CARDIOVASCULAR SYSTEM ASSESSMENT AND MULTIPLE ORGAN

DYSFUNCTION IN HORSES WITH ACUTE GASTROINTESTINAL DISEASE

by

ERIN LYNN MCCONACHIE

(Under the Direction of Michelle Henry Barton)

ABSTRACT

Acute gastrointestinal disease, or colic, is a common condition afflicting horses of

all ages, breeds and disciplines. Substantial morbidity and mortality for horses treated

surgically persists despite improvements in the management, diagnosis and surgical

correction of colic over the past few decades. While the causes for morbidity and

mortality are multifactorial, horses with clinical evidence of endotoxemia or the systemic

inflammatory response syndrome (SIRS) are at an increased risk for complications, such

as the development of the multiple organ dysfunction syndrome (MODS) and mortality.

The main objectives of the studies presented herein were to elucidate the role the

cardiovascular system plays in systemic inflammatory conditions, such as SIRS, in horses

with colic in the post-operative period and to develop criteria to describe MODS in

horses with colic. First, a non-invasive 2-Dimensional echocardiographic method for

cardiac output estimation was validated in healthy adult horses. This study revealed that

three 2-D echocardiographic methods had acceptable agreement with the reference

measurement. Second, heart rate variability (HRV) analysis was performed in the post-

operative period in horses with acute surgical colic and healthy horses that underwent an

elective procedure to assess the difference between HRV in these groups and a potential

association between HRV and survival. This study illustrated that horses with colic have

reduced HRV compared to healthy horses and that time domain measures of HRV were

associated with non-survival.

A multifaceted approach to cardiovascular system assessment, consisting of

hemodynamic monitoring, electrocardiography, and cardiac troponin (cTnI)

measurement, was then performed on healthy horses and those with acute surgical colic

to detect cardiovascular dysfunction. The results from this investigation demonstrated

cardiovascular system abnormalities in horses with colic, particularly those with ischemic

gastrointestinal lesions, characterized by reduced HRV, increased cTnI concentration,

reduced stroke volume index and increased frequency of pathologic arrhythmias. Finally,

through incorporation of what was discovered in the first three studies, a review of the

literature and the use of clinical judgment a scoring system for MODS in horses with

acute gastrointestinal disease was developed and validated.

INDEX WORDS: Equine, Critical illness, Colic, Cardiac output, Systemic

inflammatory response syndrome, MODS, Heart rate variability,

Echocardiography, Cardiac troponin

CARDIOVASCULAR SYSTEM ASSESSMENT AND MULTIPLE ORGAN

DYSFUNCTION IN HORSES WITH ACUTE GASTROINTESTINAL DISEASE

by

ERIN LYNN MCCONACHIE

BS, University of Connecticut, 2005

DVM, Oklahoma State University, 2009

A Dissertation Submitted to the Graduate Faculty of The University of Georgia in Partial

Fulfillment of the Requirements for the Degree

DOCTOR OF PHILOSOPHY

ATHENS, GEORGIA

2015

© 2015

Erin Lynn McConachie

All Rights Reserved

CARDIOVASCULAR SYSTEM ASSESSMENT AND MULTIPLE ORGAN

DYSFUNCTION IN HORSES WITH ACUTE GASTROINTESTINAL DISEASE

by

ERIN LYNN MCCONACHIE

Major Professor: Michelle Henry Barton Committee: Steeve Giguère Gregg Rapoport David J. Hurley Scott A. Brown Electronic Version Approved: Suzanne Barbour Dean of the Graduate School The University of Georgia August 2015

iv

DEDICATION

To my family and friends in Connecticut and Illinois who may not have always

understood exactly what it is that I am still doing at the University of Georgia, but have

supported and encouraged me nonetheless. To all of the horse owners that generously

allowed their horses to be enrolled in the studies that comprise this dissertation and to all

of the horses that allowed me to instrument them in the immediate post-operative period

when they must have been feeling their worst!

v

ACKNOWLEDGEMENTS

I am completely and truly grateful for the wisdom and guidance of my major

professor, Dr. Michelle Henry Barton, in both my professional career and in my life in

general. It was with extreme good fortune that my arrival as a Large Animal Internal

Medicine Resident coincided with Dr. Barton’s turn to take a new mentee under her

wing. Without her reassurance, patience and intuition I would not have accomplished a

fraction of what I accomplished during my five years as a resident and then doctoral

candidate at the University of Georgia.

I must give many thanks to Dr. Steeve Giguère for his encouragement, patience

and expertise in all aspects of my training and graduate work. Without his direction, high

standards and practical perspective the studies presented here would not be what they are.

I would also like to acknowledge the expertise, flexibility, support and

encouragement I have received from the rest of my committee members, Drs. Gregg

Rapoport, David Hurley and Scott Brown.

To my clinical mentors who have subsequently become some of my greatest

friends and supporters, Drs. Kelsey Hart, Kira Epstein and Amelia Woolums. Thank you

all for understanding the challenges of taking on graduate work while simultaneously

completing a residency training program.

To my resident-mates and house-officers past and present, particularly Lindsey

Boone, Brent Credille, Kevin and Kelley Claunch, Lisa Fultz and Harry Markwell; thank

you for not only helping me enroll cases over the past few years but for your friendship

vi

and support both in and out of the clinic. To all of the interns that helped me enroll cases

and prevented me from missing data collection points, particularly Amy Stieler, Amanda

Bergren, Jane Woodrow, Tara Shearer, Julia Miller and Jolie Demchur. Each of you went

above and beyond to help hold horses after hours for echocardiograms and assisted with

literally anything else that might have been needed to accomplish collecting a clinical

data set while I was being pulled in many different directions. I am indebted to you all for

your selfless assistance!

I am grateful for the support of my parents, grandparents and siblings as they have

encouraged me to follow my dreams even though it has taken me far from home for the

past 10 years. Their support and love knows no bounds. I also am extremely grateful for

Emma Finnegan, who came to be not only my roommate, but a kindred spirit and lifelong

friend over this final year of my doctoral candidacy. Without her, this year would have

been a struggle instead of the wonderful experience it was.

Finally, I am incredibly lucky to have my adoring and understanding husband,

Brian Beasley, who has tolerantly put up with living roughly 500 miles apart from each

other for the past few years. This endeavor wouldn’t have been worth undertaking if it

wasn’t for his love and unbelievably patient nature.

vii

TABLE OF CONTENTS

Page

ACKNOWLEDGEMENTS .................................................................................................v

CHAPTER

1 INTRODUCTION .............................................................................................1

2 LITERATURE REVIEW ..................................................................................4

SECTION I: IMPACT OF ACUTE GASTROINTESTINAL DISEASE IN

THE HORSE AND THE RATIONALE FOR THE STUDIES

PRESENTED HEREIN ...............................................................................4

SECTION II: PATHOPHYSIOLOGY OF THE SYSTEMIC

INFLAMMATORY RESPONSE SYNDROME AND MULTIPLE

ORGAN DYSFUNCTION………………………………………………. 6

SECTION III: PATHOPHYSIOLOGY OF CARDIOVASCULAR

DYSFUNCTION IN CRITICAL ILLNESS ..............................................12

SECTION IV: METHODOLOGY FOR CARDIOVASCULAR SYSTEM

ASSESSMENT IN ADULT HORSES ......................................................15

SECTION V: CURRENT EVIDENCE FOR MODS IN THE HORSE ...22

SECTION VI: THE DEVELOPMENT OF MULTIPLE ORGAN

DYSFUNCTION SCORES IN HUMANS AND SEVERITY SCORES IN

VETERINARY SPECIES .........................................................................31

REFERENCES ..........................................................................................35

viii

3 DOPPLER AND VOLUMETRIC ECHOCARDIOGRAPHIC METHODS

FOR CARDIAC OUTPUT MEASUREMENT IN STANDING ADULT

HORSES .........................................................................................................54

ABSTRACT ...............................................................................................55

INTRODUCTION .....................................................................................56

MATERIALS AND METHODS ...............................................................57

RESULTS ..................................................................................................63

DISCUSSION ............................................................................................65

FOOTNOTES ............................................................................................71

REFERENCES ..........................................................................................71

4 HEART RATE VARIABILITY IN HORSES WITH ACUTE

GASTROINTESTINAL DISEASE REQUIRING EXPLORATORY

LAPAROTOMY ..............................................................................................81

ABSTRACT ...............................................................................................82

INRODUCTION ........................................................................................83

MATERIALS AND METHODS ...............................................................85

RESULTS ..................................................................................................89

DISCUSSION ............................................................................................93

CONCLUSION ..........................................................................................99

FOOTNOTES ..........................................................................................100

REFERENCES ........................................................................................100

ix

5 ASSESSMENT OF THE CARDIOVASCULAR SYSTEM IN HORSES

WITH NATURALLY ACQUIRED ISCHEMIC INTESTNAL

DISEASE…………………………………………………………………...109

ABSTRACT .............................................................................................110

INRODUCTION ......................................................................................111

MATERIALS AND METHODS .............................................................113

RESULTS ................................................................................................120

DISCUSSION ..........................................................................................126

FOOTNOTES ..........................................................................................135

REFERENCES ........................................................................................136

6 A MULTIPLE ORGAN DYSFUNCTION SCORE FOR ADULT HORSES

WITH ACUTE GASTROINTESTINAL DISEASE .....................................151

ABSTRACT .............................................................................................152

INRODUCTION ......................................................................................153

MATERIALS AND METHODS .............................................................155

RESULTS ................................................................................................159

DISCUSSION ..........................................................................................162

FOOTNOTES ..........................................................................................167

REFERENCES ........................................................................................167

7 CONCLUSIONS............................................................................................182

REFERENCES ........................................................................................188

1

CHAPTER 1

INTRODUCTION

The purpose of the studies reported herein was to provide novel clinically relevant

information regarding the incidence and importance of the systemic inflammatory

response syndrome (SIRS) and cardiovascular system function in horses with naturally-

occurring acute gastrointestinal disease. Furthermore, criteria were proposed and

validated for assessing organ dysfunction in this population of horses expected to be at

high risk for the development of multiple organ dysfunction syndrome (MODS), filling a

gap in the current understanding of disease progression in critically ill horses.

Chapter 2 provides a comprehensive review of the literature and is organized into

six sections. Section I highlights the impact that colic has on the equine industry and

provides the rationale for performing the studies herein. Section II reviews current

concepts in the pathophysiology of SIRS and MODS and underscores the role of

endotoxin in triggering these syndromes. Endotoxemia is a commonly recognized sequela

in horses with acute gastrointestinal (GI) disease and provides a link between the current

understanding of the pathophysiology in humans and equids. Section III discusses the

pathophysiology of cardiovascular dysfunction in critical illness. Section IV reviews

studies on single organ dysfunction reported to date in critically ill horses; providing

evidence that MODS similar to that which is described in people may also exist in

critically ill horses. Section V describes the methodology employed for cardiovascular

monitoring in people and elaborates on current methodologies available to monitor the

2

cardiovascular system in adult horses. Finally, section VI describes the development of

organ dysfunction scoring systems for use in people with critical illness from a historical

perspective and compares these with severity scores thus far developed in adult horses

with acute colic.

Chapters 3 through 6 incorporate a series of manuscripts that provide the results

of the studies that are the core of this dissertation research. A validation study was

performed prior to the clinical research studies to evaluate the performance of

noninvasive echocardiographic methods of cardiac output measurement compared with

the currently accepted reference method, lithium dilution cardiac output measurement,

and is described in Chapter 3. Chapter 4 details heart rate variability (HRV) analysis in

horses with acute GI disease requiring exploratory laparotomy as compared to healthy

horses undergoing an elective surgical procedure affording a novel method for assessing

the cardiovascular system and uncovering the importance of the autonomic nervous

system in the post-operative period. Chapter 5 contains the results of a more

comprehensive assessment of the cardiovascular system in horses with naturally-

occurring acute gastrointestinal disease through comparison with healthy horses

undergoing elective surgical procedures and utilizes the cardiac biomarker, cardiac

troponin (cTnI), echocardiographic measures of cardiac output (from Chapter 3), central

venous pressure, noninvasive oscillometric mean arterial pressure, and HRV (from

Chapter 4) analysis. Finally, organ dysfunction criteria for eight organ systems for use in

a MODS scoring system in horses with acute GI disease are proposed, validated, and are

presented in Chapter 6.

3

Chapter 7 summarizes the conclusions that can be drawn from the research

presented herein and considers the clinical utility of the results drawn from this doctoral

dissertation.

4

CHAPTER 2

LITERATURE REVIEW

SECTION I. IMPACT OF ACUTE GASTROINTESTINAL DISEASE IN THE

HORSE AND THE RATIONALE FOR THE STUDIES PRESENTED HEREIN

The most common reason horses are presented to tertiary referral centers on an

emergency basis is pain associated with acute GI disease or colic.1,2 The 1998 USDA

National Animal Health Monitoring System study reported approximately 4.2 colic

episodes per 100 horses with an overall 11% mortality rate, costing the equine industry

over $115 million in losses annually.3 While the majority of horses that have colic are

treated medically, those that require surgical correction are at great risk for post-operative

morbidity and mortality related to but not limited to recurrent colic, thrombophlebitis,

ileus, surgical site infection, peritonitis, laminitis, and organ failure. The overall short-

term mortality rate of horses that have an exploratory celiotomy for surgical correction of

colic was recently reported to be 26% in one retrospective study at a referral hospital.2

Reported survival in horses following exploratory celiotomy for colic has ranged from

60-87%; however differences in case definition vary between studies. Specifically, in

horses with a large colon volvulus (a strangulating large intestinal lesion), reported rates

for survival to discharge and one and two years after discharge were 71%, 48% and 34%

respectively.4

In addition to the substantial mortality associated with colic, the long-term effects

of colic on intended use of the horse further impact the equine industry. As of a 2005

5

survey, there were an estimated 9.2 million horses in the United States.5 Approximately

37% of the equine population serves a purpose that requires them to retain a certain level

of athleticism.5,6 An additional 46% of horses in the United States are intended for

pleasure use which is invariably accompanied by an owner with a strong emotional bond.

A further 16% are used primarily for breeding and represent a subset with high economic

value. A recent retrospective study performed at North Carolina State University was

designed to provide objective data regarding the likelihood of horses to return to their

intended level of performance following exploratory celiotomy for colic.7 Sixty-eight

percent of horses returned to their previous level of performance by six months and 76%

of horses were performing for their originally intended use by one year post-operatively.

Factors related to inability to return to work by six months included in-hospital laminitis,

diarrhea, incisional hernia and history of previous celiotomy. Colic, therefore, continues

to be a frequently encountered clinical problem that is teeming with emotional, financial

and purpose driven decisions for the owner. While overall survival has improved over the

past 30 years a large number of horses continue to have post-operative complications that

are either life or career ending.

Attempts to reduce morbidity and mortality in horses with colic thus far have

been focused on improving resuscitation methods,8-10 general anesthetic practices and

surgical techniques11 with little attention given to post-operative monitoring and

treatment. Since the early 1990s, human physicians have been using clinical criteria

intended to identify patients with systemic manifestations of disease, as well as those at

risk of developing organ dysfunction upon development of acute critical illness.12

Versions of the clinical criteria used to define the systemic inflammatory response

6

syndrome (SIRS) in people have been adapted for use in the horse. In a recent clinical

study on large colon volvulus in horses, the most commonly reported cause of in-hospital

death/euthanasia (31%) was related to the development of SIRS, unrelenting pain or

colon necrosis.4 While SIRS was stated as the reason for euthanasia or death, it seems

unlikely that SIRS was the sole reason for euthanasia. Perhaps organ dysfunction

occurred in these patients but was not defined due to the current lack of criteria to

describe critical illness related organ dysfunction in horses. Through the development of

a scoring system for organ dysfunction in horses with colic, clinicians might be able to

detect horses in early stages of organ dysfunction prior to overt organ failure. Organ

dysfunction and in some instances, failure, can be reversible with appropriate

interventions. Through serial post-operative application of a scoring system for organ

failure, at risk horses will not only by identified earlier, but a validated score could serve

the clinician as an aid in directing and justifying therapy. Only then might there be an

objective measure to determine if goal-directed comprehensive monitoring and

therapeutic interventions are indeed beneficial to patient outcome. The aims of the next

few sections are to review the pathophysiology of SIRS and MODS and introduce

hemodynamic monitoring in the equine patient.

SECTION II. PATHOPHYSIOLOGY OF THE SYSTEMIC INFLAMMATORY

RESPONSE SYNDROME AND MULTIPLE ORGAN DYSFUNCTION

The innate immune system

The systemic inflammatory response syndrome is a clinical syndrome defined by

abnormalities in vital physiologic parameters (hypothermia or hyperthermia, tachycardia

7

and tachypnea) and/or a change in the total white blood cell count (leukocytosis,

leukopenia or a 10% or greater increase in band neutrophils). The presence of any two of

the above abnormalities in a patient fulfills the criteria for the clinical syndrome of

SIRS.12 Sepsis is differentiated from SIRS simply by the addition of the presence of a

confirmed infection.

When SIRS is recognized clinically, it should be assumed that there is either an

underlying infectious disease process or an abnormal response to a non-infectious

stimulus capable of initiating an immune response. Regardless of the underlying process

that initiated the innate immune response in patients with SIRS, the result is an

unbalanced activation of pro-inflammatory mediators on a systemic scale. Non-

infectious or ‘sterile’ inflammation is caused by tissue damage in the absence of a

pathogen. Examples of non-infectious inflammation include but are not limited to the

following; trauma, surgery, hemorrhage, burns, ischemia, immune mediated disease,

neoplasia and toxins.

The mechanism through which pathogens and products of damaged cells initiate

the immune response is similar. Ironically, in the valiant attempt to destroy the pathogen

and protect the body, the innate immune system can directly cause local tissue damage

and enhanced release of damage signals from the host tissues. Pattern recognition

receptors (PRRs) are stationed on the outer cell membrane or on nuclear or endosomal

membranes of immune cells, endothelial cells and parenchymal organ cells throughout

the body. The patterns recognized by these receptors are highly conserved molecular

regions of viral, bacterial, protozoal, fungal or parasitic pathogens and are collectively

referred to as pathogen associated molecular patterns (PAMPs). Similarly, there exists a

8

group of endogenously derived molecules; damage associated molecular patterns

(DAMPs) or alarmins that are recognized by PRRs when present aberrantly in the

extracellular fluid or within the cell. These molecules are typically constituents of normal

cells that are liberated as a consequence of cytokine signaling, cell membrane

permeability (injury or loss of electrochemical gradients) or necrosis. Examples of

DAMPs include heat shock proteins (HSP 70), high mobility group box-1 (HMGB-1),

adenosine triphosphate (ATP), mitochondrial DNA, histones and advanced glycation end

products.

While it is now generally accepted that cytokines with pro- and anti-inflammatory

activities are simultaneously expressed in response to tissue insult, the precise

relationship between the activation of these complementary cascades is not fully

understood in sepsis.13 Currently, the medical literature suggests that there is an equally

important syndrome of immunoparalysis or immunosuppression that might accompany

SIRS and sepsis at various stages which also contributes to mortality.14 Importantly, the

autonomic nervous system and adrenal glands both influence the inflammatory response;

with catecholamines from the sympathetic nervous system and adrenal medulla

promoting inflammation and cortisol from the adrenal cortex suppressing inflammation.15

Conceptually accepting SIRS to be the consequence of either a pro-inflammatory

‘cytokine-storm’ or a lack of the compensatory anti-inflammatory response syndrome or

the result of relative adrenal insufficiency is likely an oversimplification.

The concept of multiple, sequential, progressive organ failure in the trauma

patient was first suggested by Arthur Baue, a thoracic surgeon, in the early 1970s.16 Baue

thought of this syndrome as a consequence of modern medicine. The term multiple organ

9

dysfunction syndrome (MODS) has since been established and was first defined over 20

years ago by the American College of Chest Physicians to describe a continuum of organ

dysfunction that was clinically apparent in patients with trauma, surgery and sepsis.17

Importantly, the experts involved in the original definition of MODS recognized that

what they were observing in their patients was not an all-or-none phenomenon, but rather

a continuum of dysfunction. The multiple organ dysfunction syndrome was originally

defined as: “presence of altered dysfunction in an acutely ill patient such that homeostasis

cannot be maintained without intervention.”17 The recognition that MODS occurred on a

continuum was clinically meaningful because it implied that interventions could be

performed to halt or reverse progression before fulminant organ failure occurred.

The multiple organ dysfunction syndrome occurs as a consequence of protracted

SIRS, autonomic nervous system dysfunction, endothelial dysfunction, coagulopathy,

microvascular alterations, abnormal GI barrier function and abnormal cellular

metabolism (cytopathic hypoxia and mitochondrial failure) all of which culminate in

apoptosis or necrosis of the target organ that is well documented in the human

literature.15,18-22 A ‘second hit’ may also precede MODS, for example, in the trauma

patient in which acquisition of a nosocomial infection overwhelms the body’s ability to

maintain homeostasis and might lead to sequential organ dysfunction and failure. While

the pathophysiology of MODS is complex and multifactorial, the associated anatomical

and histopathologic abnormalities reported post-mortem are typically mild to moderate

and most often consistent with microcirculatory fibrin deposition, endothelial cell edema,

mitochondrial swelling, cytoplasmic or nuclear swelling, and in cardiomyocytes

contraction band formation and cellular necrosis.23,24 The clinical relevance of MODS is

10

highlighted by mortality rates that approach100% in people with multiple failed organs

(≥4) in the intensive care unit.25

Organ dysfunction scoring systems were first developed with the following goals

in mind: 1) to detect MODS at an early stage in the continuum of organ dysfunction such

that fulminant failure might be avoided with appropriate intervention, 2) to determine

what is the patient’s status along a continuum of dysfunction or disease severity, 3) to

determine the effects of novel interventions and medications and finally 4) to provide

prognostic information for individual patients.17

PAMPs associated with SIRS in horses with acute gastrointestinal disease

Endotoxin or lipopolysaccharide (LPS), the immunogenic component of the cell

membrane of Gram negative bacteria, is a classic example of a PAMP, which through

interaction with its PRR (TLR 4), co-receptor (CD-14) and co-stimulatory molecule

(MD2), produces a reliable cytokine profile (TNF- α, IL-1β, IL-6, IL-10) and

reproducible clinical signs (pyrexia, tachycardia, tachypnea, hypotension and abdominal

discomfort) in the horse. The mediators that are elaborated in response to interactions

between endotoxin and innate immune cells have been extensively studied in most

species in both clinical and experimental settings and are central to the current

understanding of the pathophysiology of SIRS and MODS. 26-29 Endotoxin is released

upon bacterial cell death or during logarithmic bacterial replication. Considering the vast

enteric flora of the horse, it is not surprising that the largest source of endogenous

endotoxin is that contained within the lumen of the GI tract. However, when the mucosa

is damaged by inflammation or ischemia, which frequently occurs with acute GI disease,

endotoxin can gain access to the systemic circulation where it interacts with PRRs.

11

Endotoxin can be found in the circulation and peritoneal fluid of horses with naturally-

occurring colic where plasma concentrations tend to correlate with increased

mortality.30,31 Endotoxin is not the only PAMP recognized in the pathophysiology of

equine SIRS. Other examples include: flagellin, peptidoglycan, lipoteichoic acid, double-

stranded viral RNA, and regions of unmethylated cpG DNA. The protein flagellin, a

component of flagellated bacteria (e.g. Salmonella sp.) and a PAMP, is reported to be

significantly increased in the systemic circulation of horses with acute GI disease

compared to healthy horses.32 In contrast to endotoxin which interacts with both

neutrophils and monocytes, flagellin solely induces a proinflammatory response in equine

neutrophils.33 The clinical significance of finding both of these PAMPs in the circulation

of horses with acute GI disease is that they provide evidence that mechanisms similar to

what are occurring in humans with sepsis also exist in equine colic.

SECTION III. PATHOPHYSIOLOGY OF CARDIOVASCULAR

DYSFUNCTION IN CRITICAL ILLNESS

Myocardial dysfunction in SIRS and sepsis

Discussion of the pathophysiology of cardiovascular system dysfunction in

critical illness provides an in-depth example of how SIRS and MODS affect individual

organs at a cellular level which ultimately impairs the ability of the organ to function

normally. Similarly to cells of the innate immune system, cardiomyocytes express PRRs

on the outer membrane, of which toll-like receptor 4 and 2 (TLR-4, TLR-2) remain the

best characterized, responding to circulating lipopolysaccharide and heat shock proteins,

respectively. When signal transduction occurs through these receptor mediated pathways

12

the response in the myocardium is most often NFκB activation which then signals gene

transcription and production of cytokines and other mediators such as inducible nitric

oxide synthase.34 Pro-inflammatory cytokines, specifically TNF-α, IL-1β and IL-6, alter

the normal function of the cardiomyocyte in sepsis resulting in clinically-recognized

myocardial depression.35 While there is not a universally accepted definition of

myocardial depression, reduced ejection fraction is the most commonly used clinical

correlate to describe the phenomenon.36 More recently other echocardiographic measures

of ventricular function have been used to describe myocardial depression including

reduced fractional shortening37 and diastolic dysfunction.35,38 Interestingly, in early study

of patients with septic shock, cardiac output was preserved or increased in all patients,

while survivors tended to have reduced ejection fractions and increased end-diastolic

volumes which suggested that there may be a protective role of myocardial depression.39

Proposed mechanisms of myocardial depression include cytokine mediated

negative inotropic effects36, abnormal calcium trafficking, increased nitric oxide

production from inducible nitric oxide synthase-2, mitochondrial dysfunction,

catecholamine toxicity, microcirculatory abnormalities and autonomic nervous system

dysfunction.23,35,36 While the regulatory protein, cTnI, is increased in septic patients with

myocardial depression, this does not appear to reflect widespread myocardial necrosis as

myocardial depression is reversible. In patients with acute myocardial infarct ischemic

injury can result first in membrane bleb formation40 which allows cTnI to leak out of the

cell prior to cellular necrosis. In contrast to patients with acute myocardial infarction and

eventual myocardial necrosis, patients with sepsis-related myocardial depression rarely

have evidence of necrosis, but may have inflammatory cell infiltrates, endothelial cell

13

edema, fibrin deposition in the microvasculature and mitochondrial swelling evident on

post-mortem examination.23,35 This reinforces the clinical impression that sepsis-induced

cardiac dysfunction can be fully reversible.

While myocardial depression may have evolved to decrease energy and oxygen

demands on the heart in states of severe disease, the consequence of reduced left

ventricular systolic function and in some cases reduced left ventricular diastolic function

owing to reduced ventricular compliance 38 impact the approach to therapy. Intravenous

fluid therapy administration is one of the mainstays of managing patients with SIRS,

sepsis and shock. Rapid and adequate restoration of circulating volume is one of the most

important interventions that can be applied to critically-ill patients and the benefits of

early volume replacement provide the evidence for the current recommendations in the

Surviving Sepsis Campaign.41 While, physiologic end-points for fluid resuscitation are

established in human patients, no such endpoints exist for the horse. The sequela of fluid

overload, particularly in patients with myocardial dysfunction, is an increased risk of

mortality.42

Role of the autonomic nervous system in cardiac dysfunction

In health, the sympathetic nervous system is responsible for accelerating heart rate

(positive chronotrope), maximizing contractility (positive inotropy), improving cardiac

relaxation (positive lusitropy), increasing the rate of conduction across the

atrioventricular node and causing vasoconstriction in venous capacitance and cutaneous

vessels.43 The complimentary branch of the autonomic nervous system, the

parasympathetic system, conversely causes a reduction in heart rate but has little to no

influence over contractility or ventricular relaxation owing to a difference in the

14

distribution of cholinergic nerve fibers compared to adrenergic fibers in the

myocardium.43 In critical illness, the autonomic nervous system, myocardial adrenergic

receptors, signal transduction pathways and ion channels in the cardiomyocyte are

susceptible to the effects of endotoxin and pro-inflammatory cytokines.44 These

molecules and mediators modulate efferent autonomic nervous system transduction at the

level of the brain45 but also manipulate the response at the level of the myocardium itself

through alterations in funny current ion channels (If) in the pacemaker cells which results

in narrowing of normal heart rate variability. Endotoxin is purported to cause “heart rate

stiffness” which manifests clinically as a state of unyielding tachycardia.44 Persistent,

inappropriate tachycardia is recognized as a feature of SIRS and MODS and is associated

with poor outcome. Tachycardia is detrimental to both the heart itself and the rest of the

body as it results in increased myocardial oxygen demand, restricts diastolic filling and

has the potential to result in cardiomyopathy.35

In summary, the effects of the innate immune system, changes in the peripheral

vasculature and autonomic nervous system result in a cardiac pump rendered ineffective,

poorly adaptive to changes in volume and pressure, and quite literally, marching to the

beat of its own drum. As one might expect, the implications of a dysfunctional cardiac

pump are far-reaching, affecting virtually every tissue bed and organ system. The

development of remote and sequential organ failure is not difficult to envision once the

heart ceases to function normally. The next section details how cardiovascular

dysfunction, as just discussed, might be assessed in horses.

15

SECTION IV: THE METHODOLOGY FOR ASSESSING THE

CARDIOVASCULAR SYSTEM IN THE ADULT HORSE

It should come as no surprise that clinical cardiovascular assessment is routinely

practiced and is standard of care in the human intensive care unit. While the continuous

measurement of heart rate (HR), mean arterial blood pressure (MAP) and oxygen

saturation by pulse oximetry (SpO2) have been standardly measured in all patients for

decades, these parameters do little on their own to drive decision making.46 In recent

years more sophisticated hemodynamic monitoring has become a mainstay, partially in

keeping with the monitoring necessary to determine end-points in goal directed therapy

which has gained universal support since the introduction and incorporation of the

recommendations derived in the Surviving Sepsis Campaign.47 In the contemporary

intensive care unit, hemodynamic monitoring is viewed as an important tool to both

identify and diagnose abnormalities associated with the underlying disease process as

well as a preemptive measure to detect potential abnormalities which allows intervention

prior to the onset of complications.48

Cardiac output estimation is the best available variable to assess overall

cardiovascular function.49 Cardiac output (CO), the product of heart rate and stroke

volume, is altered more quickly and to a greater capacity by changes in heart rate

compared to stroke volume. Cardiac output measurement, historically estimated via

thermodilution with a pulmonary arterial catheter, and more recently with peripheral

arterial catheterization for lithium dilution or with non-invasive echocardiography, is the

cornerstone for recognizing cardiovascular insufficiency and monitoring the response to

therapy in critically ill patients.41 Echocardiography offers advantages over both

16

thermodilution and lithium dilution and has become a widely accepted bed-side

procedure. In addition to providing a non-invasive estimate of CO, it is useful in the

assessment of various cardiac conditions and provides as assessment of left ventricular

systolic and diastolic function.48,50 Recent appraisals for hemodynamic monitoring in the

ICU favor an integrative approach, recognizing that monitoring one aspect of

cardiovascular system status conveys only one piece of the puzzle.48 The emphasis is

placed on cost-effective, non-invasive strategies in human hospitals and the same should

be sought for veterinary species.

Cardiovascular perturbations are often clinically apparent in horses with acute GI

disease and generally manifest as tachycardia with an increased incidence of ectopic

beats; the basis for these cardiac abnormalities are not well understood and are often

interpreted as evidence of pain, hypovolemia and SIRS. In order to understand the

meaning of persistent tachycardia and ectopic foci in the post-operative colic patient, an

in-depth and multifaceted approach must be adopted to understand how the heart is

functioning in relation to the rest of the organ systems. Tachycardia is a physiologic

response to hypovolemia, fever, pain and anemia. However, when tachycardia persists in

the face of restoration of intravascular volume, normothermia, amelioration of pain, or

correction of anemia, the physiologic state becomes pathologic and potentially harmful.

Methods for monitoring cardiovascular system status in horses have for the most part

been validated and utilized under resting, exercising or general anesthetic conditions.

Few studies have reported on the application of cardiovascular monitoring techniques in

critically ill adult horses. Measurement of CO perhaps best illustrates this deficit in

equine practice. Investigators have been interested in studying CO techniques in adult

17

horses for decades, mostly in the capacity of improving hemodynamics under general

anesthesia51,52 and understanding exercise physiology.53,54 Currently validated methods of

cardiac output measurement in horses include various indicator techniques

(thermodilution and lithium dilution) based on the Frick principle and Doppler

echocardiographic techniques.55 Applying any of the validated techniques to horses in a

clinical setting is wrought with limitations. For the indicator techniques, invasive

pulmonary or peripheral arterial catheterization is required. Ideally these catheters would

be maintained indwelling for continuous or serial measurement however, both of these

techniques have the potential for life threatening complications. The use of the

pulmonary arterial catheter has been shown to cause endocardial damage in horses,56

while potential risks of arterial catheterization include inadvertent arterial administration

of drugs or air into the arterial circulation which can have devastating consequences.

While Doppler echocardiographic techniques are noninvasive they require accurate

alignment with blood flow which is not always feasible in horses of all breeds and body

condition scores. Doppler echocardiographic techniques in general are more variable

between days and echocardiographers than 2-dimensional and M-mode

echocardiography.57 Both of these short-comings of Doppler echocardiography highlight

the need for a simpler, repeatable, reliable method of measurement of CO in horses for

routine clinical assessment. In anesthetized foals58, volumetric methods for determining

CO have been validated. These methods generally require standard views of the left

ventricle to obtain measurements of length and area that can then be incorporated into

mathematical equations that account for the shape of the ventricle. In anesthetized foals

the Bullet method had the best agreement with lithium dilution CO.58 However, this

18

finding cannot be directly extrapolated to adult horses owing to major disparities in

cardiac chamber size and the inability to obtain an apical view in the adult horse. To

reiterate, measuring CO in horses in a clinical setting, particularly with echocardiography

would provide clinically useful information related to overall cardiovascular function,

volume status, and with the knowledge of the arterial oxygen content, would enable

oxygen delivery (DO2) to the tissues to be calculated. Thus determining which

echocardiographic method of CO measurement in horses is most closely correlated with a

“gold standard” measurement of CO, lithium dilution, is investigated in Chapter 2 of this

dissertation.

Cardiac troponin (cTnI)

Despite the lack of data to objectively confirm hemodynamic disturbances in

horses with acute GI disease, such disturbances appear to be frequently encountered in

clinical patients. There is a widespread clinical impression that some horses with

strangulating GI lesions have evidence of cardiovascular shock59 and cardiac

arrhythmias60 following colic surgery. This impression has led researchers to evaluate the

cardiac biomarker, cTnI,61 or cTnI plus assessment of arrhythmias62 and functional

abnormalities based on echocardiography.63 While these studies provided convincing

evidence of myocardial injury and an association between increased cTnI concentrations,

the severity of the GI lesion and survival, dysfunction of the cardiovascular system was

not demonstrated. Therefore, a significant gap remains in interpreting the significance of

cTnI concentration in horses with colic.

In critically ill humans, multiple cardiac biomarkers (cTnT, cTnI, NT-proBNP)

have been correlated not only to mortality but to echocardiographic measures of cardiac

19

dysfunction.64,65 Cardiac troponin I is a regulatory protein of the thin actin filaments of

cardiac muscle that is released into the circulation as a result of acute myocardial cell

injury.66 It is a highly sensitive and specific biomarker of myocardial health. Despite its

excellent sensitivity and specificity to the heart, its utility in people is somewhat hindered

by the high incidence of acute coronary syndrome which also causes a dramatic increase

in serum troponin. Coronary heart disease is exceptionally rare in the horse, perhaps

making it an ideal biomarker of cardiac damage in critical illness. The association

between cardiac biomarkers in horses with acute GI disease and cardiac dysfunction

warrants further investigation and is a component of the studies presented herein.

Left ventricular function

While CO measurement provides one method of assessing ventricular function it

provides little information related to contractility.67 Fractional shortening (FS) and

ejection fraction are the primary methods for assessing contractility or inotropy, an

inherent characteristic of left ventricular function. Fractional shortening, the measure of

contractility most commonly used in horses, is an M-mode derived measurement of the

difference of left ventricular volume in end diastole and end systole relative to the left

ventricular end diastolic volume FS= [(LVIDd-LVIDs)/LVIDd] X 100.68 Nath and

colleagues assessed FS post-operatively in horses with acute GI disease and found no

significant differences in FS between horses with surgical colic, medical colic or control

horses.63 However, not every horse underwent echocardiography so the lack of a

statistical difference may have been attributable to a lack of power. More recently, left

ventricular systolic and diastolic dysfunction was reported in a group of horses with acute

colic and SIRS. 69 In this study, non-surviving horses had lower left ventricular stroke

20

volume index and higher pre-ejection period to ejection time ratio (PEP/ET) of Doppler

aortic flow as evidence of systolic dysfunction. The same horses also exhibited higher

peak early diastolic filling velocity to peak early diastolic myocardial velocity ratio

(E/Em), suggesting diastolic dysfunction. The evidence for both systolic and diastolic

dysfunction in horses with SIRS is consistent with the clinical picture of septic

cardiomyopathy in people.35,38

Mean arterial pressure (MAP), Central venous pressure (CVP) and Pressure

adjusted heart rate (PAR)

Mean arterial pressure may be measured in adult horses with indirect or direct

methods. Indirect oscillometric methods are clinically accepted, although few individual

units have been critically assessed in the horse. The optimal site for measurement is the

tail head (coccygeal artery) and the importance of proper cuff circumference is

highlighted in the literature.70,71 Mean arterial pressure (MAP) provides a way to estimate

systemic vascular resistance when CO and CVP are known. Measuring MAP alone is

insufficient for global hemodynamic assessment.

The measurement of CVP has been described in horses in mostly research

applications72 and provides an estimate of preload in humans. While CVP estimation in

horses seems to be responsive to acute fluid loss and replacement73 a systematic review

of the literature in people concluded that CVP should not be used to guide clinical

decision making for fluid therapy because of its inability to detect a response to fluid

challenge.74 Despite the results of the 2008 meta-analysis, CVP has been retained as an

end-point of early-goal-directed therapy in the Surviving Sepsis Campaign.47 In adult

horses, it is important to note that the reliability and reproducibility of CVP

21

measurements are dependent on maintaining a neutral head position.75

In adult horses with naturally occurring colic, systemic hypotension occurs in

horses in the later phases of endotoxemia due to decreased systemic vascular resistance in

response to prostaglandin release.76 Hypotension, need for inotrope support and increased

plasma lactate concentration are frequently reported descriptors of circulatory failure in

critically ill humans.66 Correlation of hypotension with death in human patients was the

rationale for its inclusion in the Sequential Organ Failure Assessment (SOFA) score, used

for clinical diagnosis for MODS.ref Mean arterial blood pressure values are typically at

their worst prior to therapeutic intervention. Changes in blood pressure are highly

susceptible to transient changes in fluid therapy or inotrope support, thus, it is argued that

blood pressure is a treatment-dependent variable that does not reflect the entire spectrum

of cardiac function. From this argument, a composite measure, called pressure-adjusted

heart rate (PAR) that corrects for physiologic support by calculation of the product of the

heart rate by the ratio of the CVP to MAP was introduced. This variable has shown

incremental correlation with ICU mortality in people and is thus included in another

commonly used scoring system for organ dysfunction, the MODS score.77 The use of

PAR has not been evaluated in the horse and is investigated in Chapter 5.

Heart rate variability

Autonomic dysfunction is an important aspect of the pathogenesis of myocardial

depression in sepsis. Analysis of heart rate variability (HRV) provides a method to

estimate autonomic modulation of the heart. The HRV describes both short-term

variations between consecutive heart beats and long-term variations in cardiac cyclical

activity providing a valuable tool to characterize the influences of the sympathetic and

22

parasympathetic nervous systems on the sinoatrial node.78 In general, decreases in HRV

variables reflect a shift toward sympathetic dominance. There are numerous HRV

parameters that can be easily determined using specifically designed software that

integrates and interprets digitally stored telemetric electrocardiography data. Several

studies of HRV have been performed in the horse under various conditions including

transport, laminitis, and pregnancy and have demonstrated correlation of HRV to other

measures of stress and reduced vagal tone, such as serum cortisol concentration.79-82

With the ease of obtaining HRV data and its excellent correlation to measures of

cardiovascular health, MODS, and outcome in critically ill people in which acute

coronary disease is not the primary disease process,79,83-86 it appears to be an ideal

parameter to monitor in critically ill patients. To further advance emergency and critical

care of horses, it will be increasingly important to identify specific and sensitive

indicators of subclinical organ dysfunction, including myocardial damage. The HRV in

horses with acute GI disease is investigated herein and presented in Chapter 4.

SECTION V: CURRENT EVIDENCE FOR MODS IN THE HORSE

In the horse, acute GI disease or ‘colic’ is caused by a spectrum of physiologic,

anatomic or inflammatory abnormalities that are manifested by clinical and

clinicopathological disturbances that are most often non-specific but in some cases may

reflect the severity or duration of the insult. Among the most severe of the acute GI

diseases are those that result in strangulating or ischemic lesions of the intestine. Both the

small and large intestine can become strangulated and are at risk of necrosis of one or

more layers of the intestinal wall; usually beginning with the loss of the luminal mucosal

23

layer. Untreated, this leads to GI perforation, septic peritonitis and death. Theoretically,

horses that have mechanical GI obstructions that culminate in severe distension and

certainly those with inflammatory conditions such as enteritis or colitis, also may have

disturbances in the mucosal barrier. A breach of the mucosal barrier not only allows

translocation of bacterial products but also implies there has been architectural disruption

and loss of enterocytes (necrosis or apoptosis). This in turn results in leakage of alarmins

from damaged or necrotic cells. Indeed, as mentioned previously, circulating endotoxin

(and probably other bacterial components e.g. flagellin) are present in the plasma of some

horses that present with acute colic and are associated with increased risk of death.31,87

As mentioned previously, the most common reason for referral of horses to

tertiary care centers on an emergency basis is naturally-acquired GI disease or colic.1 In

a study by Epstein and colleagues, 27/95 (28.4%) horses that presented for acute GI

disease fulfilled the criteria for SIRS.88 Our current understanding of the pathophysiology

of MODS suggests that SIRS predisposes to MODS which highlights the need to

investigate MODS in horses. While the majority of horses that have colic are treated

medically, those that require surgical intervention are often faced with post-operative

morbidity and mortality that may or may not be related to the surgical procedure and

include but are not limited to ileus, recurrent colic, surgical site infection, and rarely

septic peritonitis. Less commonly reported are dysfunctions in whole body metabolism,

organs (liver, kidney) and systems (respiratory, endocrine, musculoskeletal and

cardiovascular). The remainder of this section will provide data that supports organ

dysfunction in horses with colic.

24

Respiratory system

Post-operative complications in horses related to the respiratory tract include

pulmonary edema, aspiration pneumonia, hematogenous pneumonia, pleuropneumonia

and pulmonary thromboembolism.89-91 Aspiration pneumonia was a reported

complication in 6.8 % of horses with enteritis undergoing exploratory laparotomy.92

Experimentally TLR-4 activation via endotoxin induces inflammation and increased

pulmonary vascular permeability in mice and in horses.93-95 Under the proper set of

circumstances any one of the above mentioned post-operative complications could result

in respiratory distress due to pulmonary dysfunction. This phenomenon is best described

in adult horses by the syndromes of acute lung injury and acute respiratory distress

syndrome which are clinical manifestations of acute severe lung disease.96 The incidence

of acute lung injury and acute respiratory distress syndrome in adult horses with acute GI

disease is currently unknown. The pathogenesis of pneumonia related to acute GI disease

may be secondary to aspiration of gastric contents under general anesthesia, iatrogenic

administration of fluids or therapies from nasogastric intubation, ventilator injury, SIRS

or hematogenous seeding of enteric bacteria into the pulmonary tree.

Renal

Acute kidney injury is common in humans with severe SIRS, sepsis and

MODS.97,98 and has been documented with some frequency (11%) in septic dogs.99

Azotemia is a common admission finding for horses that present with GI disease and is

considered an independent risk factor for survival in horses with strangulating small

intestinal lesions.100 In a retrospective study examining horses that presented to a tertiary

referral center for acute GI disease, the incidence of azotemia at presentation, defined as a

25

serum creatinine concentration > 3 mg/dL, was 7.9% (79/1000 horses) with persistent

azotemia (> 72 hours duration) occurring in 2.6% (26/1000) of horses. The

pathophysiology for renal injury in horses with acute GI disease is multifactorial, and

pre-renal azotemia was proposed as the cause of azotemia at admission in the majority of

the horses examined. However, almost 3% of the horses had persistent azotemia which

suggests intrinsic renal dysfunction. The finding that horses with persistent azotemia

were three times as likely to die or be euthanized compared to those with azotemia that

resolved by 72 hours supports this assumption.101 At the time of examination, horses with

acute GI disease are often hypovolemic which results in reduced renal perfusion. This is

further complicated by the routine use of nephrotoxic drugs in equine practice, such as

flunixin meglumine and gentamicin. However, the pathogenesis of renal injury in sepsis

is complex. A cecal-ligation and puncture model of polymicrobial sepsis in mice

demonstrated the role of signaling through MyD88 (myeloid differentiation factor 88) in

the pathogenesis of acute kidney injury in sepsis whereby knockout MyD88 -/- mice had

improved survival, complete protection of renal tubular cells with absence of acute

tubular necrosis, maintained vascular permeability similar to that of controls and had

decreased migration of neutrophils into renal tissue.102

Coagulopathy

Reports of hemostatic dysfunction in horses with acute GI disease are

commonplace in the equine literature, while clinical manifestations of coagulopathy are

less common.88,103-105 The coagulopathy in horses with acute GI disease is believed to

involve activation of tissue factor on blood monocytes and the endothelium that occurs in

response to pro-inflammatory cytokines.106 Prolongation in prothrombin time occurs in

26

horses with colic and this finding is correlated with mortality.107 At admission for colic,

prolongation of prothrombin time, activated partial thromboplastin time and

thrombocytopenia occurs in approximately 25-80%, 43-100% and 29% of horses

respectively, presented for colic.103,107-109 The prevalence of clinicopathologic

coagulopathy is not surprising since hemostasis is part of the defense mechanism of the

host and its activation is intimately linked with the inflammatory response. However,

pathologic sequelae of coagulopathy exist, as evidenced by the finding of microvascular

thrombi in the tissues of foals with organ failure and sepsis110 and adult horses with

severe GI disease.111

Gastrointestinal

Gastrointestinal dysfunction is common in critically-ill horses and in fact acute GI

disease is often the cause of hospitalization.1 Post-operative ileus, is second only to post-

operative pain, as the most commonly reported short-term GI related complication in

horses that undergo exploratory laparotomy.59 The GI tract was not included in human

scoring systems for MODS because of the observation that it was difficult to find an

objective measure of GI function.112 The pathophysiology of ileus is complex and

multifactorial. At present the development of nasogastric reflux is typically the sole

criteria used to define ileus which may be misleading and result in over-diagnosis in

horses. This is because nasogastric reflux also occurs as a result of other conditions such

as enteritis or complete or partial mechanical obstruction of the small intestine.113,114

Despite these concerns, post-operative ileus continues to be defined based on the

presence of nasogastric reflux. In a recent survey of Diplomates of the European Colleges

of Veterinary Internal Medicine and Veterinary Surgeons, post-operative ileus was most

27

often defined in practice as the presence of any nasogastric reflux in the post-operative

period, however, many responses also included specific volumes that they regarded as

significant.115 Other reports have included even more stringent definitions including >8L

at any one intubation or > 20 L in a 24 hour period.116,117 Large intestinal ileus occurs as

well, but presents an even greater challenge to define in the horse. Large intestinal ileus

might be suspected upon palpation per rectum, absence of GI sounds, gross abdominal

distension, or transabdominal ultrasonographic findings that rule out other causes of

distension.

Hepatic

Measurement of serum bile acid (SBA) concentrations is considered a test of

hepatic function, in contrast to measurement of liver-specific enzyme activities. In the

horse, unlike total bilirubin, SBA are not affected by short-term fasting but will increase

after prolonged fasts of at least 3 days.118 SBA concentrations have been evaluated in

horses with acute naturally-occurring GI disease and markedly increased admission SBAs

were associated with non-survival.119 Plasma ammonia concentrations, while also a

measure of hepatic function, are a less attractive candidate as a criterion of liver

dysfunction since they require special handling and immediate measurement. In one

study in horses with colic, plasma ammonia concentration was normal in all but two

horses, in which it was mildly increased. No association was detected for plasma

ammonia concentration and survival.119 In addition, intestinal hyperammonemia occurs

despite a functional liver, making it non-specific for liver function in horses with GI

disease.120

28

Neurologic/Mentation

In the human MODS and SOFA score, the Glasgow Coma Scale is used for the

evaluation of neurologic function. Developed by Teasdale and colleagues in 1974, the

Glasgow Coma Scale assesses eye, verbal and motor responses on a scale from a fully

awake, aware person to a person in a comatose state.121 Unfortunately the Glasgow Coma

Scale is inappropriate for equine patients due to the lack of the verbal component and the

inability to perform repeatable provocative testing to assess pain responses. In addition,

the scale was originally developed for head trauma patients, and while found to be

applicable to critically ill human patients, the scale in many ways is inappropriate for

equine patients. In the literature, there are few clinical reports of neurologic evaluation or

neurologic dysfunction in horses with colic. By far, reports of neurologic signs related to

intestinal hyperammonemia predominate.120,122 While the majority of the horses

presented in the case series have inflammatory gastrointestinal lesions such as colitis or

enteritis, horses with medically and surgically treated colic were also represented. The

main clinical signs associated with hyperammonemia in these cases were depressed

mentation, head pressing, ataxia, central blindness and erratic behavior.120 In large animal

species, demeanor, and in particular clinical signs of depressed mentation, are among the

most commonly encountered evidence of abnormal neurologic function. In fact,

demeanor and posture correlate well to strong ion gap and degree of D-lactic acidosis in

calves with diarrhea123,124 while in horses with experimentally induced SIRS depression

is one of the most consistent clinical signs reported.125,126 Therefore, in critically ill

animals assessing general demeanor and behavior might provide useful information

related to neurologic function. A behavior or posture assessment might be a useful

29

alternative to a complete neurologic evaluation when attempting to determine neurologic

function related to systemic inflammation. A numerical rating scale of behavior was

previously described and applied to horses recovering from exploratory laparotomy and

not only provides a way to assess pain behaviors but also specifically rates behavior and

posture which reflect the degree of the animals awareness to their surroundings.127 While

the limitations of this scale are the lack of specific neurologic deficits such as ataxia,

cranial nerve signs or seizure like activity, the score would indirectly reflect the more

subtle changes in mentation and certainly would account for abnormal responses to

external stimuli which would coincide with seizure like activity or central blindness.

Musculoskeletal

In horses, normal musculoskeletal function is vital. The development of laminitis

and muscle injury have been the subjects of investigation in horses with acute GI disease.

Despite the fact that the pathophysiology of laminitis remains incompletely understood,

the majority of what is known is based on models that induce severe systemic

inflammation through carbohydrate overload in the GI tract.128 Experimental endotoxin

infusion129,130 fails to cause complete lamellar failure leading to studies of other factors as

possible initiators. 131,132 While the incidence of laminitis in the post-operative period

appears low, ranging between 0.4% and 3%,7,59,88 endotoxemia was shown to be a risk

factor for laminitis in horses presented to referral centers in which horses with clinical

and laboratory evidence of endotoxemia had a 5-fold increased risk of developing acute

laminitis.133 Despite the low incidence of horses developing laminitis in the post-

operative period, this complication has profound implications for an individual horse as it

often at least career ending or results in euthanasia.

30

Krueger and colleagues evaluated the clinical utility of muscle enzyme activity in

horses that presented for acute colic and found that increased activity of creatine kinase

(CK) was associated with GI lesion type and outcome.134 This group found that horses

with a CK of > 470 U/L at hospital admission were at a 2.4-times increased risk of not

surviving to hospital discharge and were 2.6-times more likely to be diagnosed with a

strangulating intestinal lesion. In addition, they proposed that the increase in CK activity

may in part be due to endotoxin-mediated injury and might merely reflect muscle damage

from the trauma associated with colicky behavior, intramuscular injections or transport.

This notion is supported by a single case series in the literature that describes

myonecrosis in horses with colic not directly related to trauma.135 Despite an unclear

underlying pathophysiology, both laminitis and increased CK activity are associated with

poor outcomes in horses with acute surgical colic.

Cardiovascular System

The cardiovascular system was discussed in detail previously. In general, the

most commonly reported cardiovascular abnormalities in horses with acute GI disease

include tachycardia, increased hematocrit and abnormal mucous membranes (color,

capillary refill time)136-138 which taken together indicate hemoconcentration, hypovolemia

and perfusion deficits. The short-comings of using the above criteria for assessment of

the status of the cardiovascular system include the co-existence of many additional

factors that influence the heart rate (e.g. underlying cardiac disease, pain, stress, anemia,

inflammation, drugs) and hematocrit (e.g. splenic contraction, blood loss, anemia, breed,

sex, age, underlying chronic disease) beside hypovolemia and the subjective nature of

examining the oral mucous membranes. As previously mentioned there are more

31

sophisticated, non-invasive methods that would provide specific information regarding

cardiovascular function, these being echocardiography, electrocardiography and cTnI

measurement.

SECTION VI: THE DEVELOPMENT OF MULTIPLE ORGAN DYSFUNCTION

SCORES IN HUMANS AND SEVERITY SCORES IN VETERINARY SPECIES

In the last 30 years, critical illness severity scores have been introduced into

intensive care units with the intended purpose of providing a method to determine illness

severity and the associated risk of mortality.139 Organ failure scores, on the other hand

were designed to describe the degree of organ dysfunction rather than predict survival.139

By design, organ dysfunction scores reflect a continuum of organ performance from

functional to failure based on serial assessment of clinical and laboratory data. The

scoring systems for organ dysfunction assess multiple organs systems, in human

medicine, the cardiovascular, renal, hepatic, neurologic, hemostatic, and respiratory

systems comprise the score. Within each organ or system, specific functional criteria are

weighted in order to establish a score that can accurately reflect a range of dysfunction.

Scoring individual patients provides an estimate of organ dysfunction severity that can be

continually reassessed throughout hospitalization, enabling the physician to monitor

response to therapy. While not designed to predict survival, the MODS score devleloped

by Marshall and colleagues in 1994 does correlate with outcome.112

In order to develop organ dysfunction scoring systems various approaches were

taken to identify the best criteria to describe dysfunction in a given organ system. First,

the criteria for an ideal descriptor of organ dysfunction were decided upon with the

32

following characteristics in mind: organ function descriptors must represent an organ

system on a continuum from functional to insufficient to failure that accurately represents

the clinical syndrome, they must be easily and readily measured in a heterogenous group

of patients and they must be reproducible and responsive to clinically significant changes

in patient status.112,140 Marshall and colleagues112 selected criteria for each organ system

based on a literature review and patient data from a surgical ICU. In contrast, the sepsis

related organ failure assessment (SOFA) score relied upon empirical selection of criteria

by a panel of experts in the European Society of Critical Care Medicine.140 A third score,

the logistic organ dysfunction score (LODS), which assigns weights to the different organ

systems, was developed from multiple logistic regression analysis which selected

variables based on a large patient data base from 137 ICUs in several countries.141 Once

developed, all three scoring systems were validated prospectively on patients in the

surgical112 or both the surgical or medical ICU142 and overtime have been validated in

numerous groups of patients.143-148 Evaluation of these scores over the last couple of

decades has repeatedly demonstrated no significant differences compared to updated

physiologic based severity scores (Acute Physiology and Chronic Health Evaluation,

APACHE III) in predicting outcome in critically-ill patients. While not the original

intended purpose of the scores, their utility seems to extend beyond determination of

organ severity.149-151

In horses, a validated scoring system does not exist for describing organ

dysfunction. Proposed criteria have been published in text books but are largely

extrapolated from criteria used in the human SOFA and MODS scores.152 Extrapolation

from human criteria poses potential problems related to differences in physiology (e.g.,

33

bilirubin is used as a criterion for liver dysfunction in humans while this is increased in

fasting horses with normal liver function), primary disease processes (e.g., sepsis in

humans versus colic in horses) and treatment modalities (e.g., many human patients are

on ventilator therapy whereas adult horses are not). There have been attempts to develop

severity scores for horses.138,153 The colic severity score proposed by Furr and colleagues

was modeled after the original APACHE severity score for critically ill human patients,

with the intention of providing a simple score that could predict outcome in horses with

colic, reflect the severity of the horse’s condition and serve as a management tool for

equine clinicians.138 Data was collected prospectively at admission from horses with colic

and a logistic regression model was used to identify the variables with the greatest

association with outcome. Cut-points for these variables were then established and given

a designated score from 0 to 4. The variables that were retained for the colic severity

score included pulse rate, peritoneal total protein concentration, blood lactate

concentration and mucous membrane appearance. The score was then validated

prospectively with a group of 71 horses that presented for colic. A score of ≤ 7 was

predictive of survival while a score of ≥ 8 was predictive of death. This score did not gain

clinical acceptance and while it had excellent positive predictive value (100%) the

negative predictive value was 91% which meant that some horses predicted to live would

die. A similar attempt was made by Grulke and colleagues, however they developed both

a ‘gravity score’ and ‘shock score’ which were meant to reflect the type and severity of

the GI lesion (based on rectal palpation, abdominal distension, borborygmi, and pain) and

the severity of hemodynamic compromise (based on HR, RR, systolic arterial pressure,

packed cell volume, blood lactate, and blood urea nitrogen concentration) respectively.153

34

Interestingly, the overall survival rate in their study was 54% which is much lower than

overall colic survival rates reported more recently.154 The utility of both of these scoring

systems are questioned. Aggressive management in horses with hemodynamic instability,

improved surgical techniques and changes in medical treatment in horses with colic have

been credited with declining mortality rates. Therefore, many of the variables utilized in

the above scoring systems that reflect cardiovascular compromise at admission are

unlikely to provide accurate prognostic data in the contemporary equine ICU. As an

example, admission heart rate, a variable included in the colic severity score and shock

score, is a parameter that routinely is cited as a poor prognostic indicator in horses with

colic137 yet it is rarely retained in logistic regression models that predict outcome.155,156

Potential reasons for this are the numerous causes of tachycardia including pain,

hypovolemia, excitement, inflammation and primary cardiac disease.

The syndrome of MODS is recognized clinically in horses,69,157 however

appropriate criteria to describe this clinical phenomenon are lacking. In the absence of a

method to detect a clinical range of organ dysfunction in critically-ill equine patients,

clinicians and researchers alike are at an impasse when it comes to improving outcome in

post-operative and critically-ill equine cases. Similar to what was observed by Baue, who

recognized MODS as a consequence of the advances of medical care in people in the

early 1970’s,16 veterinarians are now faced with similar challenges related to organ

dysfunction in equine patients that would not have survived 20 years ago. Chapter 6

provides an approach to describe MODS in horses with acute GI disease.

35

REFERENCES

1. Viljoen A, Saulez MN, Donnellan CM, et al. After-hours equine emergency

admissions at a university referral hospital (1998-2007): causes and interventions. J S Afr

Vet Assoc 2009;80:169-173.

2. Southwood LL, Dolente BA, Lindborg S, et al. Short-term outcome of equine

emergency admissions at a university referral hospital. Equine Vet J 2009;41:459-464.

3. aphis.usda.gov. Trends in Equine Mortality, 1998-2005. 2007.

4. Suthers JM, Pinchbeck GL, Proudman CJ, et al. Survival of horses following

strangulating large colon volvulus. Equine Vet J 2013;45:219-223.

5. USDA. Equine 2005, Part I: Baseline Reference of Equine Health and Management,

2005.

6. USDA. Equine 2005, Part II: Changes in the U.S. Equine Industry, 1998-2005. In:

USDA-APHIS-VS C, ed. Fort Collins, CO. 2006.

7. Davis W, Fogle CA, Gerard MP, et al. Return to use and performance following

exploratory celiotomy for colic in horses: 195 cases (2003-2010). Equine Vet J

2013;45:224-228.

8. Hallowell GD, Corley KT. Preoperative administration of hydroxyethyl starch or

hypertonic saline to horses with colic. J Vet Intern Med 2006;20:980-986.

9. Pantaleon LG, Furr MO, McKenzie HC, 2nd, et al. Cardiovascular and pulmonary

effects of hetastarch plus hypertonic saline solutions during experimental endotoxemia in

anesthetized horses. J Vet Intern Med 2006;20:1422-1428.

36

10. Fielding CL, Magdesian KG. A comparison of hypertonic (7.2%) and isotonic

(0.9%) saline for fluid resuscitation in horses: a randomized, double-blinded, clinical

trial. J Vet Intern Med 2011;25:1138-1143.

11. Brown JA, Holcombe SJ, Southwood LL, et al. End-to-Side Versus Side-to-Side

Jejunocecostomy in Horses: A Retrospective Analysis of 150 Cases. Vet Surg

2015;44:527-533

12. Bone RC, Balk RA, Cerra FB, et al. Definitions for sepsis and organ failure and

guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus

Conference Committee. American College of Chest Physicians/Society of Critical Care

Medicine. Chest 1992;101:1644-1655.

13. Iskander KN, Osuchowski MF, Stearns-Kurosawa DJ, et al. Sepsis: multiple

abnormalities, heterogeneous responses, and evolving understanding. Physiol Rev

2013;93:1247-1288.

14. Osuchowski MF, Craciun F, Weixelbaumer KM, et al. Sepsis Chronically in MARS:

Systemic Cytokine Responses Are Always Mixed Regardless of the Outcome,

Magnitude, or Phase of Sepsis. J Immunol 2012;189:4648-56.

15. Rittirsch D, Flierl MA, Ward PA. Harmful molecular mechanisms in sepsis. Nat Rev

Immunol 2008;8:776-787.

16. Baue AE. Multiple, progressive, or sequential systems failure. A syndrome of the

1970s. Arch Surg 1975;110:779-781.

17. Bone RC, Sibbald WJ, Sprung CL. The ACCP-SCCM consensus conference on

sepsis and organ failure. Chest 1992;101:1481-1483.

18. Mizock BA. The multiple organ dysfunction syndrome. Dis Mon 2009;55:476-526.

37

19. De Backer D, Orbegozo Cortes D, Donadello K, et al. Pathophysiology of

microcirculatory dysfunction and the pathogenesis of septic shock. Virulence 2014;5:73-

79.

20. Mehta NJ, Khan IA, Gupta V, et al. Cardiac troponin I predicts myocardial

dysfunction and adverse outcome in septic shock. Int J Cardiol 2004;95:13-17.

21. Balk RA. Pathogenesis and management of multiple organ dysfunction or failure in

severe sepsis and septic shock. Crit Care Clin 2000;16:337-352, vii.

22. Acharya SP, Pradhan B, Marhatta MN. Application of "the Sequential Organ Failure

Assessment (SOFA) score" in predicting outcome in ICU patients with SIRS. Kathmandu

Univ Med J (KUMJ) 2007;5:475-483.

23. Celes MR, Prado CM, Rossi MA. Sepsis: Going to the Heart of the Matter.

Pathobiology 2012;80:70-86.

24. Lucas S. The Autopsy Pathology of Sepsis-Related Death, Severe Sepsis and Septic

Shock-Understanding a Serious Killer. In: Fernandez R, ed. InTech; 2012.

25. Mayr VD, Dunser MW, Greil V, et al. Causes of death and determinants of outcome

in critically ill patients. Crit Care 2006;10:R154.

26. Niederbichler AD, Hoesel LM, Ipaktchi K, et al. Burn-induced heart failure:

lipopolysaccharide binding protein improves burn and endotoxin-induced cardiac

contractility deficits. J Surg Res 2011;165:128-135.

27. Xiong J, Wang Y, Zhu Z, et al. Expression and significance of toll-like receptor 2,4

of peripheral blood mononuclear cells in acute abdomen patients associated with

systemic inflammatory response syndrome. J Huazhong Univ Sci Technolog Med Sci

2006;26:570-572.

38

28. Wolfard A, Kaszaki J, Szabo C, et al. Prevention of early myocardial depression in

hyperdynamic endotoxemia in dogs. Shock 2000;13:46-51.

29. Liu S, Zhu X, Liu Y, et al. Endotoxin tolerance of adrenal gland: attenuation of

corticosterone production in response to lipopolysaccharide and adrenocorticotropic

hormone. Crit Care Med 2011;39:518-526.

30. Barton MH, Collatos C. Tumor necrosis factor and interleukin-6 activity and

endotoxin concentration in peritoneal fluid and blood of horses with acute abdominal

disease. J Vet Intern Med 1999;13:457-464.

31. Senior JM, Proudman CJ, Leuwer M, et al. Plasma endotoxin in horses presented to

an equine referral hospital: correlation to selected clinical parameters and outcomes.

Equine Vet J 2011;43:585-591.

32. Kwon S. Flagellin and antibodies directed to flagellin in horses and foals. In:

Differential sensitivity of equine leukocyte populations to TLR ligands. University of

Georgia 2011. http://purl.galileo.usg.edu/uga_etd/kwon_so-young_201112_phd

33. Moore JN, Vandenplas ML. Is it the systemic inflammatory response syndrome or

endotoxemia in horses with colic? Vet Clin North Am Equine Pract 2014;30:337-351,

vii-viii.

34. Rudiger A, Dyson A, Felsmann K, et al. Early functional and transcriptomic changes

in the myocardium predict outcome in a long-term rat model of sepsis. Clin Sci (Lond)

2013;124:391-401

35. Rudiger A, Singer M. Mechanisms of sepsis-induced cardiac dysfunction. Crit Care

Med 2007;35:1599-1608.

39

36. Antonucci E, Fiaccadori E, Donadello K, et al. Myocardial depression in sepsis: from

pathogenesis to clinical manifestations and treatment. J Crit Care 2014;29:500-511.

37. Dickinson AE, Rozanski EA, Rush JE. Reversible myocardial depression associated

with sepsis in a dog. J Vet Intern Med 2007;21:1117-1120.

38. Landesberg G, Gilon D, Meroz Y, et al. Diastolic dysfunction and mortality in severe

sepsis and septic shock. Eur Heart J 2012;33:895-903.

39. Parker MM, Shelhamer JH, Bacharach SL, et al. Profound but reversible myocardial

depression in patients with septic shock. Ann Intern Med 1984;100:483-490.

40. Hickman PE, Potter JM, Aroney C, et al. Cardiac troponin may be released by

ischemia alone, without necrosis. Clin Chim Acta 2010;411:318-323.

41. Dellinger RP, Levy MM, Rhodes A, et al. Surviving Sepsis Campaign: international

guidelines for management of severe sepsis and septic shock, 2012. Intensive Care Med

2013;39:165-228.

42. Sirvent JM, Ferri C, Baro A, et al. Fluid balance in sepsis and septic shock as a

determining factor of mortality. Am J Emerg Med 2015;33:186-189.

43. Lymperopoulos A. Physiology and pharmacology of the cardiovascular adrenergic

system. Front Physiol 2013;4:240.

44. Werdan K, Schmidt H, Ebelt H, et al. Impaired regulation of cardiac function in

sepsis, SIRS, and MODS. Can J Physiol Pharmacol 2009;87:266-274.

45. Sharshar T, Gray F, Lorin de la Grandmaison G, et al. Apoptosis of neurons in

cardiovascular autonomic centres triggered by inducible nitric oxide synthase after death

from septic shock. Lancet 2003;362:1799-1805.

40

46. Pinsky MR. Hemodynamic evaluation and monitoring in the ICU. Chest

2007;132:2020-2029.

47. Dellinger RP, Levy MM, Rhodes A, et al. Surviving sepsis campaign: international

guidelines for management of severe sepsis and septic shock: 2012. Crit Care Med

2013;41:580-637.

48. Vincent JL, Rhodes A, Perel A, et al. Clinical review: Update on hemodynamic

monitoring--a consensus of 16. Crit Care 2011;15:229-237.

49. Tibby SM, Murdoch IA. Measurement of cardiac output and tissue perfusion. Curr

Opin Pediatr 2002;14:303-309.

50. Noritomi DT, Vieira ML, Mohovic T, et al. Echocardiography for hemodynamic

evaluation in the intensive care unit. Shock 2010;34 Suppl 1:59-62.

51. Muir WW, Skarda RT, Milne DW. Estimation of cardiac output in the horse by

thermodilution techniques. Am J Vet Res 1976;37:697-700.

52. Linton RA, Young LE, Marlin DJ, et al. Cardiac output measured by lithium

dilution, thermodilution, and transesophageal Doppler echocardiography in anesthetized

horses. Am J Vet Res 2000;61:731-737.

53. Wilkins PA, Boston RC, Gleed RD, et al. Comparison of thermal dilution and

electrical impedance dilution methods for measurement of cardiac output in standing and

exercising horses. Am J Vet Res 2005;66:878-884.

54. Durando MM, Corley KT, Boston RC, et al. Cardiac output determination by use of

lithium dilution during exercise in horses. Am J Vet Res 2008;69:1054-1060.

55. Corley KT, Donaldson LL, Durando MM, et al. Cardiac output technologies with

special reference to the horse. J Vet Intern Med 2003;17:262-272.

41

56. Schlipf JW DC, Getzy DM et al. . Lesions associated with cardiac catheterization

and thermodilution cardiac output determination in horses. In: Proceedings of the 5th

International Congress of Veterinary Anesthesia; 1994:71-74.

57. Young LE, Scott GR. Measurement of cardiac function by transthoracic

echocardiography: day to day variability and repeatability in normal Thoroughbred

horses. Equine Veterinary Journal 1998;30:117-122.

58. Giguere S, Bucki E, Adin DB, et al. Cardiac output measurement by partial carbon

dioxide rebreathing, 2-dimensional echocardiography, and lithium-dilution method in

anesthetized neonatal foals. J Vet Intern Med 2005;19:737-743.

59. Mair TS, Smith LJ. Survival and complication rates in 300 horses undergoing

surgical treatment of colic. Part 2: Short-term complications. Equine Vet J 2005;37:303-

309.

60. Reimer JM, Reef VB, Sweeney RW. Ventricular arrhythmias in horses: 21 cases

(1984-1989). J Am Vet Med Assoc 1992;201:1237-1243.

61. Radcliffe RM, Divers TJ, Fletcher DJ, et al. Evaluation of L-lactate and cardiac

troponin I in horses undergoing emergency abdominal surgery. J Vet Emerg Crit Care

(San Antonio) 2012;22:313-319.

62. Diaz OM, Durando MM, Birks EK, et al. Cardiac troponin I concentrations in horses

with colic. J Am Vet Med Assoc 2014;245:118-125.

63. Nath LC, Anderson GA, Hinchcliff KW, et al. Clinicopathologic evidence of

myocardial injury in horses with acute abdominal disease. J Am Vet Med Assoc

2012;241:1202-1208.

42

64. Landesberg G, Jaffe AS, Gilon D, et al. Troponin elevation in severe sepsis and

septic shock: the role of left ventricular diastolic dysfunction and right ventricular

dilatation*. Crit Care Med 2014;42:790-800.

65. Landesberg G, Levin PD, Gilon D, et al. Myocardial dysfunction in severe sepsis and

septic shock - no correlation with inflammatory cytokines in real-life clinical setting.

Chest 2015;148:93-102.

66. Hunter J, Doddi M. Sepsis and the Heart. Brit J Anaesthesia 2010;104:3-10.

67. Reef VM, C. Cardiology of the Horse Second Edition Toronto: Saunders; 2010.

68. Reef VB. Equine Diagnostic Ultrasound. In: Cardiovascular Echocardiography.

Philadelphia: WB Saunders; 1998:215-272.

69. Borde L, Amory H, Grulke S, et al. Prognostic value of echocardiographic and

Doppler parameters in horses admitted for colic complicated by systemic inflammatory

response syndrome. J Vet Emerg Crit Care (San Antonio) 2014;24:302-310.

70. Latshaw H, Fessler JF, Whistler SJ, et al. Indirect measurement of mean blood

pressure in the normotensive and hypotensive horse. Equine Vet J 1979;11:191-194.

71. Tearney CC, Guedes AG, Brosnan RJ. Equivalence between invasive and

oscillometric blood pressures at different anatomic locations in healthy normotensive

anaesthetised horses. Equine Vet J 2015; doi: 10.1111/evj.12443.

72. Nolen-Walston RD, Norton JL, Navas de Solis C, et al. The effects of hypohydration

on central venous pressure and splenic volume in adult horses. J Vet Intern Med

2011;25:570-574.

43

73. Magdesian KG, Fielding CL, Rhodes DM, et al. Changes in central venous pressure

and blood lactate concentration in response to acute blood loss in horses. J Am Vet Med

Assoc 2006;229:1458-1462.

74. Marik PE, Baram M, Vahid B. Does central venous pressure predict fluid

responsiveness? A systematic review of the literature and the tale of seven mares. Chest

2008;134:172-178.

75. Norton JL, Nolen-Walston RD, Underwood C, et al. Repeatability, reproducibility,

and effect of head position on central venous pressure measurement in standing adult

horses. J Vet Intern Med 2011;25:575-578.

76. Barton MH, Magdesian, K.G. Equine Emergency and Critical Care. In: Southwood

LL, Wilkins, P.A, ed. The Systemic Inflammatory Response. Boca Raton, FL: CRC

Press; 2015.

77. Marshall JC, Cook DJ, Christou NV, et al. Multiple organ dysfunction score: a

reliable descriptor of a complex clinical outcome. Critical Care Medicine 1995;23:1638-

1652.

78. Thayer JF, Hahn AW, Sollers JJ, et al. Heart rate variability in the horse by

ambulatory monitoring. Biomedical Sciences Instrumentation 1997;33:482-485.

79. Pontet J, Contreras P, Curbelo A, et al. Heart rate variability as early marker of

multiple organ dysfunction syndrome in septic patients. J Crit Care 2003;18:156-163.

80. Nagel C, Erber R, Bergmaier C, et al. Cortisol and progestin release, heart rate and

heart rate variability in the pregnant and postpartum mare, fetus and newborn foal.

Theriogenology 2012;78:759-767.

44

81. Schmidt A, Hodl S, Mostl E, et al. Cortisol release, heart rate, and heart rate

variability in transport-naive horses during repeated road transport. Domest Anim

Endocrinol 2010;39:205-213.

82. Rietmann TR, Stauffacher M, Bernasconi P, et al. The association between heart rate,

heart rate variability, endocrine and behavioural pain measures in horses suffering from

laminitis. J Vet Med A Physiol Pathol Clin Med 2004;51:218-225.

83. Xu L, Li C-s. [Relationship between heart rate variability and serum levels of thyroid

hormones, cortisol and prognosis in patients with systemic inflammatory response

syndrome]. Zhongguo Wei Zhong Bing Ji Jiu Yi Xue = Chinese Critical Care Medicine =

Zhongguo Weizhongbing Jijiuyixue 2007;19:160-164.

84. Kovatchev BP, Farhy LS, Cao H, et al. Sample asymmetry analysis of heart rate

characteristics with application to neonatal sepsis and systemic inflammatory response

syndrome. Pediatric Research 2003;54:892-898.

85. Schmidt H, Müller-Werdan U, Hoffmann T, et al. Attenuated autonomic function in

multiple organ dysfunction syndrome across three age groups. Biomedizinische Technik

Biomedical Engineering 2006;51:264-267.

86. Schmidt H, Hoyer D, Henne R, et al. Autonomic dysfunction predicts both 1 and 2

month mortality in middle aged patients with multiple organ dysfunction syndrome. Crit

Care Med 2008;36:967-970.

87. King JN, Gerring EL. Detection of endotoxin in cases of equine colic. Vet Rec

1988;123:269-271.

88. Epstein KL, Brainard BM, Gomez-Ibanez SE, et al. Thrombelastography in horses

with acute gastrointestinal disease. J Vet Intern Med 2011;25:307-314.

45

89. Ryu SH, Kim JG, Bak UB, et al. A hematogenic pleuropneumonia caused by

postoperative septic thrombophlebitis in a Thoroughbred gelding. J Vet Sci 2004;5:75-77.

90. Norman TE, Chaffin MK, Perris EE, et al. Massive pulmonary thromboembolism in

six horses. Equine Vet J 2008;40:514-517.

91. Kaartinen MJ, Pang DS, Cuvelliez SG. Post-anesthetic pulmonary edema in two

horses. Vet Anaesth Analg 2010;37:136-143.

92. Underwood C, Southwood LL, McKeown LP, et al. Complications and survival

associated with surgical compared with medical management of horses with duodenitis-

proximal jejunitis. Equine Vet J 2008;40:373-378.

93. Tauseef M, Knezevic N, Chava KR, et al. TLR4 activation of TRPC6-dependent

calcium signaling mediates endotoxin-induced lung vascular permeability and

inflammation. J Exp Med 2012;209:1953-68.

94. Aharonson-Raz K, Singh B. Pulmonary intravascular macrophages and endotoxin-

induced pulmonary pathophysiology in horses. Can J Vet Res 2010;74:45-49.

95. Singh Suri S, Janardhan KS, Parbhakar O, et al. Expression of toll-like receptor 4

and 2 in horse lungs. Vet Res 2006;37:541-551.

96. Wilkins PO, C. Baumgardener, J. et al. . Acute lung injury and acute respiratory

distress syndromes in veterinary medicine: consensus definitions: The Dorothy Russell

Haveymeyer Working Group on ALI and ARDS in Veterinary Medicine J Vet Emerg

Crit Care (San Antonio) 2007;17:333-339.

97. Uchino S, Kellum JA, Bellomo R, et al. Acute renal failure in critically ill patients: a

multinational, multicenter study. JAMA 2005;294:813-818.

46

98. Bagshaw SM, George C, Bellomo R. Early acute kidney injury and sepsis: a

multicentre evaluation. Crit Care 2008;12:R47.

99. Kenney EM, Rozanski EA, Rush JE, et al. Association between outcome and organ

system dysfunction in dogs with sepsis: 114 cases (2003-2007). J Am Vet Med Assoc

2010;236:83-87.

100. Archer DC, Pinchbeck GL, Proudman CJ. Factors associated with survival of

epiploic foramen entrapment colic: a multicentre, international study. Equine Vet J Suppl

2011:56-62.

101. Groover ES, Woolums AR, Cole DJ, et al. Risk factors associated with renal

insufficiency in horses with primary gastrointestinal disease: 26 cases (2000-2003). J Am

Vet Med Assoc 2006;228:572-577.

102. Castoldi A, Braga TT, Correa-Costa M, et al. TLR2, TLR4 and the MYD88

signaling pathway are crucial for neutrophil migration in acute kidney injury induced by

sepsis. PLoS One 2012;7:e37584.

103. Dolente BA, Wilkins PA, Boston RC. Clinicopathologic evidence of disseminated

intravascular coagulation in horses with acute colitis. J Am Vet Med Assoc

2002;220:1034-1038.

104. Dallap Schaer BL, Epstein K. Coagulopathy of the critically ill equine patient. J Vet

Emerg Crit Care (San Antonio) 2009;19:53-65.

105. Dallap BL, Dolente B, Boston R. Coagulation profiles in 27 horses with large colon

volvulus. J Vet Emerg Crit Car 2003;13:215-225.

47

106. Opal SM, Esmon CT. Bench-to-bedside review: functional relationships between

coagulation and the innate immune response and their respective roles in the pathogenesis

of sepsis. Crit Care 2003;7:23-38.

107. Henry MM, Moore JN. Clinical relevance of monocyte procoagulant activity in

horses with colic. J Am Vet Med Assoc 1991;198:843-848.

108. Welch RD, Watkins JP, Taylor TS, et al. Disseminated intravascular coagulation

associated with colic in 23 horses (1984-1989). J Vet Intern Med 1992;6:29-35.

109. Johnstone IB, Crane S. Hemostatic abnormalities in equine colic. Am J Vet Res

1986;47:356-358.

110. Cotovio M, Monreal L, Armengou L, et al. Fibrin deposits and organ failure in

newborn foals with severe septicemia. J Vet Intern Med 2008;22:1403-1410.

111. Cotovio M, Monreal L, Navarro M, et al. Detection of fibrin deposits in tissues

from horses with severe gastrointestinal disorders. J Vet Intern Med 2007;21:308-313.

112. Marshall JC, Cook DJ, Christou NV, et al. Multiple organ dysfunction score: a

reliable descriptor of a complex clinical outcome. Crit Care Med 1995;23:1638-1652.

113. Freeman DE. Post operative ileus (POI): another perspective. Equine Vet J

2008;40:297-298.

114. Merritt AM, Blikslager AT. Post operative ileus: to be or not to be? Equine Vet J

2008;40:295-296.

115. Lefebvre D, Pirie RS, Handel IG, et al. Clinical features and management of equine

post operative ileus: Survey of diplomates of the European Colleges of Equine Internal

Medicine (ECEIM) and Veterinary Surgeons (ECVS). Equine Vet J 2015;

DOI: 10.1111/evj.12355.

48

116. Cohen ND, Lester GD, Sanchez LC, et al. Evaluation of risk factors associated with

development of postoperative ileus in horses. J Am Vet Med Assoc 2004;225:1070-1078.

117. Roussel AJ, Jr., Cohen ND, Hooper RN, et al. Risk factors associated with

development of postoperative ileus in horses. J Am Vet Med Assoc 2001;219:72-78.

118. Hoffmann WE, Baker G, Rieser S, et al. Alterations in selected serum biochemical

constituents in equids after induced hepatic disease. Am J Vet Res 1987;48:1343-1347.

119. Underwood C, Southwood LL, Walton RM, et al. Hepatic and metabolic changes in

surgical colic patients: a pilot study. J Vet Emerg Crit Care (San Antonio) 2010;20:578-

586.

120. Dunkel B, Chaney KP, Dallap-Schaer BL, et al. Putative intestinal

hyperammonaemia in horses: 36 cases. Equine Vet J 2011;43:133-140.

121. Teasdale G, Jennett B. Assessment of coma and impaired consciousness. A

practical scale. Lancet 1974;2:81-84.

122. Sharkey LC, DeWitt S, Stockman C. Neurologic signs and hyperammonemia in a

horse with colic. Vet Clin Pathol 2006;35:254-258.

123. Gomez DE, Lofstedt J, Stampfli HR, et al. Contribution of unmeasured anions to

acid-base disorders and its association with altered demeanor in 264 calves with neonatal

diarrhea. J Vet Intern Med 2013;27:1604-1612.

124. Lorenz I, Vogt S. Investigations on the association of D-lactate blood

concentrations with the outcome of therapy of acidosis, and with posture and demeanour

in young calves with diarrhoea. J Vet Med A Physiol Pathol Clin Med 2006;53:490-494.

49

125. Tadros EM, Frank N, Newkirk KM, et al. Effects of a "two-hit" model of organ

damage on the systemic inflammatory response and development of laminitis in horses.

Vet Immunol Immunopathol 2012;150:90-100.

126. Vinther AM, Skovgaard K, Heegaard PM, et al. Dynamic expression of leukocyte

innate immune genes in whole blood from horses with lipopolysaccharide-induced acute

systemic inflammation. BMC Vet Res 2015;11:134.

127. Pritchett LC, Ulibarri C, Roberts MC, et al. Identification of potential physiological

and behavioral indicators of postoperative pain in horses after exploratory celiotomy for

colic. Appl Anim Behav Sci 2003;80:31-43.

128. Belknap JK, Giguere S, Pettigrew A, et al. Lamellar pro-inflammatory cytokine

expression patterns in laminitis at the developmental stage and at the onset of lameness:

innate vs. adaptive immune response. Equine Vet J 2007;39:42-47.

129. Tadros EM, Frank N. Effects of continuous or intermittent lipopolysaccharide

administration for 48 hours on the systemic inflammatory response in horses. Am J Vet

Res 2012;73:1394-1402.

130. Menzies-Gow NJ, Bailey SR, Katz LM, et al. Endotoxin-induced digital

vasoconstriction in horses: associated changes in plasma concentrations of

vasoconstrictor mediators. Equine Vet J 2004;36:273-278.

131. Asplin KE, Sillence MN, Pollitt CC, et al. Induction of laminitis by prolonged

hyperinsulinaemia in clinically normal ponies. Vet J 2007;174:530-535.

132. Suagee JK, Corl BA, Crisman MV, et al. Effects of acute hyperinsulinemia on

inflammatory proteins in horses. Vet Immunol Immunopathol 2011;142:141-146.

50

133. Parsons CS, Orsini JA, Krafty R, et al. Risk factors for development of acute

laminitis in horses during hospitalization: 73 cases (1997-2004). J Am Vet Med Assoc

2007;230:885-889.

134. Krueger CR, Ruple-Czerniak A, Hackett ES. Evaluation of plasma muscle enzyme

activity as an indicator of lesion characteristics and prognosis in horses undergoing

celiotomy for acute gastrointestinal pain. BMC Vet Res 2014;10 Suppl 1:S7.

135. Valentine BA, Lohr CV. Myonecrosis in three horses with colic: evidence for

endotoxic injury. Vet Rec 2007;161:786-789.

136. Orsini JA, Elser AH, Galligan DT, et al. Prognostic index for acute abdominal crisis

(colic) in horses. Am J Vet Res 1988;49:1969-1971.

137. Proudman CJ, Dugdale AH, Senior JM, et al. Pre-operative and anaesthesia-related

risk factors for mortality in equine colic cases. Vet J 2006;171:89-97.

138. Furr MO, Lessard P, White NA, 2nd. Development of a colic severity score for

predicting the outcome of equine colic. Vet Surg 1995;24:97-101.

139. Vincent JL, Moreno R. Clinical review: scoring systems in the critically ill. Crit

Care 2010;14:207.

140. Vincent JL, Moreno R, Takala J, et al. The SOFA (Sepsis-related Organ Failure

Assessment) score to describe organ dysfunction/failure. On behalf of the Working

Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine.

Intensive Care Med 1996;22:707-710.

141. Le Gall JR, Klar J, Lemeshow S, et al. The Logistic Organ Dysfunction system. A

new way to assess organ dysfunction in the intensive care unit. ICU Scoring Group.

JAMA 1996;276:802-810.

51

142. Vincent JL, de Mendonca A, Cantraine F, et al. Use of the SOFA score to assess the

incidence of organ dysfunction/failure in intensive care units: results of a multicenter,

prospective study. Working group on "sepsis-related problems" of the European Society

of Intensive Care Medicine. Crit Care Med 1998;26:1793-1800.

143. Antonelli M, Moreno R, Vincent JL, et al. Application of SOFA score to trauma

patients. Sequential Organ Failure Assessment. Intensive Care Med 1999;25:389-394.

144. Janssens U, Graf C, Graf J, et al. Evaluation of the SOFA score: a single-center

experience of a medical intensive care unit in 303 consecutive patients with

predominantly cardiovascular disorders. Sequential Organ Failure Assessment. Intensive

Care Med 2000;26:1037-1045.

145. Ceriani R, Mazzoni M, Bortone F, et al. Application of the sequential organ failure

assessment score to cardiac surgical patients. Chest 2003;123:1229-1239.

146. Cholongitas E, Senzolo M, Patch D, et al. Risk factors, sequential organ failure

assessment and model for end-stage liver disease scores for predicting short term

mortality in cirrhotic patients admitted to intensive care unit. Aliment Pharmacol Ther

2006;23:883-893.

147. Graciano AL, Balko JA, Rahn DS, et al. The Pediatric Multiple Organ Dysfunction

Score (P-MODS): development and validation of an objective scale to measure the

severity of multiple organ dysfunction in critically ill children. Crit Care Med

2005;33:1484-1491.

148. Khwannimit B. Serial evaluation of the MODS, SOFA and LOD scores to predict

ICU mortality in mixed critically ill patients. J Med Assoc Thai 2008;91:1336-1342.

52

149. Chen YC, Tian YC, Liu NJ, et al. Prospective cohort study comparing sequential

organ failure assessment and acute physiology, age, chronic health evaluation III scoring

systems for hospital mortality prediction in critically ill cirrhotic patients. Int J Clin Pract

2006;60:160-166.

150. Khwannimit B. A comparison of three organ dysfunction scores: MODS, SOFA

and LOD for predicting ICU mortality in critically ill patients. J Med Assoc Thai

2007;90:1074-1081.

151. Cui HJ, Xin WQ, Tan YF. [Analysis of prognosis assessment with acute physiology

and chronic health evaluation III, multiple organ dysfunction score and sequential organ

failure assessment for the postoperative patients in cardiovascular surgery]. Zhongguo

Wei Zhong Bing Ji Jiu Yi Xue 2004;16:673-676.

152. Hart KA, MacKay, R. J. Endotoxemia and Sepsis In: Smith BP, ed. Large Animal

Internal Medicine, 5th Edition. St. Louis: Mosby; 2013:684.

153. Grulke S, Olle E, Detilleux J, et al. Determination of a gravity and shock score for

prognosis in equine surgical colic. J Vet Med A Physiol Pathol Clin Med 2001;48:465-

473.

154. Mair TS, Smith LJ. Survival and complication rates in 300 horses undergoing

surgical treatment of colic. Part 1: Short-term survival following a single laparotomy.

Equine Vet J 2005;37:296-302.

155. Hassel DM, Hill AE, Rorabeck RA. Association between hyperglycemia and

survival in 228 horses with acute gastrointestinal disease. J Vet Intern Med

2009;23:1261-1265.

53

156. Ellis CM, Lynch TM, Slone DE, et al. Survival and complications after large colon

resection and end-to-end anastomosis for strangulating large colon volvulus in seventy-

three horses. Vet Surg 2008;37:786-790.

157. Sheats MK, Cook VL, Jones SL, et al. Use of ultrasound to evaluate outcome

following colic surgery for equine large colon volvulus. Equine Vet J 2010;42:47-52.

54

CHAPTER 3

DOPPLER AND VOLUMETRIC ECHOCARDIOGRAPHIC METHODS FOR

CARDIAC OUTPUT MEASUREMENT IN STANDING ADULT HORSES1

_________________________

1 E.L. McConachie, M.H. Barton, G. Rapoport, S. Giguère. 2013. Journal of Veterinary

Internal Medicine. (27): 324-330.

Reprinted here with permission of the publisher

55

ABSTRACT

Background: Cardiac output (CO) is not routinely measured in critically ill adult horses

due to invasiveness of currently validated methods. Non-invasive CO monitoring would

complement clinical assessment of hemodynamic status in adult horses.

Hypothesis: Volumetric methods for measuring CO will have better agreement with

lithium dilution than Doppler-based methods.

Animals: Eight healthy adult horses

Methods: CO was manipulated from baseline with continuous rate infusions of

dobutamine and romifidine to achieve high and low CO states, respectively. At each

level, CO was measured by lithium dilution and various echocardiographic methods.

Images stored as video loops were reviewed by an individual blinded to the lithium

dilution results.

Results: Lithium dilution determinations of CO ranged between 16.6 and 63.0 L/min.

There was a significant effect of method of CO measurement (P < 0.001) but no

significant effect of CO level (P = 0.089) or interaction between level and method (P =

0.607) on the absolute value of the bias. The absolute values of the bias of the right

ventricular outflow tract (RVOT) Doppler, Simpson, 4-chamber area-length, and bullet

methods [5.5, 6.1, 6.5, 8.8 L/min, respectively] were significantly lower than that of the

left ventricular outflow tract (LVOT) Doppler or cubic methods [14.8, 24.3 L/min,

respectively].

Conclusions and clinical importance: The 4-chamber area-length, Simpson, bullet and

RVOT Doppler provided better agreement with lithium dilution than the other methods

56

evaluated. These methods warrant further investigation for use in critically ill adult

horses.

INTRODUCTION

Hemodynamic monitoring in critically ill veterinary subjects continues to become

more sophisticated with standards of care modeled after examination of critically ill

humans. Cardiac output (CO) is the best available variable to assess overall

cardiovascular function.1 Measurement of CO, along with blood hemoglobin

concentration and oxygen saturation of hemoglobin, allows calculation of global tissue

oxygen delivery and consumption, thereby providing useful information in individuals

with primary cardiac disease or secondary cardiovascular derangements associated with

systemic illness. Following trends in CO in individual animals in the intensive care unit

might allow for both earlier detection of cardiovascular derangements and optimization

of clinical interventions. To date, CO monitoring in adult horses has been limited to the

research setting.2-4 Cardiac or peripheral artery catheterization is required for indicator

dilution methods of CO measurement and is generally not suitable in a clinical setting.

Transthoracic echocardiography has been used to measure CO in people and in small

animals with various 2-dimensional (2-D) volumetric or Doppler methods.5,6

Ultrasonography is widely available in equine hospitals, and most ultrasound units have

software packages allowing calculation of various cardiovascular parameters including

CO. Therefore, transthoracic echocardiography may represent a convenient and

noninvasive means of measuring CO in equine patients. In one study, Doppler

echocardiographic measurement of CO was found to agree well with the thermodilution

57

method in adult horses.7 However, difficulties in aligning the ultrasound beam parallel to

the blood flow and individual variability in the cardiac window can make this method

difficult to use. In addition, indices of cardiac function derived from Doppler

echocardiography have been found to be less repeatable than indices derived from 2-D or

M-mode echocardiography in horses.8 It was recently shown that some volumetric

echocardiographic methods provide an accurate and noninvasive estimate of CO in

anesthetized neonatal foals.9 However, because of major differences in cardiac chamber

sizes and inability to obtain apical views of the heart in adult horses, data generated from

neonatal foals under general anesthesia cannot be directly extrapolated to adult horses.

The purpose of this study was to assess and validate various transthoracic

echocardiographic methods of measuring CO in standing adult horses over a range of CO

by comparing results to the lithium dilution CO (LiDCO) method. The hypothesis of the

study reported herein was that volumetric methods would have better agreement with

lithium dilution than Doppler-based methods.

MATERIALS AND METHODS

Animals

Eight horses (4 geldings and 4 mares) ranging from 4-20 years of age (mean of 9

years) with a mean body weight of 498 kg (range 425-660 kg) were included in this

study. Breeds included 2 Thoroughbreds, 3 Quarter horses, and 1 Lippizzaner,

Saddlebred and Warmblood. Horses were deemed healthy based on thorough physical

examination and echocardiography. The horses were housed in box stalls and were fed ad

58

lib hay and water. The study was approved by the University of Georgia Institutional

Animal Care and Use Committee.

Instrumentation

Horses were manually restrained in the stall with a halter and lead rope for

instrumentation and throughout the study period. A 14 gauge 5 ½ inch Teflon cathetera

was aseptically inserted into each jugular vein, one for injection of lithium chlorideb and

the other for dobutaminec and romifidined infusion, respectively. Under ultrasonographic

guidance, a 14 gauge 5 ½ inch Teflon cathetera was placed aseptically in either the right

or left carotid artery for invasive blood pressure monitoring and for lithium chloride

detection. A three-way valve was connected to the arterial line and was fitted with non-

compliant, pressure monitoring tubinge on one port and the lithium chloride sensorf on

the other. The lithium chloride sensor was attached on one side to a closed system

consisting of a peristaltic pump and blood collection bag and sensor interface on the other

end according to the manufacturer’s instructions. An electronic pressure transducer,g

calibrated with a mercury manometer, was placed on a surcingle at the level of the point

of the shoulder to approximate the base of the heart. A single-lead ECG was placed in

base-apex fashion for continuous monitoring during echocardiography.

Measurement of CO by Lithium Dilution

A LiDCO computeri was used to determine CO according to the manufacturer’s

instructions. Prior to the first LiDCO measurement, venous blood was collected for

determination of plasma sodium and hemoglobin concentrations by a blood gas analyzerj

which was necessary for calculation of CO by the LiDCO software. Lithium chloride (2.4

mmol/8 mL) was injected intravenously as a smooth, rapid bolus followed by saline flush

59

while a peristaltic pump withdrew a continuous arterial sample past the lithium sensor at

a rate of 4 mL/min. A lithium dilution curve was generated and stored if the quality

control criteria of an acceptable curve were met by the software.

Measurement of CO by Echocardiography

Echocardiography was performed using an ultrasound unitk with a built in

algorithm for stroke volume (SV) and CO determination and simultaneous display of the

ECG. Views were obtained from either the right or left parasternal window with a 2.5

MHz sector cardiac ultrasound transducerl by the same experienced clinician (MHB).

Echocardiographic images obtained from the right parasternal window included: (1) long-

axis 4-chamber view (modified slightly to include the apex of the left ventricle); (2) long-

axis left ventricular outflow tract view for measurement of the diameter of the aorta at the

sinotubular junction in systole; (3) long-axis right ventricular outflow tract view for

measurement of the diameter of the ascending pulmonary artery in systole and the

pulmonary artery velocity time integral (PVTI) in systole using pulsed-wave Doppler; (4)

short-axis 2-D view of the left ventricle at the level of the papillary muscles just below

the mitral valve; and (5) M-mode of the left ventricle at the level of the papillary muscles

just below the mitral valve. Images obtained from the left parasternal window included

the left ventricular outflow tract view for measurement of the aortic velocity time integral

(AoVTI) in systole using pulsed-wave Doppler. 6,10,11 Three video loops for each view

were stored for subsequent tracings and measurements. Tracings and measurements were

performed retrospectively by an investigator who was unaware of LiDCO results (EM).

For each method evaluated, the average of the three separate measurements was used for

60

calculating stroke volume (SV) as previously described.12-14 Determination of CO was

obtained from the following equation: CO (L/min) = SV (L/beat) x heart rate (beats/min).

2-D Volumetric Measurements

Bullet- Standard short axis views of the left ventricle at the level of the papillary

muscle just below the mitral valve were obtained for determination of the left ventricular

area in end diastole (LVASAd) and end systole (LVASAs) by tracing the blood-

endocardium interface.15 End diastole was defined as the onset of the QRS complex. End

systolic measurements were taken from the frame with the smallest left ventricular

diameter. The 4-chamber view was modified to include the apex of the left ventricle and

was used for measuring left ventricular length at end diastole (LVLRd) and end systole

(LVLRs).15 Left ventricular length was determined with electronic calipers set at the

midpoint of the line between the septal and left ventricular free wall origins of the mitral

valve to the blood-endocardium interface at the apex of the left ventricle. In this view,

end diastole was defined as the first frame in which the mitral valve was closed and end

systole was defined as the smallest left ventricular chamber size.10 The heart rate (HR)

was determined from the R-R interval measured with electronic calipers. The built-in

algorithm derived SV based on the following equation: SV = (5/6 x LVASAd x

LVLRd) – (5/6 x LVASAs x LVLRs).

Teichholz and Cubic- Standard M-mode images at the level of the papillary

muscle just below the mitral valve were obtained. A conventional left ventricular study

derived the left ventricular internal diameter in end diastole (LVIDd) and end systole

(LVIDs) and calculated SV using a built in algorithm based on the following formulas:

SV = [7 x (LVIDd)3/(LVIDd + 2.4)] – [7 x (LVIDs)3/(LVIDs + 2.4)] and SV=(LVIDd3 –

61

LVIDs3) for the Teichholz and Cubic methods, respectively.15 End diastole was defined

by the onset of the QRS complex and end systolic measurements were taken at the

smallest LV chamber diameter. The R-R interval was measured with electronic calipers

to determine HR and CO was reported based on this measurement.

4-Chamber Area-Length (4C AL) and 4-Chamber Modified Simpson’s (4C

MOD)- 4-chamber long axis views of the left ventricle were obtained from the right

parasternal view. The blood-endocardium interface was traced with electronic calipers

starting at the attachment of the mitral valve on the septum and ending at the attachment

of the mitral valve on the left ventricular free wall in diastole (LVARd) and systole

(LVARs).15 End diastole was defined as the first frame in which the mitral valve was

closed and end systole was defined as the smallest LV chamber size. The LVLRd and

LVLRs were obtained as described above under the Bullet method. HR was determined

by measuring an R-R interval with electronic calipers. A built in algorithm was used to

derive SV based on the following formulas for 4C AL and 4C MOD, respectively: SV =

[5/6 (LVARd)2/ LVLRd] – [5/6 (LVARs)2/ LVLRs]; SV= [(π/4 Σ(ai2/16) x LVLRd) –

(π/4 Σ(ai2/16) x LVLRs)].19-21

Doppler Echocardiography

Cross-sectional vessel areas (CSA) or [π x (0.5 x diameter)2] for the pulmonary

artery and aorta were calculated manually. The velocity time integral (VTI) was acquired

by placing the cursor in the vessel of interest while aligning the ultrasound beam as close

to parallel with flow as possible to obtain a full spectral envelope during systole.16-18 The

gate was positioned in the aorta or the pulmonary artery just past the valve. The VTI was

traced with electronic calipers as described previously and was reported as area under the

62

curve (cm2).19 Stroke volume was calculated from the following formula: SV=VTI X

CSA. The R-R interval just preceding or following the traced VTI was measured with

electronic calipers to determine HR.

Study Design

In order to investigate all methods over a range of CO, three levels were studied:

baseline (no treatment), high CO, and low CO. After baseline measurements, high CO

was induced with an intravenous continuous rate infusion (CRI) of dobutamine (3

μg/kg/min), started 10 minutes prior to CO measurements and was discontinued after

high CO data was collected.20 After a 45-minute wash-out period, low CO was induced

with an intravenous romifidine bolus (80 μg/kg), immediately followed by a CRI at 30

μg/kg/hour.21,22 Low CO data were collected 15 min after initiating the romifidine CRI

which coincided with expected steady state plasma concentrations and clinical effects.22

For each level, lithium dilution CO was measured twice. If any two LiDCO

measurements had more than 25% variation with respect to each other, a third

measurement was taken and the measurement outside the allowed 25% variation was

eliminated. Echocardiography was performed between LiDCO measurements. The

average of duplicate (lithium dilution) or triplicate (echocardiography) measurements

was calculated and used for data analysis. As a result, three readings were available from

each horse for a total of 24 observations. Concurrent heart rate and systolic, diastolic and

mean direct arterial blood pressure were recorded at the time of cardiac output

measurement. Cardiac index (CI) was obtained by dividing CO by body weight in

kilograms.

63

Statistical analyses

Agreement between each echocardiographic method of CO measurement and

lithium dilution was determined using the method for repeated measurements on a given

subject reported by Bland and Altman.23,24 For each observation, bias was calculated as

follows: (COLiDCO – COECHO), where COLiDCO and COECHO are the CO values measured

concurrently by lithium dilution and a given echocardiographic method. Normality of the

bias data and equality of variances were assessed using the Shapiro-Wilk and Levene’s

tests, respectively. A two-way ANOVA with repeated observations was conducted to

assess the effect of the method of CO measurement and level of CO (low, intermediate

and high) on bias. This model was used to estimate the mean bias and SD for each

method evaluated. A positive bias reflected underestimation of LiDCO-derived CO

whereas a negative bias indicated overestimation of the LiDCO-derived CO by

echocardiography. The limits of agreement were reported as bias ± 1.96 × SD.

A similar two-way ANOVA with repeated observations was used to assess

differences in the absolute value of the bias between methods. For effects found to be

significant by an overall F-test, multiple pairwise comparisons were made using the

Student-Newman-Keuls test. All analyses were done using statistical software.m,n

Statistical significance was set at P < 0.05.

RESULTS

Twenty-four pairs of lithium dilution/echocardiography CO measurements were

taken from the eight horses. Adverse effects in this study ranged from mild to moderate

hematoma formation at the site of arterial catheterization to transient collapse secondary

64

to a suspected air embolus in one horse. Adverse reactions attributable to the dobutamine

infusion included irritability and kicking out violently without cause in two horses,

synchronous diaphragmatic flutter in three horses and intermittent ventricular premature

contractions in one horse.

Lithium dilution measurements of CO ranged between 16.6 and 63.0 L/min

(mean ± SD = 30.5 ± 9.5 L/min), resulting in cardiac indices ranging between 17.5 and

129.6 mL/kg/min (59.9 ± 21.5 mL/kg/min). Lithium dilution determinations of CO

during administration of dobutamine (37.7 ± 11.7 L/min) were significantly higher than

determinations obtained at baseline (28.6 ± 4.7 L/min) or during administration of

romifidine (25.1 ± 6.3 L/min). Cardiac output during administration of romifidine was

not significantly different from CO obtained at baseline. Heart rates during administration

of dobutamine (37 ± 6 beats/min) were significantly higher than heart rates obtained at

baseline (32 ± 2 beats/min) or during administration of romifidine (29 ± 5 beats/min).

Heart rate during administration of romifidine was not significantly different from heart

rate obtained at baseline.

The analysis of bias indicated a significant effect (P < 0.001) of method of CO

measurement but no significant effect of level (low, intermediate or high) of CO (P =

0.938) on bias, indicating that the performance of each echocardiographic method was

not influenced by the magnitude of CO. The mean bias and limits of agreements for each

echocardiographic method of CO measurement are presented in Table 1. The overall

performance of each method was assessed by comparing the absolute value of their bias.

Analysis of variance revealed a significant effect of method of CO measurement (P <

0.001) but no significant effect of level of CO (P = 0.089) and no significant interactions

65

between level and method (P = 0.607). The absolute values of the bias of the 4C AL, 4C

MOD, RVOT Doppler, and Bullet methods were significantly lower than that of LVOT

Doppler or cubic methods (Table 1). Bland-Altman plots for the 4C AL, 4C MOD,

RVOT Doppler and Bullet methods are presented in Figure 1. The mean CO (± SD) for

each method of measurement at each level of CO is reported in Table 2.

DISCUSSION

Monitoring trends in CO would complement current hemodynamic monitoring

tools available for use in critically ill adult horses. Cardiac output measurement is

routinely employed in human critical care facilities and has become a common modality

for hemodynamic monitoring in foals in some neonatal intensive care units.25 Currently

validated and accepted methods for estimation of cardiac output in adult horses are not

suitable for routine clinical use as they are invasive, requiring maintenance of a

peripheral arterial catheter or a pulmonary artery catheter for lithium dilution or

thermodilution techniques, respectively. In this study, the carotid artery was catheterized

owing to difficulties in maintaining patency in smaller peripheral arteries in conscious

standing horses. An experimental study by one of the current authors reported

maintenance of a carotid artery catheter for up to 24 hours without adverse effects.26

Although placement of a carotid artery catheter under ultrasound guidance was not

technically challenging, the authors do not recommend carotid artery catheterization in

client-owned horses. The use of pulmonary artery catheters has been reported to increase

mortality and predispose to pulmonary thromboembolism in people.27 In addition,

pulmonary artery catheterization has been associated with traumatic endocardial lesions

66

in the right heart and pulmonary artery of adult horses.28 Thus, in order to eliminate the

need for arterial catheterization in a clinic setting, the main purpose of this study was to

determine which echocardiographic measures of CO most robustly correlate with an

indicator dilution technique in standing horses.

Three of the volumetric echocardiographic methods investigated (4C AL, 4C

MOD and Bullet) had significantly better agreement with the LiDCO reference method

than the other echocardiographic methods evaluated. The 4C AL and 4C MOD methods

overestimated and underestimated CO, respectively, each by a mean of approximately 1

L/min, while the Bullet method overestimated the CO by approximately 7 L/min, relative

to the LiDCO reference. When considering performance of the Doppler methods,

agreement between the RVOT Doppler and the LiDCO was not significantly different

from the best volumetric measures of CO, disproving the hypothesis of this study that the

volumetric measures of CO would be superior to the Doppler methods. The RVOT

Doppler method underestimated CO by a mean of 2 L/min relative to LiDCO while the

LVOT Doppler method overestimated CO by a mean of 14 L/min relative to LiDCO.

Although the 4C AL, 4C MOD, and RVOT Doppler methods had a significantly

lower bias than the other methods evaluated, their limits of agreement with LiDCO were

wide (± 17 L/min), representing up to ± 50% of the measured CO. Two methods of

measurement can usually be considered interchangeable if the limits of agreement fall

within a clinically acceptable range.23 A meta-analysis of studies using bias and

precision statistics to compare CO measurement techniques in people underscored that

considerable diversity exists in how the results of bias and precision are interpreted

between studies.29 In the aforementioned study, the authors proposed that acceptance of a

67

new technique for measurement of CO in people should rely on limits of agreement up to

± 30%.29 The limits of agreement that would be considered acceptable for the

measurement of CO in a horse have not been defined and might even vary depending on

the clinical situation. Despite the wide limits of agreement, 4C AL, 4C MOD, and RVOT

Doppler might prove to be of value to detect changes in magnitude and direction of CO in

a clinical setting.

Excellent image quality is essential to be able to make accurate and repeatable

echocardiographic measurements.30 Poor image quality due to subject factors (poor

compliance, body condition score, differences in the cardiac window) was encountered to

some extent in the present study. This may have influenced the performance of some of

the volumetric methods that required tracing the blood endocardium interface (Bullet, 4C

AL, 4C MOD). However, the Teichholz and Cubic methods derived from standard M-

mode images and left ventricular study measurements had poor agreement with the

LiDCO method, significantly underestimating and overestimating the CO respectively.

Poor agreement was attributed to geometric assumptions that rely on a smooth circular or

elliptical shaped ventricle and uniform contraction which may not be applicable to the

equine left ventricle.31,32 Results from the Teichholz and Cubic based methods were

similar to results in anesthetized foals.10

In this study, the LVOT Doppler method in the left parasternal window

significantly overestimated the LiDCO reference. This is in contrast to results of a

previous study comparing Doppler echocardiographic methods in standing adult horses in

which LVOT Doppler in the left parasternal window had better agreement with

thermodilution than RVOT Doppler.11 The discrepancy between the aforementioned and

68

the present study may be attributable to differences in the population of horses studied or

poor repeatability of pulsed-wave Doppler measures in general.11,33,34 Pulsed-wave

Doppler measurement of aortic VTI was found to have more day-to-day and intercardiac

cycle variation than pulsed-wave Doppler measurement of pulmonary artery VTI in one

equine study.33 While it is possible that the aortic diameter was overestimated, this was

felt to be unlikely as measurements were made at the sinotubular junction and were

within normal reference ranges for adult standing horses.34-36 It is more likely that the

pulsed-wave Doppler traces were too broad owing to artifact from poor alignment with

blood flow. Overestimation of CO by pulsed wave Doppler methods has been previously

reported in both human and animal studies.10,34,37,38

Inherent limitations exist with all CO monitoring methods available for clinical

use. While there is no universally accepted gold standard for CO determination in human

or veterinary medicine, thermodilution has traditionally been the most commonly used

method in people.16,39 Studies comparing LiDCO to thermodilution CO have been

performed in several species including the horse. 40-44 As a result of the relatively narrow

limits of agreement between the two methods, the use of the LiDCO method as the

reference standard method is widely accepted across many species.41-45 Potential sources

of error with the lithium dilution method include intracardiac shunts, which were ruled

out by our inclusion criteria, and lithium accumulation. Lithium accumulation creates

background “noise” and can result in overestimation of cardiac output. Accumulation was

unlikely in the current study as the mean cumulative lithium dose was 0.038 mmol/kg

and no more than 8 lithium injections were administered to any horse (19.2 mmol LiCl

total). Considerably higher cumulative dosages of LiCl (0.8 mmol/kg; 69.3 mmol LiCL)

69

were used in a study in exercising horses and did not result in any adverse effects,

however, overestimation of CO was documented as the number of lithium injections and

exercise intensity increased.3

Lithium determinations of CO in the present study ranged between 16.6 and 63.0

L/min which represent a wide physiological range of CO and are similar to ranges

achieved in comparable studies.11,46 Echocardiographic derivation of CO was not

influenced by level of CO. Though a statistically significant decrease in CO from

baseline was not achieved with the romifidine CRI, our purpose of measuring CO over a

wide range of values was met.

Limitations of the study design included a small and diverse sample population in

terms of breed. Although, the inclusion of various breeds may be viewed as a limitation

and indeed may have impacted the performance of some of the transthoracic

echocardiographic methods, the use of a monomorphic sample would be unrepresentative

of most clinic populations, making it difficult to extrapolate the findings of this study to

the target population. Another potential limitation included the inability to

simultaneously measure CO with lithium dilution and echocardiography. Cardiac output

is dynamic with beat-to-beat variation based on neuroendocrine input. Agreement

between the methods might have been influenced by making measurements at discrete

time points. Efforts to minimize this effect were made by performing lithium dilution

measurements just before and immediately after storage of transthoracic

echocardiographic images and by performing all measurements stall side in a quiet

environment to minimize transient excitation during measurements.

70

A potential source of error and limitation of transthoracic echocardiography is

that the Doppler-derived techniques and some of the 2-D techniques require

measurements derived by tracing velocity spectra or blood-endocardium interfaces,

which are subject to observer interpretation. In the study herein, only one person derived

measurements from the stored video clips, thus inter-observer agreement was not

evaluated. Other studies will be necessary to assess the reproducibility of these methods

and to determine the effect of the observer on variability.

In the hands of adequately trained clinicians, transthoracic echocardiography of

critically ill adult horses has advantages beyond CO measurement. Previous work has

shown that volume depletion can be recognized rapidly based on chamber morphology.47

In addition, detection of diastolic or systolic dysfunction, valvular regurgitation, chamber

dilation, regional wall abnormalities or pericardial effusion would impact therapeutic

decisions.5 The disadvantages of transthoracic echocardiography pertain to initial capital

costs of purchasing an ultrasound unit and acquiring the training necessary to

appropriately apply these techniques and interpret the results.

In conclusion, transthoracic echocardiography using the 4-chamber area length

method, 4-chamber modified Simpson method, bullet method or Doppler of the RVOT

have significantly lower biases than all other methods evaluated in standing healthy adult

horses. These non-invasive methods of estimating cardiac output provide a non-invasive

clinically accessible method for serial hemodynamic monitoring and warrant further

investigation in critically ill adult horses.

71

FOOTNOTES

a. Abbocath®-T, Hospira, Lake Forest, IL

b. Lithium Chloride 99.995%, VWR International LLC, Batavia, IL

c. Dobutamine, Hospira Lake Forest, IL

d. Sedivet® Vetmedica, St. Joseph, MO

e. Medex™, Smiths Medical, Dublin, OH

f. LiDCO™ London, UK

g. Edwards Lifesciences LLC, Irvine, CA

h. Spacelabs Medical, WA

i. LiDCO™plus, cardiac computer, LiDCO Group PLC, London, UK

j. Nova Biomedical, Waltham, MA

k. Vivid 7, GE Medical Systems, Milwaukee, WI

l. M4S transducer, GE Medical Systems, Milwaukee, WI

m. MedCalc version 12.1.4.1, Mariakerke, Belgium

n. GraphPad Prism version 5.0 La Jolla, CA

REFERENCES

1. Tibby SM, Murdoch IA. Measurement of cardiac output and tissue perfusion. Curr

Opin Pediatr 2002;14:303-309.

2. Wilkins PA, Boston RC, Gleed RD, et al. Comparison of thermal dilution and

electrical impedance dilution methods for measurement of cardiac output in standing and

exercising horses. Am J Vet Res 2005;66:878-884.

72

3. Durando MM, Corley KT, Boston RC, et al. Cardiac output determination by use of

lithium dilution during exercise in horses. Am J Vet Res 2008;69:1054-1060.

4. Corley KT, Donaldson LL, Durando MM, et al. Cardiac output technologies with

special reference to the horse. J Vet Intern Med 2003;17:262-272.

5. Brown JM. Use of echocardiography for hemodynamic monitoring. Crit Care Med

2002;30:1361-1364.

6. Uehara Y, Koga M, Takahashi M. Determination of cardiac output by

echocardiography. J Vet Med Sci 1995;57:401-407.

7. Blissitt KJ, Young LE, Jones RS, et al. Measurement of cardiac output in standing

horses by Doppler echocardiography and thermodilution. Equine Vet J 1997;29:18-25.

8. Young LE, Scott GR. Measurement of cardiac function by transthoracic

echocardiography: day to day variability and repeatability in normal Thoroughbred

horses. Equine Vet J 1998;30:117-122.

9. Shih A, Giguere S, Sanchez LC, et al. Determination of cardiac output in neonatal

foals by ultrasound velocity dilution and its comparison to the lithium dilution method. J

Vet Emerg Crit Care (San Antonio) 2009;19:438-443.

10. Giguere S, Bucki E, Adin DB, et al. Cardiac output measurement by partial carbon

dioxide rebreathing, 2-dimensional echocardiography, and lithium-dilution method in

anesthetized neonatal foals. J Vet Intern Med 2005;19:737-743.

11. Blissitt KJ, Young LE, Jones RS, et al. Measurement of cardiac output in standing

horses by Doppler echocardiography and thermodilution. Equine Vet J 1997;29:18-25.

73

12. Folland ED, Parisi AF, Moynihan PF, et al. Assessment of left ventricular ejection

fraction and volumes by real-time, two-dimensional echocardiography. A comparison of

cineangiographic and radionuclide techniques. Circulation 1979;60:760-766.

13. Schiller NB, Shah PM, Crawford M, et al. Recommendations for quantitation of the

left ventricle by two-dimensional echocardiography. American Society of

Echocardiography Committee on Standards, Subcommittee on Quantitation of Two-

Dimensional Echocardiograms. J Am Soc Echocardiogr 1989;2:358-367.

14. Pombo JF, Troy BL, Russell RO, Jr. Left ventricular volumes and ejection fraction

by echocardiography. Circulation 1971;43:480-490.

15. Reef V. Equine Diagnostic Ultrasound. Philadelphia: WB Saunders; 1998.

16. Pugsley J, Lerner AB. Cardiac output monitoring: is there a gold standard and how

do the newer technologies compare? Semin Cardiothorac Vasc Anesth 2010;14:274-282.

17. Blissitt KJ, Bonagura JD. Pulsed wave Doppler echocardiography in normal horses.

Equine Vet J 1995;27:38-46.

18. Blissitt KJ, Young LE, Jones RS, et al. Measurement of cardiac output in standing

horses by Doppler echocardiography and thermodilution. Equine Vet J 1997;29:18-25.

19. Nyland TG MJ. Small Animal Diagnostic Ultrasound. In: Keinle RD TW, ed.

Echocardiography. Philadelphia, PA.: WB Saunders; 2002:354-424.

20. Corley KT. Inotropes and vasopressors in adults and foals. Vet Clin North Am

Equine Pract 2004;20:77-106.

21. Freeman SL, Bowen IM, Bettschart-Wolfensberger R, et al. Cardiovascular effects of

romifidine in the standing horse. Res Vet Sci 2002;72:123-129.

74

22. Ringer SK, Portier KG, Fourel I, et al. Development of a romifidine constant rate

infusion with or without butorphanol for standing sedation of horses. Vet Anaesth Analg

2012;39:12-20.

23. Bland JM, Altman DG. Measuring agreement in method comparison studies. Stat

Methods Med Res 1999;8:135-160.

24. Bland JM, Altman DG. Agreement between methods of measurement with multiple

observations per individual. J Biopharm Stat 2007;17:571-582.

25. Corley KTT. Monitoring and treating haemodynamic disturbances in critically ill

neonatal foals. Part 1: Haemodynamic monitoring. Equine Vet Educ 2002;14:270-279.

26. Barton MH, Moore JN, Norton N. Effects of pentoxifylline infusion on response of

horses to in vivo challenge exposure with endotoxin. Am J Vet Res 1997;58:1300-1307.

27. Sandham JD, Hull RD, Brant RF, et al. A randomized, controlled trial of the use of

pulmonary-artery catheters in high-risk surgical patients. N Engl J Med 2003;348:5-14.

28. Schlipf JW DC, Getzy DM et al. Lesions associated with cardiac catheterization and

thermodilution cardiac output determination in horses. In: Proceedings of the 5th

International Congress of Veterinary Anesthesia; 1994:71.

29. Critchley LA, Critchley JA. A meta-analysis of studies using bias and precision

statistics to compare cardiac output measurement techniques. J Clin Monit Comput

1999;15:85-91.

30. Marr CM, Bowen, I.M. Cardiology of the Horse. In: Marr CM, Patteson, M. , ed.

Echocardiography. Philadelphia: Saunders 2010.

75

31. Kuroda T, Seward JB, Rumberger JA, et al. Left ventricular volume and mass:

Comparative study of two-dimensional echocardiography and ultrafast computed

tomography. Echocardiography 1994;11:1-9.

32. Teichholz LE, Kreulen T, Herman MV, et al. Problems in echocardiographic volume

determinations: echocardiographic-angiographic correlations in the presence of absence

of asynergy. Am J Cardiol 1976;37:7-11.

33. Buhl R, Ersboll AK, Eriksen L, et al. Sources and magnitude of variation of

echocardiographic measurements in normal standardbred horses. Vet Radiol Ultrasound

2004;45:505-512.

34. Blissitt KJ, Bonagura JD. Pulsed wave Doppler echocardiography in normal horses.

Equine Vet J Suppl 1995:38-46.

35. Slater JD, Herrtage ME. Echocardiographic measurements of cardiac dimensions in

normal ponies and horses. Equine Vet J Suppl 1995:28-32.

36. Patteson MW, Gibbs C, Wotton PR, et al. Echocardiographic measurements of

cardiac dimensions and indices of cardiac function in normal adult thoroughbred horses.

Equine Vet J Suppl 1995:18-27.

37. Garcia J, Kadem L, Larose E, et al. Comparison between cardiovascular magnetic

resonance and transthoracic Doppler echocardiography for the estimation of effective

orifice area in aortic stenosis. J Cardiovasc Magn Reson 2011;13:25.

38. Tournoux F, Petersen B, Thibault H, et al. Validation of noninvasive measurements

of cardiac output in mice using echocardiography. J Am Soc Echocardiogr 2011;24:465-

470.

76

39. Leibowitz AB, Oropello JM. The pulmonary artery catheter in anesthesia practice in

2007: an historical overview with emphasis on the past 6 years. Semin Cardiothorac Vasc

Anesth 2007;11:162-176.

40. Kurita T, Morita K, Kato S, et al. Comparison of the accuracy of the lithium dilution

technique with the thermodilution technique for measurement of cardiac output. Br J

Anaesth 1997;79:770-775.

41. Corley KT, Donaldson LL, Furr MO. Comparison of lithium dilution and

thermodilution cardiac output measurements in anaesthetised neonatal foals. Equine Vet J

2002;34:598-601.

42. Mason DJ, O'Grady M, Woods JP, et al. Assessment of lithium dilution cardiac

output as a technique for measurement of cardiac output in dogs. Am J Vet Res

2001;62:1255-1261.

43. Linton RA, Jonas MM, Tibby SM, et al. Cardiac output measured by lithium dilution

and transpulmonary thermodilution in patients in a paediatric intensive care unit.

Intensive Care Med 2000;26:1507-1511.

44. Linton RA, Young LE, Marlin DJ, et al. Cardiac output measured by lithium

dilution, thermodilution, and transesophageal Doppler echocardiography in anesthetized

horses. Am J Vet Res 2000;61:731-737.

45. Linton RA, Band DM, Haire KM. A new method of measuring cardiac output in man

using lithium dilution. Br J Anaesth 1993;71:262-266.

46. Lepiz ML, Keegan RD, Bayly WM, et al. Comparison of Fick and thermodilution

cardiac output determinations in standing horses. Res Vet Sci 2008;85:307-314.

77

47. Underwood C, Norton JL, Nolen-Walston RD, et al. Echocardiographic changes in

heart size in hypohydrated horses. J Vet Intern Med 2011;25:563-569.

78

Table 3.1. Summary statistics of the agreement between cardiac output measurements (L/min) by lithium dilution and various echocardiographic methods.

LVOT: left ventricular outflow tract; RVOT: right ventricular outflow tract †Negative value indicates overestimation of CO relative to LiDCO *Least squares means; Standard deviation = 6.3 L/min a,b,c different letters between rows indicate a statistically significant difference in the absolute value of the mean relative bias (P < 0.05). When at least 1 superscript letter is common between 2 values, the difference is not statistically significant.

Cardiac output methods Bias (L/min ± s.d.)†

Limits of agreement

(L/min)

Absolute value of bias* (L/min)

Doppler LVOT -14.2 ± 10.3 -34.3 to 5.9 14.8b

Doppler RVOT 2.2 ± 7.9 -13.4 to 17.7 5.5a

4 chamber area-length -1.1 ± 8.4 -17.6 to 15.4 6.5a 4 chamber modified Simpson 0.8 ± 7.8 -14.4 to 16.0 6.1a

Bullet -6.9 ± 9.2 -25.1 to 11.2 8.8a

Teichholz 7.6 ± 7.14 -6.5 to 21.6 10.4a,b

Cubic -24.3 ± 18.0 -59.6 to 11.0 24.3b

79

Table 3.2. Mean cardiac output (L/min ± SD) as determined concurrently by lithium dilution and various Doppler or volumetric echocardiographic methods. Measurements were obtained at baseline, after administration of dobutamine (high CO) and after administration of romifidine (low CO).

Method Baseline High Low LiDCO

28.7 ± 4.7

36.6 ± 10.03

25.1 ± 6.3

Doppler LVOT 46.1 ± 12.1 52.7 ± 9.6 35.4 ± 8.1 Doppler RVOT 29.9 ± 5.6 30.4 ± 4.4 26.1 ± 6.6 4C AL 29.9 ± 9.7 39.8 ± 10.8 25.6 ± 7.1 4C MOD 28.3 ± 8.8 37.3 ± 9.9 24.9 ± 3.3 Bullet 34.6 ± 7.1 46.2 ± 11.3 23.4 ± 3.8 Teichholz 20.5 ± 6.0 28.5 ± 8.4 19.8 ± 6.5 Cubic 47.6 ± 17.7 67.9 ± 22.3 48.9 ± 21.9

80

Figure 3.1. Bland-Altman plots of CO values measured concurrently by lithium dilution and echocardiography by the 4-chamber modified Simpson (A), 4-chamber area-length (B), Doppler RVOT (C), or Bullet (D) methods in standing adult horses. The solid line represents the mean bias and the dashed lines represent the upper and lower limits of agreements (1.96 9 SD). Three measurements were obtained from each of 8 horses for a total of 24 observations. Each symbol represents an individual horse.

81

CHAPTER 4

HEART RATE VARIABILITY IN HORSES WITH ACUTE

GASTROINTESTINAL DISEASE REQUIRING EXPLORATORY

LAPAROTOMY2

______________________________

2E.L. McConachie, S. Giguère, G. Rapoport, M.H Barton. Accepted by the Journal of

Veterinary Emergency and Critical Care.

Reprinted here pending permission of the publisher.

82

ABSTRACT

Objective- To describe heart rate variability (HRV) in horses with acute gastrointestinal

disease that undergo exploratory laparotomy. We hypothesized that horses with ischemic

gastrointestinal disease will have reduced HRV compared to horses with non-ischemic

lesions. We further hypothesized that a reduction in HRV will be associated with non-

survival.

Design - Prospective, clinical, observational study.

Setting - Veterinary Teaching Hospital.

Animals - Horses presented for acute colic (n=57) or elective surgical procedures (n=10)

were enrolled.

Interventions - Admission heart rate was recorded and continuous telemetry was placed

within 2 h of recovery from anesthesia, monitored and recorded for 48 hours post-

operatively. Stored electrocardiograms were manually inspected and R-to-R intervals

were extracted and uploaded into HRV software for analysis. Time domain and

frequency spectral analysis were investigated at Times 1 (2-10 hours), 2 (16-24 hours), 3

(30-38 hours), and 4 (44-52 hours) post-operatively. A 2-way ANOVA for repeated

measures was used for group comparisons. Logistic regression analysis was used to

detect potential associations between admission HR, time and frequency domain

variables, and non-survival.

Measurements and Main Results - Horses diagnosed with an ischemic gastrointestinal

lesion (n=22) at the time of surgery had significantly higher post-operative heart rates and

reduced time domain derived measures of HRV than horses with non-ischemic

gastrointestinal lesions (n=35) or control horses (n=10). Horses that survived to discharge

83

had significantly lower post-operative heart rates, higher time domain, and lower low

frequency spectral measures of HRV compared to non-survivors. The multivariable

logistic regression model had a ROC AUC of 0.95 and was significantly better at

predicting non-survival than admission HR (P=0.0124).

Conclusions - Reduced HRV was strongly associated with ischemic gastrointestinal

disease and non-survival. HRV analysis is a noninvasive technique that may provide

diagnostic and prognostic information pertinent to the management of post-operative

horses with severe gastrointestinal disease.

INTRODUCTION

Heart rate variability (HRV) analysis provides a non-invasive method to measure

fluctuations in the autonomic input to the sinoatrial node and reflects activity of the

individual components of the autonomic nervous system (ANS).1 This provides clinically

useful information because variation in beat-to-beat intervals is characteristic of healthy

cardiac function.2 There are many methods of HRV analysis; two of the simplest and

most widely used are time domain and frequency spectral analysis. Time domain

methods of HRV analysis, notably the standard deviation of the duration of R-to R-

intervals (so-called “normal-to-normal” or NN intervals; SDNN) and root mean square

differences of successive NN intervals (RMSSD), reflect overall autonomic modulation

and predominately vagally mediated cardiac modulation, respectively. Frequency spectral

analysis is unique in that the components of the ANS can be separated out by distinct

high-frequency (HF) and low-frequency (LF) bands which approximate parasympathetic

activity and combined sympathetic and parasympathetic activity, respectively.

84

During periods of distress which occur with serious illness, trauma, or surgery,

neurohormonal input is critical to maintain homeostasis through an appropriate

cardiovascular response. Reduced HRV reflects poor cardiovascular health and may be

due, in part, to overwhelming sympathetic tone, withdrawal of one or both branches of

the ANS, or primary cardiac disease.3 Various HRV parameters have proven to be early

predictors of major adverse clinical events, such as acute myocardial infarction in cardiac

patients and multiple organ dysfunction syndrome (MODS) in septic patients admitted to

the intensive care unit.4-7 Reduced HRV is predictive of mortality in people that suffer

acute myocardial infarction8 as well as in-hospital mortality in septic patients.9 Thus

HRV analysis can provide early predictive patient-side information for those at risk for

organ dysfunction, allowing for prompt implementation of preventative strategies.

While HRV has been evaluated in horses under a variety of physiologic and

pathologic conditions,10-14 to the authors’ knowledge, HRV has not been investigated in

horses with naturally acquired gastrointestinal disease. The main objectives of the study

described herein are to describe HRV parameters in horses that underwent exploratory

laparotomy for acute gastrointestinal disease and compare them to those of healthy

control horses that underwent elective surgical procedures. We hypothesized that horses

with ischemic lesions of the gastrointestinal tract have reduced HRV, as compared to

horses with non-ischemic gastrointestinal lesions or horses undergoing elective surgical

procedures requiring general anesthesia. We further hypothesized that a reduction in

HRV will be associated with non-survival.

85

MATERIALS AND METHODS

In this prospective, clinical study, client-owned horses > 1 year of age that were

presented to the Veterinary Teaching Hospital between November 2011 and February

2014 for either acute gastrointestinal disease and underwent exploratory laparotomy or

for elective surgery were enrolled. Owner consent was obtained at the time the horse was

determined to require surgery or upon hospital admission for healthy controls. Horses

were enrolled as controls if they required general anesthesia for an elective surgical

procedure lasting at least 1 hour, and were considered healthy on the basis of normal pre-

operative vital parameters, general physical examination and pre-operative complete

blood count, fibrinogen concentration and plasma venous blood gas analysis. Horses

were removed from the study if they were euthanized solely for financial reasons or if

they were not recovered from general anesthesia. This study was approved by the Clinical

Research Committee.

Data collection

Age, breed and sex were recorded for each horse, as was the gastrointestinal

lesion diagnosed and corrected at surgery, the elective procedure, and whether or not the

horse survived to discharge. Heart rate was recorded at admission by the attending

clinician. Horses requiring gastrointestinal surgery were grouped as ischemic or non-

ischemic based on surgical findings. Specifically, horses grouped as ischemic included

those with small or large intestinal strangulating lesions that required resection and

anastomosis and those with a large colon volvulus greater than or equal to 360 degrees,

regardless of whether or not a resection was performed. The decision for resection and

anastomosis was at the discretion of the surgeon and was based on surgical findings

86

consistent with devitalized intestine. Records were searched retrospectively to record

analgesic agent administration during the study period.

Telemetric electrocardiography placement

Within 2 hours of recovering from general anesthesia, a telemetry unita was

placed for 48-hour continuous electrocardiographic (ECG) recording. Electrodesb were

placed according to manufacturer instructions. A surcingle was placed to ensure the

telemetry unit and leads remained in place and to promote contact between the electrodes

and the horse’s skin.

Heart rate variability analysis

Telemetric ECG recordings were visually inspected for arrhythmia and artifact.

Artifact-free ECG recordings of sinus rhythm were cut and saved at approximately 12-

hour intervals, defined as Time 1 (2-10 hours post-operative), Time 2 (16-24 hours post-

operative), Time 3 (30- 38 hours post-operative) and Time 4 (44-52 hours post-

operative). Normal-to-normal intervals were extracted using the Televet software with a

30% artifact filter and saved as text files for offline analysis. NN intervals were manually

inspected and were removed if they were less than <75% or > 125% of the previous

interval.10,15 The file was then opened in the HRV softwarec for time domain and

frequency spectral analysis. The data were detrended using cubic spline transformation

with smoothness parameters set at 500 ms, as previously reported.16 The artifact filter

was set at low, which corresponds to 0.35 sec. The interpolation rate was set at 4 Hz and

the Fast Fourier Transform (FFT) spectrum window was set at 512 s (interpolated

sampling of 512 equidistant points) with an overlap of 40%. For frequency analysis,

bands of interest for LF and HF were 0.01-0.07 and 0.07-0.6 Hz, respectively, based on

87

previous studies.17 At each time point, 5- and 30-minute periods of ECG recording were

analyzed for time domain methods and 5-minute, 30-minute and 1024 NN intervals were

used for frequency spectral analysis based on previous work.15,18 Parameters of interest

for the time domain method included mean HR, SDNN, RMSSD, triangular index (the

integral of the density distribution of all NN intervals divided by the maximum of the

density distribution), and pNN50 (the proportion derived by dividing the number of

interval differences of successive NN intervals greater than 50 ms by the total number of

NN intervals). Parameters of interest for frequency spectral analysis included LF, HF and

LF/HF ratio.

Statistical Methods

Descriptive statistics for patient factors (age, breed, sex, analgesic administration,

time of euthanasia or death) were performed and are reported as mean ± SD where

applicable. Associations between breed or sex and group or outcome were evaluated with

a Chi-squared test. Normality of the data and equality of variances were assessed using

the Shapiro-Wilk and Levene’s tests, respectively. A Student t-test or Mann-Whitney U

test for normally and non-normally distributed data, respectively, was used to compare

age and HR on admission between survivors and non-survivors. A one-way ANOVA or

Kruskal-Wallis ANOVA on ranks was used to compare age or admission HR between

groups, respectively. When warranted, multiple pairwise comparisons were performed

using Dunn’s method. A two-way ANOVA for repeated measures was used to assess the

effect of disease category (ischemic, non-ischemic, control; or survivors and non-

survivors), time [HRV: Time 1(2-10 hours), Time 2 (16-24 hours), Time 3 (30-38 hours)

and Time 4 (44-52 hours)] and the interactions between disease category and time. Data

88

that were not normally distributed were rank-transformed prior to analysis. When

warranted, multiple pairwise comparisons were performed using the method of Holm-

Sidak.

Potential associations between individual variables and non-survival were first

screened by use of univariable logistic regression. For variables with a significant

association with non-survival, the best cut-off to predict non-survival was determined

using receiver operating characteristic (ROC) curve analysis. Variables for which the

screening P value was < 0.10 were considered for inclusion in the multivariable model.

Variables with a variance inflation factor > 2.50 were deleted to avoid multicollinearity.

The multivariable model was a backward stepwise model, whereby variables were

removed sequentially starting with that having the largest P value and until only those

variables with P < 0.05 remained. Goodness of fit of the final model was evaluated using

the Hosmer and Lemeshow test. Odds ratios (OR) and 95% confidence intervals (CI)

were calculated. An OR greater than 1 corresponds to a positive association with non-

survival whereas a ratio less than 1 corresponds to a negative association. The overall

performance of the multivariable regression model in predicting non-survival was

assessed by use of ROC curve analysis. The relationship between SDNN and HR was

investigated by calculating correlation coefficients using the method described by Bland

and Altman19 to account for repeated observations from individual animals. For all

analyses, P < 0.05 was considered statistically significant. Statistical analyses were

performed with commercially available statistical software.d,e

89

RESULTS

Animals

Sixty-seven horses were enrolled and included 45 geldings, 20 mares, and 2

stallions with a weight of 518.8 ± 85.9 kilograms. Horses ranged in age from 2 to 28

years (12.5 ± 6.5 years). Breeds represented included 19 Quarter Horse-type, 18

Warmbloods, 13 Thoroughbreds, 6 Arabians, 3 Saddlebreds, 3 pony breeds, 2 Morgans,

and 1 each Irish Sport horse, Connemara and Lusitano. Groups were comprised of 35

horses in the non-ischemic group, 22 horses in the ischemic group and 10 horses in the

control group. Control horses underwent general anesthesia for the following procedures:

arthroscopy (n=3), arthrodesis (n=1), laryngotomy (n=1), laryngoplasty (n=1), corneal

laceration repair (n=1), neurectomy (n=1) and exploratory laparotomy for an unrelated

research project (n=2). Lesions for horses in the non-ischemic group were as follows:

right dorsal displacement of the large colon (n=8), ileal impaction (n=6), left dorsal

displacement of the large colon (n=6), non-strangulating lipoma (n=3), small intestinal

mesenteric volvulus (n=3), cecal impaction (n=3), enterolithiasis (n=2), generalized small

intestinal distension (n=1), gastrosplenic entrapment (n=1), small colon impaction (n=1)

and large colon impaction (n=1). Lesions for horses in the ischemic groups were as

follows: strangulating lipoma (n=12), large colon volvulus without resection (n=5), large

colon volvulus with resection (n=2), epiploic foramen entrapment (n=1), gastrosplenic

entrapment (n=1) and inguinal hernia (n=1). 58 horses survived to discharge and 9 were

euthanized prior to discharge. Horses that died had a diagnosis of colonic volvulus (n=4),

strangulating lipoma (n=3) or non-strangulating small intestinal lesion (n=2). Mean time

at euthanasia was 5.0 ± 3.8 days after surgery. There was no breed or sex predisposition

90

for lesion type or survival. Horses in the ischemic group (15.9 ± 6.9) were significantly

older than the non-ischemic (11.7 ± 5.8; P = 0.024) and control (6.8 ± 3.5; P = 0.001)

groups, and the non-ischemic group was significantly older than the control group (P =

0.037). However, age of survivors (12.3 ± 6.6) was not significantly (P = 0.715) different

from that of non-survivors (13.2 ± 6.8).

Post-operative analgesia administration

All horses in the colic group (n=57) received flunixin meglumine, at 1.1 mg/kg

intravenously every 12 hours for 48-72 hours for pain management attributable to the

abdominal incision. Horses in the control group received either flunixin meglumine (1.1

mg/kg IV or PO q 12-24 hours) or phenylbutazone (2.2 mg/kg IV or PO q 12-24 hours) in

the first 24 hours post-operatively. Use of other analgesics varied at the discretion of the

attending veterinarian, with intravenously infused lidocaine being the next most

commonly used agent (39/67 including 22/22 ischemic, 17/35 non-ischemic, and 0/10

control horses).

Heart rate and heart rate variability between groups and over time

Admission HR was significantly (P < 0.001) greater for horses in the ischemic

group (61.6 ± 14.3 beats/minute) compared to the non-ischemic (49.6 ±12.4

beats/minute) and control groups (38.8 ± 5.9 beats/minute), but the difference between

non-ischemic and control horses was not significant.

There was no significant interaction between group and time for any of the post-

operative HR, time (SDNN, RMSSD, pNN50, and triangular index) or frequency

(LF/HF, LF, HF) domain variables. Post-operative mean HR for 5-minute telemetric

ECG recordings (mean HR5 min) was significantly different between groups irrespective

91

of time (P < 0.001) with the ischemic group having the highest mean HR5 min at any time

point (Figure 4.1A). The mean HR5 min at Time 1 (2-10 hours) was significantly greater

than mean HR5 min at all other post-operative time points (P < 0.003) and mean HR5 min at

Time 2 was significantly greater than mean HR5min at Time 4 (Figure 4.1A). A

significant group effect (P < 0.001) was found for time domain variables, wherein horses

in the ischemic group had significantly decreased values for SDNN5 min (Figure 4.1B),

RMSSD5 min (Figure 4.1C), and pNN505 min (Figure 4.1D) compared to both non-ischemic

and control groups, and horses in the non-ischemic group had significantly decreased

values compared to the control group. Heart rate variability parameters derived from 30-

minute recording lengths yielded the same results with the exception that for pNN5030 min,

ischemic horses had significantly lower pNN5030 min values compared to non-ischemic

and control horses; however, the non-ischemic horses were not significantly different

than control horses. For the time domain variable of triangular index, there was a group

effect for 30-min (p < 0.009) ECG recordings, wherein the ischemic group values were

significantly lower than the non-ischemic and control groups. There was a significant

negative correlation between HR5min and SDNN5 min (r = -0.518; P < 0.001). There was no

effect of time for any of time domain-derived HRV parameters. There was no effect of

group or time for any of the frequency domain variables.

Heart rate and heart rate variability and survival to discharge

Admission HR was lower in horses that survived to discharge (50.4 ± 14.3

beats/minute) compared to those euthanized in the post-operative period (61.2 ± 12.2

beats/minute; P = 0.013). There was no interaction between survival group and time for

any of the dependent variables. Compared to non-survivors, horses that survived had

92

significantly lower mean HR5 min (P < 0.001) and higher SDNN5 min (P < 0.001) (Figure

4.2A and 4.2B, respectively). Post-operative mean HR5 min was significantly greater in

non-survivors (65.3 ± 3.4 beats/minute (P <0.001) compared to horses that survived to

discharge (45.3 ± 1.3 beats/minute). An effect of time for survival groups (P < 0.001) was

detected where post-operative mean HR5 min was greater at Time 1 when compared to

Time 3 or Time 4 (Figure 4.2A). For HRV parameters derived by time domain methods

there was an effect of survival group, but no effect of time for SDNN and RMSSD for

both 5- and 30-minute ECG recordings. Horses that survived had significantly greater

SDNN5 min (52.8 ± 3.1 ms; P < 0.001) and RMSSD5 min (59.5 ± 4.2 ms; P < 0.001) than

non-survivors (23.9 ± 8.0 ms and 18.9 ± 10.9 ms, for SDNN5 min and RMSSD5 min,

respectively) (Figure 4.2B and 4.2C). The pNN505 min and pNN5030 min (P < 0.001) and

triangular index30 min (P = 0.003) were significantly greater in surviving horses. A

significant effect of time was detected for survival data for pNN50 and triangular index,

wherein pNN505 min (P = 0.01; Figure 4.2D) and triangular index30 min (P = 0.023) were

greater at Time 4 compared to Time 1 or 2 and pNN5030 min was significantly (P = 0.016)

greater at Time 4 compared to Time 1.

There was no effect of time on any of the frequency domain measures of HRV

when data were stratified by survival vs. non-survival. However, there was a significant

group effect for the LF/HF5 min and LF/HF 30 min where survivors had a lower LF/HF ratio

than non-survivors [(LF/HF5 min survivor: 0.90 ± 0.07/non-survivor: 1.20 ± 0.18, P =

0.045),( LF/HF30 min survivor: 0.88 ± 0.05/ non-survivor: 1.27 ± 0.13, P = 0.023)]. In

addition, LF power in normalized units (n.u.) was lower in survivors compared to non-

survivors for both 5- and 30-minute ECG recordings [(LF power5 min: survivor: 42.67 ±

93

1.38 n.u./ non-survivor: 50.76 ± 3.43 n.u.; P = 0.031), (LF power30 min: survivor: 40.3 ±

1.4 n.u./non-survivor: 50.0 ± 3.77 n.u.; P = 0.018)]. Finally HF power5 min was

significantly greater in survivors (59.7 ± 1.4 n.u.) compared to non-survivors (51.6 ± 3.7

n.u., P = 0.043). No significant effects of group or time were found for the remaining

frequency parameters when stratified by survival (LF/HF1024, LF power1024, HF

power1024, HF power 30 min).

Variables significantly associated with non-survival by univariable logistic

regression are presented in Table 4.1. Variables retained in the multivariable model

(overall significance of the model p < 0.0001) included peak mean HR5 min, SDNN5 min

when heart rate ≤ 55 bpm, and time period at peak mean HR5 min (Table 4.2). The model

correctly predicted non-survival in 94% of cases with a ROC curve AUC of 0.95 ± 0.029

(p < 0.0001). The sensitivities, specificities and predictive values (at the observed

prevalence of 13.4% non-survival) for possible outcomes of the multivariable model

corresponding to cut-off values to predict non-survival are available in supplemental

information (Table 4.3). The ROC AUC of the multivariable logistic regression model to

predict non-survival (0.95 ± 0.029) was significantly (P = 0.0124) higher than that of

admission HR (0.762 ± 0.075).

DISCUSSION

In this prospective, clinical, observational study, HRV parameters measured by

time and frequency domain methods were described for horses with acute gastrointestinal

disease that underwent exploratory laparotomy. The null hypothesis was rejected as

horses with ischemic gastrointestinal lesions had reduced HRV compared to those with

94

non-ischemic lesions and control cases. To the authors’ knowledge, this is the first

reported study evaluating HRV in the post-operative period for horses undergoing

surgical exploratory laparotomy for colic. In the post-operative period, time domain

variables that estimate either overall HRV (SDNN and triangular index) or high

frequency variation in heart rate (RMSSD and pNN50) were decreased in horses that

underwent exploratory laparotomy for colic, and the magnitude to which this occurred

corresponded to lesion category (Fig. 4.1). In addition, non-survivors had a significantly

reduced SDNN5 min, increased LF/HF ratio, increased LF power, and decreased HF power

compared to survivors, corresponding to a reduction in overall HRV, and sympathovagal

imbalance marked by sympathetic dominance and parasympathetic withdrawal.

The finding that horses with ischemic gastrointestinal disease, and specifically

those that did not survive, have altered HRV is novel and may have important

pathophysiologic and clinical implications related to cardiovascular function. Altered

HRV may be a function of reduced vagal tone, increased sympathetic tone, withdrawal of

both branches of the ANS or due to abnormalities at the level of the cardiac pacemaker

cells.20 Horses with ischemic gastrointestinal disease appeared to have reduced vagal tone

while non-survivors had both vagal withdrawal and relatively increased sympathetic

modulations of HR period. Reduced HRV reflects a loss of complexity in the autonomic

input to the sinoatrial node. It implies autonomic dysfunction and may occur with either

primary or secondary cardiac dysfunction. This, in turn, may perpetuate organ

insufficiency remote from the heart itself.20-23 Therefore, monitoring HRV in horses in

the post-operative period may provide a basic understanding of the role of the ANS in

95

critical illness while identifying horses at an increased risk of non-survival at an earlier

stage.

It is well documented that parasympathetic tone prevails in the normal resting

horse.17,24 Horses with acute colic are expected to have increased sympathetic tone, a

finding which has been supported based on reports of increased blood concentrations of

cortisol and catecholamines in conjunction with an increased admission HR.25,26 Several

studies investigating horses with acute colic have reported a significant association

between admission heart rate and increased likelihood of mortality.8,27-29 While there is a

repeatable association between admission HR and post-operative mortality in the equine

literature, HR should not be used as the sole prognostic indicator in horses with colic.

This is based in part on the finding, in the present study and others,30,31 that admission

HR is rarely retained in final survival models. In the study presented here, admission HR

of >48 beats per minute was associated with a 5.6 times odds of non-survival. Use of this

low HR cut-off would provide excellent sensitivity and negative predictive value to

predict non-survival, but this would come at the expense of abysmal specificity and

positive predictive value. There is no cut-off for admission HR that provides adequate

sensitivity and specificity for non-survival in horses with colic.

In the current study, post-operative peak mean HR5 min (> 66 bpm and > 67 bpm

for 30 min recording; OR 31 and 23, respectively) occurring after 24 hours post-

operatively (corresponding with Time at peak mean HR5 min > 2), was highly associated

with non-survival. This finding is not surprising as a persistently increased post-operative

heart rate (> 60 bpm) has been used as part of the criteria to define post-operative

shock.32 An increased post-operative heart rate may be indicative of ongoing

96

hemodynamic disturbances, persistent pain or primary cardiac insult. While there is an

inverse curvilinear relationship between HR and NN interval, variations in HR in the

present study accounted for only 26.8% of the variation in SDNN, which is similar to

what was reported in people.33 Though it would have been ideal to have case control

matches for HR, this was unrealistic. The independent value of HRV on understanding

autonomic dysfunction was highlighted by the finding that SDNN measured at the time

when the HR was less than or equal to 55 bpm at a cut-off of <39.5 ms, was highly

associated with non-survival (OR 16.42), a finding which remained significant in the

final regression model. From a clinical perspective, this finding underscores a relevant

use of HRV. In the post-operative period, horses with a normal or mildly increased heart

rate (i.e. tachycardia ≤ 55 bpm), that might otherwise be perceived to be clinically

improved, would remain at risk for non-survival or conceivably other major post-

operative clinical events when they have a concurrent reduction in HRV. The duration

and intensity of post-operative monitoring and preventative strategies might be adjusted

if autonomic dysfunction is detected.

Frequency-derived measures of HRV in the present study were not significantly

different when the groups were stratified by lesion type. These results were similar to

those previously reported for the perioperative period in a small group of horses that

underwent general anesthesia and retrobulbar analgesia for enucleation.34 However, the

LF/HF5 min , LF/HF30 min, LF power5 min, LF power 30 min, and HF power5 min were

significantly different between survivors and non-survivors, suggesting sympathovagal

imbalance, increased sympathetic tone, and reduced vagal modulation of heart rate in

non-survivors, respectively. In human medicine there is still debate as to what role

97

baroreceptor function and other physiologic mechanisms play in determining the various

frequency bands, making it unclear what proportion of the LF band is solely reflective of

sympathetic activity.35,36 The time domain methods RMSSD and pNN50 reflect

parasympathetic activity15,37,38 rather than overall autonomic tone. These variables were

lower in non-survivors, which in accordance with significantly lower HF power5 min for

non-survivors, suggests the alteration of HRV in this study was at least in part a reflection

of attenuated vagal tone.

The multivariable logistic regression model out-performed admission HR,

accurately predicting non-survival in 94% of cases with an excellent AUC of 0.95;

however, the practicality of the model is questionable. The variables retained in the

model, peak HR5 min, time of peak mean HR5 min and SDNN when ≤ HR 55 bpm, would

have to be collected after monitoring the horse for a 48-hour period either continuously or

for 5 minutes at approximately 12-hour intervals to obtain adequate information to use

the regression model. Therefore, its potential usefulness in predicting survival to

discharge is limited to horses that survive at least 48 hours after surgery.

Though it was not possible to control analgesic administration, all horses received

perioperative NSAIDs. Horses in the colic group received an NSAID for the duration of

the study period, however, some control cases were not administered an NSAID on Day 2

(Time 3 and 4). Provision of pain control in any form would be expected to reduce

sympathetic tone and thereby reduce LF power and indirectly increase HF power.14 In

this study there was no difference in frequency spectrum HRV parameters between the

colic and control group, suggesting analgesic drug administration was not an important

manipulator of autonomic tone. The finding that non-survivors had increased LF power

98

and reduced HF power would imply that a subset of horses experienced sympathovagal

imbalance despite analgesic administration.

Several limitations exist for this study, including the lack of pre-operative HRV

analysis and the inability to control for time of day when HRV was analyzed. While pre-

operative measurements are ideal for use as diagnostic criteria or prognostic indicators,

this was not practical owing to variability in the time from admission to surgery, and in

some cases obtaining a 30-minute diagnostic-quality ECG recording would have delayed

surgery and could be considered unethical. Finally, at the time of admission the vast

majority of horses that presented for acute gastrointestinal disease received an alpha-2

agonist or an anticholinergic spasmolytic agent, making interpretation difficult owing to

the pharmacological influence on HRV.11,39

Additionally, the post-operative time at which HRV analysis was assessed varied

between horses. This was a result of the need for uninterrupted, artifact- and arrhythmia-

free telemetric ECG recordings of 30 minutes duration. While the ECG sampling times

were limited to an 8-hour window representing the immediate post-operative period (2-10

hours) and then 24, 36 and 48 hours post-operatively without overlap, the exact time of

day that this occurred varied from horse to horse, and was dependent on the time of

recovery from surgery. For each individual horse, however, samples were obtained at

approximately the same time of day (morning and evening, or afternoon and mid-night).

It is important to note that there was no significant interaction between group and time

suggesting that the significant differences detected between ischemic, non-ischemic and

control groups and between survivors and non-survivors occurred irrespective of post-

operative time period. On the other hand, a lack of a significant effect of post-operative

99

time period could be attributable to the sampling method. For practical purposes,

identifying significant HRV measures based on the post-operative time period may be

more relevant and more convenient than sampling at a set time of day. However, the

authors acknowledge that additional studies might be warranted to determine the effect

that time of day alone has on HRV in critically ill horses.

CONCLUSION

In conclusion, HRV analysis is a non-invasive, easily attainable measure of

cardiovascular health that provides information about ANS function, and which might

prove useful as a clinical monitoring tool. Horses with colic that incur ischemic

gastrointestinal lesions have decreased time domain measures of HRV compared to

horses that undergo general anesthesia for non-ischemic gastrointestinal lesions or

elective surgical procedures. Horses that did not survive to discharge had altered HRV

with evidence of sympathovagal imbalance and vagal attenuation compared to those that

survived to discharge. Additional work is needed to determine the relationship between

alterations in HRV, other measures of cardiovascular health, and measures of the

systemic inflammatory response in horses.

100

FOOTNOTES

a: Televet Version 3 Kruuse Denmark

b: Kruuse ECG Electrodes; Jorgenson Medical

c: Kubios HRV Version 2.1 Kuopio, Finland kubios.uef.fi/

d:Sigmaplot 12.5 Systat Software, Inc. San Jose, CA

e: MedCalc 14.8.1;Ostend, Belgium

REFERENCES

1. Malik M. Heart rate variability. Standards of measurement, physiological

interpretation, and clinical use. Task Force of the European Society of Cardiology and the

North American Society of Pacing and Electrophysiology. Eur Heart J 1996;17:354-381.

2. von Borell E, Langbein J, Despres G, et al. Heart rate variability as a measure of

autonomic regulation of cardiac activity for assessing stress and welfare in farm animals -

- a review. Physiol Behav 2007;92:293-316.

3. Malik M, Camm AJ. Components of heart rate variability--what they really mean and

what we really measure. Am J Cardiol 1993;72:821-822.

4. Coviello I, Pinnacchio G, Laurito M, et al. Prognostic role of heart rate variability in

patients with ST-segment elevation acute myocardial infarction treated by primary

angioplasty. Cardiology 2013;124:63-70.

5. Huikuri HV, Makikallio TH. Heart rate variability in ischemic heart disease. Auton

Neurosci 2001;90:95-101.

6. Borovikova LV, Ivanova S, Zhang M, et al. Vagus nerve stimulation attenuates the

systemic inflammatory response to endotoxin. Nature 2000;405:458-462.

101

7. Pontet J, Contreras P, Curbelo A, et al. Heart rate variability as early marker of

multiple organ dysfunction syndrome in septic patients. J Crit Care 2003;18:156-163.

8. Balanescu S, Corlan AD, Dorobantu M, et al. Prognostic value of heart rate variability

after acute myocardial infarction. Med Sci Monit 2004;10:CR307-315.

9. Chen WL, Chen JH, Huang CC, et al. Heart rate variability measures as predictors of

in-hospital mortality in ED patients with sepsis. Am J Emerg Med 2008;26:395-401.

10. Ohmura H, Boscan PL, Solano AM, et al. Changes in heart rate, heart rate

variability, and atrioventricular block during withholding of food in Thoroughbreds. Am

J Vet Res 2012;73:508-514.

11. Sundra TM, Harrison JL, Lester GD, et al. The influence of spasmolytic agents on

heart rate variability and gastrointestinal motility in normal horses. Res Vet Sci

2012;93:1426-1433.

12. Nagel C, Aurich J, Palm F, et al. Heart rate and heart rate variability in pregnant

warmblood and Shetland mares as well as their fetuses. Anim Reprod Sci 2011;127:183-

187.

13. Schmidt A, Hodl S, Mostl E, et al. Cortisol release, heart rate, and heart rate

variability in transport-naive horses during repeated road transport. Domest Anim

Endocrinol 2010;39:205-213.

14. Rietmann TR, Stauffacher M, Bernasconi P, et al. The association between heart rate,

heart rate variability, endocrine and behavioural pain measures in horses suffering from

laminitis. J Vet Med A Physiol Pathol Clin Med 2004;51:218-225.

15. Bowen IM. Ambulatory electrocardiography and heart rate variability In: Marr CM,

Bowen, I.M., ed. Cardiology of the horse, Second ed. Toronto: Saunders Elsevier; 2010.

102

16. Schmidt A, Mostl E, Wehnert C, et al. Cortisol release and heart rate variability in

horses during road transport. Horm Behav 2010;57:209-215.

17. Kuwahara M, Hashimoto S, Ishii K, et al. Assessment of autonomic nervous function

by power spectral analysis of heart rate variability in the horse. J Auton Nerv Syst

1996;60:43-48.

18. Malik M. Heart rate variability: standards of measurement, physiological

interpretation and clinical use. Task Force of the European Society of Cardiology and the

North American Society of Pacing and Electrophysiology. Circulation 1996;93:1043-

1065.

19. Bland JA, DG. Calculating correlation coefficients with repeated observations: Part

1--Correlation within subjects. BMJ 1995:446.

20. Werdan K, Schmidt H, Ebelt H, et al. Impaired regulation of cardiac function in

sepsis, SIRS, and MODS. Can J Physiol Pharmacol 2009;87:266-274.

21. Hoyer D, Friedrich H, Zwiener U, et al. Prognostic impact of autonomic information

flow in multiple organ dysfunction syndrome patients. Int J Cardiol 2006;108:359-369.

22. Xu L, Li CS. [Relationship between heart rate variability and serum levels of thyroid

hormones, cortisol and prognosis in patients with systemic inflammatory response

syndrome]. Zhongguo Wei Zhong Bing Ji Jiu Yi Xue 2007;19:160-164.

23. Chiu IM, von Hehn CA, Woolf CJ. Neurogenic inflammation and the peripheral

nervous system in host defense and immunopathology. Nat Neurosci 2012;15:1063-1067.

24. Hamlin RL, Klepinger WL, Gilpin KW, et al. Autonomic control of heart rate in the

horse. Am J Physiol 1972;222:976-978.

103

25. Hinchcliff KW, Rush BR, Farris JW. Evaluation of plasma catecholamine and serum

cortisol concentrations in horses with colic. J Am Vet Med Assoc 2005;227:276-280.

26. Mair TS, Sherlock CE, Boden LA. Serum cortisol concentrations in horses with

colic. Vet J 2014.

27. Mair TS, Smith LJ. Survival and complication rates in 300 horses undergoing

surgical treatment of colic. Part 1: Short-term survival following a single laparotomy.

Equine Vet J 2005;37:296-302.

28. Hassel DM, Hill AE, Rorabeck RA. Association between hyperglycemia and

survival in 228 horses with acute gastrointestinal disease. J Vet Intern Med

2009;23:1261-1265.

29. Proudman CJ, Dugdale AH, Senior JM, et al. Pre-operative and anaesthesia-related

risk factors for mortality in equine colic cases. Vet J 2006;171:89-97.

30. Proudman CJ, Edwards GB, Barnes J, et al. Factors affecting long-term survival of

horses recovering from surgery of the small intestine. Equine Vet J 2005;37:360-365.

31. van der Linden MA, Laffont CM, Sloet van Oldruitenborgh-Oosterbaan MM.

Prognosis in equine medical and surgical colic. J Vet Intern Med 2003;17:343-348.

32. Mair TS, Smith LJ. Survival and complication rates in 300 horses undergoing

surgical treatment of colic. Part 2: Short-term complications. Equine Vet J 2005;37:303-

309.

33. Billman GE. The effect of heart rate on the heart rate variability response to

autonomic interventions. Front Physiol 2013;4:222.

104

34. Oel C, Gerhards H, Gehlen H. Effect of retrobulbar nerve block on heart rate

variability during enucleation in horses under general anesthesia. Vet Ophthalmol

2014;17:170-174.

35. Houle MS, Billman GE. Low-frequency component of the heart rate variability

spectrum: a poor marker of sympathetic activity. Am J Physiol 1999;276:H215-223.

36. Rahman F, Pechnik S, Gross D, et al. Low frequency power of heart rate variability

reflects baroreflex function, not cardiac sympathetic innervation. Clin Auton Res

2011;21:133-141.

37. Heart rate variability: standards of measurement, physiological interpretation and

clinical use. Task Force of the European Society of Cardiology and the North American

Society of Pacing and Electrophysiology. Circulation 1996;93:1043-1065.

38. Campos LA, Pereira VL, Jr., Muralikrishna A, et al. Mathematical biomarkers for the

autonomic regulation of cardiovascular system. Front Physiol 2013;4:279.

39. Ehara T, Ogawa Y, Kato J, et al. The effect of dexmedetomidine on arterial-cardiac

baroreflex function assessed by spectral and transfer function analysis. J Anesth

2012;26:483-489.

105

Table 4.1. Ideal cut off as assessed by ROC curve analysis for HRV parameters potentially (P < 0.1) associated with non-survival and corresponding odds ratios. OR = odds ratio; CI = confidence interval; Peak= highest single value of the four post-operative time points analyzed; Lowest= nadir value of the four post-operative time points analyzed

Variable Cut off ROC Logistic regression AUC ± SE P OR (95% CI) P

Admission HR (bpm) > 48 0.762 ± 0.075 0.0005 5.63 (1.08 to 29.48) 0.0407 Peak mean HR30 min (bpm) > 67 0.876 ± 0.058 < 0.0001 31.33 (3.57 to 275.4) 0.0019 Peak mean HR5 min (bpm) > 66 0.883 ± 0.052 < 0.0001 23.45 (2.71 to 203.5) 0.0042 Time at peak mean HR30 min (1-4) > 1 0.739 ± 0.101 0.0174 5.85 (1.08 to 31.66) 0.0403 Time at peak mean HR5 min (1-4) > 2 0.760 ± 0.099 0.0089 8.17 (1.67 to 39.84) 0.0094 HR at lowest SDNN30 min (bpm) > 67 0.885 ± 0.052 < 0.0001 28.32 (3.24 to 247.5) 0.0025 HR at lowest SDNN5 min (bpm) > 67 0.850 ± 0.056 < 0.0001 19.44 (3.47 to 109.1) 0.0070 Lowest SDNN30 min (ms) < 28.0 0.846 ± 0.052 < 0.0001 7.97 (1.51 to 42.20) 0.0146 Lowest SDNN5 min (ms) < 26.7 0.874 ± 0.051 < 0.0001 13.45 (1.57 to 114.9) 0.0175 Lowest RMSSD30 min (ms) < 16.7 0.907 ± 0.041 < 0.0001 39.20 (4.40 to 349.3) 0.0010 Lowest RMSSD5 min (ms) < 17.2 0.912 ± 0.042 < 0.0001 31.33 (3.57 to 275.4) 0.0019 Lowest pNN5030 min (%) < 1.4 0.907 ± 0.042 < 0.0001 51.00 (5.60 to 464.2) 0.0005 Lowest pNN505 min (%) < 0.6 0.904 ± 0.040 < 0.0001 51.00 (5.50 to 464.2) 0.0005 Peak LF/HF30 min > 1.9 0.809 ± 0.072 < 0.0001 13.50 (2.72 to 67.04) 0.0015 Peak LF/HF1024 > 1.5 0.739 ± 0.086 0.0053 8.17 (1.54 to 43.20) 0.0135 Peak LF power30 min (n.u.) > 63.2 0.829 ± 0.073 < 0.0001 15.27 (2.78 to 83.83) 0.0017 Lowest HF power30 min (n.u) < 36.6 0.783 ± 0.081 0.0004 8.73 (1.88 to 40.42) 0.0056 Lowest HF power5 min (n.u.) < 42.5 0.670 ± 0.071 0.0164 7.73 (0.91 to 65.77) 0.0611 SDNN5 min lowest mean HR5 min (ms) ≤ 39.2 0.868 ± 0.049 < 0.0001 16.42 (1.91 to 141.0) 0.0107

106

Table 4.2. Results of a multivariable logistic regression analysis for HRV variables associated with non-survival.

Variable Coefficient SE P value OR 95% CI

Constant -6.9847 N/A N/A N/A N/A

Peak mean HR5 min (bpm)

0.0802 0.0361 0.0262 1.0835 1.010 to 1.163

Time at peak mean HR5

min (1-4) 1.6230 0.7810 0.0377 5.0683 1.097 to 23.43

SDNN5 min at lowest mean HR5 min (ms)

-0.1053 0.0536 0.0498 0.9001 0.810 to 0.999

107

Figure 4.1. Least squares mean and standard error of the mean derived from the 5-minute ECG recordings at each post-operative time period [Times 1 (2-10 hours), 2 (16-24 hours), 3 (30-38 hours), and 4 (44-52 hours) ] are presented for mean heart rate (A) and time domain heart rate variability parameters [SDNN (B), RMSSD (C) and pNN50 (D)] for control horses (n=10), horses with non-ischemic gastrointestinal lesions (n=35) and horses with ischemic gastrointestinal lesions (n=22). Within a time period, a significant difference between groups is indicated by different letters. Significant differences across time periods are indicated by different symbols.

108

Figure 4.2. Least squares mean and standard error of the mean for 5-minute ECGs derived from post-operative time points [Times 1 (2-10 hours), 2 (16-24 hours), 3 (30-38 hours), and 4 (44-52 hours)] for mean heart rate (A), and time domain variables [SDNN (B), RMSSD (C), and pNN50 (D)] for horses when grouped as survivors (n=58) or non-survivors (n=9). Within a time period, a significant difference between groups is indicated by different letters. Significant differences across time periods are indicated by different symbols.

109

CHAPTER 5

ASSESSMENT OF THE CARDIOVASCULAR SYSTEM IN HORSES WITH

NATURALLY ACQUIRED ISCHEMIC INTESTINAL DISEASE3

______________________________

2E.L. McConachie, S. Giguère, G. Rapoport, S. A. Brown, M.H Barton. To be

submitted to the Journal of Veterinary Cardiology.

110

ABSTRACT

Objectives- To compare indices of cardiovascular system status between horses with

acute surgical colic and those undergoing elective surgical procedures at admission, and

at days 1 and 2 post-operatively.

Design- Prospective clinical study

Animals- Adult horses presented to a Veterinary Teaching Hospital for acute

gastrointestinal (GI) disease requiring exploratory laparotomy (n=62) or for elective

surgical procedures (control, n=12).

Procedures- Horses were categorized by specific surgical GI lesion, presence or absence

of hypotension under anesthesia, presence or absence of SIRS, and survival or failure to

survive to discharge. A complete blood count, blood lactate, electrolytes, and serum

cardiac troponin I (cTnI) concentrations were measured at admission and at Day 1 and 2

post-operatively. The cardiovascular status was assessed post-operatively by telemetric

electrocardiography for determination of rate, rhythm and heart rate variability (HRV),

and by 2-D and M-mode echocardiography.

Results- Horses with ischemic GI disease had significantly higher cTnI and lactate

concentrations, higher heart rate, lower HRV, greater mean systemic arterial blood

pressure, greater left ventricular relative wall thickness, were more frequently

hypotensive under anesthesia, and had more ectopic beats per hour than horses with non-

ischemic lesions or controls. Horses with an ischemic GI lesion (18/29; 62%) or those

that did not survive (8/10; 80%) were more likely to fulfill SIRS criteria. In a

multivariable logistic regression (MLR) model, non-survival was best explained by cTnI

(> 0.15 ng/mL) and lowest stroke volume index.

111

Conclusions and Clinical Relevance- Horses with ischemic GI lesions and non-

survivors are more likely to have evidence of cardiovascular system dysfunction.

INTRODUCTION

As many as 60% of septic human patients admitted to the intensive care unit have

evidence of cardiac dysfunction and, in these patients, mortality rates approach 70%.1-3 It

should not be too surprising that cardiac dysfunction plays a central role in the

development of multiple organ dysfunction syndrome (MODS) for it is in part the

inability of the cardiovascular system to meet tissue oxygen demands that leads to remote

organ failure. The clinical importance of detecting cardiac dysfunction and preventing

MODS is highlighted by the fact that once a patient develops multiple organ failure,

mortality in people approaches 100%.4,5

On the basis of perioperative increases in serum cardiac troponin I (cTnI)

concentrations, recent literature provides evidence that myocardial injury occurs in

critically ill horses with naturally occurring acute ischemic or inflammatory conditions of

the gastrointestinal tract.6-8 In critical illness, the pathophysiology of myocardial injury is

multifactorial with proposed mechanisms for myocardial dysfunction that include

hypoxic or ischemic insult, sepsis or endotoxin-induced cytokine expression, over

expression of myocardial nitric oxide, alterations in intracellular calcium signaling, and a

blunted response or decrease density of β-adrenergic receptors to catecholamines.9,10

Horses with naturally acquired ischemic gastrointestinal disease commonly have

clinical evidence of hypovolemia and up to 40% have circulating endotoxin. 11-13

Experimental infusion of endotoxin in horses was associated with a rise in serum cTnI

112

concentration that preceded development of ventricular ectopic beats, suggesting a direct

or indirect role for endotoxemia in inducing myocardial injury.14 Despite the association

between an increased serum cTnI concentration and poor outcome in horses with acute

gastrointestinal disease,6-8 the implications of a rise in this cardiac biomarker are

incompletely understood in horses.

Criteria utilized in people to assess cardiovascular function in severe illness

include hypotension, hypotension that is non-responsive to inotropic therapy, increased

pressure-adjusted heart rate (PAR= [Heart Rate (HR) x central venous pressure (CVP)/

mean arterial pressure (MAP)]),15 plasma lactate or serum cTnI concentrations and

incidence of cardiac arrhythmias as well as decreased FS, left ventricular ejection fraction

and HRV, and increased incidence of cardiac arrhythmias.2,5,15-20 Current methods for

evaluating cardiac function in horses often follow the criteria defined for use in humans.21

There is no consensus for what constitutes complete assessment of cardiovascular system

function in healthy or critically ill horses. With the realization that there are important

differences between species, it seems logical to adopt a multifaceted, comprehensive

approach to cardiovascular examination. Until sufficient data are available to dictate

which measurements are the most useful to guide intervention or prognosis, assessment

of left ventricular systolic function, presence or absence of cardiac arrhythmia, systemic

arterial blood pressure, and measures of autonomic modulation of cardiac function may

be useful. Therefore, the objectives of the study described herein were to obtain indices

of cardiovascular system function in horses undergoing surgical correction of GI disease

and horses undergoing elective surgical procedures and to compare these indices with

measures of gastrointestinal disease severity and short-term survival. The null hypotheses

113

were that horses undergoing general anesthesia (GA) for surgical treatment of ischemic

GI disease are not more likely to develop evidence of myocardial dysfunction compared

to horses undergoing surgical treatment of non-ischemic GI disease or elective surgical

procedures, and that indices of myocardial function are not correlated with the presence

of the systemic inflammatory response syndrome (SIRS) or non-survival in horses with

acute GI disease.

MATERIALS AND METHODS

Animals

Client-owned horses > 1 year of age that were presented to the University’s

Veterinary Teaching Hospital for acute GI disease that required exploratory laparotomy

were enrolled, as well as those presented for elective surgery. Owner consent was

obtained for all enrolled animals. Horses were enrolled as controls if they required GA

for an elective surgical procedure and were healthy on the basis of normal pre-operative

vital parameters, general physical examination and normal pre-operative complete blood

count, fibrinogen concentration and plasma venous blood gas analysis. Horses were

removed from the study if they were euthanized solely for financial reasons or if they

were not recovered from GA. Post-operative therapy was at the discretion of the

attending clinician. This study was approved by the University’s Clinical Research

Committee.

Data collection

Age, breed, sex, the nature of the GI lesion diagnosed and corrected at surgery,

elective procedure and survival to discharge were recorded for each horse. Horses were

114

grouped as ischemic or non-ischemic based on surgical findings, or control for horses

undergoing elective surgical procedures. Specifically, horses with small or large intestinal

lesions requiring resection and anastomosis were grouped as ischemic. Horses with a

colon volvulus greater than or equal to 360 degrees were placed in the ischemic group

regardless of whether or not a resection was performed. Hypotension (yes or no) under

GA, defined as a direct MAP < 65 mmHg for at least five minutes, and total duration of

hypotension were recorded.

For assessment of SIRS, vital parameters (rectal temperature, HR, respiratory

rate) were recorded at admission and at 12 hour intervals post-operatively. Blood was

also collected into plastic coated EDTA tubes via venipuncture at admission or through a

jugular venous catheter, for immediate determination of complete blood count with an

automated analyzera at admission and at approximately 24 (Day 1) and 48 hours (Day 2)

post-operatively. Blood smears were made for manual description of leukocyte

morphologic characteristics and were processed by the University’s clinical pathology

laboratory at admission and on Day 1 and 2. Horses that fulfilled criteria for SIRS at

admission, or on Day 1 or Day 2 post-operatively were categorized as having SIRS based

on criteria used for adult horses in similar clinical studies and included two or more of the

following: temperature ≥ 101.5°F or ≤ 98.5°F; HR ≥ 60 bpm; respiratory rate ≥ 30 bpm;

white blood cell count ≥ 14,500 cells/µL or ≤ 4,500 cells/µL and or ≥ 10% band

neutrophils.22 23

115

Plasma and serum biochemical analysis for cTnI, lactate and electrolyte

concentrations

Blood was collected into serum tubes via venipuncture at admission or through a

jugular venous catheter after discarding 12 mL of waste blood, at approximately 24 and

48 hours post-operatively for serum cTnI measurement. Blood was allowed to clot at

room temperature and was then centrifuged for 30 min at 3000 rpm in a temperature-

controlled centrifugeb prior to separating serum into plastic bullet tubes and storing at -

80°C for batch analysis. All cTnI samples were processed within three months of

collection using an ultrasensitive assay.c Heparinized whole blood was collected for

immediate measurement of lactate and electrolyte (Na+, K+, iCa2+, iMg2+) concentrations

by a rapid critical care analyzer.d

Telemetric ECG placement for HR, HRV, and rhythm analysis

Within two hours of recovering from GA a veterinary telemetry unite was placed

for a minimum of 48 hours and maximum of 52 hours of continuous electrocardiographic

recording. Electrodesf were placed according to manufacturer instructions. A surcingle

was placed to ensure the telemetry unit and leads remained in place and to promote

contact between the electrodes and the horse’s skin. All ECG analysis was performed by

a board certified internal medicine clinician (EM). The HR (bpm), measured from real-

time telemetric electrocardiography (ECG) monitoring, was recorded in triplicate at the

time of pressure measurements at 12, 24, 36 and 48 hours post-operatively.

Telemetric ECG recordings were processed as previously described24 with

universally available softwarel for HRV analysis. Based on previous work,27 5- or 30-

minute artifact- and arrhythmia-free ECG recording was used for determination of the

116

standard deviation of normal-to-normal intervals (SDNN, 5 minute) and the root mean

square of successive differences (RMSSD, 5 minute), and low frequency (LF, 30 minute)

power , respectively. The HRV results were recorded for each horse during the following

post-operative time periods 2-10 hours (Time 1), 16-24 hours (Time 2), 30-38 hours

(Time 3) and 44-52 hours (Time 4).

The number of ectopic beats (ectopic beats/24 hour), their origin, and

morphologic characteristics were recorded for each horse. The presence of any clinically

significant arrhythmia (CSA, yes or no) during the study period was recorded for each

horse. A CSA was defined as having > 1 single ventricular premature complex/hour,

ventricular tachycardia (≥ 4 consecutive ventricular premature complexes in a row, rate ≥

60 bpm), > 1 supraventricular premature complex/ hour, supraventricular tachycardia,

polymorphic ventricular premature complexes, accelerated idioventricular rhythm25,26 or

second degree atrioventricular (AV) block at a HR of ≥ 60 beats/minute.

Central venous catheterization and blood pressure monitoring

A 19-gauge 90-cm long line catheterg was placed in the left or right jugular vein

using the previously placed 14-gauge anesthesia catheter as an introducer catheter after

sterile preparation of the skin and catheter hub. When the catheter placed by the

anesthesiologist was not available to use as an introducer, the introducer catheter

included in the commercially available equine central venous pressure catheter kith was

placed after clipping and sterile preparation of the skin over the jugular groove. The tip of

the catheter was placed within the thoracic inlet at the level of the junction of the cranial

vena cava and right atrium. This was achieved by first measuring the distance from the

catheter insertion site to the mid-point of the triceps muscle at the level of the point of the

117

shoulder prior to inserting the catheter. Positioning was further confirmed by optimizing

the pressure wave form by incrementally backing the catheter out of the right ventricle

while the pressure transduceri was set at the level of the point of the shoulder and

connected to a stall side electronic pressure monitork as previously reported.27 The

catheter was sutured in place, flushed with heparinized saline and maintained indwelling

for the duration of the post-operative observation period. Central venous pressure

(mmHg) measured by a pressure transduceri and electronic pressure monitork was

recorded in triplicate at 12, 24, 36 and 48 hours post-operatively. Horses were restrained

with a halter and lead rope and the head was maintained in a neutral position at the level

of the withers. The MAP (mmHg) measurements were acquired in triplicate using an

oscillometric pressure monitork and tail pressure cuff with a width that was

approximately 50% of the tail circumference.28 The head of each horse was maintained in

a constant position during blood pressure measurement. Noninvasive blood pressure was

measured simultaneously with CVP and HR. The pressure adjusted heart rate (PAR =

HR X CVP/MAP)15 was calculated at 12, 24, 36 and 48 hours post-operatively from the

average HR, MAP and CVP measured at each time point. Therefore, one PAR was

recorded for each post-operative time point.

Echocardiography

For 2-D and M-mode echocardiography, all measurements were obtained on Day

1 and Day 2 post-operatively. Horses were examined stall side or in a quiet examination

room with minimal restraint (halter and lead rope). A single-lead ECG was placed in

base-apex fashion for continuous monitoring during echocardiography.

Echocardiography was performed using an ultrasound unitj with a built-in algorithm for

118

stroke volume (SV) and cardiac output (CO) determination and simultaneous display of

the ECG. All views were obtained from the right parasternal window with a 2.5 MHz

sector cardiac ultrasound transducerk by the same experienced clinician (EM).

Echocardiographic images obtained included: (1) 2-D long-axis 4-chamber view

(modified slightly to include the apex of the left ventricle); (2) 2-D short-axis view of the

left ventricle at the level of the papillary muscles just below the mitral valve; and (3) M-

mode of the left ventricle at the level of the papillary muscles just below the mitral valve.

Three video loops for each view were stored for subsequent tracings and measurements.

Tracings and measurements were performed retrospectively by the same investigator

(EM). Stroke volume (mL) and CO (L/min) were derived from the 4-chamber area-length

method as previously described.29 Fractional shortening (FS%) was measured in triplicate

based on published methods.30,31 The relative wall thickness (RWT) was determined from

M-mode measurements of the left ventricular free wall in diastole (LVFWd),

interventricular septum in diastole (IVSd) and the left ventricular internal diameter in

diastole (LVIDd). RWT = [(LVFWd + IVSd) / LVIDd]. The LVIDd/body weight (kg)

was also determined as a method to estimate relative left ventricular chamber size. Stroke

volume and CO were manually converted to stroke volume index (SVI) (mL/kg) and CI

(mL/kg/min), respectively, by dividing the volume or rate by body weight (kg).

Statistical analysis

Descriptive statistics for patient factors (age, breed, sex) were performed and are

reported as proportions or mean ± SD. Associations between SIRS and group (surgical

lesion category) or survival status were evaluated with a Chi-squared test or Fishers exact

test, with Bonferroni adjustment for multiple comparisons between groups. In addition,

119

associations between CSA and hypotension under GA with SIRS, group or survival status

were evaluated with a Chi-squared test or Fishers exact test, with Bonferroni adjustment

for multiple comparisons between groups. Normality of the data and equality of variances

were assessed using the Shapiro-Wilk and Levene’s tests, respectively. A two-way

ANOVA for repeated measures was used to assess the effect of disease category

(ischemic, non-ischemic, control; or survivors and non-survivors), time, and the

interactions between disease category and time for individual variables. Data that were

not normally distributed were rank-transformed prior to analysis. When warranted,

multiple pairwise comparisons were performed using the method of Holm-Sidak. Results

for two-way ANOVA analyses are reported as the least squares mean and standard error

of the mean (SEM) and median and interquartile range (IQR).

Potential associations between individual variables and survival, SIRS, or CSA,

were first screened by use of univariable logistic regression. For logistic regression,

hypotension under GA (Y or N) or duration (min), CSA (Y or N) and the trough [iCa2+,

iMg2+, SDNN, RMSSD, CVP, SVI, CI, LVIDd/kg] or the peak [cTnI, lactate, HR, LF,

PAR] value from the corresponding observation periods were used for analysis. In

addition, potential associations between the above individual variables measured only at

24 hours (excluding hypotension or CSA), rather than peak or trough values, and non-

survival were screened similarly. The potential association between electrolyte

concentrations measured closest to the time of detected arrhythmia, peak cTnI, or peak

lactate and CSA were screened with univariable regression analysis. Continuous

variables that did not meet the assumption of log linearity for regression analysis were

dichotomized based on the best cut-off as assessed by receiver operating characteristic

120

(ROC) curve analysis. Variables for which the screening P value was < 0.20 were

considered for inclusion in the multivariable logistic regression (MLR) model. Variables

with a variance inflation factor > 5.0 were deleted to avoid multicollinearity. The MLR

was a backward stepwise model, whereby variables were removed sequentially starting

with that having the largest P value and until only those variables with P < 0.05

remained. Goodness of fit of the final model was evaluated using the Hosmer and

Lemeshow test. Odds ratios (OR) and 95% confidence intervals (CI) were calculated. The

overall performance of the MLR models in predicting non-survival, SIRS or CSA were

assessed by use of ROC curve analyses. Finally, associations between RWT on Day 1

and lactate, cTnI, HR and CVP all measured on Day 1(at approximately 24 hours post-

operatively) were assessed with least squares multiple regression analysis. For all

analyses, P < 0.05 was considered statistically significant. Statistical analyses were

performed with commercially available statistical software.m,n

RESULTS

Animals

Seventy-four horses were enrolled between November 2011 and August 2014 and

included 50 geldings, 21 mares and 3 stallions with a weight of 515.3 ± 84.9 kg. Horses

ranged in age from 2 to 28 years [12.8 ± 6.6 years]. Breeds represented included 23

Quarter Horse-type, 18 Warmbloods, 14 Thoroughbreds, 7 Arabians, 3 Saddlebreds, 3

pony breeds, 2 Morgans, and 1 each Irish Sport horse, Connemara, Belgian draft-cross

and Lusitano. Associations between breed or sex and GI lesion group or short-term

survival were not statistically significant. There were 12 horses in the control group, 36

121

horses in the non-ischemic group and 26 horses in the ischemic group. Control horses

underwent GA for the following procedures: arthroscopy (n=3), arthrodesis (n=1),

laryngotomy (n=2), laryngoplasty (n=2), corneal laceration repair (n=1), neurectomy

(n=1) and exploratory laparotomy for an unrelated research project (n=2). Lesions for

horses in the non-ischemic group were as follows: right dorsal displacement of the large

colon (n=8), ileal impaction (n=6), left dorsal displacement of the large colon (n=6), non-

strangulating lipoma (n=3), small intestinal mesenteric volvulus (n=3), cecal impaction

(n=3), enterolithiasis (n=2), generalized small intestinal distension (n=1), gastrosplenic

entrapment (n=1), small colon impaction (n=1), focal infarction of the left dorsal colon

(n=1) and large colon impaction (n=1). Lesions for horses in the ischemic groups were as

follows: strangulating lipoma (n=14), large colon volvulus without resection (n=5), large

colon volvulus with resection (n=2), epiploic foramen entrapment (n=2), gastrosplenic

entrapment (n=2) and inguinal hernia (n=1). Sixty four horses survived to discharge and

10 were euthanized prior to discharge. All horses in the control group survived whereas

52 of 62 (84%) of horses with colic, 34 of 36 (94%) of horses with non-ischemic lesions

and 18 of 26 (69%) of horses with ischemic lesions, survived to discharge. All horses that

did not survive to discharge were euthanized and had a diagnosis of colonic volvulus

(n=4), strangulating lipoma (n=3), epiploic foramen entrapment (n=1) or non-

strangulating small intestinal lesion (n=2). The proportion of horses in the ischemic group

that did not survive (8/26; 31%) was significantly higher than that of horses in the non-

ischemic group (2/36; 5.6%) or in the control group (0/0) (P = 0.005).

122

Systemic Inflammatory Response Syndrome

Twenty-nine horses had clinical and/or clinicopathologic findings that fulfilled

the criteria for SIRS during the study period at a minimum of one time from admission to

through Day 2 post-operatively. Of these 29 horses with SIRS, 18 (62%) horses were in

the ischemic group and 11 (38%) horses were in the non-ischemic group; no horse in the

control group met the criteria for SIRS during the study period. A significantly greater

proportion of horses with ischemic gastrointestinal disease had SIRS compared to the

other groups (P < 0.001). Systemic inflammatory response syndrome was observed more

frequently in horses that did not survive to discharge (8/10; 80% horses) compared to

those that did survive to discharge (21/64; 32.8% horses) (P = 0.011).

Clinically Significant Arrhythmias

Clinically significant arrhythmias detected included: monomorphic ventricular

tachycardia (n=4), polymorphic ventricular tachycardia (n=1) accelerated idioventricular

rhythm (n= 1), > 1 single ventricular premature complex/hr (n=7), second degree AV

block at a HR of 70-80 bpm with ventricular premature complexes (n=1), and > 1

supraventricular premature complex/hr (n=1). Horses with ischemic GI lesions had a

significantly higher (P = 0.002) frequency of CSA (9/26; 34.6%) compared to horses with

non-ischemic GI lesions (2/36; 5.6%) or control (0/12; 0%) horses. Horses that fulfilled

the criteria for SIRS had a significantly higher (P = 0.003) frequency of CSA (9/29; 31%)

compared to horses without SIRS (2/43; 4.4%). In addition, there was a significantly

higher frequency of horses with a CSA that did not survive (4/10; 40%) compared to

those that survived (7/64; 11%); P = 0.036. Four horses had more than one type of CSA

during the observation period.

123

Cardiac troponin (cTnI)

Seventy-one horses had cTnI measured at admission and of these, 16 (22.5%)

were abnormal (> 0.03 ng/ml). Thirty-seven horses had an abnormal cTnI measured at

least once during the study period. Only 6/37 (16%) horses had peak cTnI at admission.

One non-survivor had a normal cTnI concentration (< 0.03 ng/mL) at all time-points.

When the horse that was euthanized prior to the 24-hours post-operatively was excluded,

only 2 non-survivors had an abnormal admission cTnI. Abnormal cTnI concentrations

were measured in the remaining eight horses that were non-survivors in which the peak

serum cTnI concentration occurred on either Day 1 (n= 4) or Day 2 (n= 4).

Hypotension under General Anesthesia

Fifteen horses experienced hypotension under GA. The median duration of

hypotension was 10 min [5-20] with a maximum total duration of 80 non-consecutive

minutes. Hypotension under GA occurred more frequently in horses with ischemic GI

lesions (12/26; 46%), horses with SIRS (12/29; 41.4%), and those that did not survive

(6/10: 60%) compared to their counterparts [horses with non-ischemic lesions (3/36;

8.3%), control (0%) or those without SIRS (3/29; 6.7%) or survivors (9/64; 14.1%)] P <

0.001, P < 0.001, and P = 0.003, respectively.

Comparisons of cardiovascular variables by gastrointestinal lesion group and time

Variables with significant effects of group or time measured at admission, Day 1

and Day 2 are summarized in Table 5.1, while variables with significant effects of group

or time measured only post-operatively are summarized in Table 5.2. Significant effects

of group and time were found for cTnI, lactate, potassium, and iCa2+ concentrations, post-

operative HR, and RMSSD. Variables that had an effect of group only included sodium

124

and iMg2+ concentrations, ectopic beats, RWT, SDNN and MAP. An effect of time only

was significant for CI. The effect of group, time, and interactions between group and

time were not statistically significant for SVI, FS, LVIDd/kg, LF, CVP or PAR.

Comparison of cardiovascular variables by survival status and time

Significant effects of survival status and or time on variables measured at

admission and on Day 1 and Day 2 are summarized in Table 5.3, while those measured

only post-operatively are summarized in Table 5.4. A significant effect of group and time

were found for cTnI, lactate, and iCa2+ concentrations. An effect of group only was

detected for SDNN, RMSSD, SVI, LVIDd/kg, and RWT. An effect of time only was

detected for potassium concentration. There was no significant effect of group or time for

sodium or iMg2+ concentrations, CI, FS, ectopic beats, LF, CVP, MAP or PAR.

Outcome analysis by MLR analysis

Survival to discharge

Complete data sets for 67 horses were available and consisted of 8 positive cases

(non-survivors) and 59 negative cases (survivors). Variables retained in the multivariable

model for non-survival were trough SVI and cTnI (> 0.15 ng/mL) (overall significance P

< 0.0001) (Table 5.5). The final model was highly significant (overall significance P <

0.0001) and correctly classified 91% of the cases. The area under the ROC curve for the

ability of the variables in the final model to predict non-survival was 0.95 ± 0.027; 95%

CI: 0.86 to 0.99.

Survival to discharge (Day 1 data)

Complete data sets for variables obtained at 24 hours were available for 63 horses

including 8 non-survivors and 55 survivors. Variables obtained at 24 hours that were

125

retained in the MLR model for non-survival included SVI, LF, and RWT > 0.63 (P <

0.0001; Table 5.5). This model correctly classified 96.3% of cases. The area under the

ROC curve for the ability of the variables in the final model to predict non-survival was

0.97 ± 0.019; 95% CI: 0.90-1.00.

Systemic inflammatory response syndrome (SIRS)

Complete data sets for 67 horses were available and included 25 with SIRS and

42 horses without SIRS. The MLR model that best explained SIRS included peak cTnI

concentration, peak PAR and trough SVI (P < 0.001; Table 5.5). This model correctly

classified 79% of the cases with an AUC from the ROC curve of 0.86 ± 0.046; 95% CI:

0.75-0.93.

Clinically Significant Arrhythmias

Data were available for all horses for detection of a CSA and included 11 horses

with CSA and 63 horses without CSA. The MLR model that best explained CSA

included peak cTnI concentration and an incremental decrease in iMg2+ (0.1 mg/dL=

0.041 mmol/L decrease in iMg2+ increases odds of CSA 2.5 times) concentration at the

time of the CSA (P < 0.0001; Table 5). This model correctly classified 91% of cases with

an AUC from the ROC curve of 0.93 ± 0.033; 95% CI: 0.85-0.98.

Relative wall thickness on Day 1

In an attempt to determine if hypovolemia was related to RWT a linear regression

model was generated. Complete data sets for 66 horses were available. Variables

measured at approximately 24 hours were entered into the least squares regression model

and included cTnI and lactate concentrations, CVP and HR. The overall significance

level of this model was P = 0.020. The adjusted R2 was 0.118. All variables had an

126

inflation factor of less than 1.5. The rpartial for individual variables were as follows: CVP

(0.012), cTnI (0.068), lactate (0.221) and HR (0.196).

DISCUSSION

In this study, parameters to assess the cardiovascular system of horses undergoing

surgical correction of intestinal disease and horses undergoing elective surgical

procedures were obtained and compared across groups, over time, and by short-term

survival outcome. Horses with ischemic gastrointestinal disease had increased admission

blood lactate concentrations, increased serum cTnI concentrations, higher HR, increased

MAP, lower HRV, greater RWT, and more frequent hypotension under GA and CSA

compared to horses with non-ischemic causes of colic and control horses. The null

hypothesis that horses undergoing GA for surgical treatment of ischemic gastrointestinal

disease are not more likely to develop myocardial dysfunction compared to horses

undergoing surgical treatment of non-ischemic gastrointestinal disease or elective

surgical procedures, was rejected on the basis of evidence of cardiovascular system

dysfunction characterized by alterations in HRV, cTnI concentration, RWT, MAP,

increased incidence of CSA and hypotension.

Furthermore, the null hypothesis that myocardial dysfunction would not be correlated

with measures of SIRS and survival in horses with acute colic was rejected on the basis

that peak cTnI concentration, lowest SVI, and peak PAR were retained in the model that

best classified horses with SIRS and that cTnI concentration >0.15 ng/mL and lowest

SVI were retained in the MLR model that best explained non-survival to hospital

discharge.

127

While previous studies6,32 have demonstrated an association between an increased

cTnI concentration and disease severity among horses with acute gastrointestinal disease,

to the authors’ knowledge only two studies7,8 have attempted to bridge the gap by

evaluating the relationship between cTnI concentration, the incidence of CSA or the

occurrence of myocardial dysfunction. In the study by Nath et al,7 a decreased left

ventricular ejection time, an index of systolic function, and increased HR were associated

with increased cTnI concentration. However, left ventricular ejection time was no longer

significant when the effect of HR was taken into consideration. Other parameters

evaluated in that study, but not found to be significantly associated with cTnI

concentration, were FS and cardiac rhythm. Horses were evaluated at only one clinically

relevant time point and telemetric ECG recordings were performed for one hour which

may have precluded the detection of intermittent arrhythmias. In the second study, Diaz

et al8 recorded continuous telemetric ECG for 24 hours and detected a significant

association between an increased admission cTnI concentration, ventricular arrhythmia,

surgical treatment, and outcome.8 Similarly, in the study herein, CSA occurred more

commonly in horses with ischemic GI disease and was explained by a MLR model that

included peak cTnI concentration. Cardiac troponin I concentration also contributed to

the MLR model that best explained survival. However, an important difference between

these studies is that admission serum cTnI concentration was not significantly different

between horses that survived and those that did not survive to hospital discharge in the

present study. In fact, the majority of non-survivors (7/10; 70%) had a normal admission

cTnI concentration. However, with the exception of one horse, all horses that were

euthanized had a serum cTnI of > 0.15 ng/mL for at least one of the two post-operative

128

measurements. This suggests that admission cTnI concentration is not an appropriate

biomarker for prognostication of short-term survival; alternatively a post-operative rise in

serum cTnI may be more clinically relevant. Importantly, both cTnI concentration and

SVI were retained in the MLR models that explained both survival to discharge and the

presence of SIRS, suggesting that there is a relationship between a marker of myocardial

injury, systemic inflammation and cardiac dysfunction.

A number of horses in this study, including some with and without an ischemic

gastrointestinal lesion diagnosed at surgery, demonstrated evidence of global perfusion

deficits at admission based on tachycardia and increased blood lactate concentrations;

unfortunately echocardiography was not performed at admission owing to concerns that it

would delay surgery and, as such is a limitation of the study. Admission

echocardiography would have been useful as an additional time point of comparison and

may have offered a more clinically relevant time to obtain information that could enhance

current prognostic capabilities. However, the intentions of the study were not to build

upon current prognostic determinants, but rather to more completely understand the basis

of cardiovascular abnormalities in horses with acute gastrointestinal disease. In a recent

study, Borde et al33 performed echocardiography in horses with evidence of SIRS that

presented for acute GI disease prior to undergoing exploratory laparotomy or euthanasia,

and documented both systolic and diastolic dysfunction. However, of the 41 horses

enrolled at admission, only 12 survived to discharge which suggests that the population

of horses studied may have been in an advanced stage of disease at the time of admission,

and likely do not represent the majority of horses that present for acute gastrointestinal

disease. Additionally, cTnI concentrations, or other suitable cardiac biomarkers were not

129

measured in these horses, perhaps missing an opportunity to evaluate a relationship

between cTnI concentration and cardiac function.

In the present study, 62% of horses with evidence of SIRS had ischemic GI

disease, compared to 38% of horses with non-ischemic disease and 0% of control horses.

The pathophysiology of systemic inflammation in horses with acute GI disease is in large

part attributed to GI mucosal barrier dysfunction which permits endotoxin and other

pathogen-associated molecular patterns access to the lymphatic and portal systems.34

When these systems are overwhelmed or are themselves dysfunctional, bacterial derived

pathogen associated molecular patterns may reach the systemic circulation and incite a

cascade of inflammatory cytokines (e.g., TNF-alpha, IL-6, IL-1B, and IL-10) and other

mediators (e.g., tissue factor, complement, prostaglandins) that perpetuate the

inflammatory response on a systemic level. In septic cardiomyopathy it is hypothesized

that inflammatory mediators, such as TNF-alpha and IL-6 may cause myocardial cells to

increase membrane permeability and leak proteins such as cTnI. 35 An increased serum

cTnI, therefore, can likely arise from both reversible and irreversible myocardial cellular

damage as evidenced by stress-induced cardiomyopathy in people.36 Although few

patients with sepsis have severe histopathologic myocardial lesions at necropsy,

contraction band necrosis, attributable to dysregulation and influx of calcium and

catecholamine toxicity,35 has been documented in septic human patients with left

ventricular dysfunction.37 Cardiac troponin I, a sensitive and specific biomarker of

myocardial injury, proved to be an important explanatory variable in MLR models for

short-term survival to discharge, the presence of a CSA and the presence of SIRS. This

suggests that myocardial injury occurs and is associated with adverse outcomes and SIRS

130

in horses with acute gastrointestinal disease. While the underlying etiology for an

increase in cTnI remains incompletely understood, the results of the study herein suggests

systemic inflammation plays a role.

The time domain HRV parameters, SDNN and RMSSD, were significantly lower

in horses with ischemic lesions and those that did not survive to discharge, however, only

the frequency domain parameter, LF power, measured on Day 1 was an important

explanatory variable in the survival to discharge model from cardiovascular assessment

on Day 1 post-operatively. In septic human patients, a lower LF power and reduced

RMSSD measured at admission were strongly associated with the development of MODS

in the hospital.19 In the study herein, the peak LF power was chosen for univariable

analysis on the basis that this would reflect increased sympathetic tone and based on its

performance in a previous study.24 The peak LF power measured on Day 1 was positively

associated with non-survival, which would suggest that having a higher LF power is a

negative finding in horses with colic. This is in contrast to what is recognized in septic

humans and rabbits with experimentally induced septic shock.41 Potential reasons for this

disparity might be a function of different times of analysis (human patients were studied

at admission versus the horses herein were evaluated post-operatively), distinct etiologies

of the underlying disease (sepsis in humans versus acute gastrointestinal disease in the

horse) and other species differences, especially in regard to endotoxin sensitivity.

Specifically, endotoxin sensitizes cardiac pacemaker cells to sympathetic stimuli42 and

horses are uniquely sensitive to endotoxin when compared to humans.37 Despite these

differences, reduced modulation of total HRV and the parasympathetic branch of the

ANS, corresponding to the decreased SDNN and RMSSD, respectively, are consistent

131

findings in critically-ill people and correspond to an increased risk for developing MODS

and mortality.43,44 Studying HRV at admission may be more appropriate for comparing to

the results of human studies and predicting which horses may be more likely to develop

complications in the post-operative period.

Increased post-operative HR, reduced SVI, low LVIDd/kg and greater RWT were

found in horses that did not survive to discharge which might suggest that preload was

the more pressing issue rather than left ventricular systolic dysfunction. Relative wall

thickness was the only echocardiographic derived parameter that was increased in both

horses with ischemic GI disease and non-survivors. An increase in RWT of greater than

0.55 was found to be reasonably accurate for detecting pseudohypertrophy in horses with

experimentally induced hypohydration.40 Pseudohypertrophy, from hypovolemia, could

potentially explain the increase in RWT in horses in this study; however, the multiple

regression model for RWT herein revealed that collectively HR, CVP, lactate and cTnI

concentration measured at the same time as RWT only explained 11.8% (adjusted R2

0.118) of the variability in RWT. This would suggest that hypovolemia was not the sole

reason for the increased RWT. In addition, the horses in the experimental model37 were

estimated to be 8-10% clinically dehydrated, which was not the case with the horses in

this study in the post-operative period, the majority of whom were receiving at least

maintenance rate of IV fluid therapy at the time of measurement. The effect of HR on

RWT in horses is not well documented, however other M-mode derived measurements

are altered by HR.41,42 The post-operative HR was significantly greater in the horses in

the ischemic group and non-survivors at all time-points. Tachycardia, which results in a

shorter diastolic period and limits preload might explain this finding along with the lower

132

SVI and smaller LVIDd/kg measurements in non-survivors compared to survivors. These

echocardiographic measurements were not statistically different between gastrointestinal

lesion groups but were different between survivors and non-survivors and are preload

dependent.

Low SVI was an important variable retained in the MLRs to explain non-survival

and presence of SIRS. In humans with sepsis, diastolic dysfunction is detected in the

absence of a reduced ejection fraction or other markers of systolic dysfunction, and is

more strongly correlated to cTnI.43 In fact, left ventricular end-diastolic volume index,

SVI and tissue Doppler imaging of the ratio of early mitral inflow velocity to early mitral

annular motion (E/eˊ ratio) all were significant decreased in people that died versus those

that survived. 44 Cardiac index, the product of SV and HR divided by body weight, was

significantly increased on Day 1 in horses when grouped by lesion category but was not

significantly associated with SIRS or outcome in the MLR models. This finding likely

can be explained by the quantitatively more efficient effect HR has on determining CO in

comparison to stroke volume.45 Echocardiographic measures of systolic function in this

study included FS, CI and SVI. With severe acute systolic dysfunction one would expect

a reduced FS while more chronic systolic dysfunction might be accompanied by an

increase in the LVIDd/kg that would indicate poor contractility and a dilated LV

chamber, respectively. However, with the exception of SVI, these parameters were not

retained in the MLR models and only LVIDd/kg was statistically different between

survivors and non-survivors, where survivors had larger left ventricular dimensions at

end diastole. When considering these findings together, it is not possible to draw a single

133

logical conclusion regarding left ventricular systolic function in horses with acute

surgical GI disease.

As previously reported, hypokalemia, ionized hypomagnesemia and

hypocalcemia were common findings in horses with colic.46-49 Interestingly, none of

these electrolyte derangements were associated with outcome or the presence of SIRS.

However, for every 0.1 mg/dL (0.04 mmol/L)50 unit decrease in ionized magnesium

concentration there was a 2.5 times increased risk of having a CSA. Further work is

needed to determine whether magnesium supplementation in the peri-operative period

would reduce the occurrence of CSAs in horses with colic.

A limitation in this study was that only echocardiographic methods were used to

assess systolic function and more accurate, gold-standard invasive measures were not

used. In addition, left ventricular diastolic function was not assessed with

echocardiography which may have precluded detecting diastolic dysfunction.

Furthermore, investigators were not blinded in this study, however, all measurements

were performed retrospectively and horse information entered for echocardiographic and

telemetric ECG was minimal (including only the medical record number and research

number in the study) preventing the investigators from determining which GI lesion

category the horse was in, or its SIRS or survival status. Missing data for individual

horses limited the power of the regression analysis. Despite this limitation, the numbers

in each group are comparable or greater than horses used in similar studies6-8 and the

addition of a control group that also underwent GA for a surgical procedure provided

novel information. Unfortunately, a complete necropsy was unable to be performed on

134

the majority of the horses that were euthanized; therefore, gross and histopathological

characterization of myocardial lesions was not possible.

In conclusion, on the basis of increased serum cTnI concentration, reduced HRV,

hypotension under GA, increased incidence of CSA, greater HR, RWT, or reduced SVI,

or LVIDd/kg, horses with ischemic GI lesions or horses with GI lesions requiring surgery

that do not survive have evidence of myocardial dysfunction. Although a causal

explanation for these relationships could not be definitively determined, the parameters

described herein for assessing myocardial health should be considered in the assessment

of critically ill horses. Arrhythmias in horses in the post-operative period were associated

with cTnI and iMg2+ concentration. Clinically significant arrhythmias were also more

frequent in non-survivors and horses with SIRS, further research is warranted to

determine if therapeutic intervention with magnesium supplementation, anti-

inflammatories or anti-arrhythmic drugs would improve outcome. Peak cTnI

concentration was detected after admission in non-survivors; therefore measuring serum

cTnI concentration at the single time point of admission is unlikely to be a useful

prognostic tool for short-term survival.

135

FOOTNOTES

a. Heska CBC-Diff, Heska Corp, Loveland, CO

b. Sorvall Legend X1, Thermo Fischer Scientific Inc, Suwanee, GA

c. ADVIA Centaur cTnI Ultra Assay, Immulite 1000 Siemens, Deerfield, IL

d. Nova Biomedical, Critical Care Xpress, Waltham, MA.

e. Televet Version 3, Kruuse, Denmark

f. Kruuse ECG Electrodes, Jorgenson Medical

g. Mila International, Inc., Erlanger, KY.

h. Edwards Lifesciences LLC, Irvine, CA

i. SurgiVet Vital Signs Monitor, V9203; Smiths Medical, St. Paul, MN

j. Vivid 7, GE Medical Systems, Milwaukee, WI

k. M4S transducer, GE Medical Systems, Milwaukee, WI

l. Kubios HRV Version 2.1 Kuopio, Finland kubios.uef.fi/

m. Sigmaplot 12.5 Systat Software, Inc. San Jose, CA

n. MedCalc 14.8.1,Ostend, Belgium

136

REFERENCES

1. Parrillo JE, Parker MM, Natanson C, et al. Septic shock in humans. Advances in the

understanding of pathogenesis, cardiovascular dysfunction, and therapy. Ann Intern Med

1990;113:227-242.

2. Romero-Bermejo FJ, Ruiz-Bailen M, Gil-Cebrian J, et al. Sepsis-induced

cardiomyopathy. Curr Cardiol Rev 2011;7:163-183.

3. Vieillard-Baron A, Caille V, Charron C, et al. Actual incidence of global left

ventricular hypokinesia in adult septic shock. Crit Care Med 2008;36:1701-1706.

4. Balk RA. Pathogenesis and management of multiple organ dysfunction or failure in

severe sepsis and septic shock. Crit Care Clin 2000;16:337-352, vii.

5. Mayr VD, Dunser MW, Greil V, et al. Causes of death and determinants of outcome

in critically ill patients. Crit Care 2006;10:R154.

6. Radcliffe RM, Divers TJ, Fletcher DJ, et al. Evaluation of L-lactate and cardiac

troponin I in horses undergoing emergency abdominal surgery. J Vet Emerg Crit Care

(San Antonio) 2012;22:313-319.

7. Nath LC, Anderson GA, Hinchcliff KW, et al. Clinicopathologic evidence of

myocardial injury in horses with acute abdominal disease. J Am Vet Med Assoc

2012;241:1202-1208.

8. Diaz OM, Durando MM, Birks EK, et al. Cardiac troponin I concentrations in horses

with colic. J Am Vet Med Assoc 2014;245:118-125.

9. Hunter JD, Doddi M. Sepsis and the heart. Br J Anaesth 2010;104:3-11.

137

10. Chagnon F, Bentourkia M, Lecomte R, et al. Endotoxin-induced heart dysfunction in

rats: assessment of myocardial perfusion and permeability and the role of fluid

resuscitation. Crit Care Med 2006;34:127-133.

11. Rudiger A, Singer M. Mechanisms of sepsis-induced cardiac dysfunction. Crit Care

Med 2007;35:1599-1608.

12. Steverink PSA, Rutten V, et al. Endotoxin, interleukin-6 and tumor necrosis factor

concentrations in equine acute abdominal disease: relation to clinical outcome. Journal of

Endotoxin Research 1995;2:289-299.

13. Barton MH, Collatos C. Tumor necrosis factor and interleukin-6 activity and

endotoxin concentration in peritoneal fluid and blood of horses with acute abdominal

disease. J Vet Intern Med 1999;13:457-464.

14. Senior JM, Proudman CJ, Leuwer M, et al. Plasma endotoxin in horses presented to

an equine referral hospital: correlation to selected clinical parameters and outcomes.

Equine Vet J 2011;43:585-591.

15. Nostell K, Brojer J, Hoglund K, et al. Cardiac troponin I and the occurrence of

cardiac arrhythmias in horses with experimentally induced endotoxaemia. Vet J

2012;192:171-175.

16. Marshall JC, Cook DJ, Christou NV, et al. Multiple organ dysfunction score: a

reliable descriptor of a complex clinical outcome. Crit Care Med 1995;23:1638-1652.

17. Werdan K, Oelke A, Hettwer S, et al. Septic cardiomyopathy: hemodynamic

quantification, occurrence, and prognostic implications. Clin Res Cardiol 2011;100:661-

668.

138

18. Vincent JL, de Mendonca A, Cantraine F, et al. Use of the SOFA score to assess the

incidence of organ dysfunction/failure in intensive care units: results of a multicenter,

prospective study. Working group on "sepsis-related problems" of the European Society

of Intensive Care Medicine. Crit Care Med 1998;26:1793-1800.

19. Muller-Werdan U, Buerke M, Ebelt H, et al. Septic cardiomyopathy - A not yet

discovered cardiomyopathy? Exp Clin Cardiol 2006;11:226-236.

20. Pontet J, Contreras P, Curbelo A, et al. Heart rate variability as early marker of

multiple organ dysfunction syndrome in septic patients. J Crit Care 2003;18:156-163.

21. Maeder M, Fehr T, Rickli H, et al. Sepsis-associated myocardial dysfunction:

diagnostic and prognostic impact of cardiac troponins and natriuretic peptides. Chest

2006;129:1349-1366.

22. Marr CM, Bowen, I.M. Cardiology of the Horse In: Marr CM, Patteson, M. , ed.

Echocardiography. Philadelphia: Saunders 2010.

23. Epstein KL, Brainard BM, Gomez-Ibanez SE, et al. Thrombelastography in horses

with acute gastrointestinal disease. J Vet Intern Med 2011;25:307-314.

24. Hart KA, MacKay, R. J. Endotoxemia and Sepsis In: Smith BP, ed. Large Animal

Internal Medicine, 5th Edition. St. Louis, MO: Mosby, 2013;684.

25. McConachie EG, S. Rapoport, G. Barton, M. . Heart rate variability in horses with

acute gastrointestinal disease requiring exploratory laparotomy. Journal of Veterinary

Emergency and Critical Care 2015; accepted.

26. Reef VM, C. Cardiology of the Horse Second Edition Toronto: Saunders, 2010.

27. Reimer JM, Reef VB, Sweeney RW. Ventricular arrhythmias in horses: 21 cases

(1984-1989). J Am Vet Med Assoc 1992;201:1237-1243.

139

28. Wilsterman S, Hackett ES, Rao S, et al. A technique for central venous pressure

measurement in normal horses. J Vet Emerg Crit Care (San Antonio) 2009;19:241-246.

29. Parry BW, McCarthy MA, Anderson GA, et al. Correct occlusive bladder width for

indirect blood pressure measurement in horses. Am J Vet Res 1982;43:50-54.

30. Tearney CC, Guedes AG, Brosnan RJ. Equivalence between invasive and

oscillometric blood pressures at different anatomic locations in healthy normotensive

anaesthetised horses. Equine Vet J 2015.

31. Branson KR. A clinical evaluation of an oscillometric blood pressure monitor on

anesthetized horses. Journal of Equine Veterinary Science 1997;17:537-540.

32. McConachie E, Barton MH, Rapoport G, et al. Doppler and volumetric

echocardiographic methods for cardiac output measurement in standing adult horses. J

Vet Intern Med 2013;27:324-330.

33. Patteson MW, Gibbs C, Wotton PR, et al. Echocardiographic measurements of

cardiac dimensions and indices of cardiac function in normal adult thoroughbred horses.

Equine Vet J Suppl 1995:18-27.

34. Bonagura JD, Blissitt KJ. Echocardiography. Equine Vet J Suppl 1995:5-17.

35. Hallowell G BI. Cardiac troponin I in equine surgical colic patients: myocaridal

damage due to endotoxemia or hypoperfusion. J Vet Intern Med 2007; 21.

36. Borde L, Amory H, Grulke S, et al. Prognostic value of echocardiographic and

Doppler parameters in horses admitted for colic complicated by systemic inflammatory

response syndrome. J Vet Emerg Crit Care (San Antonio) 2014;24:302-310.

140

37. Moore JN, Vandenplas ML. Is it the systemic inflammatory response syndrome or

endotoxemia in horses with colic? Vet Clin North Am Equine Pract 2014;30:337-351, vii-

viii.

38. Favory R, Neviere R. Significance and interpretation of elevated troponin in septic

patients. Crit Care 2006;10:224-230.

39. Boland TA, Lee VH, Bleck TP. Stress-Induced Cardiomyopathy. Crit Care Med

2015. 2015/01/08 Ed, 2015;1-8.

40. ver Elst KM, Spapen HD, Nguyen DN, et al. Cardiac troponins I and T are biological

markers of left ventricular dysfunction in septic shock. Clin Chem 2000;46:650-657.

41. Garrard CS, Kontoyannis DA, Piepoli M. Spectral analysis of heart rate variability in

the sepsis syndrome. Clin Auton Res 1993;3:5-13.

42. Werdan K, Schmidt H, Ebelt H, et al. Impaired regulation of cardiac function in

sepsis, SIRS, and MODS. Can J Physiol Pharmacol 2009;87:266-274.

43. Schmidt H, Muller-Werdan U, Hoffmann T, et al. Autonomic dysfunction predicts

mortality in patients with multiple organ dysfunction syndrome of different age groups.

Crit Care Med 2005;33:1994-2002.

44. Schmidt H, Hoyer D, Hennen R, et al. Autonomic dysfunction predicts both 1- and

2-month mortality in middle-aged patients with multiple organ dysfunction syndrome.

Crit Care Med 2008;36:967-970.

45. Houle MS, Billman GE. Low-frequency component of the heart rate variability

spectrum: a poor marker of sympathetic activity. Am J Physiol 1999;276:H215-223.

141

46. Rahman F, Pechnik S, Gross D, et al. Low frequency power of heart rate variability

reflects baroreflex function, not cardiac sympathetic innervation. Clin Auton Res

2011;21:133-141.

47. Nolen-Walston RD, Norton JL, Navas de Solis C, et al. The effects of hypohydration

on central venous pressure and splenic volume in adult horses. J Vet Intern Med

2011;25:570-574.

48. Sandersen C, Detilleux J, Art T, et al. Exercise and pharmacological stress

echocardiography in healthy horses. Equine Vet J Suppl 2006:159-162.

49. Sandersen CF, Detilleux J, de Moffarts B, et al. Effect of atropine-dobutamine stress

test on left ventricular echocardiographic parameters in untrained warmblood horses. J

Vet Intern Med 2006;20:575-580.

50. Bouhemad B, Nicolas-Robin A, Arbelot C, et al. Isolated and reversible impairment

of ventricular relaxation in patients with septic shock. Crit Care Med 2008;36:766-774.

51. Landesberg G, Gilon D, Meroz Y, et al. Diastolic dysfunction and mortality in severe

sepsis and septic shock. Eur Heart J 2012;33:895-903.

52. Klabunde RE. Cardiovascular physiology concepts. Philadelphia: Lippincott

Williams & Wilkins, 2005.

53. Johnson PJ. Electrolyte and acid-base disturbances in the horse. Vet Clin North Am

Equine Pract 1995;11:491-514.

54. Garcia-Lopez JM, Provost PJ, Rush JE, et al. Prevalence and prognostic importance

of hypomagnesemia and hypocalcemia in horses that have colic surgery. Am J Vet Res

2001;62:7-12.

142

55. Johansson AM, Gardner SY, Jones SL, et al. Hypomagnesemia in hospitalized

horses. J Vet Intern Med 2003;17:860-867.

56. Navarro M, Monreal L, Segura D, et al. A comparison of traditional and quantitative

analysis of acid-base and electrolyte imbalances in horses with gastrointestinal disorders.

J Vet Intern Med 2005;19:871-877.

57. Stewart AJ. Magnesium disorders in horses. Vet Clin North Am Equine Pract

2011;27:149-163.

143

Table 5.1 Least squares mean ± SEM; Median [IQR] for variables measured over time in horses with acute GI disease grouped by surgical lesion category and time.

Variable Group Time

P value

Group Time Group

× time

Admission Day 1 Day 2

cTnI

(ng/mL)

I 1.7 ± 0.17; 0.02 [0.01 - 0.39]aA 2.8 ± 0.18; 0.14 [0.03 - 2.78]aB 2.0 ± 0.19; 0.11 [0.04 - 1.13]aB <0.001 0.001

0.07

NI 0.03 ± 0.15; 0.01 [0.01 -0.01]bA 0.15 ± 0.14; 0.02 [0.01 - 0.78]bB 0.19 ± 0.15; 0.02 [0.01- 0.07]bB

C 0.02 ± 0.25; 0.01 [0.01 - 0.01]cA 0.03 ± 0.25; 0.01 [0.01-0.02]cB 0.02 ± 0.25; 0.01 [0.01- 0.01]cB

Lactate

(mmol/L)

I 5.5 ± 0.30; 4.2 [1.5 - 8.0]a* 2.1 ± 0.31; 0.8 [0.6 - 1.45]a† 1.6 ± 0.37; 0.7 [0.53 - 0.98]a† < 0.001

< 0.001

0.007

NI 1.84 ± 0.26; 1.4 [0.9 - 2.2]b* 0.73 ± 0.26; 0.6 [0.5 - 0.8]b† 0.73 ± 0.26; 0.7 [0.6 - 0.9]ab†

C 0.72 ± 0.55; 0.7 [0.4 - 1.15]c 0.61 ± 0.45; 0.6 [0.3 - 0.85]b 0.55 ± 0.45; 0.45 [0.4 - 0.7]b

Na+

(mmol/L)

I 135 ± 0.41; 134.3 [133.4-136.7] 136 ± 0.45; 135.8 [134.2- 137.9] 135 ± 0.48; 134.6 [132.2-136.3] 0.032# 0.411 0.082

NI 134 ± 0.35; 134.4 [133.1-136.3] 134 ± 0.35; 133.7 [132.5- 134.9] 134 ± 0.37; 133.9 [132.2-135.6]

C 136 ± 0.97; 135.7 [133.4-138.2] 135 ± 0.70; 135.9 [133.7-137.2] 135 ± 0.64; 134.1 [133.6-136.2]

K+

(mmol/L)

I 3.4 ± 0.06; 3.4 [3.2 - 3.7]a* 3.5 ± 0.07; 3.4 [3.2 - 3.8]a* 3.8 ± 0.07; 3.8 [3.5 - 3.9]a† 0.001 < 0.001

0.04

NI 3.7 ± 0.05; 3.6 [3.5 - 3.9]b* 3.7 ± 0.05; 3.7 [3.5 - 4.0]ab* 4.1 ± 0.06; 4.0 [3.8 - 4.3]b†

C 3.6 ± 0.15; 3.5 [3.4 - 3.6]ab* 4.0 ± 0.11; 3.9 [3.7 - 4.3]b† 3.9 ± 0.10; 3.9 [3.6 - 4.0]ab†

144

I: ischemic; NI: non-ischemic; C: control. Within a column a lower case letter denotes difference between groups within a row an upper case letters denote a difference between time. Within a row different symbols (*†‡) indicate a difference within a group over time. cTnI: cardiac troponin I; Na+ sodium; K+: potassium, iCa2+: ionized calcium; iMg2+: ionized magnesium. # no significant difference after multiple pairwise comparisons.

iCa2+

(mmol/L)

I 1.29 ± 0.02; 1.32 [1.18-1.36]a* 1.41 ± 0.02; 1.38 [1.29-1.53]a† 1.51 ± 0.02; 1.51 [1.43-1.58]‡ <0.001 <0.001 0.007

NI 1.38 ± 0.02; 1.4 [1.32-1.44]b* 1.44 ± 0.02; 1.46 [1.40-1.51]a† 1.53 ± 0.02; 1.54 [1.52-1.58]‡

C 1.47 ± 0.04; 1.45 [1.39-1.50]b* 1.57 ± 0.03; 1.55 [1.53-1.61]b† 1.53 ± 0.03; 1.53 [1.48-1.58]*†

iMg2+

(mmol/L)

I 0.42 ± 0.01; 0.41[0.37- 0.44]a 0.42 ± 0.01; 0.43 [0.39 - 0.46]a 0.45 ± 0.01; 0.44 [0.39 - 0.50]a < 0.001 0.124

0.988

NI 0.47 ± 0.01; 0.47 [0.43 - 0.53]b 0.45 ± 0.01; 0.46 [0.39 - 0.50] b 0.47 ± 0.01; 0.46 [0.42 - 0.53]b

C 0.50 ± 0.02; 0.51 [0.47 - 0.53]c 0.50 ± 0.02; 0.54 [0.45 - 0.56]c 0.50 ± 0.02; 0.51 [0.49 - 0.54]c

Table 5. 1 continued

145

Table 5.2. Least squares mean ± SEM; Median [IQR] for variables measured post-operatively and with significant differences in horses with acute GI disease grouped by surgical lesion category and time.

Variable Group Time* P value

Group Time Group × time

Time 1 Time 2 Time 3 Time 4

SDNN (ms)

I 29.1 ± 3.8; 19.9 [8.6 - 49.4]a

27.9 ± 3.8; 29.1 [12.7 - 44.1]a

36.3 ± 4.1; 32.7 [19.7 - 38.2]a

40.2 ± 4.4; 42.5 [26 - 59.2]a < 0.001 0.076 0.438

NI 46.7 ± 3.3; 41.2 [24.3 - 56.6]b

44.7 ± 3.4; 38.9 [27.7 - 52.4]b

54.6 ± 3.3; 46.9 [35.6 - 60.6]b

56.5 ± 3.3; 46.4 [39.1 - 65.0]b

C 66.7 ± 6.0; 63.0 [32.9 - 79.3]c

75.9 ± 5.7; 78.0 [48.5 - 96.2]c

69.6 ± 5.7; 64.2 [46.1 - 96.5]c

68.0 ± 7.3; 62.0 [28.7 - 97.0]c

RMSSD (ms)

I 30.3 ± 5.1; 14.4 [6.1 - 46.2]a

27.6 ± 5.1; 23.8 [11.1 - 41.9]a

40.8 ± 5.4; 30.7 [18.7 - 43.0]a

42.5 ± 5.9; 42.5 [27.1 - 71.3]a

< 0.001 0.037# 0.212

NI 51.6 ± 4.4; 37.9 [21.0 - 61.8] b

48.2 ± 4.5; 43.5 [25.1 - 56.9] b

59.7 ± 4.3; 54.8 [36.1 - 71.6] b

62.8 ± 4.3; 50.0 [42.6 - 70.0]b

HR (bpm)

I 56 ± 1.16; 57 [49 - 75]a* 59 ± 1.16; 51 [45 - 68]a* 55 ± 1.75; 50 [41 - 65]a*† 50 ± 1.75; 46 [39 - 58]a† <0.001 < 0.001 0.009

NI 46 ± 1.31; 50 [44 - 63]b* 43 ± 1.35; 43 [38 - 51]b† 39 ± 1.40; 41 [36 - 48]b‡ 39 ± 1.37; 38 [33 - 42]ab†‡

C 41 ± 2.47c; 34 [32 - 44]b* 34 ± 2.54; 38 [32 - 49]c† 36 ± 2.54; 38 [31 - 43]b† 34 ± 2.54; 32 [29 - 37]c†

146

I: ischemic; NI: non-ischemic; C: control. Time*: for SDNN, RMSSD Time 1: 2-10 hours; Time 2: 16- 24 hours; Time 3: 30-38 hours ;Time 4: 44-52 hours. For MAP, HR, CI, RWT, ectopic beats: Time 1: 12 hours; Time 2: 24 hours; Time 3: 36 hours; Time 4: 48 hours. Within a column a lower case letter denotes difference between groups within a row an upper case letters denote a difference over time. Within a row different symbols (*†‡) indicate a difference within a group over time. SDNN: standard deviation of normal to normal intervals; RMSSD: root mean square of successive differences; HR: Heart rate; MAP: mean arterial blood pressure; CI: cardiac index; RWT: relative wall thickness. # indicates no significant difference after multiple pairwise comparisons.

MAP (mmHg)

I 86 ± 2.6; 88 [73 - 99] 84 ± 2.7; 86 [76 - 92] 82 ± 3.0; 83 [73 - 90] 87 ± 2.9; 89 [79 - 96] 0.047# 0.986 0.546

NI 83 ± 2.2; 82 [75 - 89] 79 ± 2.3; 78 [74 - 85] 81 ± 2.5; 81 [72 - 88] 77 ± 2.6; 79 [68 - 87]

C 74 ± 4.5; 72 [63 - 83] 82 ± 4.3; 75 [69 - 98] 79 ± 4.6; 79 [71 - 85] 79 ± 4.3; 76 [69 - 85]

Ectopic beats (#/ 24hr)

I --- 23.3 ± 6.8; 2.0 [0.0 - 13.0]a --- 31.4 ± 6.8; 3.0 [0.0 - 9.0]a 0.006 0.157 0.297

NI --- 2.2 ± 5.8; 0.0 [0.0 - 1.8]b --- 2.3 ± 5.8; 0.0 [0.0 - 1.0]b

C --- 1.7 ± 10.1; 0.5 [0.0 - 2.0]b --- 0.6 ± 10.1; 0.0 [0.0 - 1.5]b

CI (mL/kg/min)

I --- 66.7 ± 3.1; 62.0 [45.5 - 76.8]A

--- 59.8 ± 3.1; 51 [42.5 - 70.5]B

0.089 0.031 0.933

NI --- 60.4 ± 2.5; 59.0 [49.0 - 65.0]A

--- 52.5 ± 2.5; 51 [44.0 - 61.5]B

C --- 50.7 ± 4.5; 51.0 [38.0 - 60.0]A

--- 46.8 ± 4.5; 44 [37.0 - 50.0]B

RWT I --- 0.63 ± 0.02; 0.63 [0.53 - 0.70]a

--- 0.62 ± 0.02; 0.59 [0.48 - 0.70]a

0.004 0.668 0.483

NI --- 0.54 ± 0.02; 0.52 [0.49 - 0.59]b

--- 0.54 ± 0.02; 0.53 [0.49 - 0.56]b

C --- 0.51 ± 0.03; 0.51 [0.48 - 0.55]b

--- 0.52 ± 0.03; 0.52 [0.48- 0.59]b

Table 5.2. continued

147

Table 5.3. Least squares mean ± SEM; Median [IQR] for significant variables measured in horses with acute GI disease grouped by survival outcome and time.

Variable Group Time

P value

Group Time Group

× time

Admission Day 1 Day 2

cTnI

(ng/mL)

S 0.39 ± 0.11; 0.01 [0.01 - 0.02]* 0.77 ± 0.11; 0.02 [0.01 - 0.1]a† 0.51 ± 0.11; 0.02 [0.01 - 0.08]a† 0.01 < 0.001

0.001

NS 2.1 ± 0.28; 0.01 [0.01 - 0.02]* 2.8 ± 0.31; 0.23 [0.04 - 1.6]b† 2.9 ± 0.36; 0.17 [0.02 - 1.3]b†

Lactate

(mmol/L)

S 2.6 ± 0.22; 1.5 [0.9 - 3.4]aA 0.73 ± 0.21; 0.6 [0.5 - 0.8]aB 0.71 ± 0.22; 0.7 [0.5 - 0.9]aB 0.001

< 0.001

0.311

NS 5.5 ± 0.54; 4.6 [1.3 -7.2]bA 4.3 ± 0.60; 1.6 [0.6 - 2.3]bB 3.3 ± 0.87; 0.9 [0.5 - 2.0]bB

K+

(mmol/L)

S 3.6 ± 0.04; 3.6 [3.3 - 3.8]A 3.7 ± 0.04; 3.7 [3.3 - 4.0]AB 3.9 ± 0.04; 3.9 [3.7 - 4.2]B 0.835 0.002

0.648

NS 3.5 ± 0.1; 3.6 [3.4 - 3.7]A 3.7 ± 0.1; 3.7 [3.4 - 3.9]AB 3.9 ± 0.2; 3.8 [3.7 - 4.1]B

iCa2+

(mmol/L)

S 1.36 ± 0.1; 1.39 [1.28 - 1.44]aA 1.45 ± 0.1; 1.47 [1.38 - 1.53]aB 1.53 ± 0.1; 1.54 [1.51 - 1.58]aB 0.002 <0.001

0.081

NS 1.33 ± 0.2; 1.36 [1.26 - 1.37]bA 1.40 ± 0.3; 1.43 [1.34 - 1.51]bB 1.41 ± 0.2; 1.43 [1.40 - 1.49]bB

S: survivor; NS: non-survivor. Within a column a lower case letter denotes difference between groups within a row an upper case letters denote a difference between time. Within a row different symbols (*†‡) indicate a difference within a group over time. cTnI: cardiac troponin I; Na+: sodium; K+: potassium, iCa2+: ionized calcium.

148

Table 5.4. Least squares mean ± SEM; Median [IQR] for significant variables measured post-operatively in horses with acute GI disease grouped by survival outcome and time.

Variable Group Time* P value

Group Time Group x

Time

Time 1 Time 2 Time 3 Time 4

SDNN

(ms)

S 47.4 ± 2.5; 43.8

[23.0 - 61.7]a

47.4 ± 2.5; 40.0

[27.5 - 58.0]a

55.5 ± 2.5; 47.5

[34.5 - 75.9]a

57.0 ± 2.6; 49.4

[36.2 - 68.4]a

< 0.001

0.151

0.354

NS 20.6 ± 6.2; 12.5

[5.4 - 30.7]b

21.6 ± 6.2; 22.5

[9.2 - 33.9]b

18.4 ± 6.7; 19.6

[9.7 - 30.3]b

27.5 ± 7.8; 32.4

[9.3 - 46.6]b

RMSSD

(ms)

S 54.0 ± 3.3; 40.4

[19.8 - 70.0]a

52.3 ± 3.3; 45.2

[27.2 - 69.1]a

62.6 ± 3.3; 52.6

[34.8 - 77.6]a

64.1 ± 3.5; 52.1

[41.5 - 81.8]a

< 0.001 0.053 0.924

NS 13.7 ± 8.2; 7.6

[5.7- 25.0]b

16.7 ± 8.2; 16.7

[10.4 - 24.6]b

17.7 ± 8.9; 17.2

[5.0 - 33.9]b

23.9 ± 10.4; 32.2

[12.3 - 44.0]b

HR

(bpm)

S 46 ± 1.0; 49 [43 - 61]aA 44 ± 1.0; 43 [38 - 51]aAB 41 ± 1.0; 42 [36 - 48]aB 39 ± 1.0; 38 [33 - 43]aB <0.001 0.002

0.371

NS 63 ± 2.6; 65 [52 - 77]bA 70 ± 2.9; 62 [50 - 72]bAB 64 ± 3.5; 73 [43 - 96]bB 61 ± 3.5; 56 [41 - 76]bB

149

S: survivor; NS: non-survivor. Time*: for SDNN, RMSSD Time 1:2-10 hours; Time 2: 16- 24 hours; Time 3: 30-38 hours ;Time 4: 44-52 hours. For HR, SVI, LVIDd/kg, RWT: Time 1: 12 hours; Time 2: 24 hours; Time 3: 36 hours; Time 4: 48 hours. Within a column a lower case letter denotes difference between groups within a row an upper case letters denote a difference between time. SDNN: standard deviation of normal to normal intervals; RMSSD: root mean square of successive differences; SVI: stroke volume index; LVIDd/kg: left ventricular internal diameter in diastole/kilogram; RWT: relative wall thickness.

SVI (mL/kg)

S --- 1.4 ± 0.04; 1.3 [1.2 - 1.6]a --- 1.3 ± 0.04; 1.3 [1.1 - 1.5]a < 0.001 0.689 0.933

NS --- 0.9 ± 0.11; 0.9 [0.7 - 1.2]b --- 0.9 ± 0.11; 1.0 [0.6 - 1.2]b

LVIDd/kg S --- 0.021 ± 0.0004; 0.02 [0.019 - 0.022]a

--- 0.021 ± 0.0004; 0.02 [0.019 - 0.022]a

0.004 0.702 0.532

NS --- 0.018 ± 0.0011; 0.017 [0.016 - 0.02]b

--- 0.017 ± 0.0011; 0.016 [0.016 - 0.02]b

RWT S --- 0.55 ± 0.118; 0.52 [0.49 - 0.60]a

--- 0.54 ± 0.118; 0.53 [0.48 - 0.57]a

< 0.001 0.809 0.405

NS --- 0.75 ± 0.033; 0.72 [0.65 - 0.84]b

--- 0.77 ± 0.033; 0.70 [0.69 - 0.81]b

Table 5.4. continued

150

Table 5.5. Results of backward stepwise multivariable logistic regression analyses for survival to discharge, survival to discharge for data collected at 24 hours, systemic inflammatory response syndrome (SIRS) and clinically significant arrhythmias in horses that underwent exploratory laparotomy for colic and elective surgical cases.

Model n= Variable Coefficient SE P value OR 95% CI

Survival to discharge 67 Constant 0.612 N/A N/A N/A N/A

cTnI > 0.15 ng/mL 3.043 1.175 0.010 20.982 2.097 - 209.89

Trough SVI (mL/kg) -3.992 1.783 0.025 0.0185 0.001 - 0.608 Survival to discharge (24 hr) 63 Constant -2.748 N/A N/A N/A N/A

24 hr LF (n.u) 0.165 0.067 0.014 1.179 1.034 - 1.344 24 hr SVI (mL/kg) -8.286 3.391 0.015 0.0003 0.000 - 0.194

24 hr RWT >0.63 2.664 1.417 0.060 14.35 0.893- 230.485 Systemic inflammatory response syndrome

67 Constant 2.723 N/A N/A N/A N/A Peak cTnI (ng/mL) 0.955 0.347 0.006 2.599 1.316 - 5.136

Peak PAR 0.512 0.250 0.041 1.669 1.022 - 2.723 Trough SVI (mL/kg) -4.423 1.462 0.003 0.012 0.0007 - 0.211

Clinically significant arrhythmias

74 Constant 6.917 N/A N/A N/A N/A

Peak cTnI (ng/mL) 1.017 0.316 0.0013 2.766 1.488 to 5.141

iMg2+ (0.1 mg/dL) -0.933 0.418 0.0256 0.393 0.173 to 0.893

151

CHAPTER 6

A MULTIPLE ORGAN DYSFUNCTION SCORE FOR ADULT HORSES WITH

ACUTE GASTROINTESTINAL DISEASE5

______________________________

5E.L. McConachie, S. Giguère, M.H. Barton. To be submitted to the Journal of

Veterinary Internal Medicine.

152

ABSTRACT

Background: The incidence of multiple organ dysfunction syndrome (MODS) in horses

with acute gastrointestinal disease is unknown. At present there are no validated criteria

to confirm MODS in adult horses.

Objectives: Develop criteria for a MODS score for use in horses with acute

gastrointestinal disease (MODS GI) and evaluate its association with six-month survival.

Compare the newly developed MODS GI scoring system to another recently proposed

MODS score that was extrapolated from human criteria for use in equids (MODS EQ).

Animals: Adult horses (> 1 year of age) presented for colic that required exploratory

laparotomy (n=62). Healthy adult horses (> 1 year of age) presented for an elective

surgical procedure (n=12) were used to establish the reference range of some variables.

Methods: A MODS GI scoring system was proposed based on organ system-specific

criteria that were developed from a literature review, data collected from healthy animals,

and clinical judgment. Based on data prospectively collected from Day 1 and Day 2 post-

surgery, horses with acute surgical colic were scored retrospectively using both the

MODS GI and the MODS EQ scoring criteria. The total number of organs affected and

the total number of organs failed were recorded for each horse. Receiver operating

characteristic (ROC) curve analysis was used to assess the diagnostic performance of the

MODS GI scoring system and to compare its overall performance to MODS EQ.

Results: The MODS GI score proposed herein had excellent performance post-

operatively with an area under the ROC curve (AUC) of 0.95 [0.87 - 0.99]). The area

under the ROC curve for the MODS GI score was significantly higher than that of the

MODS EQ (AUC: 0.76 [0.63-0.86].

153

Conclusions and clinical significance: The MODS GI score proposed herein predicts

six-month survival to discharge in horses with acute surgical gastrointestinal lesions. The

MODS GI score derived from equine specific criteria performed better than a score

extrapolated from human scoring systems.

INTRODUCTION

Critically-ill patients in the medical or surgical intensive care units often develop

progressive organ dysfunction unrelated to their underlying condition. This clinical

phenomenon, aptly coined multiple organ dysfunction syndrome (MODS), was first

recognized in human intensive care units in the 1970’s and ironically coincided with an

improved capacity to save trauma patients from what had previously been life-ending

injuries.1 The clinical significance of identifying sequential organ dysfunction is two-

fold. First, as multiple organs fail the risk of death increases accordingly.2 Secondly,

since the introduction of MODS scoring systems for critically-ill patients, overall MODS

severity and mortality rates have declined in the surgical ICU.3 In part, this is due to

earlier recognition of changes in patient status in combination with supportive and goal-

directed interventions.3,4

Various approaches were used to develop scoring systems for sequential organ

dysfunction in humans. In 1995, Marshall and colleagues published their MODS score

based on a review of the literature and retrospective data collected from critically-ill

patients that enabled the authors to determine organ-specific criteria that reflected a range

of clinical dysfunction from normal to failure. A separate group determined organ

dysfunction criteria empirically based on the consensus of experts and created the sepsis-

154

related organ failure assessment (SOFA) score5 (the name was later changed to sequential

organ failure assessment). Both groups applied their scoring systems prospectively to

critically-ill patients to assess the association of the score and all-cause in-hospital

mortality and the incidence of development of multiple organ failure.2,6 The purpose of

both of these scoring systems was to provide a way to objectively describe the continuum

of organ dysfunction with simple criteria that could easily be compared across a

heterogenous group of critically-ill patients and across hospitals with varied protocols.

In horses, individual organ dysfunction of the renal, hepatic, cardiovascular and

hemostatic systems has been reported most frequently in horses with acute

gastrointestinal disease.7-12 Criteria for MODS in horses, referred to herein as the MODS

EQ score, have been previously proposed based on the human MODS criteria, but remain

unvalidated.13 Multiple organ dysfunction syndrome is a dynamic process that can be

reversible if detected and managed prior to end-stage disease, highlighting the need for

equine-specific organ dysfunction criteria and a scoring system validated for clinical use.

Anecdotally, the clinical phenomenon of multiple organ system failure (MOF) is

recognized in horses, however, comprehensive reports in the equine literature are lacking.

Two studies refer to MODS as an outcome or cause for death/euthanasia,11,14 however,

the criteria used in these studies vary, and are essentially adaptations from the human

scoring systems that may not be an appropriate representation of organ dysfunction in

critically-ill adult horses.13 Additionally, the current criteria provide a dichotomous

outcome of organ dysfunction, thereby failing to reflect a continuum of organ

dysfunction. No studies have critically evaluated organ dysfunction criteria in a clinical

setting and as such it is debatable if these criteria are appropriate for describing MODS in

155

critically-ill adult horses. Similarly, the incidence of MODS and relevance of detecting

MODS in critically-ill equids are currently unknown.

The purpose of the study reported herein was to develop criteria for individual

organ dysfunction and a MODS scoring system for use in horses with acute surgical

gastrointestinal disease (MODS GI) reflecting a range of clinical severity. The

hypotheses tested in this study were that 1) the MODS GI score developed herein would

be associated with long-term outcome (six-month survival), 2) the mortality rate would

increase in correspondence with an increasing number of dysfunctional organs, and 3) the

MODS GI score would be associated with the systemic inflammatory response syndrome

(SIRS). In addition we tested the null hypothesis that the MODS GI score proposed

herein would not be inferior to the MODS EQ score.

MATERIALS AND METHODS

Score Development

A literature search was conducted to identify studies or reports of single organ

dysfunction in horses with naturally-occurring acute gastrointestinal disease, primary

organ failure or organ dysfunction following anesthetic events. The organ systems

described most commonly included the cardiovascular,a,9,10,15 renal,7 hepatic,8

gastrointestinal,16-18 musculoskeletal19 (including laminitis),20 respiratory21,22 and

hemostatic23-26 (coagulation) systems. Organ system specific criteria were then chosen

from this literature search and included serum cardiac troponin concentration (cTnI),

stroke volume index (SVI), standard deviation of normal-to-normal intervals (SDNN),

creatinine and serum bile acids (SBA) concentrations, nasogastric reflux volume,

156

abdominal distension, serum creatine kinase activity (CK), Obel grade lameness,

PaO2/FiO2 ratio, respiratory rate and effort, and platelet count or prothrombin time (PT).

Neurologic system evaluation is routinely assessed in critically-ill human patients, though

it is infrequently performed or reported in horses with colic.27 Therefore, in place of a

neurologic score, attitude was assessed using a modified pain score (Supplement 6.1)28

that incorporated both postural and social behaviors with the purpose of recording a

general sense of demeanor and awareness (lassitude, responsive or agitated) similar to the

purpose of the Glasgow Coma Score in critically-ill people.

In order to develop a range of organ dysfunction criteria, the normal reference

range from the hospital’s clinical pathology laboratory, data extracted from the literature

review, or data collected on three consecutive days from 12 healthy adult horses that

underwent general anesthesia of at least one hour duration for an elective surgical

procedure (control group), were used as aides for determining ranges for each organ

criterion. Raw data was assessed for normality with visual inspection of the histogram

and with a Shapiro-Wilks test. For normally distributed data the mean and standard

deviation were used to establish a reference interval. When data was non-normally

distributed the 95% reference interval was derived using a non-parametric approach. A

range of values that corresponded to a score of 0 to 3 (0 = normal, 1 = mildly abnormal, 2

= moderately abnormal, 3 = severely abnormal) were assigned based on the references

interval for normal values (score of 0) and abnormal values reported in the literature

associated with outcome when available for scores of 1 to 3. Ultimately, the collective

clinical judgment of the three authors was also used to empirically propose cut-off

designations which followed a similar approach to what has been done for previous

157

scoring systems in both human and veterinary medicine.5,29 The criteria used to develop

the individual scores for each organ system are presented in Supplement 6.2.

Score Assessment

The score developed herein was then retrospectively evaluated in 62 horses that

presented for colic and required exploratory laparotomy. The outcomes of interest were

survival to six months and the presence of SIRS. Horses that were euthanized solely due

to financial constraints were not included in the data set. Horses were categorized as

having SIRS if they fulfilled criteria for SIRS on Day 1 or 2 post-operatively based on

criteria used for adult horses in similar clinical studies and included two or more of the

following: temperature ≥ 101.5°F or ≤ 98.5°F; heart rate ≥ 60 bpm; respiratory rate ≥ 30

bpm; white blood cell count ≥ 14,500 cells/µL or ≤ 4,500 cells/ µL and or ≥ 10% band

neutrophils.12,13 Horses were scored based on data collected at Day 1 and Day 2 post-

operatively with possible scores ranging from 0 to 24 for the MODS GI score created

herein (Table 6.1). The total number of organs affected (score > 1) and the total number

of organs failed (score = 3) on each day were recorded for each horse for association with

outcome. The Delta MODS GI score was recorded (Day 1 MODS GI score – Day 2

MODS GI score) for each horse. Finally, the individual organ system scores were also

recorded for each horse on both days. Additionally, a total score was also given for the

MODS EQ score, with a possible range of 0 to 7 for comparison to the MODS GI score

developed in this study (Table 6.2).

Data collection

Clinical data were collected both prospectively and retrospectively from 62 horses

with acute gastrointestinal disease from Day 1 and Day 2 after surgery. All horses were

158

then scored retrospectively on the two consecutive days. If two measurements were

recorded in a 24-hour period the measurement corresponding to the worst score was used

for calculation of both MODS score. For some organ systems, multiple criteria are

proposed; in which case the variable that corresponded to the highest score was applied.

For organ systems with multiple criteria, only one score was given per organ system,

meaning that scores for individual variables within an organ system were not tallied.

Data collection has been described previously for cardiovascular parameters

evaluated herein and included measuring indirect mean arterial blood pressure (MAP),

continuous telemetry for heart rate variability (HRV) and rhythm analysis, and

echocardiography for stroke volume index as previously described.a,15 For

clinicopathologic data blood was collected in serum, calcium EDTA and citrate tubes for

measurement of serum bile acids, total bilirubin, and cTnI concentrations, GGT and CK

activities, platelet count, and PT, respectively, at admission and on Day 1 and Day 2.

Serum and citrated plasma were separated in a temperature-controlled centrifugeb at 3000

rpm for 30 minutes and stored at -80°C for batch analysis at the University’s clinical

pathology laboratoryc,d with the exception of cTnI which was analyzed with an

ultrasensitive assaye at a regional hospital. Platelet counts were run immediately on an

automated analyzerf and were confirmed with manual platelet estimates. Platelet counts

accompanied by a morphology comment indicating platelet clumping were not used in

the score. In addition, arterial blood was collected into a heparinized blood gas syringe

from the transverse facial artery for immediate measurement of PaO2 and heparinized

blood creatinine concentration on a critical care analyzerf on Days 1 and 2 post-

operatively. Attitude was scored prospectively by a single evaluator (ELM) at Day 1 and

159

2 hours post-operatively as a proxy for assessment of the neurologic system (see

Supplement 6.1). Retrospective data collection included searching the record for

respiratory rate and effort and the presence or absence of gastrointestinal sounds at

approximately 24 and 48 hours after recovering from surgery as well as measurement of

total nasogastric reflux/24 hours on Day 1 and Day 2 post-operatively. Delta values for

creatinine and cTnI concentrations were derived from the difference between admission

and Day 1 values, and from the difference between Day 1 and Day 2 values.

Statistical analysis

The overall performance of MODS total scores, total number of organs affected

(score ≥ 1) and total number of organs failed (score = 3) in predicting six-month survival

was assessed using receiver operator characteristic (ROC) curve analysis. The optimal cut

point to maximize sensitivity and specificity was selected based on the Youden index.

Logistic regression was used to calculate the odds ratio at the optimal cut point.

Multivariable logistic regression was used to investigate the association between scores

for individual organs and six-month survival. The significance of the difference between

the AUC of MODS GI and that of MODS EQ was assessed using the method described

by DeLong et al.30 A similar approach was used to assess the overall performance of

MODS GI in predicting SIRS. For all analyses, significance was set at P < 0.05.

RESULTS

Of the 62 horses with colic that required exploratory laparotomy and that were

evaluated in this study, 49 horses survived to six months and 13 horses were euthanized

prior to six months. Ten of these horses were euthanized prior to hospital discharge due

160

to reasons related to their primary complaint or development of severe complications

[clinical evidence of multiple organ failure (n=4), post-operative ileus (n=2), adhesions

diagnosed with a repeat laparotomy (n=2), septic peritonitis and abdominal incision

dehiscence (n=1), and hemoabdomen (n=1)]. Three additional horses were euthanized at

their respective farms within 45 days [14 (10-45) days] of hospital discharge due to

repeat colic episodes. Field necropsies were not performed, however all three of these

horses were moderately to severely painful, had spontaneous nasogastric reflux and

evidence of small intestinal distension upon palpation per rectum.

Surviving horses had the following diagnoses: strangulating lipoma (n= 14), right

dorsal displacement (n=10), ileal impaction (n=6) , left dorsal displacement (n=5), cecal

impaction (n=3), large colon volvulus ≥ 360 degrees with partial resection (n=2),

mesenteric volvulus (n=2), enterolith (n=2), large colon volvulus ≥ 360 degrees with no

resection (n=1), epiploic foramen entrapment (n=1), gastrosplenic ligament entrapment

(n=1), inguinal hernia with small intestinal incarceration (n=1), focal infarction of the left

dorsal colon (n=1). Two horses that survived > six months had repeat exploratory

laparotomies which revealed a right dorsal displacement in one horse and a non-

functional jejunoileostomy which was subsequently revised.

Horses that were euthanized had the following diagnosis at initial exploratory

laparotomy: strangulating lipoma (n=5), large colon volvulus ≥ 360 degrees (n=4),

epiploic foramen entrapment (n=1), omental entrapment (n=1), mesenteric volvulus (n=1)

and a right dorsal displacement with small intestinal distension (n=1). Four horses in the

non-surviving group had repeat exploratory laparotomies at which time two horses were

euthanized due to adhesion formation, one horse had a decompression of the small

161

intestine with no revision of the original jejunoileostomy, and one horse had a region of

necrotic colon resected and over-sewn.

Scores were calculated on Day 1 for all 62 horses. A score could not be calculated

for three horses on Day 2 as these three horses were euthanized prior to 36 hours post-

operatively. Horses with a MODS GI score of > 8 on Day 1 and > 6 on Day 2 had OR of

105.6 and 46.1, respectively (Table 6.3). When horses had > 3 organs affected on Day 1

or Day 2 they were 14.4 and 35.1 times more likely to not survive to six months,

respectively (Table 6.3). In addition, horses with > 1 organ failing on Day 1 were 25.3

times more likely to not survive to six months and those with > 1 organs failing on Day 2

were 22.5 times more likely to not survive at six months (Table 6.3). The overall

diagnostic performance of the MODS GI score on Day 1 (AUC: 0.93 ± 0.04) was similar

and not significantly different (P = 0.90) from that obtained on Day 2 (AUC: 0.94 ±

0.03). Therefore, the average of Day 1 and 2 was used in subsequent analyses. The

average of the MODS GI score from Day 1 and Day 2 had the best overall performance

(AUC: 0.95 ± 0.03). The best sensitivity (92%) and specificity (87%) for the MODS GI

score was at a score of > 7. Figure 6.1 demonstrates sensitivity and specificity based on

total MODS GI score at various cut points. In addition, horses with a score of > 7 were

10.7 times more likely to have SIRS. A MODS GI score excluding the cardiovascular

criteria SDNN was evaluated and had similar performance compared to the total MODS

GI score (AUC: 0.91 95% CI = 0.81- 0.97). The Delta MODS score was not

significantly associated with outcome. No single organ system was significantly

associated with outcome. The frequency of organ systems affected and the corresponding

MODS GI score on Day 1 and Day 2 are presented in Table 6.4.

162

There was a tendency for lower diagnostic performance of the MODS EQ score

on Day 1 (AUC: 0.61 ± 0.11) versus Day 2 (AUC: 0.82 ± 0.07); P = 0.058. The MODS

EQ score was subsequently averaged which resulted in fair test performance (AUC: 0.76

± 0.08). The MODS GI score performed significantly better for predicting six-month

survival compared to the MODS EQ score (P = 0.008; Figure 6.2).

DISCUSSION

The MODS GI score proposed herein had excellent test performance for

determining six-month survival. In addition, there was an association with the number of

organs affected, and the number of organs failed with six-month survival where horses

with > 3 affected organs on Day 1 or 2, or > 1 failed organ on Day 1 and > 1 failed organ

on Day 2 were significantly less likely to be alive at six months. Finally, the null

hypothesis that there would be no difference between the MODS GI score developed

herein and the MODS EQ score extrapolated from human criteria was rejected based on

the significantly better performance of the MODS GI score compared to the MODS EQ

score (Figure 6.2). A statistically significant difference was not found for the

performance of either score on Day 1 versus Day 2. However, there was a trend towards

inferior performance of the MODS EQ score on Day 1 versus Day 2, which suggests that

the performance of the MODS EQ score might vary from day to day. When considering

the variation in the MODS EQ this might limit its usefulness as a system to score organ

dysfunction on consecutive days. In addition, with the overall inferior performance of the

score when compared to the MODS GI score, the MODS EQ score is most likely

inappropriate for describing organ dysfunction in adult equids.

163

Several of the individual organ system criteria in the proposed MODS GI score

incorporate a choice from multiple possible criteria rather than limiting assessment to a

single variable. As an example the criteria for the cardiovascular system includes the

choice to score cTnI concentration, SVI, SDNN (a variable from HRV) or clinically

significant arrhythmias. Despite the possibility that multiple variables within an organ

system might be abnormal, a single score was given for each organ system with the

purpose of avoiding organ score inflation. In this study when data were available for

more than one modality for an individual organ, the criterion which gave the highest

possible score was used. It could be argued that tallying the scores for each variable

within an organ system might be another reasonable approach. The rationale for

providing multiple methods or criterion for scoring an individual organ system was

centered on the principle that the organ could be scored even if all variables were not

measured in an individual horse. Providing clinicians with a choice of measurement

modalities for individual organs offers more flexibility in a clinical setting. Recognizing

that HRV is not routinely performed in the clinic, the total score was evaluated when

SDNN was excluded from the cardiovascular criteria and the performance of the MODS

GI score remained similar (AUC: 0.91; 95% CI: 0.81-0.97).

The range of the score 0 to 24 is similar to the range established in human critical

care and provides a continuum of organ dysfunction rather than a dichotomous outcome

of failed or not failed. In this study a score of > 7 had the best sensitivity (92%) and

specificity (88%) for the outcome of survival to six months. The overall mortality rate in

the study herein was 13/62 (21%) this corresponds well to scoring and mortality rates in

164

the human literature where a MODS score between 6 and 10 corresponds with a mortality

rate between 7 and 26% and with two or fewer failed organs.2

While the MODS GI score developed herein shows potential to be useful in the

assessment of critically-ill patients, prospective studies are needed to test the performance

of this score in horses with various acute disease etiologies. The MODS and SOFA

scores were originally validated in patients in the surgical ICU2 or in the a mixed group

of medical and surgical ICU patients.6 Since the initial inception of the multiple organ

dysfunction scoring systems in those specific groups, both scores have been successfully

validated in patients with a broad spectrum of medical and surgical conditions.31-33

The limitations in this study include the lack of a specific definition for what

constitutes organ dysfunction and failure in each evaluated organ system, determination

of specific cut-off criteria based on data collected from horses at one center, and score

criteria that were in part limited to the opinion of three board-certified clinicians. A

fundamental problem with creating or “validating” a severity score of any kind is

choosing the most appropriate outcome parameter. Ideally a MODS score should be

correlated to absolute organ failure. As such the best outcome parameter would

seemingly be organ failure confirmed histopathologically. The inherent problem with that

approach is that there is not a universally accepted pathognomonic histopathologic lesion

that corresponds with organ failure.34 Instead the definition of dysfunction and

subsequent failure of organs is a clinical entity that is to a large extent, opinion

dependent. Therefore, a seemingly appropriate outcome measure for the total score is

survival to a clinically relevant time point. Survival to six months was chosen as the

outcome of interest rather than survival to discharge because in people with MODS,

165

organ dysfunction is not only related to short-term mortality but also with long-term

mortality, where typically 28-day mortality and out-of-hospital complications are typical

outcomes.35 In the group of horses that survived to hospital discharge but were

euthanized prior to six months, horses were euthanized at day 28, 39 and 68 post-

operatively, which more closely reflects the 28-day and 90-day mortality rates often used

as outcomes in human critical care.36,37

The criteria proposed herein were developed primarily from the normal reference

ranges of the clinical pathologic laboratory at this institution, normal horses from the

region and the clinical judgment of the authors. Criteria chosen empirically were used to

establish the original SOFA score in humans and have been validated repeatedly. In fact

the cardiovascular criteria used in the SOFA score performed more robustly than the

MODS score.38 Whenever available, data in the literature collected from other institutions

were incorporated into determining the cut-offs for each score within an organ system.8,19

Furthermore, the use of ROC analysis on the data set to define cut-points for organ scores

was avoided to prevent choosing criteria that were only associated with non-survival

since the goal of the score was to reflect a range of organ dysfunction. Instead, the organ

scores were proposed first and retrospectively applied to a clinical data set. Despite the

excellent performance of the MODS GI score in the horses that comprised this data set, it

will be necessary to validate the MODS GI score in multiple centers before it can be

employed as a routine clinical assessment or as a definition for MODS in clinical

research.

The proposal of this MODS GI scoring system provides an initial step for

studying and understanding the pathophysiology and incidence of multiple organ

166

dysfunction in critically-ill equids. Prospective studies will be needed to determine the

clinical utility of using a system to score MODS with the goal of determining if the

MODS GI score enhances a clinician’s ability to recognize organ insufficiency at an

earlier stage. This ultimately might provide objective criteria to monitor and measure

responses to both well-established and novel therapies thereby justifying the cost and

labor required in measuring the criteria needed to formulate the MODS GI score. At

present the MODS GI score provides an objective method to measure disease severity

and assess risk of mortality. Finally, similarly to what was one of the original purposes of

severity scoring systems in people, a reliable equine MODS score could be applied across

various equine critical care units facilitating the comparison of the performance of goal-

directed therapy. This might assist equine clinicians in meeting the ultimate target of

improving outcomes in critically-ill equine patients.

In conclusion, the MODS GI score provides prognostic information in horses with

acute surgical colic when scored post-operatively. A MODS GI score of > 7 provided

good test sensitivity and specificity and was associated with the presence of SIRS. The

MODS GI score performed better at predicting six-month survival compared to the

MODS EQ score. Future studies will be necessary to test the validity of this scoring

system prospectively on critically-ill horses in different centers and with different disease

etiologies.

167

FOOTNOTES

a. McConachie EL, Giguère S, Rapoport G, Brown S, Barton MH. Assessment of

cardiovascular status in horses with naturally acquired ischemic intestinal disease.

J Vet Cardio. Submitted.

b. Sorvall Legend X1, Thermo Fischer Scientific Inc, Suwanee, GA

c. Hatachi P-module biochemical analyzer, Roche Inc., Florence, SC

d. Trinity AMAX Destiny Coagulation analyzer, Diamond Diagnostics

e. ADVIA Centaur cTnI Ultra Assay, Immulite 1000 Siemens, Deerfield, IL

f. Heska CBC-Diff, Heska Corp, Loveland, CO

g. Nova Biomedical, Critical Care Xpress, Waltham, MA.

REFERENCES

1. Baue AE. Multiple, progressive, or sequential systems failure. A syndrome of the

1970s. Arch Surg 1975;110:779-781.

2. Marshall JC, Cook DJ, Christou NV, et al. Multiple organ dysfunction score: a reliable

descriptor of a complex clinical outcome. Crit Care Med 1995;23:1638-1652.

3. Barie PS, Hydo LJ, Shou J, et al. Decreasing magnitude of multiple organ dysfunction

syndrome despite increasingly severe critical surgical illness: a 17-year longitudinal

study. J Trauma 2008;65:1227-1235.

4. Dellinger RP. The surviving sepsis campaign: 2013 and beyond. Chin Med J (Engl)

2013;126:1803-1805.

5. Vincent JL, Moreno R, Takala J, et al. The SOFA (Sepsis-related Organ Failure

Assessment) score to describe organ dysfunction/failure. On behalf of the Working

168

Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine.

Intensive Care Med 1996;22:707-710.

6. Vincent JL, de Mendonca A, Cantraine F, et al. Use of the SOFA score to assess the

incidence of organ dysfunction/failure in intensive care units: results of a multicenter,

prospective study. Working group on "sepsis-related problems" of the European Society

of Intensive Care Medicine. Crit Care Med 1998;26:1793-1800.

7. Groover ES, Woolums AR, Cole DJ, et al. Risk factors associated with renal

insufficiency in horses with primary gastrointestinal disease: 26 cases (2000-2003). J Am

Vet Med Assoc 2006;228:572-577.

8. Underwood C, Southwood LL, Walton RM, et al. Hepatic and metabolic changes in

surgical colic patients: a pilot study. J Vet Emerg Crit Care (San Antonio) 2010;20:578-

586.

9. Radcliffe RM, Divers TJ, Fletcher DJ, et al. Evaluation of L-lactate and cardiac

troponin I in horses undergoing emergency abdominal surgery. J Vet Emerg Crit Care

(San Antonio) 2012;22:313-319.

10. Diaz OM, Durando MM, Birks EK, et al. Cardiac troponin I concentrations in horses

with colic. J Am Vet Med Assoc 2014;245:118-125.

11. Borde L, Amory H, Grulke S, et al. Prognostic value of echocardiographic and

Doppler parameters in horses admitted for colic complicated by systemic inflammatory

response syndrome. J Vet Emerg Crit Care (San Antonio) 2014;24:302-310.

12. Epstein KL, Brainard BM, Gomez-Ibanez SE, et al. Thrombelastography in horses

with acute gastrointestinal disease. J Vet Intern Med 2011;25:307-314.

169

13. Hart KA, MacKay, R. J. Endotoxemia and Sepsis In: Smith BP, ed. Large Animal

Internal Medicine, 5th Edition. St. Louis, MO: Mosby; 2013:684.

14. Sheats MK, Cook VL, Jones SL, et al. Use of ultrasound to evaluate outcome

following colic surgery for equine large colon volvulus. Equine Vet J 2010;42:47-52.

15. McConachie EG, S. Rapoport, G. Barton, M. . Heart rate variability in horses with

acute gastrointestinal disease requiring exploratory laparotomy. J Vet Emerg Crit Car

2015; In press.

16. Lefebvre D, Pirie RS, Handel IG, et al. Clinical features and management of equine

post operative ileus: Survey of diplomates of the European Colleges of Equine Internal

Medicine (ECEIM) and Veterinary Surgeons (ECVS). Equine Vet J 2015. Early view.

DOI: 10.1111/evj.12355.

17. Roussel AJ, Jr., Cohen ND, Hooper RN, et al. Risk factors associated with

development of postoperative ileus in horses. J Am Vet Med Assoc 2001;219:72-78.

18. Cohen ND, Lester GD, Sanchez LC, et al. Evaluation of risk factors associated with

development of postoperative ileus in horses. J Am Vet Med Assoc 2004;225:1070-1078.

19. Krueger CR, Ruple-Czerniak A, Hackett ES. Evaluation of plasma muscle enzyme

activity as an indicator of lesion characteristics and prognosis in horses undergoing

celiotomy for acute gastrointestinal pain. BMC Vet Res 2014;10 Suppl 1:S7.

20. Parsons CS, Orsini JA, Krafty R, et al. Risk factors for development of acute

laminitis in horses during hospitalization: 73 cases (1997-2004). J Am Vet Med Assoc

2007;230:885-889.

21. Ryu SH, Kim JG, Bak UB, et al. A hematogenic pleuropneumonia caused by

postoperative septic thrombophlebitis in a Thoroughbred gelding. J Vet Sci 2004;5:75-77.

170

22. Wilkins PO, C. Baumgardener, J. et al. . Acute lung injury and acute respiratory

distress syndromes in veterinary medicine: consensus definitions: The Dorothy Russell

Haveymeyer Working Group on ALI and ARDS in Veterinary Medicine J Vet Emerg

Crit Care (San Antonio) 2007;17:333-339.

23. Dallap BL, Dolente B, Boston R. Coagulation profiles in 27 horses with large colon

volvulus. J Vet Emerg Crit Car 2003;13:215-225.

24. Welch RD, Watkins JP, Taylor TS, et al. Disseminated intravascular coagulation

associated with colic in 23 horses (1984-1989). J Vet Intern Med 1992;6:29-35.

25. Dallap Schaer BL, Epstein K. Coagulopathy of the critically ill equine patient. J Vet

Emerg Crit Care (San Antonio) 2009;19:53-65.

26. Dolente BA, Wilkins PA, Boston RC. Clinicopathologic evidence of disseminated

intravascular coagulation in horses with acute colitis. J Am Vet Med Assoc

2002;220:1034-1038.

27. Sharkey LC, DeWitt S, Stockman C. Neurologic signs and hyperammonemia in a

horse with colic. Vet Clin Pathol 2006;35:254-258.

28. Pritchett LC, Ulibarri C, Roberts MC, et al. Identification of potential physiological

and behavioral indicators of postoperative pain in horses after exploratory celiotomy for

colic. Appl Anim Behav Sci 2003;80:31-43.

29. Brewer BD, Koterba AM. Development of a scoring system for the early diagnosis

of equine neonatal sepsis. Equine Vet J 1988;20:18-22.

30. DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or

more correlated receiver operating characteristic curves: a nonparametric approach.

Biometrics 1988;44:837-845.

171

31. Ceriani R, Mazzoni M, Bortone F, et al. Application of the sequential organ failure

assessment score to cardiac surgical patients. Chest 2003;123:1229-1239.

32. Lorente JA, Vallejo A, Galeiras R, et al. Organ dysfunction as estimated by the

sequential organ failure assessment score is related to outcome in critically ill burn

patients. Shock 2009;31:125-131.

33. Graciano AL, Balko JA, Rahn DS, et al. The Pediatric Multiple Organ Dysfunction

Score (P-MODS): development and validation of an objective scale to measure the

severity of multiple organ dysfunction in critically ill children. Crit Care Med

2005;33:1484-1491.

34. Lucas S. The Autopsy Pathology of Sepsis-Related Death, Severe Sepsis and Septic

Shock-Understanding a Serious Killer. In: Fernandez R, ed. InTech; 2012.

35. Mizock BA. The multiple organ dysfunction syndrome. Dis Mon 2009;55:476-526.

36. Serpa Neto A, Veelo DP, Peireira VG, et al. Fluid resuscitation with hydroxyethyl

starches in patients with sepsis is associated with an increased incidence of acute kidney

injury and use of renal replacement therapy: a systematic review and meta-analysis of the

literature. J Crit Care 2014;29:185 e181-187.

37. Schmidt H, Muller-Werdan U, Hoffmann T, et al. Autonomic dysfunction predicts

mortality in patients with multiple organ dysfunction syndrome of different age groups.

Crit Care Med 2005;33:1994-2002.

38. Peres Bota D, Melot C, Lopes Ferreira F, et al. The Multiple Organ Dysfunction

Score (MODS) versus the Sequential Organ Failure Assessment (SOFA) score in

outcome prediction. Intensive Care Med 2002;28:1619-1624.

172

Supplement 6.1. Modified Attitude score1

1. Pritchett LC, Ulibarri C, Roberts MC, et al. Identification of potential physiological

and behavioral indicators of postoperative pain in horses after exploratory celiotomy for

colic. Appl Anim Behav Sci 2003;80:31-43.

Criteria 1 2 3 4 Score Gross pain: (e.g. flank watching, rolling, pawing, teeth grinding)

None Occasional Continuous

Head position Above withers

At withers Below withers

Ear position Forward, frequent moving

Slightly back little movement

Location in stall

At door watching environment

Standing in middle facing door

Standing in middle facing sides of walls

Standing in middle facing back of the stall

Spontaneous locomotion

Moves freely Occasional steps

No movement

Response to open door

Moves to door

Looks at door

No response

Response to approach

Moves to observer, ears forward

Looks at observer ears forward

Moves away from observer

Does not move, ears back

Total Attitude score

173

Supplement 6.2. Rationale for MODS GI score

URL: upper reference limit; cTnI: cardiac troponin I; SVI: stroke volume index; SDNN: standard deviation of normal-to-normal intervals; CK: creatine kinase;*serum sample measured with an ultrasensitive assay; †Measured with echocardiography using the 4 chamber area-length method; ‡Measured from a 5-min artifact- and arrhythmia-free ECG recording using Kubios HRV software, γmeasured with a rapid critical care biochemical analyzer (Nova Biomedical),# the difference between two consecutive creatinine concentrations measured ≥ 24 hours apart (or at least 12 hours apart) when patient is on intravenous fluid therapy. °see Supplement 6.1

Organ system Criteria Rationale for Score criteria 0 1 2 3 Cardiovascular cTnI (ng/mL)* URL Literature review1,2 (McConachie 2015)

Delta cTnI Clinical judgment SVI (ml/kg/min)†

Literature review3 (McConachie 2015)

SDNN (ms)‡ Literature review4 Renalγ Creatinine

(mg/dL) Upper 95% CI from EC

Clinical judgment and literature review5

Delta creatinine#

Clinical judgment

Hepatic Serum bile acids (µmol/L)

URL Clinical judgment and literature review6

Respiratory PaO2/FiO2 Lower limit of 95% CI from EC

Clinical judgment, reflects PaO2 63-84 mmHg in unventilated animal breathing room air

Literature review7

Respiratory rate/effort

Clinical judgment

Musculoskeletal CK (U/L) URL Clinical judgment and literature review8 Laminitis Clinical judgment and literature review

Coagulation Platelet count (x 103 cells/µL)

Lower limit of laboratory RR

Clinical judgment and literature review9,10

Prothrombin time (sec)

Upper limit of laboratory RR

Clinical judgment and literature review10

Gastrointestinal Nasogastric reflux (L/24h)

Clinical judgment and literature review11-13

Abdominal distension

Clinical judgment

Neurologic Modified attitude score°

Clinical judgment

174

1. Nath LC, Anderson GA, Hinchcliff KW, et al. Clinicopathologic evidence of

myocardial injury in horses with acute abdominal disease. J Am Vet Med Assoc

2012;241:1202-1208.

2. Radcliffe RM, Divers TJ, Fletcher DJ, et al. Evaluation of L-lactate and cardiac

troponin I in horses undergoing emergency abdominal surgery. J Vet Emerg Crit Care

(San Antonio) 2012;22:313-319.

3. Borde L, Amory H, Grulke S, et al. Prognostic value of echocardiographic and Doppler

parameters in horses admitted for colic complicated by systemic inflammatory response

syndrome. J Vet Emerg Crit Care (San Antonio) 2014;24:302-310.

4. McConachie EG, S. Rapoport, G. Barton, M. . Heart rate variability in horses with

acute gastrointestinal disease requiring exploratory laparotomy. J Vet Emerg Crit Car

2015; In press.

5. Groover ES, Woolums AR, Cole DJ, et al. Risk factors associated with renal

insufficiency in horses with primary gastrointestinal disease: 26 cases (2000-2003). J Am

Vet Med Assoc 2006;228:572-577.

6. Underwood C, Southwood LL, Walton RM, et al. Hepatic and metabolic changes in

surgical colic patients: a pilot study. J Vet Emerg Crit Care (San Antonio) 2010;20:578-

586.

7. Wilkins PO, C. Baumgardener, J. et al. . Acute lung injury and acute respiratory

distress syndromes in veterinary medicine: consensus definitions: The Dorothy Russell

Haveymeyer Working Group on ALI and ARDS in Veterinary Medicine J Vet Emerg

Crit Care (San Antonio) 2007;17:333-339.

175

8. Krueger CR, Ruple-Czerniak A, Hackett ES. Evaluation of plasma muscle enzyme

activity as an indicator of lesion characteristics and prognosis in horses undergoing

celiotomy for acute gastrointestinal pain. BMC Vet Res 2014;10 Suppl 1:S7.

9. Dolente BA, Wilkins PA, Boston RC. Clinicopathologic evidence of disseminated

intravascular coagulation in horses with acute colitis. J Am Vet Med Assoc

2002;220:1034-1038.

10. Dallap BL, Dolente B, Boston R. Coagulation profiles in 27 horses with large colon

volvulus. J Vet Emerg Crit Car 2003;13:215-225.

11. Roussel AJ, Jr., Cohen ND, Hooper RN, et al. Risk factors associated with

development of postoperative ileus in horses. J Am Vet Med Assoc 2001;219:72-78.

12. Cohen ND, Lester GD, Sanchez LC, et al. Evaluation of risk factors associated with

development of postoperative ileus in horses. J Am Vet Med Assoc 2004;225:1070-1078.

13. Lefebvre D, Pirie RS, Handel IG, et al. Clinical features and management of equine

post operative ileus: Survey of diplomates of the European Colleges of Equine Internal

Medicine (ECEIM) and Veterinary Surgeons (ECVS). Equine Vet J 2015. Early view.

DOI: 10.1111/evj.12355

176

Table 6.1. Equine MODS GI criteria

Organ system Criteria Score criteria Organ score

0 1 2 3 Cardiovascular cTnI (ng/mL)* ≤ 0.03 0.04 - 0.14 0.15 to

0.25 > 0.25

Delta cTnI Positive Negative SVI (ml/kg/min)†

≥1.4 1.2-1.3 1.0-1.1 ≤0.9

SDNN (ms)‡ >56 40-56 26.7-39 < 26.7 Renalγ Creatinine

(mg/dL) ≤ 1.9 1.9-2.2 2.3-3 > 3

Delta creatinine# Positive or 0 (when RV <1.9) OR > 1.9 prior to fluid therapy that is within RR within 24h

0 to ≤ 0.2 (when RV 1.9 to 2.2) OR ≤ -0.3 when RV < 1.9

≤ -0.1 (when RV ≥ 1.9); OR 0 to ≤ 0.2 (when RV ≥ 2.3)

< -0.1 to OR < -0.2 OR ≤ -0.3 when RV > 2.3

Hepatic Serum bile acids (µmol/L)

≤15 16 – 30 30 - 50 > 50

Respiratory PaO2/FiO2 > 400 300 – 400 200 - 300 < 200 Respiratory rate/effort

Normal Abnormal (RR > 30

bpm, nostril flare,

increased abdominal

effort)

Musculoskeletal CK (U/L) < 343 343 – 643 644 - 943 > 943 Laminitis None Obel grade I Obel

grade II or >

Coagulation Platelet count (x 103 cells/µL)

≥ 104 88 – 103 55 - 88 < 55

Prothrombin time (sec)

< 11.6 11.6 - 13.6 13.7 - 14.3 > 14.3

Gastrointestinal Nasogastric reflux (L/24 h)

< 10 10 – 36 36 - 50 > 50

Abdominal distension

No Yes

Neurologic Attitude score [7-27]

7 - 12 13 - 18 19 - 23 24 - 27

Total Score: RV: reference value. *serum measured with an ultrasensitive assay; †Measured using the 4 chamber area-length method; ‡Measured from a 5-min artifact- and arrhythmia-free ECG (Kubios HRV software); γmeasured with a rapid critical care biochemical analyzer;# the difference between two consecutive creatinine concentrations measured ≥ 24 hours apart (or at least 12 hours apart) when patient is on intravenous fluid therapy.

177

Table 6.2. MODS EQ1 criteria

GGT: gamma glutamyl transferase; aPTT: activated partial thromboplastin time 1. Hart KA, MacKay, R. J. . Endotoxemia and Sepsis In: Smith BP, ed. Large Animal

Internal Medicine, 5th Edition. St. Louis: Mosby; 2013:684.

Organ system Criteria Organ Score

Cardiovascular (Hemodynamic)

Mean arterial pressure < 65 mm Hg after ≥ 20 mL/kg IV crystalloid fluids

Renal* Creatinine > 2 mg/dL after ≥ 20 mL/kg IV crystalloid fluids, or increase of ≥ 0.5 mg/dL since last measurement

Hepatic Bilirubin concentration >6 mg/dL; GGT > 60 U/L with no other explanation

Respiratory PaO2 < 65 mm Hg, or < 75 mm Hg with oxygen supplementation or mechanical ventilation

Musculoskeletal (Laminitis) Bounding digital pulses, sensitivity to digital pressure over the coronary band, sensitivity to hoof tester pressure over the sole, Obel grade >1

Coagulation Platelet count < 100,000/μL or aPTT > 70 seconds

Gastrointestinal Absent gut sounds, or absent motility on ultrasound examination

Neurologic Severe obtundation (stupor, semicoma, coma)

Total Score:

178

Table 6.3. Ideal score cut-off as assessed by ROC curve analysis by Day for MODS GI total score, number of organs affected and number of failed organs associated with six-month survival (P < 0.05) and corresponding odds ratios.

Score Score Sensitivity Specificity ROC Logistic regression AUC ± SE P OR (95% CI) P

Day 1 MODS GI > 8 92.3 89.8 0.94 ± 0.03 < 0.0001 105.6 (11.2 to 991.9) <0.0001 Total No. organs affected > 3 92.3 65.3 0.84 ± 0.05 < 0.0001 14.4 (1.7 to 122.9) 0.0015 Total No. failed organs > 1 69.2 91.8 0.90 ± 0.04 < 0.0001 25.3 (5.3 to 120.4) < 0.0001 Day 2 MODS GI > 6 90.0 83.7 0.94 ± 0.03 < 0.0001 46.1 (5.1 to 416.5) 0.0006 Total No. organs affected > 3 90.0 79.6 0.90 ± 0.04 < 0.0001 35.1 (4.0 to 310.4) 0.0014 Total No. failed organs ≥ 1 90.0 71.4 0.87 ± 0.07 < 0.0001 22.5 (2.6 to 194.5) 0.0047

179

Figure 6.1. Sensitivity and Specificity for the MODS GI score at various cut-points

180

Figure 6.2. Receiver operator characteristic curve analysis for MODS GI and MODS EQ

scores

181

Table 6.4. Number of horses (n) with organ systems affected according to MODS GI on Day 1 and Day 2

Organ System Time SCORE 0 1 2 3

Cardiovascular Day 1 n= 5 n= 10 n= 18 n= 29

Day 2 n= 7 n= 20 n= 15 n= 19 Renal Day 1 n= 54 n= 5 n= 0 n= 2

Day 2 n= 48 n= 8 n= 0 n= 0 Hepatic Day 1 n= 53 n= 5 n= 1 n= 0

Day 2 n= 54 n= 1 n= 1 n= 0 Respiratory Day 1 n= 50 n= 10 n= 2 n= 0

Day 2 n= 54 n= 8 n= 0 n= 0 Coagulation Day 1 n= 15 n= 32 n= 4 n= 10

Day 2 n= 26 n= 26 n= 3 n= 4 Musculoskeletal Day 1 n= 17 n= 15 n= 15 n= 12

Day 2 n= 22 n= 26 n= 4 n= 7 Neurologic Day 1 n= 28 n= 19 n= 9 n= 6

Day 2 n= 46 n= 9 n= 5 n= 0 Gastrointestinal Day 1 n= 47 n= 10 n= 0 n= 5

Day 2 n= 46 n= 7 n= 1 n= 8

182

CHAPTER 7

CONCLUSION

In previous chapters, data were presented to demonstrate the importance of colic

as a common disease entity in the horse to highlight the significant loss of use and

mortality caused by this condition.1-5 Despite the fact that acute GI disease has been the

focus of equine research for decades, substantial morbidity and mortality persists,

particularly in the post-operative period. One reason for this, similar to what previously

has been recognized in human medicine, is the occurrence of multiple organ dysfunction

and failure in the post-operative period. This phenomenon has gradually become apparent

in horses as we have improved our ability to diagnose and treat horses in the acute stages

of GI disease. These improvements in care have led to a new set of challenges in the post-

operative period related to remote and sequential organ dysfunction.6,7

The cardiovascular system was the focal point of the research presented herein for

the following key reasons: 1) cardiovascular system dysfunction and failure contributes

to the pathogenesis of MODS by promoting injury in other organs as a consequence of

reduced oxygen delivery 2) there is a significant association between cardiovascular

dysfunction, development of MODS and mortality in humans8 and 3) apart from the GI

tract itself, the cardiovascular system is perhaps the next most extensively studied organ

system in the horse due to the interest in athletic performance.9,10

Consequently, the overall aims of the studies presented herein were to describe

cardiovascular system function in horses with acute GI disease in comparison to healthy

183

adult horses and to develop a criterion based definition for MODS in horses. These goals

were achieved through a series of studies that comprised the work in this dissertation.

First, a non-invasive, 2-Dimensional echocardiographic method for CO estimation was

validated in standing adult horses. Secondly, HRV was investigated in horses with acute

GI disease and compared to healthy control horses, providing a novel approach to assess

overall cardiac health and the role of the autonomic nervous system in horses with colic.

The association between HRV, the autonomic nervous system and the cardiovascular

system, as discussed in previous chapters, is highlighted by the finding that reduced

overall HRV and increased LF power in septic humans is predictive of the development

of MODS and 28-day mortality.8,11

In the third study, the cardiovascular system was assessed comprehensively in

horses with acute GI disease by incorporation of electrocardiography, serum cTnI

concentration measurement and hemodynamic assessments including a proof of concept

measure, pressure adjusted heart rate (PAR). Finally in the fourth paper, MODS criteria

were developed and validated utilizing a group of horses with acute surgical GI disease.

Important, clinically relevant conclusions were drawn from the studies presented

herein. First, three 2-D volumetric methods for estimating CO were validated in adult

standing horses. The 4-chamber area length, the 4-chamber modified Simpson’s and the

Bullet method all had acceptable agreement with the reference method, lithium dilution.

While the Doppler method utilizing the right ventricular outflow tract was not

significantly different from the 2-D volumetric methods, the Doppler method from the

left ventricular outflow tract had a larger relative bias highlighting its inaccuracy and lack

of utility in a clinical setting. In the study herein, the use of healthy horses of different

184

breeds and ages to compare methods of CO measurement enhanced the applicability of

our results to a clinical setting in comparison to previous studies that used a specific

breed of horse, typically Standardbreds or Thoroughbreds. In addition to having a higher

degree of variability than most 2-D derived measurements, Doppler echocardiography

methods are more challenging to obtain in horses of various breeds and sizes which

becomes a factor independent of clinical proficiency. The 2-D echocardiographic

methods provide a way to recognize changes in the magnitude and direction of CO,

which is critical to the overall assessment of CV system status. Further studies will be

needed to determine the utility of non-invasive CO measures in the assessment of

common clinical practices, such as intravenous fluid administration and to define factors

that impact their accuracy and precision, such as variation over time and between

observers.

The second study highlighted a critically important difference in overall

cardiovascular health between horses with colic that required surgical exploration and

healthy horses undergoing elective surgical procedures, namely HRV. The HRV was

significantly reduced in horses with acute GI disease of any cause compared to the

healthy control horses. Furthermore, reduced HRV was associated with ischemic GI

lesions and non-survival, which suggests value for HRV in monitoring post-operative

colic cases. Specifically, the time domain method, SDNN was particularly useful when

assessed at a heart rate that was < 55 beats per minute, in which horses with an SDNN of

< 39.5 ms were 16.4 times more likely to not survive to discharge. The clinical

significance of this finding is apparent when one considers the fact that in the post-

operative period surviving horses had a median heart rate of 45 + 1.3 beats per minute in

185

contrast to non-survivors which had a median heart rate of 65 + 3.4 beats per minute.

Therefore, horses with heart rates that are in the range of 45- 55 beats per minute, that

may otherwise appear to be improving, may benefit from having HRV assessed routinely.

In such cases, when the SDNN is < 39.5 ms, further monitoring and addition or

continuation of supportive interventions in the post-operative period should be

considered. In contrast to many studies in humans, the frequency domain methods,

LF/HF ratio, LF power, and HF power were not associated with outcome. There were,

however, significant differences between horses when they were grouped according to

survival status where non-survivors had an increased LF/HF ratio, increased LF power

and reduced HF power. Taken together, these results indicate that there was concurrent

sympathetic overdrive and parasympathetic withdrawal in non-surviving horses.

Similarly, the SDNN and RMSSD variables that approximate overall HRV and

parasympathetic modulation, respectively, were reduced in horses with ischemic GI

disease and non-survivors offering further support to the proposal that autonomic

imbalance occurs in these cases. Perhaps one of the most clinically useful results of this

study was the finding that there was no significant difference between time domain

parameters when they were derived from 5- or 30-minute ECG recordings. Therefore, 5-

minute duration ECG recordings are sufficient to obtain time domain HRV indices

making this a useful clinical stall side tool.

In the third study, horses with acute GI disease had evidence of myocardial injury,

clinically significant arrhythmias, reduced stroke volume index (SVI), and reduced HRV

compared to healthy control horses. Specifically, multivariable logistic regression

analysis revealed a significant association between non-survival and both cTnI

186

concentration and low SVI. In contrast to what is observed in people, PAR was not a

useful composite measure in horses, as it was not significantly different between healthy

horses and those with acute GI disease and was not associated with outcome. However,

peak PAR, along with peak cTnI and lowest SVI were retained in a multivariable logistic

regression model that was associated with SIRS. Incremental increases in PAR are used

to define cardiovascular dysfunction in the original MODS score developed by Marshall

and colleagues.12 The PAR was initially proposed as a method to correct cardiovascular

function for physiologic support. The finding that PAR was retained as an important

explanatory variable in the SIRS multivariable logistic regression model indicates that

PAR might be a useful alternative for assessment of peripheral vascular hemodynamics

rather than measurement of MAP or CVP alone in horses. Fractional shortening, a

measure of contractility, was not different between horses with colic and control cases,

which is similar to results reported by Nath and colleagues.13 Although stroke volume

index was decreased in horses that did not survive, a single conclusion regarding the state

of left ventricular systolic function in horses with acute GI disease could not be made.

The results of this study provided a more complete understanding of the

significance of serum cTnI concentrations in horses with acute GI disease. Of note, mild

increases in cTnI are prevalent in horses that undergo exploratory laparotomy for acute

GI disease, but not in healthy horses that undergo surgery for elective procedures. The

magnitude of the cTnI increase was associated with outcome wherein a cTnI

concentration of > 0.15 ng/mL was an important explanatory variable for non-survival in

a multivariable logistic regression model. In addition, cTnI concentration was associated

with the presence of clinically significant arrhythmias and SIRS in horses with acute GI

187

disease. While non-survivors tended to have increased cTnI concentrations compared to

survivors post-operatively, there was not a significant difference between survivors and

non-survivors in the admission cTnI concentrations calling to attention that use of this

biomarker as a prognostic tool at admission would be inappropriate.

Finally, a MODS score based on equine data was developed, herein designated

the MODS GI score, and was associated with survival in horses with acute GI disease. In

addition, the MODS GI score performed equally well on both days of analysis and

performed better than a previously proposed score, designated the MODS EQ score,

which is based on organ function criteria extrapolated from human scoring systems.

Further investigation is necessary to determine if this score is significantly associated

with MODS in horses with other disease etiologies.

In summary, significant mortality is associated with dysfunction of the

cardiovascular system in horses characterized primarily by a syndrome of low SVI,

increased occurrence of clinically significant arrhythmias, reduced HRV and increased

serum cTnI concentration. The HRV was found to be a useful tool in horses with acute

GI disease and can be measured simply and rapidly with an artifact-free, 5-minute ECG

recording. Finally, a MODS GI score was developed and tested in a group of horses with

acute GI disease revealing that indeed horses with a higher total MODS GI score and

those with incrementally more organs deemed dysfunctional or failed were less likely to

survive. The criteria for the MODS GI score were based off established reference

intervals, a review of the literature and clinical judgment and the resultant score was

associated with six-month survival rather than confirmed organ failure. This is

188

appropriate as MODS is a functional, rather than a structural, syndrome which cannot be

adequately defined by pathology in many if not the majority of cases.

REFERENCES

1. aphis.usda.gov. Trends in Equine Mortality, 1998-2005. 2007.

2. Proudman CJ, Edwards GB, Barnes J, et al. Factors affecting long-term survival of

horses recovering from surgery of the small intestine. Equine Vet J 2005;37:360-365.

3. Suthers JM, Pinchbeck GL, Proudman CJ, et al. Survival of horses following

strangulating large colon volvulus. Equine Vet J 2013;45:219-223.

4. Mair TS, Smith LJ. Survival and complication rates in 300 horses undergoing surgical

treatment of colic. Part 3: Long-term complications and survival. Equine Vet J

2005;37:310-314.

5. Davis W, Fogle CA, Gerard MP, et al. Return to use and performance following

exploratory celiotomy for colic in horses: 195 cases (2003-2010). Equine Vet J

2013;45:224-228.

6. Sheats MK, Cook VL, Jones SL, et al. Use of ultrasound to evaluate outcome

following colic surgery for equine large colon volvulus. Equine Vet J 2010;42:47-52.

7. Borde L, Amory H, Grulke S, et al. Prognostic value of echocardiographic and

Doppler parameters in horses admitted for colic complicated by systemic inflammatory

response syndrome. J Vet Emerg Crit Care (San Antonio) 2014;24:302-310.

8. Werdan K, Oelke A, Hettwer S, et al. Septic cardiomyopathy: hemodynamic

quantification, occurrence, and prognostic implications. Clin Res Cardiol 2011;100:661-

668.

189

9. Durando MM, Corley KT, Boston RC, et al. Cardiac output determination by use of

lithium dilution during exercise in horses. Am J Vet Res 2008;69:1054-1060.

10. Blissitt KJ, Young LE, Jones RS, et al. Measurement of cardiac output in standing

horses by Doppler echocardiography and thermodilution. Equine Vet J 1997;29:18-25.

11. Pontet J, Contreras P, Curbelo A, et al. Heart rate variability as early marker of

multiple organ dysfunction syndrome in septic patients. J Crit Care 2003;18:156-163.

12. Marshall JC, Cook DJ, Christou NV, et al. Multiple organ dysfunction score: a

reliable descriptor of a complex clinical outcome. Crit Care Med 1995;23:1638-1652.

13. Nath LC, Anderson GA, Hinchcliff KW, et al. Clinicopathologic evidence of

myocardial injury in horses with acute abdominal disease. J Am Vet Med Assoc

2012;241:1202-1208.