exercise echocardiography: detection wall motion...

Exercise Echocardiography: Detectionof Wall Motion Abnormalities During IschemiaSTEVEN J. MASON, M.D., JAMES L. WEISS, M.D., MYRON L. WEISFELDT, M.D.,

JOHN B. GARRISON, PH.D., AND NICHOLAS J. FORTUIN, M.D.

SUMMARY To assess the feasibility of detecting wall motion abnormalities with echocardiography duringexercise-induced ischemia, we performed echocardiograms on 13 patients with angiographically documentedcoronary artery disease at rest and during supine bicycle exercise at increasing work loads until angina orischemic electrocardiographic changes appeared. We analyzed echocardiographic indices of regional left ven-tricular function in these patients and 11 age- and heart rate-matched normal volunteers. In the 13 patients, 22of 25 echocardiographically defined wall segments (13 septa, nine posterior left ventricular walls) weresupplied by coronary arteries with > 70% stenosis and were compared with the corresponding 22 segmentsfrom the 11 normals. Mean systolic septal thickening increased in the 22 segments of normals from 56 ± 3%(SEM) at rest to 77 ± 7% in exercise (p < 0.01) while in the patients' 22 wall segments supplied by stenoticvessels the mean value fell during peak exercise from 59 6 to 35 ± 3% (p < 0.005). Mean systolic posteriorleft ventricular wall thickening rose similarly in normals from 89 ± 9 to 115 ± 8% (p < 0.005) but fell duringpeak exercise from 75 ± 9 to 54 ± 9% (p < 0.01) in the patients' nine abnormally perfused segments. Max-imal velocity of diastolic wall thinning rose from rest to exercise in the septa and posterior left ventricular wallsof normals from 5.5 ± 0.3 to 7.7 ± 0.6 cm/sec (p < 0.005) and from 8.4 ± 0.8 to 11.8 ± 1.2 cm/sec(p < 0.001), respectively. In patients, these same indices fell at peak exercise from 5.9 ± 0.5 to 4.3 ± 0.4cm/sec (p < 0.05) and from 8.3 ± 1.2 to 6.2 ± 1.4 cm/sec (p < 0.005 vs normals in exercise), respectively.The percent systolic change in left ventricular internal diameter increased from rest to exercise in normalsfrom 38 ± 2 to 44 ± 2% (p < 0.001), but fell in patients during peak exercise from 35 ± 2 to 28 ± 2%(p < 0.001).

Exercise echocardiography appears in these patients to be sensitive in detecting wall motion abnormalitiesduring exercise-induced ischemia, and may be applicable in patients in whom exercise electrocardiography isequivocal, or where the functional significance of a coronary arterial lesion is uncertain.

IN PATIENTS WITH coronary artery disease thefunction of the regions of myocardium supplied by astenosed artery depends on the balance between oxy-gen demand of the muscle and the limited capacity ofthe narrowed artery to increase its delivery of oxy-genated blood. In the absence of previous infarction orongoing ischemia, many patients with even severelystenosed coronary arteries have normal ventricularfunction at rest. During exercise, however, as myocar-dial oxygen demands increase, flow through thestenosed artery can increase only to a certain level;beyond that level, the supplied myocardium becomesischemic.

Studies in animals indicate that regional left ven-tricular wall motion and thickening decreasedramatically within seconds after the onset ofischemia.", 2 Although decreases in regional endocar-

From the Cardiology Division, The Johns Hopkins MedicalInstitutions, and The Johns Hopkins University Applied PhysicsLaboratory.

Presented in part at the 50th Scientific Session of the AmericanHeart Association, Miami Beach, Florida, November 28, 1977.

Supported in part by NIH grant HL-19232 from the US PublicHealth Service, the Frank T. McClure Fellowship in Cardiovas-cular Research at the Johns Hopkins University, and a researchfellowship from the Maryland Affiliate of the American HeartAssociation.

Dr. Mason's present address: 60 Temple Street, New Haven,Connecticut.

Address for reprints: James L. Weiss, M.D., Harvey 502, TheJohns Hopkins Hospital, 601 North Broadway, Baltimore,Maryland 21205.

Received June 7, 1978; revision accepted September 6, 1978.Circulation 59, No. 1, 1979.

dial motion and global function have been demon-strated during exercise-induced ischemia by cine-ventriculography3' and radionuclide techniques,5accurate measurement of the time course of changesin wall thickness is not possible with these methods.Changes in wall thickness may be more independentof translational movement of the heart than arechanges in endocardial motion relative to an arbitraryreference point. In addition, recent studies have shownthat areas of the left ventricular endocardium thatmove poorly on a resting ventriculogram may containnormal myocardium histologically.6' A techniquethat examines regional indices of thickening and thin-ning might give a more accurate and objective assess-ment of viability. Accordingly, we assessed the accu-racy of echocardiography in detecting and localizingabnormalities in regional left ventricular wall thicken-ing and thinning during exercise-induced ischemia in agroup of patients with coronary artery disease.

Methods

Study Population

Eighty-three persons were examined initially byM-mode echocardiography at rest. The normal sub-jects were derived from two sources: 1) volunteers age-matched to the patient population who had nohistorical, physical, electrocardiographic, or echo-cardiographic findings suggestive of disease in the car-diovascular system and negative stress ECGs; and 2)individuals with chest pain of uncertain etiologyreferred to The Johns Hopkins Hospital for coronary

50

by guest on June 23, 2018http://circ.ahajournals.org/

Dow

nloaded from

http://circ.ahajournals.org/

DETECTION OF ISCHEMIA BY EXERCISE ECHO/Mason et al.

TABLE 1. Study Population

Angiographic dataPropranolol Exercise data* No.

Patient dose Resting Maximum HR for ECG leads EF vessels % Stenosisno. Sex Age (mg/day) HR HR echo positive (%) narrowed LAD PD

Ischemic heart disease1 M 33 0 78 123 123 2,3.F,Vs 6 77 1 95 02 M 55 0 86 97 97 None 37 3 100 903 M 57 0 68 73 73 2,3,F 61 3 90 904 M 50 320 64 70 70 None 51 3 70 70-905 M 56 80 64 101 101 2,3,F,V3-6 74 3 70-90 706 M 54 0 54 107 107 2,3,F,V3-6 71 2 70 307 M 39 0 72 117 117 V5 6 50 3 70 1008 M 59 0 64 103 103 1,2,V3 5 80 3 70-90 09 M 46 160 70 90 90 2,3,F,V6 70 3 70-90 70-9010 M 55 80 84 106 106 V5-6 61 2 70 011 M 50 0 72 95 95 2,V3.5 70 2 95 50-7012 M 48 0 54 85 85 1,2,3,F,V3-6 70 3 70-90 7013 M 57 0 67 84 84 1,2,V2-6 66 3 100 100

51 =i=8 96 =t 16

Normal14 F 50 160 71 89 89 None 76 0 0 015 F 40 40 69 125 100 None 82 0 0 016 M 51 0 85 140 116 None 64 0 0 017 M 58 0 69 140 99 None18 M 55 0 63 125 97 None19 M 64 0 56 110 96 None20 M 54 0 66 140 97 None21 F 50 0 56 155 98 None22 F 46 0 56 140 91 None23 M 62 0 64 140 96 None24 M 30 0 74 157 127 None

53 9 101 13

*ECG positive if 1 mm or more J point depression with planar or downsloping ST segment 1 mm or more depressed 80 msecafter J point.

Abbreviations: HR = heart rate; EF = ejection fraction; LAD = left anterior descending; PD = posterior descending.

arteriography who had no evidence of cardiac diseaseor coronary artery spasm at catheterization. Thepatients had angiographically documented significantcoronary artery disease (. 70% stenosis of one ormore coronary arteries) and were exercised within 2weeks of coronary arteriography.

High-quality resting echocardiograms in whichboth sides of the septum and posterior left ventricularwall were seen continuously throughout the cardiaccycle were obtained on 54 persons who, after givinginformed consent, exercised according to the protocolbelow. Echocardiograms of similar quality during ex-ercise were obtained on 24 of these (eight normalvolunteers, three persons with chest pain of uncer-tain etiology and normal coronary arteriograms andleft ventriculograms, and 13 patients with documentedcoronary artery disease) and form the basis for thisreport. The first four patients studied were consecu-tive patients with angiographically documented cor-onary artery disease who had high-quality restingechocardiograms. Two of these had a history ofmyocardial infarction. Subsequently, patients whowere chosen had no history of myocardial infarction,normal left ventriculograms and at least one coronaryartery with . 70% stenosis. Details of the study pop-ulation are given in table 1. The 13 patients with cor-

onary artery disease had a mean age of 51 ± 8 yearsand exercised until typical angina or ischemic ECGchanges appeared. Ischemic ECG changes weredefined at . 1 mm planar ST depression for at least80 msec. Because of the severity of the coronary dis-ease in the patient population and propranolol therapyin four, the mean maximal heart rate was 96 ± 16beats/min. The average duration of exercise in thisgroup was 5.3 minutes (range 3-8 min). The 11 nor-mal volunteers had a mean age of 53 ± 9 years and ex-ercised to exhaustion. Values of echocardiographic in-dices in the normals were selected at pre-exhaustionheart rates to match the maximal rates achieved bythe patients with coronary artery disease, with a meanrate of 101 ± 13 beats/min. The average duration ofexercise until exhaustion in the normals was 8.2minutes (range 6-12 min); the average duration of ex-ercise at which pre-exhaustion measurements wereselected was 3.5 minutes (range 3-7 min).

Exercise EchocardiographyPatients and normal volunteers exercised on a

mechanically-braked bicycle ergometer (Monark,model 850) while lying on a foam rubber wedge in a200 semi-supine or slightly left-lateral decubitus posi-

5 1


Dow

nloaded from


VOL 59, No 1, JANUARY 1979

tion. A graded exercise protocol was followed withpatients pedalling at 18 km/hr at a work load of 300kpm/min for the first 3 minutes. Thereafter the workload was increased by 300 kpm/min every 3 minutes,with speed held constant. Twelve-lead electrocardio-grams (with V2 and/or V3 in the third left intercostalspace so as to allow room for the echocardiographictransducer) and M-mode echocardiograms (Smith-Kline Ekoline 20A ultrasonoscope with a 2.25 MHzinternally focused transducer interfaced with a

Honeywell-1856 Visicorder strip chart recorder set ata paper speed of 100 mm/sec) were recorded at rest,during every minute of exercise, and every minute for5 minutes thereafter.The echocardiographic transducer was hand-held,

without the need for special equipment, in the stan-dard left parasternal location in the fourth intercostalspace and aimed along the left ventricular minor axisjust inferior to the mitral valve leaflets.

The right and left sides of the interventricular sep-tum and the endocardial and epicardial surfaces of theposterior left ventricular wall were digitized bymanual tracking on a magnetic digitizing board (Digi-grid, Computer Equipment Corp) throughout severalcardiac cycles at rest and during every minute of exer-

cise and the post-exercise recovery period. Thethickness of both walls was measured from leadingedge to leading edge in the usual manner. The per-son performing digitization was blinded to the resultsof coronary angiography. Indices of regional functionwere calculated by computer from the digitizeddata according to the following formulas:

% systolic wall thickening (%zAT) -

end-systolic thickness - end-diastolic thicknessend-diastolic thickness

% systolic change of diameter (%AD) =

end-diastolic diameter - end-systolic diameterend-diastolic diameter

End-diastolic measurements were taken at the onset ofthe QRS complex; end-systolic measurements weremade at the greatest wall thickness or smallest ventric-ular diameter. The maximum velocities of wallthickening in systole (VT), of wall thinning in diastole(VTh), of decrease of diameter in systole (max dD/dt)and of increase of diameter in diastole (max - dD/dt)were calculated using a 5-point numerical differentia-tion equation.8 The derivative was computed for eachdigitized point (one every 2.5 msec), after which thederivatives were smoothed by taking an 11-pointaverage.

Data were tested for statistical significance by thepaired or unpaired t test, as appropriate, and 2 X 2 chisquare analysis with Yates' continuity correction.9 Weexamined the reproducibility of our digitizing tech-nique by analysis of variance between eight repeatdigitizations of the same in-exercise cardiac cycle by a

single observer and between eight independent

digitizations of the same cycle by each of twoobservers (table 2). There were no significantdifferences between independent digitizations by thesame observer or between observers with the exceptionof a small (5%) difference in the means of %ADbetween the two observers. These data are comparableto those of Gibson and co-workers,10, 11 who demon-strated the reproducibility of digitizing resting M-mode echocardiograms and the validity of suchmeasurements in reference to digitized cine-ventriculograms.

ResultsWall Thickening

One patient had single vessel coronary artery dis-ease with near-total obstruction of the proximal leftanterior descending coronary artery (table 1, patient1). Examination of sample beats from his exerciseechocardiogram allows an internal comparisonbetween an abnormally perfused segment and a nor-mally perfused one in the same patient at the sametime (fig. 1). At rest, the systolic thickening of the sep-tum and posterior wall are equal. During exercise,after the onset of ischemia, however, septal systolicthickening decreases, while posterior wall thickeningincreases. Mean percent change in ventriculardiameter (%AD) for seven cardiac cycles at rest was42% and fell slightly to 38% for three cycles duringischemia.

Figure 2 shows sample beats at rest and exercise ina normal person at similar heart rates to the beatsshown in figure 1. At rest, the %AT of the septum ancposterior wall are also equal. During exercise systolicthickening of both walls increases equally. Systolicthickening is the increase in wall thickness from end-diastole, expressed as a percent of end-diastolicthickness, and is an index of segmental function.Figure 3 shows the mean %AT ifi systole of the septumand posterior wall for several beats at rest, in early ex-ercise, and after the development of ischemia inpatient 1. The %AT of both segments is similar at restand increases early in exercise, from 76 ± 6 to83 ± 4% (mean ± SEM) in the septum and from71 ± 5 to 101i± 9% in the posterior wall. Whenischemia develops, however, septal thickening falls to38 ± 12%, while the posterior wall maintains the rate-related rise in segmental thickening, to 102 ± 10%.The values at rest and in exercise for percent of

regional wall thickening in all the patients comparedwith the normal volunteers are shown in figure 4.Panels A and B are a comparison of segmentalthickening of the septum (%ATs) and posterior wall(%ATp) in patients with . 70% stenosis of the arterysupplying that wall, with the same indices in normals.The septum is supplied by the left anterior descendingcoronary artery, which was . 70% stenosed proximalto the first septal perforator in all 13 patients. Theposterior wall is supplied by the posterior descendingcoronary artery, a branch of the right coronary arteryin a right-dominant circulation and of the left cir-

52 CIRCULATION


Dow

nloaded from



TABLE 2. Reprodlucibiltty of Digitizing Method: Analysis of Variance

Variance

WithinBetween observers observers

Index Observer* X SEM cd p p

%AT8 1 0.81 0.02 4 NS NS2 0.84 0.03

%ATp 1 1.04 0.07 6 NS NS2 1.10 0.09

V Th, (cm/see) 1 3.02 0.11 3 NS NS2 3.12 0.13

V Thp (cm/sec) 1 5.45 0.27 8 NS NS2 5.01 0.33

%AD 1 36.9 0.5 5 <0.005 NS2 38.7 0.2

Max dD/dt (cm/see) 1 14.2 0.4 6 NS NS2 15.1 0.7

Max -dD/dt (cm/sec) 1 22.0 0.9 8 NS NS2 23.9 0.9

*Observers 1 and 2 each independently digitized the same echocardiogram of a single in-exercise cycleeight times for comparison betweeni anid withini observers.

Abbreviations: S, ATS, %,AT,, =percent change in thickness of septum and posterior wall; V Th,, V Thpmaximum velocity of thinniing of septum and posterior wall; '%AI) percent chan-ge in diameter; Max

dD/dt = maximum velocity of decrease in diameter, systole; Max -d)/dt = maximum velocity of increasein diameter, diastole; X meanl of eight observations by each observer; %d = percent difference betweenmeans of the two observers; NS = not significaint.

cumflex coronary artery in a left-dominant circula-tion. The proximal posterior descending, or the vesselgiving rise to it, was stenosed > 70% in nine of our 13patients. In the remaining four patients the posteriorwalls were supplied by arteries with < 30% stenosis.While it is possible that some segments perfused by anartery with a significant (> 70%) stenosis receivedadequate collateral flow from another artery during

EKGCHESTWALL

exercise, we considered all segments perfused by sucha stenosed artery to be at risk for inadequate perfusionduring exercise-induced ischemia.The mean values for %oATs and %,ATp at rest were

similar in patients and normals. In exercise the mean%ATs of the normals rose significantly, from 56 ± 3to 77 ± 7% (p < 0.01) while the mean for patientswith coronary disease fell from 59 + 6 to 35 ± 3%

-___

RV 0.9 120 ~~ 1.6 0- Lc,

SEPTUM

LV_~~~~~ v.6~

POSTERIOR ~WALL ;-=-

nA .,.%c:-P

LX U.. trr>S..: it

REST(HR=75) EXERCISE- ISCHEMIA (HR=C1O)FIGURE 1. Echocardiograms at rest (left) andpeak exercise (right)from a patient with near total occlusionofleft anterior descending coronary artery and normal left circumflex and right coronary arteries. The scaleandpaper speed (100 mm/sec) are the samefor both. Both septal andposterior wall thickening are the sameat rest. In exercise, at the time of ischemia, posterior wall systolic thickening has increased while septalthickening falls.

53


Dow

nloaded from



CHEST WALL__ __

RVr +=.~~~~p

SEPTtUM -L =§?t~ -

LV 16 tQ-::AnG ,cm

POSTERIOR X

Qn~TFInPA t M-~t7 >w ;>- -t| {,f> '5^vSWALL__

f -i

REST (HR60)

4-- N.,.~ es_. :,,

m 0,9 *½< / A.<w-x

Fof1a;:: C-.'-"*_T--'K"W

s'@Sctt.7AS

EXRCS (HR-JIG) S

FIGURE 2. Echocardiograms at rest (left) and exercise (right) in a normal person. The scale and paperspeed (100 mm/sec) are the samefor both. Systolic thickening ofboth walls is the same at rest and increasesin exercise.

(p < 0.005). Only one of the 13 patients' valuesoverlapped the range of values for the normals, andthe mean value for patients in exercise wassignificantly lower than the mean for normals in exer-cise (p < 0.001). Similarly, in exercise, normalsshowed a significant rise in mean %ATp from 89 ± 9to 115l1 8% (p < 0.005), while the mean for patientswith coronary disease fell from 75 ± 9 to 54 ± 9%(p < 0.01), a value significantly below that for nor-mals in exercise (p < 0.001).The regional wall thickening data were also

analyzed looking at the change in %AT from rest toexercise. The 11 normal volunteers had a total of 22wall segments with normal perfusion. The 13 patientswith coronary disease had a total of 22 segments withabnormal perfusion: 13 septa and nine posterior leftventricular walls were supplied by a coronary arterywith > 70% stenosis. Figure 5 shows the change in%AT from rest to exercise. Twenty-one of the 22segments in normal volunteers thickened the same ormore at exercise than at rest. In contrast, %AT fellfrom rest to exercise in 19 of 22 segments in thepatients with coronary artery disease. By chi squareanalysis these differences were highly significant(x2 - 26.5, p < 0.001). This suggests that, in ourpatients, a fall in %AT in a given segment predictssignificant stenosis of the coronary artery perfusingthat segment.

In order to determine if changes in wall thickeningoccurred before clinical signs of ischemia in thepatients with coronary artery disease, we comparedthe values for %AT at rest with %AT during theminute before the onset of angina or ischemic ECGchanges (pre-ischemic %AT). Both %ATs and %ATpfell insignificantly, from 59 ± 6 to 47 ± 5% in the sep-

tum and from 75 1 9 to 64 ± 9% in the posterior wall.The maximum velocities of thickening of the sep-

tum (VTs) and posterior wall (VTp) increasedsignificantly from rest to exercise in normals (table 3).The mean VTs for patients fell significantly in exercise

L-zw

w-J(-z(0)

zzwyCL)I

-JI4zwC,

REST EXERCISE EXERCISE-ISCHEMIA

(-Al (3)

10OL

801

60F

40[

20%

iz/,

(3)>

(7)

A NO. BEATS

E POSTERIOR WALL (±SEM)

A SEPTUM (tSEM)

75 105HEART RATE

A

(3)

120

FIGURE 3. Mean percent segmental thickening for severalbeats at rest, in early exercise, and at peak exercise with theonset of ischemia in the same patient asfigure 1. Segmentalthickening increases in both septum and posterior wall earlyin exercise, but with ischemia in the septum, regionalthickening falls markedly. Values represent mean + SEM.

f -1

54 CIRCULATION

120rtil


Dow

nloaded from



A

0 NORMALS

* CAD (LAD)(I I)

t a p<O.OI vs. REST

P<0.005 vs. RESTt p<o.oOI vs. NORMALS

(13) IN EXERCISE

EXERCISE

(11)

* NORMALS

A CAD(PD)

a P<0.X5 vs. REST

pP<O.OI vs. RESTt P <O.OOI vs. NORMALS

IN EXERCISE

(9)

REST EXERCISE

FIGURE 4. Mean percent regional systolic wall thickeningat rest and exercise in all the patients with coronary arterydisease (CAD) compared with the normal volunteers. PanelA, % septal thickening; panel B, % posterior wall thicken-ing. All values are mean + SEM. A highly significantdifference between normals and patients with coronaryartery disease developed in exercise for both parameterswhere none existed at rest. LAD = left anterior descending;PD = posterior descending.

to a value significantly lower than the normals in exer-cise, while the mean VTp for patients did not changesignificantly.

(U)

LIxLi

"

+60

zZ +40

-J

F- -20

4

z

)

z

0 -60

ENORMALS-IISEPTA,IIPOSTERIOR LV WALLSECHO SEGMENT (CAD- 13 SEPTA. 9 POSTERIOR LV WALLS

II.

I.

IL

0

000

000

000co

0

Va

AL.

> 0

<0

LA

NORMALS (o) CAD('

2 1 5

1 1 9

2

X = 26.5 p < 0.0001FIGURE 5. Change in percent segmental thickening fromrest to exercise in all segments from normal volunteers(circles) and all segments from patients with coronary arterydisease (CAD, triangles) that were perfused by an arterywith . 70% stenosis. A fallfrom rest to exercise in this pop-ulation was sensitive for (19 of 22) and predictive of (I 9 of20) the presence of significant coronary stenosis.

sion to blood flow is relieved in experimental models,13we examined a parameter which may, in part, reflectthe rate of relaxation, the VTh. Panels A and B offigure 6 show mean rest and exercise values for themaximum velocity of septal and posterior wall thin-ning, VThS and VThp, respectively. The mean valuesat rest were similar for normals and patients. In exer-cise VThS increased from 5.5 ± 0.3 to 7.7 + 0.6cm/sec in the normal subjects, (p < 0.005) but fell inthe ischemic segments from 5.9 ± 0.5 to 4.3 + 0.4cm/sec (p < 0.05; p < 0.001 vs normals in exercise).Similarly, during exercise VThp increased in normalsfrom 8.4 ± 0.8 to 11.8 ± 1.2 cm/sec (p < 0.001) butfell in the patients from 8.3 ± 1.2 to 6.2 ± 1.4 cm/sec(p < 0.005 vs normals in exercise). This indicates thatslower diastolic thinning, as well as decreased systolicthickening, occurs with ischemia.We also compared rest with pre-ischemic VTh, and

VThp, but found no significant fall in either before theonset of clinical ischemia. At rest VTh8 and FThp were5.9 ± 0.5 and 8.3 + 1.2 cm/sec, respectively, and inexercise before ischemia were 5.8 ± 0.4 and 7.2 + 1.0cm/sec, respectively.

Wall Thinning

Because abnormal relaxation may occur withoutabnormalities of systolic function in the presence ofischemial2 and persist for many minutes after occlu-

Cavity Dimension

Figure 7 shows the changes from rest to exercise forthe %AD. In the normals there was a significant rise in

lOOr

80

(9zz

uJ7-)I

Q-

LiJ(U) 401

REST

1251B

100zLLi0y

UI

-J

0

1/)0CL

75

501

55

)r

)r-%L

N,


Dow

nloaded from


CIRCULATION

TABLE 3. Velocity of Change in Thickness and Dimension During Rest and Exercise

Normals PatientsIndex Rest Exercise p Rest Exercise p

VT, (cm/sec) 5.2 - 0.5 6.6 - 0.8 <0.05 4.8 - 0.3* 3.9 - 0.3t <0.05

VT, (cm/sec) 5.3 - 0.6 7.1 - 1.4 <0.05 4.8 i 0.4 4.1 - 0.5 NSMax dD/dt (cm/sec) 10.2 i 0.7 13.0 = 1.4 <0.01 10.2 - 0.7 9.8 - 0.71 NSMax -dD/dt (cm/sec) 14.9 - 1.1 20.2 - 2.0 <0.05 13.8 1.6 14.4 - 1.0$ NS

*No significant difference between patients and normals in any resting values.tP <0.0005 vs normals in exercise.$p <0.05 vs normals in exercise.Values represent mean- SEM for 11 normals and 13 patients with the exception of VT,, which is based on

the nine patients who had > 70/ stenosis of the artery supplying the posterior wall.Abbreviations: VT, = maximum velocity of septal thickening in systole; VTp = maximum velocity of

posterior wall thickening in systole; Max dD/dt = maximum velocity of decrease of diameter in systole;Max -dD/dt = maximum velocity of increase of diameter in diastole.

mean %AD from 38 ± 2 to 44 ± 2% (p < 0.001), and10 of 11 individual volunteers showed increases. Incontrast, 12 of 13 patients with coronary diseaseshowed a fall in %AD at peak exercise, and only oneexercise value in the ischemic group (patient 1, table 1,with single vessel disease) overlapped the normalrange. The mean value for the coronary disease groupdid not differ significantly from the normals at rest butin exercise fell from 35 ± 2 to 28 ± 2% and wassignificantly below the value for the patients at rest(p < 0.001) and the normals in exercise (p < 0.0005).The fall of %AD in exercise and the value at peak ex-

ercise thus appear sensitive in detecting ischemia inthe presence of significant disease of the coronaryarteries perfusing the septum and posterior left ven-tricular wall. During the minute before the onset ofangina or ischemic ECG changes there was also asignificant fall in %AD: mean %AD fell from 35 ± 2%at rest to 31 ± 2% in this period before clinical signsof ischemia (p < 0.01).Max dD/dt, increased from rest to exercise in nor-

mals, but not in patients (table 3). The same table alsoshows that max - dD/dt increased significantly fromrest to exercise in normals without changingsignificantly in patients. The pre-ischemic value ofmax - dD/dt, 14.5 1.4 cm/sec, did not differsignificantly from the resting or peak exercise values.However, the mean value of both parameters at peakexercise was significantly lower for patients than fornormals.

Discussion

Tennant and Wiggers'4 first recorded the rapid dis-appearance of systolic muscle shortening followed by

systolic expansion that occurs after coronary occlu-sion in the dog. Other workers have described theregional dysfunction associated with ischemia inanimal models by use of cinematography,'- straingauges, , 16 epicardial dimension gauges,17 18

zz

z

I

-C

I-

(a)

0

cr

x

10

8

z6

N4

E

-4

2

z2

I

3:-lJcr

0

w

a-en 4)0

UE

LL Q

w

ir

D

x

FIGURE 6. Maximum velocity of wall thinning in diastoleat rest and exercise for patients with coronary artery disease(CAD) as compared to normals. Panel A, velocity ofseptalthinning; panel B, velocity of posterior wall thinning. Allvalues are mean for group ± SEM. A highly significantdifference between normals and patients developed in exer-cise for both parameters where none existed at rest. LAD =left anterior descending; PD = posterior descending.

141-

A

t

((3)

REST

* NORMALSA CAD (LAD)

* p<0.005 vs. RESTp<o0.05 vs. REST

t p<O.OOI vs. NORMALSIN EXERCISE

EXERCISE

B

12F

1O-

8

6

(11)

(9)

4j_,1,

* NORMALS

A CAD (PD)

* p<O.OOI vs. REST

t p<0.005 vs. NORMALSIN EXERCISE

REIEECS

56 VOL 59, No 1, JANUARY 1979

EXERCISEREST


Dow

nloaded from



NORMALS (I I)

6oo-

50ha0-

c,w

w

IL

zI)

40

30

CAD (13)

P <O.OOI vs. REST

tP< O.OOI vs. RESTP<O.OOI vs. NORMALS

IN EXERCISE

20[

01-

REST EXERCISE REST EXERCISEFIGURE 7. The effect of exercise on systolic percent

change in ventricular diameter (%AD) in normals andpatients with coronary artery disease (CAD). The linesreflect changes in individuals and the large circles andtriangles, the mean ± SEM. One CAD value for %ADoverlaps the normal range.

cineradiography of radiopaque markers,'9 20 reflectedultrasound,2' and implanted ultrasonic crystals.2 3 22

These studies indicate that the deterioration of con-tractile function of the ventricle appears withinseconds of coronary occlusion, before ST segmentshifts.' Changes in systolic myocardial thickeningparallel deterioration in epicardial shortening.23 Thesemechanical alterations are rapidly reversible uponrelease of the stenosis, before ST segments return tobaseline, although relaxation abnormalities may per-sist for almost an hour after a brief occlusion.'3

In humans, the resting echocardiogram in patientswithout previous infarction is a poor predictor of thepresence of coronary disease.22' 24, 25 Echocardio-grams have been recorded during angina produced byhandgrip stress26-29 and have demonstrated reducedwall motion amplitude, ejection fraction, and velocityof circumferential fiber shortening in some cases. Un-fortunately, these abnormalities have only been seenin some patients in each group and all of them havehad high-grade proximal disease.28' 29 Furthermore,handgrip is frequently an inadequate stress forproducing ischemia.30 Fogelman et al.'2 recordedposterior wall endocardial echograms during handgripor supine bicycle ergometer exercise in patients withischemic heart disease. The patients he examined hadslightly but significantly lower mean and maximalvelocities of diastolic endocardial motion at rest thannormal volunteers, but during exercise-induced anginathese parameters fell markedly. The effect of exerciseon wall thickening was not examined.

Cineventriculograms and gated blood pool scanshave been performed during exercise in patients withischemic heart disease. By exercise cineventriculog-

raphy Sharma et al.3 demonstrated that the develop-ment of new areas of dyskinesis not present at rest cor-related well with the location of coronary stenoses.Borer and colleagues,5 using radionuclide stress ven-triculography, also demonstrated new areas ofregional wall motion abnormality during exercise insegments perfused by an artery with > 50% stenosis.These workers also documented a fall in ejection frac-tion in most patients during exercise, as in the presentstudy. Both regional and global cardiac functiondeteriorated in some patients in the absence of symp-toms.

In this study we have examined changes in regionalwall thickening and thinning and left ventricularcavity dimension during exercise using M-modeechocardiography. In our patients, a fall in segmentalthickening during exercise appears to be a sensitiveand predictive indicator of significant coronary arterydisease (fig. 5), as %AT fell during exercise in only oneof 22 normally perfused segments (5%), while thisparameter fell in 19 of 22 abnormally perfusedsegments (86%). In a patient with one normally andanother abnormally perfused wall (fig. 3), %AT roseinitially in both segments during low levels of exercise;at the onset of ischemia, percent thickening fell only inthe wall segment perfused by a stenosed artery, in-dicating that in the one patient examined with singlevessel disease these changes correlated with coronaryanatomy. Closer correlation with coronary anatomymust await further studies on patients with less severeor diffuse coronary disease.

Diastolic velocity of wall thinning falls in ischemicsegments (fig. 6). This is consistent with earlier data ofFogelman et al.,'2 showing a decrease in diastolicendocardial velocity during exercise-induced angina.Our patients' resting values did not differ from thenormal controls. This may be because we used age-matched controls, while Fogelman's controls were ayoung population. His patients may also have hadmore severe disease, as one had resting angina, andthe mean heart rate achieved in exercise was lower forhis patient population. The finding of slower diastolicthinning during ischemia may reflect delayed relaxa-tion which has been described previously in man.3'We found that a fall in %AD to less than 37% at

peak exercise was a useful indicator of the presence ofmyocardial ischemia (fig. 5). The systolic change ofleft ventricular dimension measured echocardio-graphically will decrease if the systolic thickening ofeither septum or posterior wall decreases. Theseregions are usually perfused by the left anterior de-scending and right coronary arteries, respectively (orin a left-dominant circulation, by the circumflex).Thus, during ischemia, changes in the performance ofthese segments may result in deterioration of %AD.We also found that %AD fell significantly during exer-cise before the onset of ischemia recognizable byangina or electrocardiographic ST segment changes.This change was the earliest detectable sign ofischemia. None of the other parameters we measuredshowed a significant fall before the onset of angina orischemic ECG changes. Nine of our 13 patients had

57


Dow

nloaded from



> 70% stenosis of the arteries supplying both the sep-tum and posterior wall. The pre-ischemic fall in %ADmay be a summation of the nonsignificant pre-ischemic decreases in %ATs and %zATp or an earlyechocardiographic index of regional ischemia. Sincecoronary disease frequently involves the left ventricleasymmetrically, resting or exercise-induced changes inwall motion in one or two areas may not reflect overallventricular performance. The echocardiogramtranssects the left ventricle at two points along a singlechord. Changes in %AD as well as Max dD/dt andMax - dD/dt measured echocardiographically duringischemia reflect changes in regional performance ofthe septum and posterolateral wall; such changes mayfalsely overestimate or underestimate changes inglobal ventricular performance because the antero-lateral and apical regions are not examined.A recent study32 has shown that Max - dD/dt in-

creases early in exercise and at low heart rates inyoung normals. Our data agree with this finding inolder normals but the patients with coronary arterydisease showed no significant change in this param-eter at similar heart rates. This lack of rise may alsobe an early sign of ischemia.Most exercise protocols use upright exercise on a

treadmill or bicycle. Patients with heart disease,however, reach higher heart rates and systolic bloodpressures for a given level of oxygen consumption inthe supine rather than the upright position.33 Further-more, heart volume is larger in the supine positionthan upright.34 Consequently, for inducing angina,supine exercise may be preferable because greaterheart rate, blood pressure and cavity size (and there-fore wall stress) produce a higher oxygen demand.From a practical point of view, moreover, reproduci-ble echocardiograms during exercise in a sitting posi-tion are very difficult to record, and we have had muchbetter success in obtaining data during supine exercise.

Four of the coronary artery disease patients andtwo normals were taking propranolol at doses up to320 mg/day at the time of exercise echocardiog-raphy. Although this drug causes resting bradycardiaand may decrease resting ventricular function, it doesnot prevent exercise-induced increases in ventricularfunction in normals or exercise-induced deteriorationof ventricular function or regional wall motion inpatients with ischemic heart disease.3' 3 All patientsinvestigated on whom exercise echocardiograms wereobtained developed ischemia with exercise, includingthe four on propranolol.The greatest limitation of this technique is the

difficulty in obtaining high quality echocardiogramsduring exercise. Of the 54 persons with good echo-cardiograms of both sides of the septum and posteriorleft ventricular wall at rest, echocardiograms ofsimilar quality could be obtained during exercise in 24,or 44%. This compares favorably with other studieswhich have sought to record echoes of the septum andposterior wall continually throughout systole anddiastole.37

There are several advantages of exercise echocar-diography in the evaluation of patients with ischemicheart disease. Echocardiography is the only techniquewidely available for measuring dynamic changes inwall thickness in different regions of the heart and islikely more independent of motion artifact thantechniques which examine endocardial movement.Resting evaluation of regional endocardial motioncorrelates poorly with histological examination of thesame regions.6' 7 Therefore, measurement of regionalthickening, thinning, and motion during exercise givesparameters of regional functional ability under con-ditions of high oxygen demand likely to precipitateischemia. In this preliminary study an exercise-induced fall in thickening or minor axis shorteningseemed sensitive for and predictive of ischemiain patients with high-grade proximal lesions. Thistechnique may help assess the relative functionalseverity of different coronary stenoses in order to plansurgery or may reveal to the angiographer the func-tional significance of a lesion which may not haveappeared severe. Further evidence of these applica-tions must await studies on a larger group of patientswith less severe coronary stenoses. In the presence ofresting regional hypokinesis, echocardiography duringexercise might detect improvement in regional func-tion and localize a potentially viable wall segmentsimilar to the effect of intervention-ventriculog-raphy.38 After coronary artery bypass surgery, ex-ercise echocardiography may, as have radionuclidetechniques,39 demonstrate improved performance dur-ing exercise of segments successfully reperfused. Inthe presence of disorders that make interpretation ofthe standard exercise test difficult, such as left bundlebranch block, digitalis therapy, left ventricular hyper-trophy, or ventricular pacing, echocardiography dur-ing exercise may be the only way of evaluating percentchange in wall thickening, which should be unaffectedby abnormal patterns of electrical depolarization ofrepolarization.Compared with exercise cineangiography, echocar-

diography has the advantage of being noninvasive,easily repeatable, and inexpensive; it looks at a largenumber of cardiac cycles before, during, and afterischemia, and does not alter the function of that whichit seeks to measure. Compared with stress radio-nuclide ventriculography, the technique reported herefollows instantaneous events in each cardiac cyclerather than combining several cycles, imposes noradiation exposure, and evaluates changes in thicken-ing as well as endocardial motion that gated bloodpool scans do not assess. The yield of this type of in-vestigation will be higher and the application morewidespread when technical advances provide betterresolution of echocardiographic signals and whenother techniques become available for measuringregional thickening in man. With the extension of thismethodology to two-dimensional echocardiography,most wall segments of the left ventricle will be accessi-ble for similar analysis.

58 CIRCULATION


Dow

nloaded from



Acknowledgments

We acknowledge the assistance of Mrs. Sue Livengood and Dr.Charles G. K. Pawsey in the performance of these studies and MissSpring Forsythe in the preparation of the manuscript.

References1. Goldstein S, DeJong JW: Changes in left ventricular wall

dimensions during regional myocardial ischemia. Am J Cardiol34: 56, 1974

2. Theroux P, Franklin D, Ross J Jr, Kemper WS: Regionalmyocardial function during acute coronary occlusion and itsmodification by pharmacologic agents in the dog. Circ Res 35:896, 1974

3. Sharma B, Taylor SH: Localization of left ventricularischaemia in angina pectoris by cineangiography during exer-cise. Br Heart J 37: 963, 1975

4. Sharma B, Goodwin JF, Raphael MJ, Steiner RE, RainbowRG, Taylor SH: Left ventricular angiography on exercise: anew method of assessing left ventricular function in ischaemicheart disease. Br Heart J 38: 59, 1976

5. Borer JS, Bacharach SL, Green MV, Kent KM, Epstein SE,Johnston GS: Real-time radionuclide cineangiography in thenon-invasive evaluation of global and regional left ventricularfunction at rest and during exercise in patients with coronaryartery disease. N EngI J Med 296: 839, 1977

6. Bodenheimer MM, Banka VS, Hermann GA, Trout RG,Pasdar H, Helfant RW: Reversible asynergy. Histopathologicand electrographic correlations in patients with coronary arterydisease. Circulation 53: 792, 1976

7. Hutchins GM, Bulkley BH, Ridolfi RL, Griffith LSC, Lohr FT,Piasio MA. Correlation of coronary arteriograms and left ven-triculograms with postmortem studies. Circulation 56: 32, 1977

8. Kelly LG: Handbook of Numerical Methods and Applications.Reading, Massachusetts, Addison Wesley, 1967, p 51

9. Dixon WJ, Massey FJ Jr: Introduction to Statistical Analysis,3rd Edition. New York, McGraw-Hill, 1969, chs 8, 13

10. Gibson DG, Brown D: Measurement of instantaneous left ven-tricular dimensional filling rate in man, using echocardiog-raphy. Br Heart J 35: 1141, 1973

11. Gibson DG, Brown DJ: Measurement of peak rates of left ven-tricular wall movement in man: Comparison of echocardiog-raphy with angiography. Br Heart J 37: 677, 1975

12. Fogelman AM, Abbasi AS, Pearce ML, Kattus AA: Echocar-diographic study of the abnormal motion of the posterior leftventricular wall during angina pectoris. Circulation 46: 905,1972

13. Theroux P, Ross J Jr, Franklin D, Kemper WS, Sasayama S:Regional myocardial function in the conscious dog during acutecoronary occlusion and responses to morphine, propranolol,nitroglycerin, and lidocaine. Circulation 53: 302, 1976

14. Tennant R, Wiggers CJ: The effect of coronary occlusion onmyocardial contraction. Am J Physiol 112: 351, 1935

15. Prinzmetal M, Schwartz LL, Corday E, Spritzler R, BergmanHC, Druger HE: Studies on the coronary circulation. VI. Lossof myocardial contractility after coronary artery occlusion.Ann Intern Med 31: 429, 1949

16. Tatooles CJ, Randall WC: Local ventricular bulging after acutecoronary occlusion. Am J Physiol 201: 451, 1961

17. Puri PS, Bing RJ: Effect of drugs on myocardial contractility inthe intact dog and in experimental myocardial infarction. Am JCardiol 21: 886, 1968

18. Hood WB Jr, Covelli VH, Abelmann WH, Norman JC: Per-sistence of contractile behavior in acutely ischemic myocar-dium. Cardiovasc Res 3: 249, 1969

19. Hekkila J, Tabakin BS, Hugenholtz PG: Quantification offunction in normal and infarcted regions of the left ventricle.

Cardiovasc Res 6: 516, 197220. Maroko PR, Libby P, Ginks WR, Bloor CM, Shell WE, Sobel

BE, Ross J Jr: Coronary artery reperfusion 1: Early effects onlocal myocardial function and the extent of myocardialnecrosis. J Clin Invest 51: 2710, 1972

21. Kerber RE, Abboud FM: Echocardiographic detection ofregional myocardial infarction. An experimental study. Cir-culation 47: 997, 1973

22. Ross J Jr, Franklin D: Analysis of regional myocardial func-tion, dimensions, and wall thickness in the characterization ofmyocardial ischemia and infarction. Circulation 53 (suppl 1): I-88, 1976

23. Sasayama S, Tomoike H, Crozatier B, McKown D, KemperWS, Franklin D, Ross J Jr: Wall thickness and endocardial seg-ment dynamics during chronic myocardial infarction in con-scious dogs. (abstr) Am J Cardiol 37: 169, 1976

24. Jacobs JJ, Feigenbaum H, Corya BC, Phillips JF: Detection ofleft ventricular asynergy by echocardiography. Circulation 48:263, 1973

25. Dortmer AC, DeJoseph RL, Schiroff RA, Zelis R: Echocar-diographic predictors of coronary artery disease. (abstr) ClinRes 24: 215A, 1976

26. Widlansky S, McHenry PL, Corya BC, Phillips JF: Coronaryangiographic, echocardiographic, and electrocardiographicstudies on a patient with variant angina due to coronary arteryspasm. Am Heart J 90: 631, 1975

27. Corya BC: Applications of echocardiography in acute myocar-dial infarction. Cardiovasc Clin 11: 113, 1975

28. Ahmad M, Watson LE: Application of stress echocardiog-raphy in diagnosis of left anterior descending coronary arterydisease. (abstr) Clin Res 24: 515A, 1976

29. Klein W, Pavek P: Belastungsechokardiographie bei koronaresHerzkrankheit (Exercise echocardiography in coronary heartdisease). Z Kardiol 66: 112, 1977

30. Lowe DK, Rothbaum DA, McHenry PL, Corya BC, KnoebelSB: Myocardial blood flow response to isometric (handgrip)and treadmill exercise in coronary artery disease. Circulation51: 126, 1975

31. McLaurin LP, Rolett EL, Grossman W: Impaired left ventric-ular relaxation during pacing-induced ischemia. Am J Cardiol32: 751, 1973

32. Weiss JL, Weisfeldt ML, Garrison JB, Livengood SV, QueenJL, Fortuin NJ: Evidence of Frank-Starling effect in manduring maximal semisupine exercise. (abstr) Circulation 54(suppl II): 11-215, 1976

33. Epstein SE, Beiser GD, Stanpfer M, Braunwald E: Exercise inpatients with heart disease: effects of body position and typeand intensity of exercise. Am J Cardiol 23: 572, 1969

34. Holmgren A, Ovenfers CO: Heart volume at rest and duringmuscular work in the supine and in the sitting position. ActaMed Scand 167: 267, 1960

35. Borkowski H, Berger HJ, Langori RA, Cohen LS, GottschalkA, Zaret BL: Rapid assessment of left ventricular performanceduring exercise-induced myocardial ischemia: first pass radio-nuclide technique. (abstr) Circulation 56 (suppl lII): 111-198,1977

36. Ehrhardt JC, Verani MS, Marcers ML: Exercise isotope ven-triculogram: use in assessing changes in left ventricular func-tion. (abstr) Circulation 55 (suppl III) 11-141, 1977

37. Paraskos JA, Grossman W, Saltz S, Dalen JE, Dexter L: Anon-invasive technique for the determination of velocity of cir-cumferential fiber shortening in man. Circ Res 29: 610, 1971

38. Helfant RH, Bodenheimer MM, Banka VS: Asynergy in cor-onary heart disease: evolving clinical and pathophysiologic con-cepts. Ann Intern Med 87: 475, 1977

39. Kent KM, Borer JS, Green MV, Bacharach SL, McIntosh CL,Conkle DM, Epstein SE: Effects of coronary-artery bypass sur-gery on global and regional left ventricular function during ex-ercise. N Engl J Med 298: 1434, 1978

59


Dow

nloaded from


S J Mason, J L Weiss, M L Weisfeldt, J B Garrison and N J FortuinExercise echocardiography: detection of wall motion abnormalities during ischemia.

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1979 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation doi: 10.1161/01.CIR.59.1.50

1979;59:50-59Circulation.

http://circ.ahajournals.org/content/59/1/50.citationthe World Wide Web at:

The online version of this article, along with updated information and services, is located on

http://circ.ahajournals.org//subscriptions/

is online at: Circulation Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answer information about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. FurtherEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculationpublished in Requests for permissions to reproduce figures, tables, or portions of articles originallyPermissions:


Dow

nloaded from

http://circ.ahajournals.org/content/59/1/50.citation

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circ.ahajournals.org//subscriptions/


exercise echocardiography: detection wall motion...

Documents