–1335.

CHAPTER 20

Pulmonary hypertension

Daniel Fox, David P. McLaughlin, George A. Stouffer

Introduction

The normal pulmonary circulation consists of the pulmonary arteries, whichcarry deoxygenated blood from the right ventricle to the capillary beds of therespiratory parenchyma, and the pulmonary veins, which deposit the newlyoxygenated blood into the left atrium (Figure 20.1). This circulation is charac-terized by high blood flow, low pressure, and low resistance. The normal adultpulmonary vascular bed is highly distensible and capable of accommodatinglarge increases in blood flow with minimal elevations of pressure. Normal peaksystolic pressures in the pulmonary arteries range from 18 to 25 mm Hg andend-diastolic pressures in the pulmonary arteries range from 6 to 10 mm Hgwith a normal mean pressure of 10–16 mm Hg.

Pulmonary hypertension may occur in response to a multitude of mech-anisms, divided into primary and secondary hypertensive states based ondisease etiology. Primary pulmonary hypertension is a rare disease charac-terized by elevated pulmonary artery pressure with no apparent cause. It hasbeen reported in patients of nearly all demographics, although two-thirds ofcases are in women with typical age of onset of 36 years. The pathophysiol-ogy of primary pulmonary hypertension is poorly understood but characteris-tic pathology includes vascular scarring, endothelial dysfunction, and intimaland medial proliferation. The diagnosis is usually made after excluding otherknown causes of pulmonary hypertension.

The vast majority of patients with elevated pulmonary pressures have sec-ondary pulmonary hypertension. The most common causes of secondary pul-monary hypertension include left-sided heart failure, valvular disease, corpumonale, and thromboembolism. Less common etiologies include systemicsclerosis, sarcoidosis, congenital heart defects with left to right shunts (e.g.ASD, VSD, PDA), hepatopulmonary hypertension, HIV infection, and pul-monary venoocclusive disease. Secondary pulmonary artery hypertension canbe divided into causes due to left atrial hypertension and causes independentof left atrial pressure (Table 20.1).

The most important and prevalent cause of pulmonary hypertension is leftventricular heart failure. Virtually any disease state that causes left-sidedheart failure can also cause secondary pulmonary hypertension. Cor pul-monale, another cause of secondary pulmonary hypertension, results from

223


Trachea

Left bronchus

Aorta

Left superior pulmonaryvein

Left inferior pulmonaryvein

Right bronchus

Superior vena cava

Right pulmonary vein

Right inferiorpulmonary vein

Inferior vena cava

Left pulmonary artery

Right pulmonary artery

Left ventricle

Left atrium

Right ventricle

Right atrium

Figure 20.1 Pulmonary arteries and veins.

Table 20.1 Causes of pulmonary artery hypertension withnormal left atrial pressure.

• Pulmonary venoocclusive disease• Parenchymal lung disease (cor pulmonale)• Chronic thromboembolic disease• Hepatopulmonary hypertension• Congenital heart diseases with systemic to

pulmonary shunts (e.g. ASD, VSD, PDA)• Collagen-vascular disease (e.g. CREST)• Sleep disordered breathing• HIV infection• Sarcoidosis• Idiopathic (primary pulmonary hypertension)• Toxins (e.g. crack cocaine)• Drugs (e.g. “fen-phen” - dexfenfluramine and phentermine)• Schistosomiasis• Sickle cell disease

pathology within the lung parenchyma. Chronic hypoxia contributes to pul-monary hypertension (a feature of the pulmonary vasculature is its ability toconstrict in conditions of low oxygen concentration) and recent studies havesuggested that other factors (e.g. chronic hypercapnia) in patients with lungdisease may also contribute. An oxygen tension in arterial blood of <60 mm Hgand/or a carbon dioxide tension in arterial blood >50 mm Hg are thought tobe thresholds for the development of pulmonary hypertension in patients withchronic obstructive pulmonary disease (COPD). Pulmonary hypertension canalso result from chronic thromboembolic disease. In a subset of patients withthis pathologic condition, the thromboemboli become lodged in the pulmonaryvasculature effectively increasing resistance and pressure. If left untreated, thisincreased pressure may lead to a remodeling of the vasculature within the pul-monary circulation resulting in increased pulmonary vascular resistance.

Chapter 20 Pulmonary hypertension 225

Also worthy of mention when considering causes of pulmonary hyperten-sion is a rare condition termed Eisenmenger’s syndrome. Eisenmenger’s syn-drome results from longstanding, untreated congenital heart malformationsthat allow left to right intracardiac shunting. The increased pulmonary bloodflow results in elevated pulmonary resistance and pressure. Eventually, resis-tance in the pulmonary arteries leads to elevation in right heart pressures suchthat right-to-left shunting occurs within the heart. Patients with Eisenmenger’sphysiology develop cyanosis and hypoxemia.

Hemodynamic changes associated withpulmonary hypertension

The hemodynamic hallmark of pulmonary hypertension is elevated pul-monary artery pressures but there is no clear consensus as to what pressurelevels constitute pulmonary hypertension. Most commonly, pulmonary hyper-tension is defined as a mean pulmonary artery pressure (MPAP) of >25 mm Hgalthough in various studies, MPAP >18 mm Hg, MPAP >20 mm Hg, a systolicPAP >30 mm Hg at rest, and/or MPAP >30 mm Hg on exercise have been used[1–3].

Pulmonary artery pressures increase in response to increases in left atrialpressure, pulmonary vascular resistance or cardiac output, although the nor-mal pulmonary vascular bed is highly distensible and able to accommodatelarge blood flows with minimal changes in pulmonary artery pressures. Themagnitude of pulmonary hypertension can provide clues concerning the eti-ology. Mean pulmonary artery pressures ≥45 mm Hg are generally found inprimary pulmonary hypertension or chronic pulmonary thromboembolismwhereas lesser elevations are more commonly found in patients with lungdisease or elevated left atrial pressure.

Associated hemodynamic changes in patients with elevated pulmonarypressures depend on the extent of the pulmonary hypertension. The normalRV is a thin-walled, highly compliant structure that can accommodate acuteincreases in volumes at physiological pressures (e.g. during exercise) but isnot designed to overcome acute increases in afterload. Thus, acute RV dilationand failure can accompany sudden increases in pulmonary pressure (e.g. withacute pulmonary embolus). In chronic pulmonary hypertension, RV preloadincreases to enable the generation of increased systolic pressures via the Frank–Starling mechanism [4]. RV wall thickness increases (i.e. hypertrophy) to nor-malize wall stress and myocardial oxygen consumption. In moderate casesof pulmonary hypertension, hemodynamic changes include increased rightventricular end-diastolic pressure and increased right atrial pressure withprominent A waves (Table 20.2). In more severe cases and especially if signifi-cant tricuspid regurgitation is present, right atrial pressure will be further ele-vated and prominent V waves may be present on right atrial pressure tracings(Figure 20.2). In severe cases of tricuspid regurgitation, RA pressure tracingstake on characteristics similar to ventricular pressure tracings (see Chapter 12).


Table 20.2 Hemodynamic findings in pulmonaryhypertension. • ↑ Pulmonary artery pressure

• ↑ Right ventricular end-diastolic pressure• ↑ Right atrial pressure• Prominent A and V waves

An example of hemodynamic tracings in a patient with Eisenmenger’ssyndrome is shown in Figure 20.3. These are from a 42-year-old male whopresented with complaints of exertional dyspnea and near syncope. A mem-branous ventricular septal defect had been diagnosed in early childhood buthad not been repaired. Physical exam revealed central cyanosis and a loudpulmonary component of the second heart sound. The hemodynamic tracingsshow severe pulmonary hypertension with equalization of pulmonary arteryand systemic systolic pressures.

Hemodynamic changes detected by physical exam

The most common symptom attributable to pulmonary hypertension is exer-tional dyspnea. Other common symptoms include chest pain similar to anginapectoris that may represent RV ischemia, fatigue, syncope, and peripheraledema. On physical exam, increased jugular venous pressure with visuallyprominent A waves are typical. A left parasternal lift produced by the impulseof the hypertrophied high-pressure right ventricle is sometimes palpable. Onauscultation, patients with pulmonary hypertension will have an increasedpulmonic component of the second heart sound and may have a right-sidedS4 heart sound and/or an early systolic ejection click due to sudden interrup-tion of pulmonary valve opening. A midsystolic ejection murmur caused byturbulent transvalvular pulmonary flow may be audible.

The Graham Steell murmur of pulmonary regurgitation, an early diastolicdecrescendo murmur at the left sternal border is not uncommon in advancedcases. Other findings in patients with severe pulmonary hypertension includefindings associated with tricuspid regurgitation (a holosystolic murmur thataugments with inspiration and prominent V waves in the jugular veins) andright ventricular failure (hepatojugular reflux, a pulsatile liver, a right ventric-ular S3 gallop, and peripheral edema).

Two-dimensional echocardiography in pulmonaryhypertension

Assessment of the tricuspid regurgitation jet velocity by 2D echocardiographyin patients with pulmonary hypertension is a useful noninvasive measure toestimate pulmonary artery systolic pressure. The velocity of the TR jet enablesestimation of the difference between right ventricular systolic pressure andright atrial pressure via use of a modified Bernoulli equation (Figure 20.4). Pul-monary artery systolic pressure can then be estimated by adding this pressure


1 mV

1 mV

1 mV

1 mV

1 mV

1 mV

1 mV

1 mV

1 mV

1 mV

1 mV

1 mV

s

ss s

ss

s s

ss

d

d

dd

d

d

100

50v

av

v

v v v v va

v vv

v

v

vv v

v

vv

a

a

0

(a)

(b)

(c)

Figure 20.2 Pressure tracings in patients with pulmonary hypertension. Panel (a) showspulmonary artery pressure waveform on a 100 mm Hg scale. Note peak systolic pressure of80 mm Hg and mean pressures of 55–60 mm Hg. Panel (b) shows pressures obtained via aballoon-tipped pulmonary artery catheter before and after balloon let-down (i.e. pulmonarycapillary wedge pressure to pulmonary artery pressure). Notice the significant change in phasicmovement and rise from mean wedge pressure to mean pulmonary artery pressure. Panel (c)shows right atrial hemodynamic tracing in a patient with cor pulmonale. Note the elevated meanright atrial pressure and prominent A waves due to a noncompliant, hypertrophied right ventricle.


1mV

1mV

1mV

1mV

1mV1mV

1mV1mV

1mV1mV

1mV1mV

1mV1mV

1mV1mV

r r r r r r r r r r r r r r

s

dd

1 2 3 4 5 6 7 8 9SERIES IV;WITH MEDICALS

d dd

d d d d dd d

dd

ss

s ss s

s

s s s s ss

I

II

V5

INormal Sinus Rhyum Diagnostic

DiagnosticNormal Sinus Rhyum

II

V5

200

100

200

100

0

r r r r r r r r r r r r r r r r r r r

e e e e e ee e e e ed d

d d d dd d

d d d d dd d dd

025 mm/s >

[-19-]

r r r r r r r r r r r r r r r r r r

sss ss s s

s

ss

s

s

ss

s

s

s

ss

s

ss

s

s

ss

s

ss

s s

s

ss s

s

ss ss

dd dddd dd d

d dd dd dd dddd d

ddd

dd dd dd dd d

d

ss ssss ss

ss ss ss ssss s ss ss ss ss

ssss

s

(a)

(b)

(c)

Figure 20.3 Pressure tracings from a patient with Eisenmenger’s syndrome. Panel (a) showspulmonary artery pressures on 200 scale. Peak systolic and mean pressures of 120 mm Hg and82 mm Hg indicative of severe pulmonary hypertension. Panel (b) shows simultaneous leftventricular and PA pressures. Panel (c) shows simultaneous measurements of right and leftventricular pressures. At this point in time, the pressures are equal in the two ventricles.


(a)

(b) Calculation of TR Jet velocity from 2D echo

RAP + 4v 2 = Pulmonary artery systolic pressure RAP = Right atrial pressure = 5–10 mm Hg

v = velocity of tricuspid regurgitation (TR jet,) ,

Figure 20.4 Doppler echocardiography showing an example of pulse wave Doppler obtainedat the level of the tricuspid valve (a) and then formulas used in estimating pulmonary arterysystolic pressure (b).

difference to right atrial pressure. The right atrial pressure that is used to obtainpulmonary artery systolic pressure can be either a standardized value (usually5–10 mm Hg) or an estimated value based on echocardiographic characteris-tics of the inferior vena cava, or jugular venous pulse on physical examination.These calculations assume the absence of pulmonic stenosis, which can beverified using echocardiography.

Knowing systolic pulmonary artery pressure, it may be possible to estimatemean pulmonary artery pressure. In a study of 31 patients, Chemla et al. [5]found that mean pulmonary artery pressure (MPAP) and systolic pulmonaryartery pressure (SPAP) were related by the following formula: Mean PAP =0.61 SPAP + 2 mm Hg.

Pulmonary vascular resistance

Pulmonary vascular resistance (PVR) can be calculated knowing the transpul-monary pressure gradient and cardiac output by using the following equation:

PVR = (MPAP − PCWP)/CO


Table 20.3 Pressures in a patient with severe systolic LVdysfunction at baseline and during nitroprusside infusion.

Baseline Nipride infusion

Mean PCWP 30 24PA (mean) 76/40 (50) 64/36 (44)Cardiac output 2.5 3.7PVR (Woods units) 8 5.4

PVR is generally expressed in Wood units (1 WU = 1 mm Hg min/L =80 dyne s/cm5).

PVR is composed of two components; the fixed component that cannotchange acutely and which involves pathological changes to the pulmonaryvasculature and a dynamic component that results from vasoconstriction. ThePVR is mainly related to the geometry of small distal resistive pulmonary arte-rioles and according to Poiseuille’s law, PVR is inversely related to the fourthpower of arterial radius.

The relative magnitude of these two components is critically importantin patients undergoing evaluation for cardiac transplantation because donorhearts do not have hypertrophied right ventricles and are unable to pumpagainst significant pulmonary resistance for any length of time. Thus if thereis a large component of fixed resistance, the cardiac allograft will fail becausethe donor right ventricle will be unable to acutely increase the contractile forcenecessary to maintain cardiac output. On the other hand, if the PVR is highbut consists primarily of a dynamic component, then the patient’s pulmonaryresistance will drop dramatically following transplantation because left heartpressures will be less in the donor heart. To determine relative contributions offixed and dynamic components, an agent that decreases the dynamic compo-nent of PVR (e.g. nitroprusside, prostaglandin E1, nitric oxide) is given to thepatient in the catheterization laboratory and the measurements are repeated.

The contribution of reversible and fixed components of pulmonary vas-cular resistance can best be illustrated by a case presentation. This patientis a 49-year-old male who suffered from endstage ischemic cardiomyopathy.He had had three myocardial infarctions and a coronary artery bypass graftoperation in the past. He presented with worsening shortness of breath andorthopnea. Physical exam and chest X-ray were consistent with fluid over-load and he received aggressive treatment for congestive heart failure. He wasreferred for cardiac catheterization as part of an evaluation for cardiac trans-plantation. Left ventriculography revealed severely depressed function withejection fraction estimated at approximately 10%. Selective coronary angiogra-phy revealed severe native coronary artery disease and occlusion of all saphe-nous vein grafts. A left internal mammary conduit anastamosed to his leftanterior descending coronary artery was patent.

His initial hemodynamic findings are shown in Table 20.3. Note that despiteaggressive diuresis, his intracardiac pressures remain markedly elevated. His


pulmonary artery mean pressure is 50 mm Hg and his pulmonary capil-lary wedge pressure is 30 mm Hg. At a cardiac output of 2.5 L/min, thepulmonary vascular resistance is 8 Wood units. During nitroprusside infu-sion, this patient’s pulmonary artery and PCWP decreased and his cardiacoutput increased. His PVR also decreased however, there remained a large,fixed component, which would preclude successful transplantation.

Some investigators have recommended using pulmonary vascular resistanceindex (PVRI) rather than PVR to determine whether an individual qualifiesfor heart transplantation. PVRI is calculated using cardiac index rather thancardiac output. This approach avoids underestimating PVR in patients withlarge BMI.

References

1 Galie N, Torbicki A, Barst R, et al, for the Task Force on Diagnosis and Treatment of Pul-monary Arterial Hypertension of the European Society of Cardiology. Guidelines on diag-nosis and treatment of pulmonary arterial hypertension. Eur Heart J 2004;25:2243–2278.

2 Rubin LJ. Diagnosis and management of pulmonary arterial hypertension: ACCP evidence-based clinical practice guidelines. Chest 2004;126:7S–10S.

3 Chemla D, Castelain V, Herve P, Lecarpentier Y, Brimioulle S. Haemodynamic evaluationof pulmonary hypertension. Eur Respir J 2002;20:1314–1331.

4 Chin KM, Kim NH, Rubin LJ. The right ventricle in pulmonary hypertension. Coron ArteryDis 2005;16:13–18.

5 Chemla D, Castelain V, Humbert M, et al. New formula for predicting mean pulmonaryartery pressure using systolic pulmonary artery pressure. Chest 2004;126:1313–1317.

CHAPTER 21

Coronary hemodynamics

David P. McLaughlin, Samuel S. Wu, George A. Stouffer

Basic principles of coronary blood flow

The coronary arteries are the first vessels to branch off the aorta and throughthem the heart receives at rest about 5% of the cardiac output, or 250 mL/min.

The heart has the highest oxygen consumption per tissue mass of all humanorgans. Under basal conditions, myocardium extracts approximately 75% ofdelivered oxygen (the myocardium has a basal metabolic requirement which isapproximately 15–20 times that of resting skeletal muscle and approximatelyequal to that of skeletal muscle under severe acidotic conditions). The hearthas the highest AVO2 difference of any major organ (10–13 mL/100 mL) andthe oxygen saturation in the coronary sinus is one of the lowest in the body(Figure 21.1).

Because there is little room for increased oxygen extraction and the heart hasminimal capacity for anaerobic metabolism, increased metabolic demands ofthe heart are met primarily via increases in coronary blood flow (i.e. flow istightly coupled to oxygen demand). Coronary blood flow, in turn, is primarilycontrolled by changes in coronary resistance.

It is useful for purposes of understanding coronary blood flow to appreciatethat the coronary circulation is functionally a two-component model. Nor-mal epicardial coronary arteries provide little if any resistance to myocardialblood flow even at maximal flow and are commonly referred to as conductancevessels. In contrast, the microcirculation, composed of an extensive arteriolarnetwork, provides resistance to flow under normal resting conditions; they arecommonly referred to as resistance vessels.

Determinants of myocardial oxygen demands include preload, afterload,heart rate, contractility, and basal metabolic rate. Other than basal metabolicrate, these are factors that influence stroke volume. Systolic wall tensionuses approximately 30% of myocardial oxygen demand. Wall tension itselfis affected by intraventricular pressure, afterload, end-diastolic volume, andmyocardial wall thickness.

Regulation of coronary blood flow

Coronary blood flow can increase by 4- to 5-fold in the normal heart. In adultpatients with angiographically normal vessels and coronary artery disease risk

233


Glycolysis

Glucose-6-PO4

Pyruvic acid

Lactic acid CO2 + H2O

Aerobicmetabolism

Anaerobicmetabolism(only used

during extreme hypoxia)

O2 debt rarely

incurred

O2 extraction from perfusing blood nearly maximal

at rest

Increased work requires greater O2consumption which must be met by

increased blood flow

Cardiac muscle

Figure 21.1 Oxygen extraction in cardiac muscle.

Chapter 21 Coronary hemodynamics 235

factors coronary blood flow increased by 2.7 ± 0.6 fold with maximal coro-nary vasodilation [1]. The difference between resting and maximal coronaryflow is labeled coronary flow reserve (CFR). The maximal achievable coronaryblood flow can be achieved experimentally using any one of three independentmechanisms. Reactive hyperemic flow is that flow which occurs following atransient period of flow cessation. Metabolites of ischemia accumulate dur-ing interruption of flow and vasodilate the microcirculation. When flow isreestablished, the resistance of the microcirculation is markedly diminishedand coronary blood flow increases markedly. Maximal achievable flow canalso be estimated using exercise or another stimulus to increase flow. Lastly,maximal coronary flow can be approximated by intracoronary administrationof potent vasodilators such as adenosine.

Coronary blood flow is primarily controlled by release of local metabolitessuch as adenosine or nitric oxide. Hypoxia is a more potent coronary vasodila-tor than either hypercapnea or acidosis. Neural influences on coronary bloodflow are relatively minor. Sympathetic activation to the heart results in transientvasoconstriction followed by coronary vasodilation due to increased metabolicactivity. Parasympathetic stimulation of the heart directly stimulates coronaryvasodilation but this effect is modest.

Coronary blood flow is unique in that it primarily occurs during dias-tole because of systolic compression of myocardial arteriolars and capillar-ies by the contracting myocardium. Systolic compressional forces are muchgreater in the subendocardial layers than in the epicardial arteries. Flowin the left coronary artery has a greater diastolic predominance than theright coronary artery because the compressive forces of the right ventricle(underlying a portion of the right coronary artery) are less than those ofthe left ventricle. At least 85% of coronary flow in the left anterior descend-ing occurs in diastole whereas the right coronary flow is more or less equalin systole and diastole. The predominance of flow during diastole exacer-bates myocardial ischemia during tachycardia. With increased heart rates,oxygen supply is reduced (because diastole is shortened) while demandincreases.

The heart has the ability to maintain coronary blood flow in the presence ofvarying perfusion pressures (termed autoregulation). Autoregulation main-tains consistent coronary flow over a range of perfusion pressures from 60to 150 mm Hg (Figure 21.2) although there is evidence that autoregulation isexhausted at a higher pressure (approximately 70 mm Hg) in patients with leftventricular hypertrophy [2]. In the setting of maximum vasodilation of coro-nary resistance vessels, coronary blood flow is no longer autoregulated andvaries linearly with perfusion pressure. One example of an autoregulatorymechanism is the Bayliss phenomenon in which increased perfusion pressurecauses reflex vasoconstriction. The ability of autoregulation to maintain flowwhen perfusion pressures are decreased is especially important in the presenceof epicardial coronary stenoses.


500 D

A

R1

R2

400

300

200

100

00 50 100 150

Flow (mL/min)

Mean pressure (mm Hg)Figure 21.2 Coronary autoregulation.(Modified from A. Polese [2].)

Clinical measurement of coronary hemodynamicsin the cardiac catheterization laboratory---Dopplerand pressure wires

There are several ways to directly measure coronary blood flow but these aregenerally difficult, time consuming, and unable to measure rapid changes incoronary flow. More commonly, angioplasty wires with Doppler probes orpressure transducers are used to make clinical decisions and to study coronaryphysiology. The Doppler wire (an angioplasty wire with a Doppler transducerplaced near the tip) is used to measure coronary flow velocity at rest and thenunder conditions of maximal flow (coronary flow reserve). This technique haslargely been supplanted in clinical medicine by use of the pressure wire (anangioplasty wire with a pressure transducer near the tip), which measuresblood pressure distal to a stenosis. By comparing distal perfusion pressurewith aortic pressure under conditions of maximal flow, fractional flow reserve(FFR) can be calculated.

Measurement of coronary blood flow

1. Clearance methods—These techniques involve introducing an inert gas (usu-ally nitrous oxide) into the circulation via the lungs and following the pro-gressive saturation of cardiac tissue. The increases in the systemic arterial and


coronary sinus concentrations of indicator are measured over the time untilarteriovenous difference reaches zero. The reciprocal of this time reflects theblood flow in milliliters per minute per 100 g of tissue.2. Thermodilution—As originally implemented in the coronary circulation,this technique requires placement of a catheter into the coronary sinus andthen a continuous infusion of cold saline through a lumen near the tip ata constant rate. The temperature of the blood at a site several centimetersback from the tip of the catheter is measured with a thermistor. The methoduses the form of the Fick equation dealing with continuous (rather thanbolus) infusion of indicator. More recently, Pijls et al. have validated a tech-nique to measure coronary blood flow whereby a bolus injection of saline isgiven into a coronary artery and temperature change distal in the artery ismeasured [12].3. Flowmeter techniques—Electromagnetic and Doppler flowmeters have beenused at surgery, where they are best suited for measurement of the flow in veingrafts. The native coronary vessels are never dissected for the sole purpose ofplacing a flowmeter.4. Doppler wire—See text for further description.

Doppler wire and coronary flow reserve

The physiologic principle underlying the use of CFR is that clinically significantatherosclerosis will impair coronary blood flow responses during hyperemicstresses. Coronary blood flow under resting conditions is generally 15–20%of maximal blood flow in patients with normal coronary arteries and is notaltered by gender or age. As atherosclerosis progresses, maximal coronaryblood flow is initially reduced followed in the severe stages by a reductionin resting coronary blood flow. CFR, the ratio of maximal blood flow to basalblood flow, will decrease with progressive obstruction of a coronary arterylumen by atherosclerosis.

In humans, CFR was originally measured with probes placed during cardiacsurgery and then subsequently with a pulsed Doppler catheter. There wereproblems with widespread use of both these techniques as epicardial coronaryartery probes were limited to patients undergoing surgery and the size ofthe Doppler catheter prevented placement distal to coronary stenoses. Both ofthese problems have been overcome with the development of a guidewire witha 12 MHz piezoelectric transducer mounted on its tip. To determine CFR, thesewires are placed in the coronary artery of interest distal to the lesion and phasicspectral blood flow velocity is measured (Figure 21.3). It is assumed that thereare minimal changes in coronary diameter thus enabling velocity to be used inplace of flow (flow = velocity × area). Measurements of blood velocity underbasal conditions are taken and then the patient is given a hyperemic stimulus,velocity is remeasured and CFR is calculated. Adenosine is commonly used to


Figure 21.3 Measurement of CFR in the left circumflex. Doppler FloWire measurement followingthe administration of adenosine. The top window represents a real time measurement of coronaryvelocity. The bottom left window is a “baseline” measurement and the bottom right window isvelocity as measured after administration of adenosine. Abbreviations: S, systole; D, diastole;APV, time-averaged distal peak velocity; DSVR, diastolic/systolic flow–velocity ratio; CFR,coronary flow reserve.

induce hyperemia as it causes less hemodynamic and ECG changes than doespapaverine [3].

Though the Doppler wire has great theoretical advantages it does not enjoywidespread use because of several important limitations. First, conditionsother than atherosclerosis can affect CFR. These include factors that raise basalcoronary blood flow (e.g. fever, hypoxia, tachycardia, anemia, or ventricularhypertrophy) and factors that impair vasodilatory responses of the microvas-culature (e.g. ventricular hypertrophy or diabetes mellitus). Second, accuratemeasurements are dependent upon correct positioning of the Doppler flowwire. The transducer should be pointing away from the vessel wall and intothe flow stream to avoid vessel wall artifacts. Gray scale signal amplitude andpeak velocity can be used as indicators of proper positioning. Third, there is alack of consensus on what value of CFR is consistent with a hemodynamicallysignificant lesion. In various clinical studies, CFR cutoff values between 1.6and 2.5 were used to determine ischemia-causing lesions.

Since CFR cannot discriminate between epicardial and microvascular lesionsthe concept of “relative CFR” was developed. This approach requires thatCFR be measured in a coronary artery without epicardial disease in order to


interpret the value of CFR in an artery with epicardial disease. If CFR is abnor-mal in the artery without disease, this result implies an impaired microvas-culature. The use of relative coronary flow reserve (rCFR) has several caveatsincluding the requirement for a vessel without significant epicardial diseaseand the assumption that microvasculature function is consistent across dif-ferent vascular beds (an obvious problem in the case of prior myocardialinfarction).

Pressure wire and fractional flow reserve

The development of a pressure transducer mounted on an angioplastyguidewire allows measurement of the pressure distal to a coronary stenosis.The concept of using pressure gradient as a technique to assess stenosis severityhas existed since the early days of endovascular intervention. Early attemptsto measure pressures distal to a stenosis were hampered by the use of fluidfilled tubes, which, because of their size, increased the translesional gradi-ent. The advent of a pressure transducer mounted on a 0.014-inch angioplastywire overcame these difficulties and enabled the introduction of the conceptof pressure-derived FFR.

The concept of pressure as a surrogate for flow requires review of Ohm’slaw:

Flow = pressure/resistance

FFR is defined as maximum myocardial blood flow in the presence of a steno-sis divided by the theoretical maximum flow in a normal vessel (i.e. theabsence of any stenosis). Under maximum arteriolar vasodilatation, the resis-tance imposed by the normal myocardial bed is minimal and blood flow isproportional to driving pressure. Thus, FFR represents that fraction of normalmaximum flow that is achievable in the presence of epicardial coronary steno-sis. Since myocardial ischemia occurs in a patient when maximum blood flowis insufficient to meet myocardial demands, FFR aims to measure perfusionpressure during maximum coronary flow to establish the physiological signif-icance of a stenosis. FFR is independent of changes in systemic blood pressure,heart rate, or myocardial contractility.

Maximal flow can be achieved with various vasodilators but for practicalpurposes in most catheterization labs intracoronary adenosine is used [4].Figure 21.4 shows pressure tracings obtained in our laboratory in a 63-year-oldmale with chest pain and an intermediate lesion on angiography. Under basalconditions, there is a 7 mm Hg difference in mean pressures between the distalleft anterior descending coronary artery (measured with the pressure wire) andthe aorta (measured with the guide catheter) and the FFR was 0.94. Followingadministration of 60 �g of adenosine, distal perfusion pressure decreases, thedifference in mean pressures increased and the FFR decreased to 0.58. FFRwith hyperemia will decrease in proportion to the hemodynamic severity ofthe coronary stenosis (Figure 21.5).


1 mV

1 mV

1 mV

1 mV

1 mV

1 mV

1 mV

1 mV

1 mV

1 mV

1 mV

1 mV

r

a

d d d d d d d

a

aa

s s s ss

ss

v v vv

vv

v

r r r r r r

200

200

100

0 1 2 3 4 5 6

100

0

r

s

d d d d dd d d

a aa a

a

s s s s ss s

r r r r

vvvv

(a)

(b)

Pressure wire

Aorta

Aorta

Pressure wire

Figure 21.4 Measurement of FFR. Simultaneous measurement of pressure in aorta and distalLAD (labeled as Pressure wire) at rest (a) and following intracoronary administration of adenosine(b).


Normal FFR

150

100

50

Mildly decreasedFFR

150 150

100

50

100

50

Moderatelydecreased FFR

Severely decreasedFFR

Figure 21.5 Examples of FFR measurements in four patients with coronary disease of differenthemodynamic severity. Simultaneous pressure recordings, following intracoronary administrationof adenosine, from the aorta and distal coronary artery in four different patients showing normalFFR and mildly, moderately, and severely decreased FFR.

Clinical studies of FFR

1. Determining the hemodynamic significance of angiographically intermediatelesions: The use of angiography alone has significant limitations when tryingto determine the hemodynamic significance of an intermediate lesion (40–80%). In a study of 45 patients with moderate coronary stenosis and chest painof uncertain origin, Pijls and colleagues showed that a cutoff value of 0.75detected ischemia as indicated by noninvasive testing modalities such as per-fusion scintigraphy, stress echocardiography, and bicycle exercise testing witha sensitivity of 88%, specificity of 100%, and diagnostic accuracy of 93% [5].To further determine the correlation of FFR with ischemia, these investigatorsassessed FFR in patients with positive stress nuclear studies rendered negativeby coronary intervention. Before percutaneous coronary intervention (PCI) allpatients had an FFR < 0.74 and post-PCI all patients had FFR > 0.75. This studyand others has led to the generally accepted cutoff for ischemia of FFR<0.75 [6].

The DEFER trial assessed the ability of FFR to predict freedom from adversecardiac events in patients with moderate coronary artery disease referred forcoronary intervention [7]. Patients with moderate stenosis and FFR > 0.75 wererandomized to PCI versus medical therapy. The patients with FFR > 0.75 whowere randomized to PCI had an event rate greater than those patients managedmedically. Moreover, those patients with FFR > 0.75 whose intervention wasdeferred had a very low event rate in follow-up.


Figure 21.6 Angiograms, IVUS, and FFR measurements from the illustrative case.

2. Determining the success of PCI: A multicenter registry found that on mul-tivariate analysis, FFR determined immediately after stenting was the mostsignificant independent variable related to future events. The event rate at6 months was 4.9%, 6.2%, 20.3%, and 29.5% in patients with FFR > 0.95, 0.90 <

FFR < 0.95, FFR < 0.90 or FFR < 0.80, respectively [8].3. Determining the significance of left main coronary lesions: In a study of 54patients with equivocal lesions of the left main, medical instead of surgicaltreatment was used in 24 patients with FFRs greater than 0.75, while coronarybypass surgery was performed in the remaining 30 patients, who had FFR


values less than 0.75. Mean follow-up was 29 months. The survival rates ofthe patients in the medical treatment and surgical groups were 100% and 97%,respectively [9].4. Determining the significance of multiple stenoses in the same coronary artery:FFR can be used to “map” the hemodynamic effects of multiple stenoses inthe same vessel. When using this approach it is important to realize that onestenosis will affect the FFR of the other when assessed in tandem. There is aninteraction between stenoses and FFR of each lesion independently cannot becalculated by the equation for isolated stenoses (Pd/Pa during hyperemia) butcan be predicted by more complex equations [10]. This phenomenon is bestexplained by the limitation of the reactive hyperemia imposed by the treatedstenosis when assessed in the baseline state. In patients with more diffusedisease a gradual pullback may reveal no focal area of pressure change. Thistype of diffuse disease is unlikely to benefit from PCI.

Case study

A 61-year-old female with diabetes and hypertension reported the recent onsetof exertional angina. An exercise treadmill test was performed during whichshe developed chest pain and 2 mm of ST depression 2 minutes into the Bruceprotocol. At angiography, her coronary arteries were unremarkable except for a60% lesion involving the ostium of the left anterior descending coronary artery(Figure 21.6a). Because of the intermediate nature of the lesion, FFR analysiswas performed using the pressure wire. FFR was 0.94 at baseline and 0.73 fol-lowing intracoronary administration of 60 �g of adenosine (Figure 21.6b). QCAof the proximal LAD revealed a size of 3.3 mm and therefore a 3.0 mm Cypherdrug-eluting stent was placed at 12 atmospheres. Angiography revealed sat-isfactory result (Figure 21.6c); however, FFR was 0.82 following intracoronaryadministration of 60 �g of adenosine. Intravascular ultrasound showed that thevessel diameter was much larger than it appeared on angiography and thatthe stent was under-deployed (Figure 21.6d). The Cypher stent was dilatedto 3.8 mm and angiography repeated (Figure 21.6e). Pressure wire analysisshowed that the FFR was now 0.98 at baseline and 0.95 following intracoro-nary administration of 60 �g of adenosine.

FFR limitations

FFR has several limitations. Although it is felt to be epicardial lesion-specific,application is critically dependent upon an intact microcirculation to respondto the hyperemic stimulus. As such FFR can be somewhat problematic in dis-ease states associated with abnormalities of the microcirculation (e.g. diabetesor ventricular hypertrophy) and some have advocated that a higher “cutoff”value for ischemia be used in these conditions. FFR should be used verycautiously in the acute phase of myocardial infarction but has been shown


to correlate with myocardial scintigraphy with a sensitivity and specificity ofapproximately 80% in the convalescent phase of myocardial infarction.

There are also technical considerations that can lead to incorrect results [11].

Definitions of coronary hemodynamic measurements used clinically

Coronary flow reserve (CFR)—Ratio of maximal coronary flow to basal coronaryflow.Relative Coronary flow reserve (rCFR)—Used to minimize the influence ofmicrovasculature on assessment of the hemodynamic severity of an epicar-dial stenosis. It is the ratio of CFR in the diseased vessel to CFR in a “normal”vessel (rCVR = CVRtarget/CVRreference).Fractional flow reserve (FFR)—Maximum myocardial blood flow in the presenceof a stenosis divided by the theoretical maximum flow in the absence of thestenosis.

References

1 Kern MJ, Bach RG, Mechem CJ, et al. Variations in normal coronary vasodilatory reservestratified by artery, gender, heart transplantation and coronary artery disease. J Am CollCardiol 1996;28:1154–1160.

2 Polese A, De Cesare N, Montorsi P, et al. Upward shift of the lower range of coronary flowautoregulation in hypertensive patients with hypertrophy of the left ventricle. Circulation1991;83:845–853.

3 Sonoda S, Takeuchi M, Nakashima Y, Kuroiwa A. Safety and optimal dose of intracoro-nary adenosine 5′-triphosphate for the measurement of coronary flow reserve. Am HeartJ 1998;135:621–627.

4 de Bruyne B, Pijls NH, Barbato E, et al. Intracoronary and intravenous adenosine 5′-triphosphate, adenosine, papaverine, and contrast medium to assess fractional flowreserve in humans. Circulation 2003;107:1877–1883.

5 Pijls NH, de Bruyne B, Peels K, et al. Measurement of fractional flow reserve to assess thefunctional severity of coronary-artery stenoses. N Engl J Med 1996;334:1703–1708.

6 Pijls NH, Van Gelder B, Van der Voort P, et al. Fractional flow reserve. A useful index toevaluate the influence of an epicardial coronary stenosis on myocardial blood flow [seecomments]. Circulation 1995;92:3183–3193.

7 Bech GJ, de Bruyne B, Bonnier HJ, et al. Long-term follow-up after deferral of percuta-neous transluminal coronary angioplasty of intermediate stenosis on the basis of coronarypressure measurement. J Am Coll Cardiol 1998;31:841–847.

8 Pijls NH, Klauss V, Siebert U, et al. Coronary pressure measurement after stenting predictsadverse events at follow-up: a multicenter registry. Circulation 2002;105:2950–2954.

9 Bech GJ, Droste H, Pijls NH, et al. Value of fractional flow reserve in making decisions aboutbypass surgery for equivocal left main coronary artery disease. Heart 2001;86:547–552.

10 Pijls NH, de Bruyne B, Bech GJ, et al. Coronary pressure measurement to assess the hemo-dynamic significance of serial stenoses within one coronary artery: validation in humans.Circulation 2000;102:2371–2377.

11 Pijls NH, Kern MJ, Yock PG, de Bruyne B. Practice and potential pitfalls of coronarypressure measurement. Catheter Cardiovasc Interv 2000;49:1–16.

12 Pijls NH, de Bruyne B, Smith L, et al. Coronary thermodilution to assess flow reserve:validation in humans. Circulation 2002;105:2482–2486.

CHAPTER 22

Hemodynamics of intra-aorticballoon counterpulsation

Richard A. Santa-Cruz

History and uses

The idea of using diastolic augmentation to treat left ventricular failure was firstproposed in the 1950s. The earliest method involved removing blood from thefemoral artery during systole and replacing this volume rapidly during dias-tole. In the early 1960s an experimental prototype of the intra-aortic balloonpump (IABP), in which inflation and deflation were triggered by the ECG, wasdeveloped. The initial use of IABP in clinical practice was in 1968. The earlydevices were 15 French in size and required surgical insertion and removal.Subsequent improvements included a dramatic reduction in size enabling per-cutaneous insertion.

Current indications for IABP include cardiogenic shock or left ventricu-lar failure, unstable ischemic syndromes, mechanical complications of acutemyocardial infarction, ischemic ventricular arrhythmias, severe mitral regur-gitation, stabilization of patients undergoing coronary artery bypass graftingwith high-risk anatomy (e.g. severe left main disease), failure to separate apatient from cardiopulmonary bypass, high-risk coronary intervention, or asa bridge to heart transplantation (Table 22.1).

Description

The IABP is a device that rapidly shuttles a gas into and out of a polyurethaneballoon placed in the patient’s descending aorta. The gases used for inflation areeither helium or carbon dioxide. Helium is most commonly used and has theadvantage of a lower density and therefore a better rapid diffusion coefficient.Carbon dioxide, on the other hand, has an increased solubility in blood andthereby reduces the potential consequences of gas embolization following aballoon rupture.

The IABP is placed with the distal tip below the origin of the left subclavianartery (a useful angiographic marker is the left main stem bronchus). Theproximal end should be positioned above the renal arteries. It is important tomatch the appropriate balloon size to patient size as mismatches can causeeither ineffective counterpulsation or possible mechanical trauma to the aorta.

245


Table 22.1 Indications and contraindications for IABP.

Indications Relative contraindications Absolute contraindications

Complicated acute myocardialinfarction

Severe peripheral vasculardisease

Patient refusal/lack ofconsent

Refractory unstable angina Active hemorrhage Moderate to severe aorticregurgitation

Cardiogenic shock (including rightventricular failure)

Severe thrombocytopenia

Mechanical complications of acutemyocardial infarction or trauma

Contraindication toanticoagulation

Severe coronary artery disease withhemodynamic compromise

Left main diseaseSupport of high risk coronary

interventionsInduction and weaning of

cardiopulmonary bypassBridge to cardiac transplantationRefractory ventricular arrhythmiasNon-cardiac surgery for high-risk

cardiovascular patientsSevere mitral regurgitation

An IABP produces hemodynamic effects through the cardiac cycle. Duringdiastole, the IABP inflates thereby displacing blood from the descending aorta.The balloon deflates immediately before systole resulting in decreased aorticimpedance (Figure 22.1).

Hemodynamic effects

The primary hemodynamic benefits of IABP treatment are diastolic pressureaugmentation and afterload reduction (Table 22.2). Diastolic augmentation canincrease coronary perfusion pressure, mean arterial blood pressure, and sys-temic perfusion although these effects have not been consistently found in allstudies. As shown in Figure 22.2, diastolic pressure augmentation by aorticcounterpulsation can be a major support of mean arterial blood pressure inpatients with cardiogenic shock. Afterload reduction can decrease left ventric-ular filling pressures and decrease myocardial oxygen demand. The decreasein afterload generally results in a decrease in left ventricular end-diastolicpressure (LVEDP) during counterpulsation, with preservation or increase ofthe left ventricular stroke volume and ejection fraction. Left ventricular workis decreased by a reduction in afterload, which results in lower myocardialoxygen demand. The decrease in systolic pressure is generally less than theincrease in diastolic pressure resulting in an increase in mean arterial bloodpressure. Because an IABP works via diastolic pressure augmentation andafterload reduction, the beneficial effects are dependent upon patient-related

Chapter 22 Intra-aortic balloon counterpulsation 247

Diastole Systole

(a) (b)

Figure 22.1 IABP inflation and deflation. During diastole, the IABP is inflated, increasing diastolicpressure thus augmenting flow not only into the coronary arteries but also the great vessels andthe renal arteries (a). Just prior to systole, the IABP is deflated, creating a void where the inflatedballoon was, thus increasing forward flow into the aorta and to the periphery (b).

Table 22.2 Hemodynamic effects of IABP therapy.

� Decrease left ventricular afterload� Decrease aortic SBP� Decrease pulmonary capillary wedge pressure� Decrease LV volume� Decrease LV end-diastolic pressure� Increase coronary blood flow� Increase cardiac output

factors such as cardiac output and blood pressure. Diastolic augmentationis limited in patients with poor cardiac output independent of how well theIABP is functioning. This is in contradistinction from cardiopulmonary bypassin which device function is independent of intrinsic cardiac activity.

Counterpulsation has many metabolic effects on both ischemic and nonis-chemic myocardium. In general, aortic counterpulsation results in decreasedmyocardial metabolic demands. Increased end-organ perfusion is probably animportant contributor to the overall improvement in patient’s status associatedwith aortic counterpulsation among those with hemodynamic compromise,such as in cardiogenic shock.


Normal sinus Monitor length:10 sec.

Figure 22.2 Aortic counterpulsation in a patient with cardiogenic shock. Left ventricle to aorticpullback in a patient with cardiogenic shock. The patient was a 48-year-old male who presentedwith an extensive anterior–lateral myocardial infarction. An IABP was placed via the left femoralartery. An end-hole catheter was inserted into the left ventricle and pressures were recorded asthe catheter was withdrawn into the aorta. Note the decreased left ventricular systolic pressureand increased left ventricular end-diastolic pressure. IABP diastolic augmentation is a majorcontributor to aortic pressure.

Intra-aortic balloon pump timing

To achieve optimal effects of counterpulsation, inflation and deflation need tobe correctly timed to the patient’s cardiac cycle. This can be accomplished byusing the patient’s ECG signal, the patient’s arterial waveform, or an intrinsicpump rate. The most common method of triggering the IABP is from the Rwave of the patient’s ECG signal. Upon identification of the R wave the balloondeflates. Balloon inflation is generally set to start automatically in the middleof the T wave. Tachyarrhythmias, cardiac pacemakers, and poor ECG signalsmay cause difficulties in obtaining synchronization when the ECG mode isused. In such cases the arterial waveform may be useful for triggering.

For optimal efficacy, the IABP should inflate during early diastole (after theaortic valve closes, which is identified by the dicrotic notch) and deflate justprior to systole (just before the aortic valve opens). IABP inflation and deflationtiming, as well as ratio of counterpulsations to cardiac cycles, can be adjustedto optimize performance (Figures 22.3 and 22.4). In general, optimal assistancefrom the IABP occurs when each cardiac cycle is augmented in a 1:1 ratio. Inpatients with tachycardia (e.g. rapid atrial fibrillation), a lower assist ratio (1:2or 1:3) can be used to allow appropriate gas shunting from the console to theballoon.

The most common cause of inadequate balloon counterpulsation is the inac-curate timing of inflation and deflation. Some errors associated with IABPtiming will simply result in poor hemodynamic response to IABP but othersare potentially dangerous for patients with already tenuous cardiovascularsituations. Understanding IABP pressure waveforms is critically important toproperly use this device. The timing errors can be divided into two groups:systolic errors (early and late inflation) and diastolic errors (early and latedeflation).


Diastolic augmentation

Unassisted systolic pressure

Assisted systolic pressure

Mean blood pressure

Unassisted diastolic pressureAssisted diastolic pressure

DN

Figure 22.3 Correct IABP timing and hemodynamic effects. Normal timing of the IABP (arrow)with inflation at the dicrotic notch (DN) and good diastolic augmentation which increases coronaryblood flow and increases mean blood pressure. Assisted systolic and end-diastolic pressures arelower than unassisted systolic and end-diastolic pressures.

DN

DN

(a)

(b)

Figure 22.4 Aortic pressure with and without IABP. Aortic pressure waveforms prior to (a) andafter initiation of IABP at 1:2 ratio (b). DN, dicrotic notch.

Early inflation

Inflation of the IABP prior to aortic valve closure will increase afterload as theheart must eject blood against an inflated balloon. This increases myocardialwall stress and LVEDP, thereby increasing LV work, reducing cardiac output,and potentially precipitating early closure of the aortic valve. On the hemo-dynamic tracing, a sharp or rapid rise in the augmented pressure will be seenprior to the dicrotic notch (Figure 22.5).



DN

Figure 22.5 Incorrect IABP timing—early inflation. Aortic pressure waveform demonstrating earlyinflation of the IABP (arrow) prior to the dicrotic notch (DN) and prior to the closure of the aorticvalve. There is a rapid rise in augmented diastolic pressure (∗) with the aortic valve still open,which causes a dramatic increase in afterload and reduces cardiac output.

Late inflation

Although probably the least dangerous error, late IABP inflation can precludethe patient from obtaining optimal hemodynamic support. Here, the IABPinflates too long after the aortic valve closes. This delay does not allow foradequate diastolic augmentation. There is a blunting or complete lack of thediastolic augmentation waveform (Figure 22.6). Associated with this is a lack ofend-diastolic pressure decrease (afterload reduction) and no increase in dias-tolic perfusion pressure.

Diastolic augmentationUnassisted systolic pressure

DN

Figure 22.6 Incorrect IABP timing—late inflation. Aortic pressure waveform demonstrating lateinflation of the IABP. The notable hemodynamic changes are the clearly visible dicrotic notch (DN)with the IABP inflation following (arrow). There is a prolonged dip or U wave (∗) then a blunted ordecreased diastolic augmentation.



Assisted and unassistedsystolic pressure

DN

Figure 22.7 Incorrect IABP timing—early deflation. Aortic pressure waveform demonstratingearly deflation of the IABP. Early deflation of the IABP in diastole does not affect the diastolicaugmentation but does prevent optimal end-diastolic pressure decrease. This is evident by aprolonged dip and U wave (∗) postaugmentation. There is no reduction in assisted systolicpressure, which is equal to the unassisted systolic pressure representing a lack of afterloadreduction.

Early deflation

Early deflation, similar to late inflation, is not a particularly dangerous sit-uation although the patient will not receive optimal balloon counterpulsa-tion. The hemodynamic tracing reveals a normal diastolic augmentation butthen a long drawn out U-shaped wave prior to the assisted systolic waveform(Figure 22.7). Notice that there is inadequate preload reduction and thereforeno change in assisted and unassisted systolic pressure. Simply prolonging theballoon inflation will quickly improve the hemodynamic support that the bal-loon pump can provide.


Unassisted systolic pressure

Assisted systolic pressure

Assisted end-diastolic pressure

Unassisted end-diastolic pressure

Figure 22.8 Incorrect IABP timing—late deflation. Aortic pressure waveform demonstrating latedeflation of the IABP. Late deflation of the IABP in diastole causes a dramatic increase in afterloadas systole begins with higher intra-aortic pressure. This causes late opening of the aortic valveand severely reduces the cardiac output. The pressure waveform reveals an assisted end-diastolicpressure that is higher than the unassisted end-diastolic pressure and a drastically reducedassisted systolic pressure.


Late deflation

Late IABP deflation is potentially the most critical timing error. If the balloonremains inflated when the left ventricle enters systole, the left ventricle is forcedto contract against the inflated balloon. The result can be acute hypotensionand cardiac arrest. The hemodynamic tracing reveals a long inflation time withan elevated end-diastolic pressure and a severely reduced assisted systolicpressure (Figure 22.8).

Conclusion

The IABP is a useful device in the management of patients suffering fromthe ill effects of acute and chronic cardiovascular disease. Understanding thecharacteristic hemodynamic tracings of balloon counterpulsation is critical forproper functioning and ensuring the maximal therapeutic benefit. In turn,recognizing the possible complication and pitfalls of IABP timing is crucial tominimizing potential adverse effects.

Review of IABP tracings

Match the tracing with the diagnosis. All tracings are taken with the IABP ona 1:2 ratio.a. Accurate timingb. Early balloon inflationc. Early balloon deflationd. Late balloon inflatione. Late balloon deflation

Tracing 1

Tracing 2


Tracing 3

Tracing 4

Tracing 5

Answers

1. Early balloon deflation2. Late balloon deflation3. Accurate timing4. Early balloon inflation5. Late balloon inflation

CHAPTER 23

Arrhythmias

Lukas Jantac, George A. Stouffer

Introduction

The cardiac rhythm is integral for development of physiologic pressure wave-forms. Disturbances in electric conduction can change pressure tracings andhence, the hemodynamic parameters. This chapter will describe changes thattake place as a result of some of the most common arrhythmias encounteredin the coronary care unit or cardiac catheterization laboratory. Although sinustachycardia and sinus bradycardia usually represent normal cardiac physi-ology, they will also be discussed in this chapter because of their importanteffects on pressure tracings and hemodynamic parameters.

Premature atrial and ventricular contractions

Perhaps the most frequently encountered cardiac arrhythmia is a prematurebeat, either a premature atrial contraction (PAC) or a premature ventricularcontraction (PVC). Although a single ectopic beat is unlikely to have a clin-ically significant hemodynamic effect, the pressure tracings associated withit are interesting. As a result of the earlier than normal ventricular contrac-tion and shortened diastolic filling time, the stroke volume and arterial pulsepressure are substantially reduced. The postectopic beat tends to produce anincreased stroke volume secondary to enhanced ventricular filling during thecompensatory pause (Figure 23.1). An increase in arterial pulse pressure willalso be evident.

A PVC can serve as an aid in diagnosing hypertrophic obstructive cardiomy-opathy (HOCM). In HOCM, a characteristic post-extrasystolic pressure changeoccurs in the aorta following a PVC. This finding was first described in 1961and is termed the Brockenbrough–Braunwald–Morrow sign (see Chapter 13for a further description of this sign). On the sinus beat immediately follow-ing a PVC, the aortic pulse pressure decreases (Figure 23.2). This finding isassociated with HOCM, whereas the other associated hemodynamic findingsincluding an increase in the left ventricular systolic pressure and an increasein the systolic gradient between the left ventricle and aorta can also be seenwith aortic valve stenosis. On palpation of the carotid pulse, a reduced impulseon the beat following an ectopic beat may be noted in patients with HOCM.Normally, an increased impulse (i.e. increased pulse pressure) would be noted.

255


1mV

1mV

1mV1mV

1mV

1mV

IGain:100

IIGain:50

V5Gain:50

r

s s ss

d

d

025mm/s >

100

200

d

r r r r

Figure 23.1 Aortic pressure tracing during PAC.

Heart block

Complete heart block causes dissociation of the electrical activity of the atriaand ventricles. The resulting ventricular rate is dependent upon the site of pace-maker activity (i.e. junctional or ventricular) that serves as an escape rhythm.Complete heart block is generally associated with a marked slowing in rate.Systolic blood pressure may be preserved but the long diastolic periods areassociated with lower diastolic blood pressure (Figure 23.3).

Chapter 23 Arrhythmias 257

1mV1mV

1mV

1mV

1mV

1mV

r r r

S

S

dd

ddd

dd d

dde

e e

e

S

SS

S

S

SS

S

v

r r

IGain:50

IIGain:50

V5Gain:50

200

100

01 2 3 4 5 61.50mm Hg/mm

...157/66 (100)

200

150

100

50

0

25mm/s >[-17-]

(a)

(b)

Figure 23.2 Brockenbrough–Braunwald–Morrow sign. Simultaneous recordings of LV and aorticpressure. Panel (a) is taken from a patient with HOCM. Note that pulse pressure falls in the normalbeat post-PVC (arrow). In contrast, pulse pressure increases post-PVC in a patient with aorticstenosis (b).

Cannon A waves

Large A waves on atrial pressure tracings that result from atrial contractionwhen the mitral or tricuspid valves are closed are termed cannon A waves.Cannon A waves result when ventricular contraction takes place during orjust prior to atrial contraction. Any arrhythmia that causes atrioventriculardissociation (e.g. complete heart block, pacemaker syndrome, or ventricular


I

II

V5

I

II

V5 200

100

01 2 3 4 5 625mm/s >

[-21-]

200

100s

d d d d d d

s s s s ss

s

01 2 3 4 51.50 mm Hg/mm

1.50 mm Hg/mm

25mm/s >[-12-]

r

r r

r

r

r r

1mV

1mV

1mV

1mV

1mV1mV

1mV

1mV

1mV

1mV

1mV1mV

r

s

d d

s s s

r r

(a)

(b)

Figure 23.3 Aortic pressure tracings in a patient who developed complete heart block. Aorticpressure during a brief period (a) and sustained episode (b) of complete heart block are shown.


Figure 23.4 Right atrial pressure in a patient with complete heart block. Note the increase in Awave amplitude when the atrium contracts against a closed mitral valve (cannon A waves; arrows).

tachycardia) can cause random cannon A waves (Figure 23.4). More rare arearrhythmias in which the atrium is activated via retrograde conduction (e.g.AV nodal reentry or some forms of ventricular tachycardia), which can causeregular cannon A waves.

Ventricular tachycardia

Much like a PVC, ventricular tachycardia originates in the ventricles andrepresents an abnormal sequence of electrical as well as mechanical events.The P wave on an electrocardiogram (ECG) representing atrial activationmay be visible during ventricular tachycardia. Atrial activation during thisarrhythmia may take place as a result of normal antegrade activation fromthe sinus node or via retrograde conduction from the ventricles. Ventriculartachycardia resulting in complete atrioventricular dissociation will producerandom cannon A waves. Less commonly, when atrial activation duringventricular tachycardia occurs via retrograde activation, cannon A waves willbe observed regularly on right atrial pressure tracings as the atria contractagainst the already closed mitral or tricuspid valves.

The rapid rate and loss of atrial contribution to ventricular filling that is seenin ventricular tachycardia can lead to hemodynamic deterioration (Figure 23.5).As shown in Figure 23.6, the sudden onset of ventricular tachycardia in thispatient resulted in the loss of systolic pressure generation and thus a steadydecay in mean blood pressure.

Junctional rhythm

During junctional rhythm, atria maybe activated via retrograde impulse origi-nating in or around the atrioventricular node. The retrograde atrial activation


1mV

1mV

1mV1mV

1mV1mV

rr

d

d

s

s

s

s

I

II

V5

rr r r r r r rrr

100

50

01 2 3 4 5 6 71.50mmHg/mm25mm/s >

[-23-]

Normal sinus rhythm Monitor Length: 10 sec.

Figure 23.5 Aortic pressure in an elderly woman with a permanent pacemaker and a transientepisode of ventricular tachycardia.

is evident in leads I and II on the standard 12-lead ECG as the P waves, if seen atall, will be inverted. Atrial systole may occur during, before, or after ventricularsystole in this setting. As seen with ventricular tachycardia, complete atrioven-tricular dissociation may also be seen during a junctional rhythm. Because atriamay be activated immediately following ventricular systole, cannon A wavescan be observed on the right atrial hemodynamic recording. Although junc-tional rhythm is in general slow (40–60 bmp), accelerated junctional rhythmmay compromise systolic pressure generation (Figure 23.7).

Atrial fibrillation and atrial flutter

Because atrial fibrillation implies the lack of organized atrial activity, the Awave will be absent from the pressure tracings. Depending on the size of theatrium, occasionally V waves will also be lacking if the atrium is large enoughto serve as a distensible, compliant reservoir (Figure 23.8a). Occasionally, atrialactivity may produce enough pressure to open and close the mitral and tricus-pid valves and give rise to a C wave on the right atrial pressure recording. Therhythm in atrial fibrillation is irregular with varying length of time spent indiastole. As a result, the stroke volume and hence arterial pulse pressure may


I

II

V5

I

II

V5

200

100

200

100

025mm/s >[-1-]

01 2 3 4 5 6 7 8 9

d

s

Cough

s

1.50mmHg/mm

1.50mmHg/mm


25mm/ s >

[-1-]

1mV

1mV

1mV

1mV1mV

1mV

Cardioversion

d

1 2 3 4 5 6 7 8 9SERIESIV:WITH MEDICAL

dd dd d d d

(b)

(a) Normal sinus rhythm Monitor Length: 10 sec.

Figure 23.6 Aortic and left anterior descending (LAD) coronary artery pressure in a patient whodeveloped ventricular tachycardia following intracoronary adenosine administration. The patientwas undergoing fractional flow reserve evaluation of a lesion in the LAD when ventriculartachycardia developed (a). The increase in aortic pressure with a cough can be seen. In panel (b),the return of pulsatile blood pressure can be seen following electrical cardioversion.


Figure 23.7 Left ventricular pressure during an episode of junctional tachycardia.

vary greatly beat to beat. In general, the longer the diastolic period, the higherthe pulse pressure (Figure 23.8b).

In atrial flutter, atrial activity is more organized than in atrial fibrillationand atrial systole can occur (usually at a rate of 300 beats/min). The regularatrial activity is reflected on right atrial pressure tracings as flutter waves(Figure 23.9). Systolic pressure will vary with left ventricular filling. Almostalways, atrial–ventricular block exists whereby not every atrial impulse is con-ducted to the ventricles. Because atrial contraction can take place against aclosed mitral or tricuspid valve, an exaggerated flutter wave similar to a can-non A wave may be seen. For a detailed description of the hemodynamicimportant of atrial contraction, please see chapter 5.

Sinus rhythm

Sinus rhythm in the form of sinus bradycardia and tachycardia may resultin hemodynamic effects and pressure tracings that mimic other arrhythmias.During sinus tachycardia, for example, the cardiac cycle length is shortened asevidenced by decreased R–R interval. On the right atrial pressure waveform,one may not be able to distinguish separate A and V waves as a result of the Awave of one cycle coming progressively closer to the V wave of the precedingcycle as the heart rate increases. As a result, the tracing may resemble that ofcardiac tamponade because the Y descent will not be evident. As a result ofthe reduced diastolic filling time, stroke volume may be significantly reduced(although cardiac output may increase because of the increase in heart rate).


I

II

V5

I

II

V5

50

25

01 2 3 4 5 6 7 8 90.38mmHg/mm

1.50mmHg/mm

25mm/s >[-25-]

1mV

1mV

1mV

1mV

1mV

1mV1mV200

100

01 2 3 4 5 6 7 8 9

SERIES IV:WITHMEDICAL

SERIES IV:WITHMEDICAL

25mm/s >[-28-]

1mV

1mV

1mV

1mV1mV

r r r r r r r r r r r r r r r r r r

v

v v vv

vv

v vv

r r r r r r r r r r r r r r r

ss

s s

ddd

(a)

(b)



Figure 23.8 Pressure tracings in a patient with atrial fibrillation. Right atrial (a) and aortic (b)pressure tracings during atrial fibrillation. Note the increase in aortic systolic and pulse pressurewhen the diastolic filling period is increased.

During sinus bradycardia, the R–R interval on an ECG in lengthened. Asa result, a right atrial pressure tracing will reveal a longer diastolic periodevidenced by lengthening the interval between the V wave of one cycle and Awave of the next cycle. On careful observation, one may be able to distinguish


1mV

1mV

1mV

1mV

1mV

1mV

I

II

V5 200

100

01.50mm Hg/mm25mm/s >

[-1-]

r

s

s

s

s s

s

s s

s

s

s

r r r r r r r

e d d dd d

d d d d de e ee

1 2 3 4 5 6

e e

Figure 23.9 Left ventricular and right atrial pressure in a patient with atrial flutter.

an additional positive deflection following the Y descent on the right atrialpressure waveform tracing. The etiology and significance of this waveform,termed the H wave, is unclear.

CHAPTER 24

Hemodynamics of pacemakers

Rodrigo Bolanos, Kimberly A. Selzman

Introduction

Optimal cardiac performance is dependent on the proper timing of atrial andventricular contraction. The electrical conduction system of the heart plays acentral role in coordinating these events. When the cardiac electrical conduc-tion system becomes diseased, typically from ischemic heart disease, calcifica-tion, or a degenerative process, bradycardia or heart block may occur. Thesepatients often require the placement of a temporary or permanent pacemaker.The first cardiac pacemaker was implanted in 1958 by Elmqvist and Senningvia thoracotomy. Since that time, pacemakers have become more reliable andmore capable of mimicking natural conduction. Despite these improvements,there remain acute and chronic hemodynamic effects of pacemakers and under-standing these changes in various patient populations is important in optimiz-ing patient care and outcomes.

The most common type of pacing is a dual-chamber system in which both theright atrium (RA) and right ventricle (RV) are paced. Single-chamber pacemak-ers pace only the RA or only the RV (Figure 24.1). With biventricular pacemak-ers, the left ventricle (LV) is also paced, usually via a posterior lateral branchof the coronary sinus (triple-chamber pacing systems). The primary goal ofpacemakers continues to be the achievement of “physiologic pacing.” The keycomponents of physiologic pacing are optimization of atrioventricular (AV)synchrony, rate responsiveness, and the evolving concept of interventricularsynchrony.

AV synchrony involves the proper timing of atrial diastole and systole inrelation to ventricular diastole and systole so ventricular preload is maxi-mized, atrial pressures are optimal, and closure of AV valves occurs priorto ventricular systole. This is adjusted by varying the AV delay, which is pro-grammable in all dual-chamber and biventricular pacemakers. Rate respon-siveness refers to the ability of a pacemaker to increase the rate appropriately tochanging physiologic demands. This is typically achieved with an accelerom-eter or minute ventilation sensor built into the pacemaker. Interventricularsynchrony addresses the coordination of contraction between RV and LV tooptimize cardiac function.

265


Subclavian vein

Pacemaker

Transvenous endocardiac electrodes with implanted pacemaker

ICD lead

RA leadRV lead

LV lead

Figure 24.1 Schematic of where a permanent pacemaker lies in the chest and a chest x-rayshowing four cardiac leads in a patient with an ICD and bi-ventricular pacemaker.

Physiology and pathophysiology of AV synchrony

Atrial contraction and AV synchrony influence cardiac output and diastolicfilling pressures (Figure 24.2). Dyssynchronous or asynchronous atrial contrac-tion results in reduced end-diastolic volume at a given end-diastolic pressure,

Chapter 24 Hemodynamics of pacemakers 267

Junctional bradycardia Temporary atrialpacemakerHR = 55 bpm

BP =120/60 mm HgCardiac = 7.6 L/minoutputRA = 9 mm Hg

HR = 80 bpmBP =160/70 mm HgCardiac = 9.2 L/minoutputRA = 4 mm Hg

ECG

40

30

20

10

0

50 mm Hg

PCWP

40

30

20

10

0

50 mm Hg

PA

PCWP = 21 mm Hg PCWP = 11 mm Hg

Figure 24.2 PCWP and PA pressure tracings during junctional bradycardia and atrial pacing(Courtesy of WG Sanders Jr.).

decreased LV end-diastolic volume, reduced LV stroke volume, and reducedcardiac output. Two mechanisms have been implicated in the shift in the LVpressure volume curve that is observed during impaired atrial activity: (1) theloss of atrial contraction directly influences ventricular diastolic filling and(2) failure of the atria to empty during ventricular diastole increases volumewithin the pericardium and thus filling pressures [1]. In a given patient, theabsence of atrial contraction and/or loss of AV synchrony may result in alteredhemodynamics with important clinical implications.

The loss of “atrial kick,” or ventricular diastolic filling from atrial contrac-tion, may be particularly detrimental to patients with decreased ejection frac-tions or noncompliant ventricles. Atrial contraction may contribute up to 25–30% of diastolic filling in these individuals. This scenario may be observed inpatients with bradycardia but preserved AV synchrony who are paced fromthe RV at a rate higher than the intrinsic rate (e.g. with a temporary pacingwire or a permanent single-chamber pacemaker). The loss of atrial contraction


may be evident as a nondistinct atrial pressure waveform and as lower RV andLV end-diastolic pressures.

If atrial contraction is present but occurs after ventricular contraction hasbegun, then the atria will contract against closed AV valves often resultingin prominent A waves or “cannon A waves” on atrial pressure tracings. Thisscenario can occur when atrial activity does not precede ventricular activity(sinus node dysfunction), atrial activity is asynchronous with ventricular sys-tole (complete heart block with AV dissociation), or when there is conductionretrograde from the ventricles to the atria (VA conduction) resulting in contrac-tion of the atria against closed AV valves. This lack of AV synchrony may leadto impaired ventricular filling and thus increase the mean right atrial pressureand pulmonary capillary wedge pressure (PCWP).

Pacemaker syndrome is used to describe patients that respond adverselyto the loss of AV synchrony during pacing. Symptoms of pacemaker syn-drome range from pulsatile sensations in the neck, to malaise, presyncope,or syncope in severe cases due to hypotension. This syndrome is most oftenseen although not limited to pacemaker systems where there is a single leadpacing the right ventricular (RV) apex (VVI pacing mode) in a patient withintact sinus node function. The pacemaker will neither sense atrial activity ortime ventricular pacing based on atrial contractions. The development of dual-chamber pacemaker systems that sense and pace both the atrium and ventriclehas decreased but not eliminated pacemaker syndrome. These symptoms canoccur in dual-chamber devices if the programmed parameters are inappropri-ate for the individual patient and allow asynchronous atrial and ventricularpacing. And similar to a single lead system, a dual-chamber device with anonfunctioning atrial lead will also lead to pacemaker syndrome symptoms.

Patients exhibit a wide range of clinical sensitivity to the loss of AV syn-chrony. Some patients are asymptomatic while others are completely intoler-ant. Likewise, the expected clinical benefit of reestablishing AV synchrony inpatients through placement of a pacemaker is quite variable. Although multi-ple invasive and noninvasive studies have shown significant improvement incardiac output with AV sequential pacing, the expected benefits have not beenconsistently borne out in clinical trials.

Patient response is heterogeneous and it appears that patients with low ornormal filling pressures benefit most from the “atrial kick” due to the theirposition on the upslope of the Starling curve. Interestingly, patients with sig-nificantly increased filling volumes may be the least likely to benefit in termsof increased cardiac output from the “atrial kick” due to their position onthe downward curve of the Starling curve. These patients still appear to ben-efit from AV synchrony but through other mechanisms such as reestablish-ment of the optimal diastolic filling period and abolition of diastolic mitralregurgitation [2].

Numerous clinical trials have examined the benefits of “physiologic pacing”;that is, atrial-based pacing systems that promote AV synchrony versus singlelead systems (VVI). Generally accepted benefits of physiologic pacing include


improved symptoms, a likely reduction in the risk of developing persistentatrial fibrillation and possible trends toward less pacemaker syndrome andheart failure. At this time there is no uniform evidence across trials showing amortality benefit or reduction in stroke with atrial-based systems.

Role of AV synchrony: key points

� Loss of AV synchrony may result in ↑ PCWP, ↓ BP, and ↓ CO. Symptomsmay include pulsatile sensations in the neck, malaise, presyncope, or syncope� Loss of atrial contraction may be evident as nondistinct atrial pressure wave-forms, ↓ LVEDP on ventricular pressure tracings, and ↓ BP on LV/aortic pres-sure waveforms� Atrial contraction occurring against closed AV valves may be evident as“cannon A waves,” ↑ mean atrial pressure, and ↑ PCWP� Pacemaker syndrome incidence can be decreased by using dual-chamberpacing rather than RV single-chamber pacing (VVI pacing)� Clinically, dual-chamber pacing versus single-chamber pacing (VVI) likelydecreases the incidence of atrial fibrillation with trends toward ↓ HF with noconsistent benefit on stroke or mortality

Pacemakers in specific patient populations

There are specific patient populations where pacemakers often have distinc-tive hemodynamic effects. These include RV myocardial infarctions requiringtemporary pacing, hypertrophic obstructive cardiomyopathy (HOCM), andsevere congestive heart failure from dilated cardiomyopathy with left ventric-ular dysfunction.

Right ventricular infarctionMyocardial infarction involving the RV is complicated by bradyarrhythmias(usually in the form of sinus bradycardia or AV block) in 10–33% of cases. Com-plete heart block due to an RV infarction is most often transient; however, it canlast several days or longer. Correction of bradycardia is occasionally necessaryto maintain hemodynamic stability. Although some patients may respond toatropine or aminophylline, most patients with RV infarction and heart blockrequire temporary pacing during the peri-infarct period. This is typically donewith a temporary transvenous pacing wire, which is placed in the RV.

Some patients may experience hemodynamic deterioration with the initia-tion of VVI pacing possibly due to loss of AV synchrony with loss of atrialcontribution to ventricular filling, development of tricuspid regurgitation, ordevelopment of asynchronous contraction between the right and left ventri-cles. These patients may be better served by temporary pacing of both theatrium and ventricle with maintenance of AV synchrony. Topol et al. reportedfour patients who had low cardiac output during RV infarctions. Ventricularpacing resulted in increased heart rates but had no effect on cardiac output.


Pre pacemaker Post pacemaker

Figure 24.3 Effect of pacing on outflow tract gradient (as determined by Doppler interrogation) ina patient with HOCM. (Courtesy of WG Sanders Jr.)

In contrast, atrial pacing increased cardiac output by 25–63% [3]. Love et al.subsequently reported seven patients with loss of AV synchrony during RVinfarction. Restoration of AV synchrony resulted in increased systolic bloodpressure, cardiac output (by an average of 50%), and stroke volume [4].

Hypertrophic obstructive cardiomyopathyDual-chamber pacemakers have been advocated as a therapeutic treatmentoption for patients with HOCM. The theory underlying the use of dual-chamber pacing with a short-programmed AV delay resulting in continuousRV pacing in these patients is that activation of the RV will cause asynchronouscontraction of the septum and LV free wall with resultant increased LV out-flow tract diameter, decreased pressure gradient and decreased systolic ante-rior motion of the mitral valve. Early studies reported marked decreases in LVoutflow tract gradients and improvement in symptoms in most patients withdual-chamber pacing (an example of a dramatic response to pacing is shown inFigure 24.3). The initial enthusiasm was dampened by subsequent randomizedcross over trials which revealed that reductions in LVOT gradients varied sig-nificantly from patient to patient and were more modest than initially reported.A significant placebo-effect emerged with subjective symptom improvementbut no difference in objective measures of exercise capacity between periodsof active pacing and periods when the pacemaker was set to backup mode.Based on available data at this time, dual-chamber pacing for HOCM is notrecommended as first-line therapy for relief of symptoms and outflow tractobstruction [5]. In patients with significant LVOT gradient and severe symp-toms (NYHA class III or IV), surgical myectomy or alcohol septal ablation areconsidered prior to pacing. Pacing is generally reserved for selected patientsubgroups such as those >65 years old who may not be good candidates forinvasive treatment or perhaps in patients who receive an implantable cardiacdefibrillator (ICD) for prevention of sudden cardiac death.


Role of pacemakers in HOCM: key points

� Dual-chamber pacemakers with continuous pacing of the RV apex mayresult in 25–40% reduction in LVOT gradient� Reduction of LVOT and symptom relief varies widely between patients� Use of permanent pacemakers in HOCM is reserved for patients with sig-nificant outflow gradient and severe symptoms who are not candidates formyectomy or alcohol septal ablation and/or who have another indication forpermanent pacing or ICD therapy

Biventricular pacing in patients with heart failure anddilated cardiomyopathyAs the LV dilates in the setting of impaired systolic function, progressiveadverse changes in cardiac geometry and structure can occur resulting inincreased wall stress, decreased mechanical performance, and worseningmitral regurgitation. Often accompanying the mechanical changes is the devel-opment of disordered electrical timing resulting in a prolonged AV delay andasynchronous contraction between the ventricles of the heart. These electricalchanges are manifest on the ECG as a prolonged PR interval and widened QRSduration or bundle branch block. Pacemaker therapy in dilated cardiomyopa-thy has thus evolved not only to correct coexisting sinus node and AV nodedysfunction, but also to directly address interventricular dyssynchrony fromdisordered electrical timing.

Role of pacemakers in heart failure from systolic dysfunction:key points

� Dual-chamber pacing can acutely increase CO and contractility� Dual-chamber pacemakers with chronic, frequent (i.e. >40%) pacing of theRV may exacerbate heart failure� Cardiac resynchronization therapy leads to ↓ PCWP, ↑ CI, and ↑ stroke work� Patients with class III/IV experience ↑ exercise capacity, improved func-tional capacity, and quality of life w/CRT (75–80% patient response rate)� Cardiac resynchronization therapy decreases hospitalizations and mortalityin patients with class III/IV HF who have evidence of dyssynchrony and whoare on optimal medical therapy

Initial pacing efforts in patients with decreased LV ejection fraction focusedon reducing AV conduction delay using dual-chamber devices pacing the RAand RV (AV synchronous pacing). Shortening the AV interval has several theo-retical advantages including lengthening diastolic filling time, changing dias-tolic filling pattern to a more physiologically balanced early and late (atrial)filling components compared to primarily early filling (resulting in lower mean


atrial pressures), and reducing mitral regurgitation. Initial studies of acuteeffects revealed improved hemodynamics, improved contractility, increasedcardiac output, and in some patients, decreased mitral regurgitation. Earlyenthusiasm for pacing strategies that reduced AV delay but only paced the RAand RV in patients with dilated cardiomyopathy was dampened by severallarge trials that failed to show improved outcomes.

In fact, there is now growing evidence that frequent RV pacing may be asso-ciated with worsening heart failure and that longer AV delays should perhapsbe tolerated to promote intrinsic ventricular contraction. Sweeney et al. showedin a population of patients with sinus node dysfunction and narrow QRS thatDDR pacing resulting in RV pacing >40% of the time was a strong predictorof heart failure hospitalization and atrial fibrillation despite maintenance ofAV synchrony [6]. In the Dual Chamber and VVI Implantable Defibrillator(DAVID) Trial, DDDR pacing at 70 bpm versus backup VVI pacing at 40 bpmresulted in a 10.6% absolute increase in the risk at 1 year of the compositeendpoint of mortality and hospitalization for heart failure likely as a resultof the ventricular desynchrony imposed by ventricular pacing even when AVsynchrony is preserved [7].

The focus has now shifted to directly addressing the interventricular asyn-chrony that exists in patients with a dilated cardiomyopathy and a wide QRScomplex. The wide QRS is felt to be an electrical marker for mechanical dyssyn-chrony between the two ventricles. The reestablishment of concerted LV andRV contraction via simultaneous electrical stimulation of the RV and LV is com-monly known as cardiac resynchronization therapy (CRT). The concept of CRTis that pacing of the LV may correct interventricular conduction delays lead-ing to improved hemodynamics and contraction. The term CRT was coinedby Cazeau and colleagues in 1994 when they placed four epicardial leads ina 54-year-old man with LBBB, dilated cardiomyopathy, and class IV conges-tive heart failure with an immediate decrease in PCWP and increase in cardiacoutput [8]. Six weeks later the patient’s symptoms had improved to class IIaccompanied by drastic weight loss and resolution of peripheral edema.

Acute hemodynamic improvements are seen with LV pacing in patients withdilated cardiomyopathy, decreased systolic function, and left bundle branchblock (Figure 24.4). For example, Kass et al. in a study of 18 patients with dilatedcardiomyopathy and prolonged QRS found that LV free-wall pacing increaseddP/dtmax (a measurement of LV contractile performance by 23.7 ± 19.0%),pulse-pressure (by 18.0 ± 18.4%) and stroke work. Biventricular pacing yieldedless improvement whereas RV apical or midseptal pacing had negligible effects[9]. AV delay had less influence on LV function than the ventricular pacing site.Nelson et al. found that LV pacing and biventricular pacing improved arterialpulse pressure, cardiac output, and dP/dtmax while simultaneously reducingenergy demands (8% reduction in myocardial oxygen consumption; MVO2). Incontrast, administration of dobutamine at a level to achieve a similar effect ondP/dtmax increased MVO2 by 22% [10]. Evidence that pacing the LV at the siteof greatest activation delay without concurrent RV pacing has hemodynamic


120 NSR

BiV

80

40

00 100

LV volume (mL)

LV p

ress

ure

(mm

Hg)

200 300

Figure 24.4 Pressure–volume loops in a patient with left bundle branch block and dilatedcardiomyopathy taken during normal sinus rhythm (NSR) and during biventricular pacing (BiV).(Adapted from Kass et al. [9])

benefits similar to biventricular pacing in some patients is intriguing becauseLV-only pacing may improve intraventricular delay, but if anything worsensinterventricular synchrony.

Large CRT trials have revealed improved functional capacity, exercise capac-ity, and quality of life for patients with dilated cardiomyopathy, severe symp-toms (NYHA class III) and LBBB. CRT has also been demonstrated to fosterbeneficial remodeling as evidenced by multiple echocardiographic parame-ters such as reduction in LV end diastolic dimensions. When compiling thedata of multiple CRT trials, a recent meta-analysis also showed a mortal-ity benefit when compared to maximal medical therapy [11]. The CardiacResynchronization—Heart Failure (CARE-HF) Study of 813 patients withsevere LV dysfunction and class III or IV symptoms reported a mortality ben-efit with CRT as well as improvement in interventricular mechanical delay,end-systolic volume index, area of the mitral regurgitant jet; LV ejection frac-tion, symptoms and quality of life [12]. The AHA/ACC pacemaker guidelinesrecommend implantation of a CRT device in patients with an LVEF ≤35%,NYHA class III symptoms, and a QRS duration ≥130 milliseconds [13].

Additional patient populations that could benefit from CRT are being eval-uated. For example, it has been postulated that patients with a right bundlebranch block may benefit from pacing in the RV apex to preexcite the RV incomparison to the left. In addition, some patients with a narrow QRS by ECGbut mechanical dyssynchrony by echocardiogram appear to benefit from CRTwith resulting increase in cardiac index and decrease in pulmonary capillarywedge pressure. These patients likely have mechanical dyssynchrony despitethe narrow QRS electrically [14].

References

1 Linderer T, Chatterjee K, Parmley WW, Sievers RE, Glantz SA, Tyberg JV. Influence of atrialsystole on the Frank–Starling relation and the end-diastolic pressure-diameter relation ofthe left ventricle. Circulation 1983;67:1045–1053.


2 Nishimura RA, Hayes DL, Holmes DR, Jr, Tajik AJ. Mechanism of hemodynamic improve-ment by dual-chamber pacing for severe left ventricular dysfunction: an acute Dopplerand catheterization hemodynamic study. J Am Coll Cardiol 1995;25:281–288.

3 Topol EJ, Goldschlager N, Ports TA, et al. Hemodynamic benefit of atrial pacing in rightventricular myocardial infarction. Ann Intern Med 1982;96:594–597.

4 Love JC, Haffajee CI, Gore JM, Alpert JS. Reversibility of hypotension and shock by atrialor atrioventricular sequential pacing in patients with right ventricular infarction. Am HeartJ 1984;108:5–13.

5 Maron BJ, McKenna WJ, Danielson GK, et al. American College of Cardiology/European Society of Cardiology clinical expert consensus document on hypertrophiccardiomyopathy. A report of the American College of Cardiology Foundation Task Forceon Clinical Expert Consensus Documents and the European Society of Cardiology Com-mittee for Practice Guidelines. J Am Coll Cardiol 2003;42:1687–1713.

6 Sweeney MO, Hellkamp AS, Ellenbogen KA, et al. Adverse effect of ventricular pacing onheart failure and atrial fibrillation among patients with normal baseline QRS duration in aclinical trial of pacemaker therapy for sinus node dysfunction. Circulation 2003;107:2932–2937.

7 Wilkoff BL, Cook JR, Epstein AE, et al. Dual-chamber pacing or ventricular backup pacingin patients with an implantable defibrillator: the Dual Chamber and VVI ImplantableDefibrillator (DAVID) Trial. JAMA 2002;288:3115–3123.

8 Cazeau S, Ritter P, Bakdach S, et al. Four chamber pacing in dilated cardiomyopathy. PacingClin Electrophysiol 1994;17:1974–1979.

9 Kass DA, Chen CH, Curry C, et al. Improved left ventricular mechanics from acute VDDpacing in patients with dilated cardiomyopathy and ventricular conduction delay. Circu-lation 1999;99:1567–1573.

10 Nelson GS, Berger RD, Fetics BJ, et al. Left ventricular or biventricular pacing improvescardiac function at diminished energy cost in patients with dilated cardiomyopathy andleft bundle-branch block. Circulation 2000;102:3053–3059.

11 McAlister FA, Ezekowitz JA, Wiebe N, et al. Systematic review: cardiac resynchronizationin patients with symptomatic heart failure. Ann Intern Med 2004;141:381–390.

12 Cleland JG, Daubert JC, Erdmann E, et al. The effect of cardiac resynchronization onmorbidity and mortality in heart failure. N Engl J Med 2005;352:1539–1549.

13 Gregoratos G, Abrams J, Epstein AE, et al. ACC/AHA/NASPE 2002 guideline updatefor implantation of cardiac pacemakers and antiarrhythmia devices: summary article: areport of the American College of Cardiology/American Heart Association Task Forceon Practice Guidelines (ACC/AHA/NASPE Committee to Update the 1998 PacemakerGuidelines). Circulation 2002;106:2145–2161.

14 Achilli A, Sassara M, Ficili S, et al. Long-term effectiveness of cardiac resynchronizationtherapy in patients with refractory heart failure and “narrow” QRS. J Am Coll Cardiol2003;42:2117–2124.

CHAPTER 25

Unknowns

George A. Stouffer

Introduction

This chapter includes hemodynamic tracings taken from 15 patients. Extract asmuch information as you can from these tracings and then check your answersat the end of the chapter. Remember that it is rare to arrive at a specific diag-nosis based on hemodynamic tracings. Thus the goal is not to come up with adiagnosis but rather to list the useful hemodynamic findings present in eachtracing.

In assessing the hemodynamic status of a patient, it is important to use asystematic approach so that the maximal useful information is obtained. Beloware nine suggested steps to use in analyzing hemodynamic data:1. Make sure that the data are accurate – It goes without saying that makingclinical decisions based on incorrect data can have disastrous consequences.The equipment should be calibrated and leveled properly and the minimumamount of tubing and stop-cocks should be used. Examine the pressure trac-ings to make sure that the pressures are not damped or distorted by incorrectcatheter position (e.g. when a pigtail catheter has a side hole above the aorticvalve during recording of left ventricular pressures). Other sources of errorsin hemodynamic measurements are discussed in Chapter 2.2. Note whether intracardiac pressures are elevated – Elevation in diastolicpressures is a sensitive indicator that pathology is present. Conversely, ‘normal’diastolic pressures may mask characteristic hemodynamic findings (e.g. inrestrictive cardiomyopathy or constrictive pericarditis) and should elicit a fluidbolus if there is a high degree of clinical suspicion.3. Examine pressure waveforms from each cardiac chamber and from the pul-monary artery – Examine A and V waves and X and Y descents in atrial tracingsand determine whether characteristic waveforms (e.g. W or M configuration)are present. Similarly in ventricular tracings, pay special attention to whethera ‘dip and plateau’ configuration is present and to the slope of the rise in pres-sure during diastole. The aortic pressure tracing can provide helpful clues tomany diseases as discussed in Chapter 4.4. Take note of the effect of respiration – In the normal heart, the decrease inintrathoracic pressure with inspiration is transmitted to the heart with char-acteristic results. If right atrial pressure does not decrease with inspiration

275


(i.e. Kussmaul’s sign), this suggests constrictive pericarditis but has also beenfound in other diseases.5. Characterize diastolic filling – Examination of the atrial and ventricularpressure tracings can give an indication of diastolic filling patterns. An exag-gerated Y descent implies that significant ventricular filling occurs in earlydiastole whereas an exaggerated X descent signifies late diastolic filling. Foreach patient determine whether diastolic filling occurs primarily in early dias-tole (e.g. constrictive pericarditis or aortic regurgitation), throughout diastoleor in late diastole (e.g. cardiac tamponade).6. Interpret cardiac output based on filling pressures – Cardiac output is a func-tion of left ventricular filling pressure. A low cardiac output in the setting of anelevated left ventricular end diastolic pressure (LVEDP) is an ominous findingwhereas low cardiac output with low LVEDP is an expected consequence ofdehydration. An elevated cardiac output is commonly a sign of anxiety butcan also be a clue to anemia, arterio-venous shunt, high-output heart failure,etc. Note also that pulmonary artery pressure and pressure gradients acrossstenotic valves increase in proportion to cardiac output. For example, exercisecan be a useful technique in unmasking symptomatic mitral stenosis.7. Compare pressures measured simultaneously from two sites within theheart or major vessels – There are numerous benefits to measuring pressuresat two sites simultaneously. Comparison of pressures measured proximal anddistal to a diseased valve provide information useful in determining the degreeof stenosis or regurgitation. Simultaneous measurement of left ventricular andleft atrial pressures (or PCWP) is useful in determining diastolic filling char-acteristics as well as mitral valve pathology. Also, simultaneous measurementof left ventricular/right ventricular pressures or left ventricular/right atrialpressures gives an indication of whether these chambers are independent orinterdependent (e.g. in constrictive pericarditis).8. Perform any necessary calculations – Calculate valve area, intracardiacshunt fraction, systemic vascular resistance, etc. as needed. Treatment deci-sions are often based on the severity of the condition as reflected in thesecalculations.9. Synthesize hemodynamic information with clinical presentation, ECG, pastmedical history, echocardiographic data, etc. to arrive at a working diagnosis.

Chapter 25 Unknowns 277

IGain:50

IIGain:50

V5Gain:50 200

100

01 2 3 4 5 6 725 mm/s > 1.50mm Hg/mm

1mV1mV

1mV

r

s

s

d d d d d

ee

e e e e

d d d d d d

s s s s s

s s s s s

r r r r r

1mV

1mV1mV

Aorta

LV

Normal sinus Monitor Length:10 sec

Figure 25.1 A 66-year-old male with exertional dyspnea and one episode of syncope.

r

1 mv1 mv

1 mv

1 mv

1 mv

1 mv

s

s

s

ee

d

0

25

50

25mm/s> 0.38mm Hg/mm SERIES IV. WITH BIOMEDICAL

d

1 2 3 4 5 6 7 8 9

d d

d

e

e e

d d

d

e

d

ee

d

e

e

s

s s

s

s

s

r r r r r r r r r r

IGain:100

IIGain:100

Gain:100V5


Figure 25.2 A 62-year-old female with acute ST elevation inferior MI. RA = 13 mm Hg andPCPW = 12 mm Hg. Does this patient have hemodynamic evidence of RV infarction?


1 mV1 mV

1 mV

1 mV

1 mV

s

d

0.38 mm/Hg/mm

d d

dd

dd

d

s

s s s

s s

s

s

01 2 3 4 5 6 7

25

50

PA

Normal sinus Monitor Length: 10 sec.

Figure 25.3 Pulmonary artery pressure in a 52-year-old female with a recent episode ofendocarditis who now presents with exertional dyspnea.

1 mV

s

a

v

a

v

a

e

2SERIES IV-WITH BIOMEDICAL11

e

1

e

r

s

r

s

r

1 mV

1 mV1 mV

1 mV1 mV

0100 mm/s>[-1-]

0.75 mmHg/mm

50

100

IGain: 200

IIGain: 50

V5Gain: 50

Figure 25.4 A 50-year-old male who is being evaluated after a recent hospital admission for“pneumonia.”


1mV1mV

1mV

1mV

1mV

r r r r r r r

a

25

20

15

10

5

00.19mm Hg/mm

a a a a avav

1 2 3 4

RA

5

Figure 25.5 A 54-year-old male with end stage renal disease admitted with congestive heartfailure. Past medical history was remarkable for an episode of pericarditis requiringpericardiocentesis 10 years ago.


Normal sinus Monitor Len 8.0/10 secI

II

V5

200

100

0

s

ee

ee e

ee

ee

e

dd d d

d d dd

d

s s s ss s

LV

ss

s

1mV

1mV

1mV

1mV

1mV

1mV

Figure 25.6 A 77-year-old female with acute anterior wall, ST elevation MI. What information isavailable from this pressure tracing?

Aorta

200

200

025 mm/s>

I

II

1 mV

r r

s s ss

s

s ss

dddd

ddd

r r r r r rr

r r r r r r r

1 mV

1 mV1 mV

1 mV

1 mVV5


1 2 3 4 5 6 7 81.50mm Hg/mm

Figure 25.7 A 56-year-old male with history of hypertension who now presents with congestiveheart failure.


60%

61%

84%

93%

77%

94%

76%

75%

57%

Figure 25.8 A 53-year-old female with cough and shortness of breath. Listed above are theoxygen saturations obtained in various chambers. What is the diagnosis?

1mV1mV

1mV1mV

1mV

1mV

I

II

V5 50

25

01 2 3 4 5 6 70.38 mmHg/mm25 mm/s >

[-12-]

r r r r r r

a

aa

a

a

av

v

v v

v v

Figure 25.9 Where is this pressure tracing obtained from?


1mV

1mV

1mV1mV

1mV1mV200

100

0

1 2 3 4 5 61.50mm Hg/mm

r r r r r r

v

v

r r

a aa

a

LV

RFA

a a

v

a

v

Figure 25.10 A 73-year-old white male with a history of atrial fibrillation who complains ofdyspnea on exertion.


Monitor Length: 10 sec

1mV

1mV1mV1mV

1mV

1mV

50

25

0

1 2 3 4 5 6 70.38mmHg/mm

r r r r r r r

a

a aa a

aa

a

v

RA

vv

v

v

v v

Figure 25.11 A 72-year-old female who has had recurrent episodes of “congestive heart failure.”

Normal sinus Moniter Length: 10 sec.1 mV

1 mV1 mV

1 mV1 mV

1 mV1 mV

1 mV

1 mV1 mV

1 mV1 mV

200

100

0 d1 2 3

d d d4 5 6 7 8

0

100

200

>1.50mmHg/mmd d d

eeeeeeee

e

d d d d d d d d

ss

ss

s ss s

s s s s s s s s ss ss s s

dddddddd

ee e e

d 1 2 3 4 5 6 7 8d d d d d d d d

eeeeee

s

s

s

s

s

s

s

ssss

r r r r r r r r rr r r r r r r r r

s

1.50mmHg/mm

Figure 25.12 These two tracings are obtained as an end-hole catheter (marked LV) is slowlywithdrawn from the LV apex to the aorta (while right femoral artery pressure is continuouslyrecorded) in a 29 year-old female who presented with dyspnea on exertion.


1mV

1mV

1mV

1mV

1mV

1mV

I

II

V5200

100

0

Aorta

d d d d d d dd d d

d

s s s s s s s s s s

Figure 25.13 What does this aortic pressure tracing represent?

200

100

025 mm/s>

[-1-]

1 mV

r

ss

s

s

s

e

1 2 3 4 5 6 7 8 9

d

ed

e

d

d

e

d

d

d

e

d

r r r r r r r r

1 mV

1 mV1 mV

1 mV1 mV

I

II

V5

Figure 25.14 An 82-year-old man with chest discomfort on exertion. What is the diagnosis?


1 mV1 mV

1 mV1 mV

1 mV

1 mV

200

100

r

s

s

s

s

s

s

s

s

s

d

d

d

ddd

r r r r r r r r r

01 2 3 4 5 6 7

s>1.50 mmHg/mm

Figure 25.15 An 81-year-old female with hypertension and diabetes mellitus who presented withcomplaints of substernal chest pain awakening her from sleep. Cardiac enzymes are elevated.What information is available from the aortic pressure tracing?

Answers

25.1. The pressure tracings show a large gradient (approximately 90 mm Hg)between the left ventricle and the aorta. These tracings could be found in a num-ber of conditions including valvular aortic stenosis, hypertrophic obstructivecardiomyopathy, subvalvular aortic stenosis, or supravalvular aortic stenosis.The contour of the aortic pressure will occasionally provide some help in thediagnosis but here there is neither bisferiens pulse (seen in hypertrophic obstru-ctive cardiomyopathy) nor pulsus parvus et tardus (seen in aortic stenosis). Sta-tistically, the most likely diagnosis in a 66-year-old patient is valvular aorticstenosis which was the case here. The aortic valve area calculated to be 0.5 cm2.

Also note in this tracing the elevated LVEDP (approximately 25 mm Hg).25.2. The pressure tracing shows elevated RV diastolic pressure. This is con-sistent with RV infarction. The diagnosis is confirmed by findings that RA pres-sure was greater than PCWP (a specificity of 100% for diagnosis of RV infarctionin one study), RA pressure greater than 10 mm Hg and within 1–5 mm Hg of the


PCWP (a specificity of 100% for diagnosis of RV infarction in another study),and a mean RA pressure of 10 mm Hg or more and an RA/pulmonary arterywedge pressure ratio of 0.8 or more (specificity of 92% in a third study).25.3. The tracing is taken from a catheter placed in the pulmonary artery. Notehowever the ventricular like appearance with a rise in pressure during diastole.This tracing is suggestive of severe pulmonary insufficiency (presumably fromendocarditis). Also consistent with this diagnosis was the elevated RVEDP at21 mm Hg. As the degree of pulmonic valve incompetence progresses, RV andPA pressures become similar in diastole.25.4. The pressure tracings show a gradient between PCWP and LV that per-sists throughout diastole. Resistance (and thus pressure gradient) betweenpulmonary artery (wedge position) and LV can occur either in the pulmonaryvenous system (e.g. pulmonary venous stenosis following atrial fibrillationablation) or at the level of the mitral valve. The PCWP tracing shows distinct Aand V waves (implying that the PCWP is a good surrogate for left atrial pres-sure) and thus the most likely diagnosis is mitral stenosis. That was the casein this patient who had rheumatic mitral stenosis with a valve area calculatedto be 0.7 cm2.25.5. RA pressure is elevated and the tracing shows characteristic findings ofconstrictive pericarditis. Note the exaggerated Y descent with a preserved Xdescent. Although the tracing is too short to make definitive conclusions, therealso appears to be minimal variation with respiration.25.6. This tracing shows elevated LV end-diastolic pressure (32 mm Hg) andreduced LV systolic pressure (95 mm Hg). In the setting of acute myocardialinfarction, this relationship can provide important information regarding LVfunction. In this case, LV systolic pressure is low despite elevated filling pres-sures suggesting that LV contractility is severely reduced (based on the Frank–Starling principle). The prognostic importance of LV systolic and end-diastolicpressures has been known since at least 1967 when Killip and Kimball demon-strated that mortality in the setting of acute MI was a function of LV fillingpressures (in their case using physical exam findings of pulmonary edema)and aortic systolic pressure. This patient would be Killip class IV (see Killip T,Kimball JT. Treatment of myocardial infarction in a coronary care unit: a twoyear experience of 250 patients. Am J Cardiol 1967;20:457–464).25.7. Systolic pressure alternates in a beat-to-beat fashion in this aortic pres-sure tracing. This is characteristic of pulsus alternans which is found in patientswith failing left ventricles. This patient had a severe nonischemic cardiomy-opathy with an ejection fraction of less than 20%. Be aware of pseudo-pulsusalternans that can occur due to atrial arrhythmias and also cause beat-to-beatvariation in systolic pressure.25.8. This patient has an atrial septal defect (ASD). There is a significant‘step-up’ in oxygen saturations between the vena cava and right ventricle.To estimate Qp/Qs using the short equation, first calculate mixed venous (MV)oxygen saturation. In the case of an ASD, MV oxygen saturation can be calcu-lated using the equation of MV O2 = (3SVC + {IVC)/4. In this case, MV O2 =


59.25 rounded to 59. Qp/Qs = (SAO2 – MVO2)/(PVO2 – PAO2). If we assumeSAO2 = PVO2 (a valid assumption in the absence of right-to-left shunting),Qp/Qs = (94 – 59)/(94 −77) = 2.1.25.9. This tracing shows left atrial pressure. Note that the mean pressure is25 mm Hg and that there is significant respiratory variation, both consistentwith atrial pressure. In the left atrium, the V wave is generally larger than theA wave; the opposite is generally true in the right atrium. These pressures wereobtained from pulmonary capillary wedge position, which explains the delaybetween the ECG and pressure deflections (compare the P wave on the ECGand A wave on the tracing).25.10. This aortic pressure tracing shows findings consistent with hyper-trophic obstructive cardiomyopathy. There is a minimal gradient between rightfemoral artery (RFA) and left ventricle (LV) at rest but a gradient >100 mm Hg ispresent post-PVC. Also present is a Brockenbrough, Braunwald, and Morrowsign. This sign, originally described in 1961, is defined as diminished pulsepressure in a post-extrasystolic beat compared to the beat immediately pre-ceding the premature ventricular contraction (PVC).25.11. The RA tracings show “ventriculization,” i.e. the atrial pressure takeson characteristics of a ventricular pressure (an exaggerated Y descent, rapidrise in pressure during ventricular diastole and lack of X descent). This patienthad severe tricuspid regurgitation.25.12. This patient had congenital subvalvular stenosis. The tracing on theleft shows a pressure difference between the LV apex and femoral artery witha mean gradient of 29 mm Hg. The tracing on the right was taken with an end-hole catheter just below the aortic valve. It identifies the obstruction as beingsubvalvular since there is no gradient between the ventricle at this location(note the ventricular contour of the tracing) and the femoral artery.25.13. This patient has an intra-aortic balloon pump. Note the diastolicaugmentation.25.14. At initial glance, this looks like aortic stenosis since there is a pressuregradient between LV and femoral artery. Note, however, that the femoral arterypressure increases during the catheter pullback and that there is no differencebetween LV systolic and aortic systolic pressure. Presumably, this patient hada thrombus in the sheath that was dislodged during pullback. The cause of thechest symptoms was coronary artery disease.25.15. At first glance this tracing looks like pulsus alternans. Note, how-ever, that the ECG shows type 1, second-degree heart block (Wenckebach).The beat-to-beat variability in aortic systolic pressure is explained by short-ened diastolic filling times in alternate beats and does not indicate severe LVfailure.

a p p e n d i x 1

Useful Hemodynamic Formulas

1. Calculation of Blood Flow

Q = �P/R

Q = V × A (or Q = V × � × r2)

Where:Q = blood flow

�P = Pressure gradient = pressure proximal – pressure distal

V = velocity

A = cross sectional area

r = radius

2. Calculation of Resistance to Flow

R = 8 × L × �/� × r4

Putting together Ohm’s law and Poiseuille’s law

Q = �P × � × r4/8 × L × �

Where:Q = blood flow

�P = Pressure gradient = pressure proximal – pressure distal

V = velocity

A = cross sectional area

r = radius

L = length

R = resistance

� = viscosity

3. Resistance of Vessels in Series and Parallel

Resistance in series: R = R1 + R2 + R3 + · · ·Resistance in parallel: R = 1/R1 + 1/R2 + 1/R3 + · · ·

289

290 Appendix

4. Mean Arterial Pressure and Pulse Pressure

MAP = 1/3 (SBP) + 2/3 (DBP)Pulse pressure = SB P − DB P

Where:DB P = diastolic blood pressure

MAP = mean arterial pressure

SB P = systolic blood pressure

5. Cardiac Index

Cardiac index = CO/BSA

Where:CO = cardiac output

BSA = body surface area

6. Stroke Volume and Ejection Fraction

SV = CO/HR

SV = LVEDV – LVESV

Ejection fraction = SV/(LVE DV)

= (LVEDV – LVESV)/(LVEDV)


HR = heart rate

LVE DV = left ventricular end diastolic volume

LVE SV = left ventricular end systolic volume of LV

SV = stroke volume

7. Vascular Resistance

SVR = (MAP − CV P)/CO

PVR = (P AP − PCWP)/CO


CV P = central venous pressure

MAP = mean arterial pressure

Appendix 291

P AP = mean pulmonary artery pressure

PCWP = pulmonary capillary wedge pressure

PVR = pulmonary vascular resistance

SVR = systemic vascular resistance

8. Simplified Formula for Calculation of a Left to RightIntracardiac Shunt using Oxygen Saturations

Qp/Qs = (arterial O2 sat –PA O2 sat)/(arterial O2 sat – CVP O2 sat)

Where:Qp = pulmonary blood flow

Qs = systemic blood flow

CV P = central venous pressure

P A = pulmonary artery

9. Valve Area

Quick estimate of valve area (Hakki formula):Valve area = CO/square root of peak to peak pressure difference

Gorlin formula (see text for more details)Valve area = CO/(flow time × HR × valve factor × square root of meanpressure gradient)

Where:valve factor = 44.5 for aortic valve and 38.0 for mitral valve.CO = cardiac output in ml/minFlow time is in secondsHR = heart rate in beats/minPressure is in mm Hg (mean pressure gradient is calculated during systole foraortic stenosis and during diastole for mitral stenosis)

APPENDIX 2

Hemodynamic Maneuvers

Inspiration causes a decrease in intrathoracic pressure resulting in decreasedcentral venous pressure, increased venous return to the right atrium, increasedright ventricle stroke volume, decreased pulmonary vascular resistance,decreased left atrial pressure and decreased aortic pressure.

Expiration associated with increased intrathoracic pressure and effects oppo-site of those observed with inspiration.

Change in posture rising from a recumbent position decreases venous return tothe heart; squatting from a standing position is associated with a simultaneousincrease in venous return and systemic vascular resistance and a rise in arterialpressure.

The Valsalva maneuver is widely used to elicit well-defined hemodynamicchanges and is a simple test of complex autonomic reflex controls of cardiovas-cular function. Correct performance of a Valsalva maneuver results in 4 distincthemodynamic phases. It can be induced by multiple mechanisms including aforced expiration against a closed glottis, straining during a bowel movementor lifting weights during a breath hold. At the bedside, the patient should beasked to attempt exhalation against a resistance or to simulate straining fora bowel movement. These efforts should be maintained for approximately 30seconds and then released.

These maneuvers result in a large rise in intrathoracic pressure which com-presses the vessels within the chest cavity. There is initially a transient increasein aortic pressure (Phase 1) due to aortic compression and increased cardiacoutput due to enhanced left atrial blood flow. This increase in aortic pressurecauses a reflex bradycardia due to baroreceptor activation. Because the tho-racic vena cava is also compressed, venous return to the heart is compromised,resulting in a large fall in cardiac output within several seconds. This leads toa secondary fall in aortic pressure (Phase 2), and as aortic pressure falls, thebaroreceptor reflex increases heart rate. When breathing is resumed, the releaseof aortic compression results in a small, transient dip in arterial pressure anda further reflex increase in heart rate (Phase 3). When compression of the venacava is removed, venous return suddenly increases causing a rapid rise in car-diac output several seconds later which leads to a transient increase in arterialpressure (Phase 4). Sympathetic activation that occurred during Phase 2 leadsto arterial pressure greater than baseline because the systemic vascular resis-tance is increased. Heart rate reflexively decreases during Phase 4 in responseto the transient elevation in arterial pressure.

293

294 Appendix

Summary of Hemodynamic Effects of Valsalva Maneuver

Phase 1 - transient increase in left ventricular output.Phase 2 (straining phase) - decreased venous return, right and left ventricu-

lar volumes, stroke volumes, mean arterial pressure, and pulse pressure;reflex increase in heart rate.

Phase 3 (release of Valsalva) – brief phase with a further reduction in leftventricular volume.

Phase 4 - increase in stroke volume and arterial pressure and reflex slowingof heart rate

Handgrip sustained for 20 to 30 seconds leads to an increase in systemic vas-cular resistance, arterial pressure, cardiac output, and left ventricular volumeand filling pressure.

Mueller’s maneuver is forced inspiration against airway resistance. Thisresults in decreased intrathoracic pressure and increased LV afterload. In astudy of 10 normal subjects [1], the hemodynamic response to the Muellermaneuver included:� no change in aortic systolic and mean pressures� decreased aortic diastolic pressure� increased pulse pressure� large decrease in mean right atrial pressure (± SE) decreased from 7 ± 1 to−17 ± 4 mm� accentuation of right atrial ”x” descent� decrease in left ventricular end-diastolic pressure� increase in systemic vascular resistance� reduction in cardiac output and stroke volume� no significant change in heart rate.

In another study of patients with systolic dysfunction, the Mueller maneuvercaused immediate increases in systolic left ventricular transmural pressure andsimultaneous reductions in blood pressure, stroke volume, and cardiac output.These changes were proportional to the magnitude of the negative intrathoracicpressure generated [2].

References

1 Condos WR, Latham RD, Hoadley SD, Pasipoularides A. Hemodynamics of the Muellermaneuver in man: right and left heart micromanometry and Doppler echocardiography.Circulation 1987;76:1020-1028

2 Hall MJ, Ando S, Floras JS, Bradley TD. Magnitude and time course of hemodynamicresponses to Mueller maneuvers in patients with congestive heart failure. J Appl Physiol1998;85:1476–1484.

Index

A wave, 67, 69, 264cannon, 259–261prominent, 70f

ACC. See American College of CardiologyAccess sites, 22Acute respiratory distress syndrome (ARDS),

17Afterload, changes in, 50fAIx. See Augmentation indexAlcoholic septal ablation, 169American College of Cardiology (ACC), 165Amyl nitrate, 166fAngiograms, 244fAorta

pigtail catheter pullback in, 107f, 108f, 110fpressure, 47, 58–60

in AR, 126in HCM, 162in heart failure, 174–175in HOCM, 164fin ventricular tachycardia, 263fmeasurement, 106frespiration and, 62–65, 64fschematic of, 60ftracing, 176f, 205f, 258f, 260fwith IABP, 251

pullback, 165ftracings, 130fwaveform, 222

Aortic regurgitation (AR), 125–126acute, 129–130

hemodynamic changes in, 129pharmacological treatment of, 130

angiographic classification of, 133case study, 131–132chronic, 126–127

aortic pressures, 126classic findings in, 128left ventricular pressures, 127pathophysiology of, 126physical exam in, 128

hemodynamic changes in, 125, 128–129physical exam of, 127–128

hemodynamic findings in, 127hemodynamic tracings of in, 130–133in aortic stenosis, 109–110stroke volume in, 125

Aortic stenosis, 101–102aortic regurgitation in, 109–110case studies in, 109–110clinical pears in, 106

echocardiographic evaluation of, 103–104etiology of, 101invasive hemodynamics, 104–105low-gradient, 108physical exam, 102–103

of severe, 103tpitfalls in, 106–107progression of, 101–102

Aortic valveclosure, 37–39cusps of, 125replacement, 125

AR. See Aortic regurgitationARDS. See Acute respiratory distress

syndromeArrhythmias, 31, 257

atrial fibrillation, 262–264atrial flutter, 262–264cannon A waves, 259–261heart block, 258junctional rhythm, 261–262premature atrial contractions, 257premature ventral contractions, 257sinus rhythm, 264–266ventricular tachycardia, 261

Arteriespressure, 57–58 (See also specific arteries)pressure-volume relationship for,

11fASD. See Atrial septal defectsAtrial contraction, 269–270Atrial fibrillation, 262–264

pressure tracings in, 265fAtrial flutter, 262–264

left ventricular pressure in, 266fright atrial pressure in, 266f

Atrial kick, 269, 270Atrial septal defects (ASD), 89

locations of, 90fAtrial waveform, 67

components of, 67–69important points, 78pressure abnormalities and, 69–76schematic of, 68f

Atrioventricular (AV) synchrony, 267pathophysiology of, 268–271physiology of, 268–271role of, 271

Augmentation index (AIx), 62, 63fAustin Flint murmur, 128AV synchrony. See Atrioventricular synchrony

295

296 Index

Bernoulli equation, 12, 103Bernoulli, Daniel, 3Biventricular pacing, 273Blood flow

basic principles of, 235case studies, 245doppler wire and, 239–241fractional flow reserve and, 241in cardiac catheterization, 238–239in constrictive pericarditis, 191flaminar, 6–9

schematic of, 7fturbulent transition from, 8

measurement of, 238–239clearance method of, 238–239doppler wire, 239flowmeter techniques, 239thermodilution, 239

models, 235pressure and, 4–5pressure wire and, 241regulation of, 235–237resistance to, 5–6schematic of, 21fturbulent, 6–9

laminar transition to, 8velocity of, 13–14

Blood pressure. See also Pulmonary capillarywedge pressure

aortic, 47, 58–60measurement, 106frespiration and, 64f

arterial, 57–58blood flow and, 4–5cardiac output, 3–4describing, 58during PA catheterization, 23ffemoral artery measurement, 106fhydrostatic, 3–4in circulation, 15fintracardiac, 171–174LA, 44–45laminar and turbulent flow in, 7fLV, 45–46, 47fmeasurement, 65–66, 276PA, 46RA

respiratory variation and, 27fwaveform, 43

RV, 47fvolume loop, 49f, 51fwaveform, 60–62

Blood samples, 91Blood stream, energy in, 3–4Blood vessels

in parallel, 14fin series, 14fincreased resistance and, 14velocity and cross-sectional

areas of, 13–14Bradycardia, 265

Breath holdPCWP and, 54fRA and, 54f

Brockenbrough-Braunwald-Morrowsign, 168f, 259f

C wave, 67Cannon A waves, 259–261Carabello’s sign, 110Cardiac chambers, 41–48

phases of, 200RA pressure tracings in, 203f

Cardiac cyclemechanical events of, 40fsteps in, 37

Cardiac index (CI), 29in heart failure, 174

Cardiac muscle, oxygen extraction in, 236fCardiac output, 5, 81, 276

as LVEDP function, 173fcatheterization and, 28–29Doppler echocardiographic measurement

of, 86–87estimating, 28Fick method for determining, 81–83formulas involving, 82tin heart failure, 174in mitral stenosis, 118increasing, 11–12thermodilution measurement of, 84–86

Cardiac Resynchronization - Heart Failure(CARE-HF), 275

Cardiac resynchronization therapy(CRT), 274, 275

Cardiac sarcoidosis, pressure tracingsfrom, 183f

Cardiac structuresduring diastole, 39fduring systole, 39f

Cardiac tamponade, 199, 201faortic pressure tracings, 205fcase studies in, 210catheterization in, 202techocardiography in, 206–208hemodynamic findings, 202–205hemodynamic pathophysiology

of, 199–202pericardial pressure tracings in, 204fphysical exam findings in, 205–206pressure tracings in, 209fpulsus paradoxus in, 206

Cardiogenic shock, 250Cardiomyopathy

dilated, biventricular pacing in, 273–275endstage ischemic, 232hypertrophic, 157–158 (See also Hypertrophic

obstructive cardiomyopathy)aortic pressure in, 162case studies, 169catheterization and, 161diastolic dysfunction in, 160

Index 297

hemodynamics in, 158–161LV pressure in, 162outflow tract gradient, 162–164PCWP and, 161–162physical exam in, 158treatment, 167–169

restrictive, 181case study, 184echocardiography in, 183–184hemodynamic principles of, 181–182

CARE-HF. See Cardiac Resynchronization -Heart Failure

Case studiesaortic stenosis, 109–110AR, 131–132blood flow, 245catheterization, 33–36effusive-constrictive pericarditis,

213HCM, 169heart failure, 176–177in cardiac tamponade, 210intracardiac shunts, 96–97restrictive cardiomyopathy, 184tricuspid stenosis, 151

Catheterizationcardiac output and, 28–29case studies, 33–36in blood flow, 238–239in cardiac tamponade, 202tin constrictive pericarditis, 192–193MR hemodynamics and, 139pulmonary artery, 17–18

accurate data from, 24–28chest X-ray of, 25fcomplications of, 22t, 31–33education project, 33phlebostatic axis in, 26physiology relevant to, 18–20placement in, 20–21pressure tracings during, 23fschematic of, 18trials, 19tuses of, 17–18vascular access, 20zeroed, 25

right heartin heart failure, 172tintracardiac shunts at, 93physiology relevant to, 18–20

RV myocardial infarction, 221–222tricuspid regurgitation and, 147–148tricuspid stenosis and, 151

Central venous pressure (CVP), 5, 190CFR. See Coronary flow reserveChest discomfort, 284fChronic obstructive pulmonary disease

(COPD), 226CI. See Cardiac indexCirculation

flow in, 15f

pressure in, 15fvelocity in, 15f

Clearance method, 238–239Closed systems, 12Compensatory mechanisms

in chronic MR, 138–139Concordance, 194fCongenital subvalvular stenosis, 287Congestive heart failure, 279, 280, 283Constrictive pericarditis, 187–188

blood flow in, 191fcase studies, 196catheterization in, 192–193characterization of, 187echocardiography in, 195hemodynamic findings in, 188t

sensitivity and specificity of, 193–195hemodynamic principles, 189–192intracardiac pressures in, 189fRA tracing in, 190frestrictive cardiomyopathy differentiated

from, 182, 191–192, 195–196ventricular pressure tracings in, 188f

Continuity equation, 13Contractility

changes in, 50findices of, 52–53

Contractions, premature, 257COPD. See Chronic obstructive pulmonary

diseaseCoronary flow reserve (CFR), 239–241

definition of, 246measurement of, 240f

Counterpulsation, 249, 250fCRT. See Cardiac resynchronization

therapyCusps, 125CVP. See Central venous pressure

Data, 275DDR pacing, 274Diabetes mellitus, 285fDiastole

cardiac structures during, 39fventricular, 45f, 48f

Dilated cardiomyopathy, 273–275Discordance, 194f

ventricular systolic, 193Diuresis, 232Dobutamine, 108, 223

effects of, 109fDoppler recordings

of LV inflow, 207fDoppler transmitral flow velocity paradoxus,

207Doppler wire, 239

limitations of, 240Dual-chamber system, 267Dyspnea, 277f, 278f, 282f, 283f

ECG. See Electrocardiogram

298 Index

Echocardiographyconstrictive pericarditis, 195doppler

in tricuspid regurgitation, 147of cardiac output, 86–87

in AR, 128–129in cardiac tamponade, 206–208in effusive-constrictive pericarditis, 212–213in HCM, 164–167in mitral regurgitation, 142in mitral stenosis, 122in restrictive cardiomyopathy, 183–184in tricuspid regurgitation, 146–147in tricuspid stenosis, 151, 152f

Effusive-constrictive pericarditis, 211case studies on, 213echocardiography in, 212–213hemodynamics of, 212pericardial imaging techniques in, 212physical exam in, 212

Eisenmenger’s syndrome, 230fElectrocardiogram (ECG), 26End hole catheters, 283End stage renal disease, 279fEnd-systolic diameter (ESD), 139Endocarditis, 32, 280fEnergy, in blood stream, 3–4ESD. See End-systolic diameterEvaluation Study of Congestive Heart Failure

and Pulmonary Artery CatheterizationEffectiveness, 171

Exercise hemodynamicsin mitral stenosis, 120f

Femoral arterypressure, 130f, 170fmeasurement, 106f

FFR. See Fractional flow reserveFick method, 81–83

in low output states, 83limitations of, 83thermodilution method v., 82t

Fick, Adolph, 28, 81Flowmeter techniques, 239Fluoroscopy, 197fFractional flow reserve (FFR), 241

clinical studies of, 243–245angiographically intermediate lesions in,

243left main coronary lesions in, 244–245multiple stenoses in, 245PCI success, 244

definition of, 246limitations, 245–246measurement of, 242f, 243f

Frank-Starling law, 9–10

Gallavardin’s phenomenon, 103Gorlin formula, 105

for calculating mitral valve area, 119–120in mitral stenosis quantification, 121

Graham Steel murmur, 122, 228Gravitational forces, 4

H wave, 69HCM. See Hypertrophic cardiomyopathyHeart block, 258

aortic pressure tracings in, 260fright atrial pressure in, 261f

Heart failure, 171aortic pressure in, 174–175biventricular pacing in, 273–275cardiac index in, 174cardiac output in, 174case studies in, 176–177etiology of, 172themodynamics in, 175intracardiac pressure, 171–174LVEDP in, 174mixed venous oxygen saturation in,

174pulmonary hypertension and, 225–226pulsus alternans, 174–175right heart catheterization in, 172t

Hemodynamicsintroduction, 37–40normal values, 41t

HOCM. See Hypertrophic obstructivecardiomyopathy

Hydraulic systems, 4Hypertension, 47, 58, 280f, 285f

classification of, 59tpulmonary, 225–227

causes of, 225–226echocardiography in, 228–231hemodynamic changes in, 227–228physical exam, 228PVR in, 231–233with normal left atrial pressure, 226f

Hypertrophic cardiomyopathy (HCM),157–158

aortic pressure in, 162case studies, 169catheterization and, 161diastolic dysfunction in, 160echocardiography in, 164–167hemodynamics of, 158–161

findings, 160tLV pressure in, 162microscopic features of, 157outflow tract gradient, 162–164PCWP and, 161–162physical exam in, 158treatment, 167–169

Hypertrophic obstructive cardiomyopathy(HOCM), 158, 257

aortic pressure in, 164feccentric hypertrophy in, 159fLV tracings in, 163foutflow tract obstruction in, 160pacemakers and, 272

Hypoxia, 237

Index 299

IABP. See Intra-aortic balloon counterpulsationICD. See Implantable cardiac defibrillatorImplantable cardiac defibrillator (ICD), 272Inferior vena cava (IVC), 95

saturation, 91Intra-aortic balloon counterpulsation (IABP)

aortic pressure with, 251description, 247–248early deflation, 253early inflation, 251hemodynamic effects, 248–249

timing, 251fhistory of, 247indications for, 248tinflation, 249flate deflation, 253f, 254late inflation, 252pump timing, 250

incorrect, 252ftherapy, 249tracings, 254–255uses, 247

Intracardiac pressure, 189f, 275Intracardiac shunts

detection of, 89–90diagnosis of, 93management, 96quantifying, 92t

formulas in, 94tsample case, 96–97

Intrathoracic pressures, 204Invasive hemodynamics, 104–105Ischemia

PA, 32right ventricle, 217–218IVC. See Inferior vena cava

Jugular venous pulse (JVP), 76–78Junctional rhythm, 261–262JVP. See Jugular venous pulse

Korotkoff sounds, 7, 206Kussmaul’s sign, 69

LA. See Left atriumLaminar flow, 6–9

in blood pressure, 7fReynold’s number for, 8schematic of, 7fturbulent flow transition from, 8

Laplace relationship, 10–11LBBB. See Left bundle branch blockLeft atrium (LA)

in mitral stenosis, 115pressure, 44–45

Left bundle branch block (LBBB), 274Left ventricle (LV), 45–46

apex, 170ffunction, 49in mitral stenosis, 117–118inflow, 207f

pigtail catheter pullback in, 107f, 108f, 110fpreload, 49–52pressure, 47f

in AR, 127in HCM, 162junctional tachycardia, 264ftracings, 75f, 105f

pullback, 165fsystolic dysfunction, 232ttracings, 46f, 130f, 132f

in HOCM, 163fLeft-to-right shunts

quantifying, 93–95Lewis, Thomas, 77LV. See Left ventricleLV end-diastolic pressure (LVEDP), 51

cardiac output as function of, 173IABP, 248in heart failure, 174LVEDV and, 175f

LV end-diastolic volume (LVEDV), 127, 138increasing, 137LVEDP and, 175f

LV end-systolic volume (LVESV), 138,139

LVEDP. See LV end-diastolic pressureLVEDV. See LV end-diastolic volumeLVESV. See LV end-systolic volume

MAC. See Mitral annular calcificationMagnetic resonance imaging (MRI),

10fMAP. See Mean arterial pressureMean arterial pressure (MAP), 5, 59–60Mean pulmonary artery pressure (MPAP), 227,

231Mitral annular calcification (MAC), 135Mitral insufficiency, 136Mitral regurgitation, 135, 160–161

acute, 136symptomatic, 140

cardiac catheterization and, 139chronic

compensatory mechanisms in,138–139

hemodynamic concepts in, 136–137echocardiography in, 142grading of, 143hemodynamics of, 142–143

acute, 142chronic compensated, 142chronic decompensated, 143

pathology, 135–136occurence of, 135

PCWP tracing in, 140fphysical examination, 139–140pressure-volume loops in, 137fpseudomitral stenosis and, 143

Mitral stenosis, 113–114cardiac hemodynamics in, 115–118

cardiac output, 118

300 Index

left atrium, 115pulmonary artery, 115–116right ventricle, 118

exercise hemodynamics in, 120fin echocardiography, 122origin of, 113physical examination in, 121–122progressive, 113quantification of, 118–119

Gorlin formula for, 121stages of, 113, 114ftransvalvular pressure gradient in,

116fMitral valve

area, 119–120Gorlin formula for, 119–120

closure, 37–39inflow, 184PBMV in, 122–123resistance, 119

Mixed venous oxygen saturation (MVO2), 95in heart failure, 174

MPAP. See Mean pulmonary artery pressureMRI. See Magnetic resonance imagingMVO2. See Mixed venous oxygen saturationMyocardial infarction

right ventricular, 217clinical presentation of, 218diagnosis of, 222–223findings at catheterization, 221–222hemodynamics of, 218–221, 223–224management, 223pacemakers and, 271–272pressure tracings, 221f, 222f, 223f

Myocardial oxygen, 235Nitroprusside, 108, 232t, 233

Outflow tract gradientin HCM, 162–164

Oximetry, 90for quantifying intracardiac shunts,

92–93Oxygen

consumption of, 83extraction, 236fsaturation run, 90–92, 281fmixed venous, 95

PAC. See Premature atrial contractionsPacemakers, 267

biventricular, 273–275HCOM and, 272history, 267in specific patient populations, 271patient response, 270permanent, 268fright ventricular infarction and,

271–272role of, 273syndrome, 270

Pansystolic murmur, 140PAP. See Pulmonary artery pressurePatent ductus arteriosus (PDA), 89PBMV. See Percutaneous balloon mitral

valvuloplastyPCI. See Percutaneous coronary interventionPCWP. See Pulmonary capillary wedge

pressurePDA. See Patent ductus arteriosusPEEP. See Positive-end expiratory pressurePercutaneous balloon mitral valvuloplasty

(PBMV), 122–123Percutaneous coronary intervention

(PCI), 243success of, 244

Pericardial friction rub, 206Pericardial imaging techniques, 192

in effusive-constrictive pericarditis, 212Pericarditis, 206

constrictive, 187–188blood flow in, 191fcase studies, 196characterization of, 187echocardiography in, 195hemodynamic findings in, 188t, 193–195hemodynamic principles, 189–192pericardial imaging techniques in, 192physical exam in, 192RA tracing in, 190frestrictive cardiomyopathy differentiated

from, 182effusive-constrictive, 211

echocardiography in, 212–213hemodynamics of, 212physical exam in, 212

Pericardium, 187fluoroscopy of, 197fnoncompliance of, 200pressure, 200pressure tracings, 204f

Peripheral amplification, 65Phlebostatic axis, 26Physical exam

aortic stenosis, 102–103cardiac tamponade in, 205–206in AR, 127–128

chronic, 128in effusive-constrictive pericarditis, 212in MR, 139–140in pulmonary hypertension, 228in restrictive cardiomyopathy, 192in tricuspid regurgitation, 146mitral stenosis, 121–122of JVP, 76–78tricuspid stenosis, 150

Physiologic pacing, 270–271Pigtail catheters, 107f, 108f, 110fPneumonia, 279fPoiseuille’s law, 5–6, 14Positive-end expiratory pressure (PEEP), 85

Index 301

Preload, 9–10changes in, 50fLV, 49–52

Premature atrial contractions (PAC), 257aortic pressure tracing during, 258f

Premature ventricular contractions(PVC), 257

Pressure waveform, 60–62Pressure waveforms, 275Pressure wire, 241Pressure-volume loops, 49f, 51f, 52

in bundle branch blocks, 275fin MR, 137f

Pressure-volume relationshipsfor arteries, 11ffor veins, 11f

Pseudomitral stenosis, 143Pulmonary artery

catheterization, 17–18accurate data from, 24–28chest X-ray of, 25fcomplications of, 22t, 31–33education project, 33phlebostatic axis in, 26physiology relevant to, 18–20placement in, 20–21

pressure tracings during, 23fschematic of, 18trials, 19tuses of, 17–18vascular access, 20zeroed, 25

in mitral stenosis, 115–117infarction, 32ischemia, 32pressure, 46

tracings, 269fpulsus alternans, 177frupture, 32tracing, 141fwaveform, 222

Pulmonary artery pressure (PAP), 30Pulmonary capillary wedge pressure (PCWP),

17, 24, 32–33, 44–45, 46f, 67, 115, 270breath hold and, 54HCM and, 161–162in heart failure, 172LV pressure tracings and, 75fpressure tracings, 269frespiratory variation and, 53ftracing, 140f, 173tV wave in, 71

Pulmonary embolus, 32Pulmonary hypertension, 225–227

causes of, 225–226echocardiography in, 228–231hemodynamic changes in, 227–228physical exam, 228pressure tracings in, 229fPVR in, 231–233with normal left atrial pressure, 226f

Pulmonary vascular resistance (PVR), 29–30,173

calculating, 30components of, 232in pulmonary hypertension, 231–233index, 233

Pulse wave velocity (PWV), 57, 58in disease states, 63

Pulsus alternansdefined, 178in heart failure, 174–175pulmonary artery, 177representation of, 176f

Pulsus paradoxus, 190–191in cardiac tamponade, 203, 206

PVC. See Premature ventricular contractionsPVR. See Pulmonary vascular resistancePWV. See Pulse wave velocity

QRS complex, 24, 41, 43, 274

R wave, 250R-R intervals, 265RA. See Right atriumRBBB. See Right bundle branch blockrCFR. See Relative coronary flow reserveRelative coronary flow reserve (rCFR), 246Resistance

increasing, 14–15pulmonary vascular, 29–30systemic vascular, 29–30

Respiration, 62–65, 275Respiratory variation, 54–55

aortic changes in, 53fPCWP and, 53fRA pressure and, 27f, 53f

Respirometers, 207fRestrictive cardiomyopathy, 181

case study, 184constrictive pericarditis differentiated from,

182, 191–192, 195–196echocardiography in, 183–184hemodynamic findings in, 196themodynamic principles of, 181–182physical exam, 192

Reynold’s number, 6–7laminar flow and, 8

Right atrium (RA), 17, 41–43aortic pressure tracings, 204fbreath hold and, 54pressure

abnormalities in, 69–76in cardiac tamponade, 202in complete heart block, 261fin tricuspid regurgitation, 150ftracings, 71t, 72t, 76f, 77f, 78f, 149f, 203f,

204f, 221f, 222f, 223fpressure waveform, 43frespiratory variation and, 53frespiratory variation on pressure, 27fschematic of, 42f

302 Index

tracingsin constrictive pericarditis, 190f

waveforms, 220Right bundle branch block (RBBB), 31Right heart catheterization, 20–21

in heart failure, 172tintracardiac shunts at, 93physiology relevant to, 18–20

Right ventricle (RV)blood supply, 219fin mitral stenosis, 118ischemia and, 217–218pacing, 274preload, 223pressure, 47f

in cardiac tamponade, 203in tricuspid regurgitation, 150ftracings, 149f, 220f

schematic of, 42fwalls of, 219fwaveform, 222

Right ventricular myocardial infarction,217

clinical presentation of, 218diagnosis of, 222–223findings at catheterization, 221–222hemodynamics of, 218–221, 223–224management, 223pacemakers and, 271–272pressure tracings, 220f, 221f, 222f, 223f

Right-to-left shunting, 96Rupture, pulmonary artery, 32RV. See Right ventricleRV end diastolic pressure (RVEDP), 193RVEDP. See RV end diastolic pressure

Saline injection, 84SAM. See Systolic anterior motionSemilunar valve, 39SEP. See Systolic ejection periodSinus bradycardia, 265Sinus rhythm, 264–266SPAP. See Systolic pulmonary artery pressureST elevation, 277f, 280fStarling’s law, 137Stroke volume, 235Superior vena cava (SVC), 95SVC. See Superior vena cavaSvO2 monitoring, 30–31SVR. See Systemic vascular resistanceSyncope, 277fSystemic vascular resistance (SVR), 29–30

calculating, 29Systole

atrial, 40cardiac structures during, 39f

Systolic anterior motion (SAM), 160Systolic ejection period (SEP)

LV-aortic pressure tracings with, 105tSystolic pulmonary artery pressure (SPAP), 231

Tachycardia, 69ventricular, 261, 262f

aortic pressure in, 263fThermodilution method, 84–86

curves, 85fFick method v., 82tfor blood flow measurement, 239

Transvalvular pressure gradientin mitral stenosis, 116f

Tricuspid regurgitation, 73f, 84, 145–148at catheterization, 147–148echocardiography, 146–147pathophysiology of, 145physical exam, 146RA/RV pressure in, 150fRA/RV tracings in, 149fsevere, 148treatment in, 148

Tricuspid stenosis, 148–153case study of, 151catheterization, 151echocardiography in, 151, 152fpathophysiology of, 148–149physical exam, 150treatment, 151

Tricuspid valve, 145Turbulent flow, 6–9

in blood pressure, 7flaminar flow transition to, 8

V wave, 69, 264important points about, 75tin PCWP, 71prominent, 70f, 74f

Velocitycross-sectional areas of blood vessels and,

13–14in circulation, 15fin closed systems, 12

Venous system, 11–12pressure-volume relationship for, 11f

Ventral septal defects (VSD), 89Ventricles. See also Left ventricle; Right

ventriclediastole, 45fforce developed by, 9–10function curve, 52fMRI image of, 10f

Ventricular tachycardia, 261, 262faortic pressure in, 263f

Ventriculogram, 91fVSD. See Ventral septal defects

Wall tension, 10–11Wenckebach, Karel, 79Wiggers diagram, 38f, 61f

X descent, 67

Y descent, 69

cardiovascular hemodynamics for the clinician

Health & Medicine