JACC Vol. 24. No. 2 February 1994:369-76 369

From the San i .ancisco Heart Institute, Seton Medical Center, California.

All editorial decisions for this article, including selection of referees, were made bv a Guest Editor. This nolicv annlies to all articLs with autbors from the I_fni&ity~~f California, Sk F~an&co.

Manuscript received March 19, 1993; revised manuscript received Sep- tember 16, 1993, accepted September 30, 1993.

: Dr. Harvey S. Hecht, San Francisco Heart Institute, Seton Medical Center, 1900 Sullivan Avenue, Daly City, California, 94015.

01994 by the American College of Cardiology

determine the ~mci~emce of truly silent i

370 HECHT ET AL. TRULY SILENT ISCHEMIA

JACC Vol. 23, No. 2 Febmry 1994:369-76

stress echocardiography and were part of a consecutive series of patients referred for evaluation of possible coronary disease who underwent both procedures. Seven patients had left bundle branch block or baseline ST segment depression that precluded accurate analysis of further ST segment changes with exercise and were excluded from further analysis. The remainder were assigned to three study groups, depending on whether or not the stress test was accompanied by any chest pain (including clear angind equivalents, e.g., jaw and throat pain in two patients) OT ECG abnormalities. Group I (symptomatic ischemia) in- cluded 33 patients with exercise-induced chest pain and

II (asymptomatic isch- sitive stress ECG find-

t ischemia) included 56 patients induced chest pain nor positive stre

Seven patients had chest pain wit and were excluded from all of the small statistical sa with Groups I to III. All patients gave written informed consent,

Exert hy. The 1Zlead EC6 during exercise was considered positive if there was 21 mm of horizontal or downsloping ST segment depression for at least Q.O!3 s after the I point, compared with the tracing

described technique (4 tem with 3.75. and 2.5-

Toshiba MA u~t~so~~d sys- transducers was utilized, with

s were obtained in the five views during each stage. A d Cineview was used for four-quad~nt display of

es for simultaneous comparison of rest and studies. Eight frames/cardiac cycie, triggered

from the R wave at intervals of 67,5O or 33 ms, depending on heart rate, were displayed. Continuous three-lead ECG monitory was utilized with a 12-lead ECC during each

The five views were consolidated into the standard 16 segmentipatient model of the American Society of Echocar- diography (8) to avoid duplication of the same areas on multiple views. The anterior, ~~terolatera~ and anteroseptal segments were assigned to the left anterior descending coronary artery, the inferior and inferoseptal segments to the right coronary artery and the inferolateral segments to the left circu~ex coronary artery.

Each segment was scored independently by two observ- ers, on the basis of visual analysis, as follows: 1 = hyperki- nesia, 2 = normal, 3 = mild hypokinesia, 4 = moderate hypokinesia, S = severe hypokinesia, 6 = akinesia, 7 =

graphic findings.

obtained by either the ~~d~~~s or the

diameter ~dMctio~

are analyses were us

variance techniques were used to study the relation between ischemic myocardium scores and bistorica~ and dnical factors. Multivariate stepwise rno~tip~e regression was per- formed using the SPSS statistical package utilizing values of 0.05 for F to enter and 0.10 to remove variables from the regression equation. No regressions exceeded the maximal number of steps, which was 14, or twice the number of independent variables used in each stepwise regression. A level of 0.05 was considered statistically significant.

1 and 2). There were no significant diffe gender, incidence of diabetes, chest pain history and an- tianginal medication among the three gro Patients with truly silent ischemia (Group III) had a h r incidence of previous myocardial infarction compared with patients with asymptomatic ischemia (Group II). Coronary arteriographic variables were similar in the three groups, with no differ-

JACC Vol. 23, No. 2 February t99~369-76

B Cli

Group ! Group 88 Grolpp 111 (PB = 33) (m = 34) (n = 56)

58 -c PO 61 29 61 h I3

28 (85%) 31 (91%) 49 (84%)

Female 5 (15%) 3 (9%) 9 (86%)

Diabetes mellitus 3 (9%) 3 6%) 6(11%%)

Previous MI 8 (24%) 6 (18%)* 23 (41%)

Chest pain history 32 (99%) 27 (79%) 46 (82%)

24 (9393 27 (80%) 46 [82%7c)

24 (93%) 25 (74%) 42 (95%)

Nitrates 5 (15%) 8 (24%) 10 ( IFi3 Beta-blockers 4 (1X) 5 (15%) 9 (16%)

.O5 versus Group II!. Values presented are mean value + SD or

number (%). Group I = positive stress e~e~t~~c~~~~~~~rarn (ECZG) with exercise

chest pain; roup II = positive stress ECG without exercise chest

Group Ill = negative stress KG without exercise &es! fXhl;

fll&lCiIKM infarction.

were no di

atic ischemisl. exercised for a shorter duraG3n to a lower

54% for stress electsocar no differences in sensitivrty amon

le .%. Coronary ar~er~o~raQ~y

Group E Group II Group II1

Coronary artery disease

Single-vesseI 10 130%) I2 (35%) 26 (46%)

Double-vessel 11 (33%) 9 (29%) 19 (34%)

Triple-vessel 12 (39%) I3 (38%) I I (20%) LAD 23 (90%) 30 (88%) 42 (75%)

RCA 23 (70%) 19 (50%) 33 (59%)

LCX 22 (67%) 18 (53%) 25 (50%)

% diameter stenosis

All vessels 85.2 + 10 83.8 ? 13 83.5 It I2

LAD 88.0 f II 84.3 i: I4 85.8 f I5

RCA 83.8 f 11 76.8 c l-l 86.5 + 19

LCX 83.4 + I2 83.9 c 15 93.9 + 17

*p = NS for all comparisons. Unless otherwise indicated, values pre- sented are mean value 2 SD. LAD = left anterior descending coronary arterY;

LCx = left circumflex coronary artery; RCA = right coronary artery; other abbreviations as in Table I.

*p < 0.05 versus Groups II and HII. tp < O.OE versus Group II, and p < 0.05 versus Group III. $p < 0.000 Group III. Values presented are m pressure: HR = heart rate: SBP = systolic blood pressure; other definitions as

in Tabk 1.

le 4. Multivariate Stepwise Regression AaaQysis of Factors

Associated With lschemic Mvocardiurn

Multiple R F P Value AU’ Value Value

No. of abnormal segments

Positive stress ECG

Previous Ml

Chest pain history

Age Exercise chest pain

Gender

Diabetes

Average score/segment

Positive stress KG

Chest pain history

Age Exercise chest pain

Previous MI

Gender

Diabetes

Summed score

Positive stress ECG

Chest pain history

Age

0.41

0.47

0.52

0.55 -

0.17

0.06

0.05

0.03 -

- -

- -

24.1

8.5

9.5

5.4

I.1

0.4

0.1

0.001

0.004

0.009

0.023

0.315

0.550

0.933

0.38

0.45

0.48 -

-

0.15

0.05 0.03 -

- -

0.001

0.005

0.024

0.099

0.157

0.861

0.998

Previous MI

Exercise chest pain

Gender

Diabetes

0.40 0.16

0.47 0.06

0.51 0.04 - -

-

23.4

9.2

6.5

I.9

I.4

0.3

0.1

0.001

0.003

0.012

0.161

0.236

0.592

- 0.825

Abbreviations as in Table I.

372 HECHTETAL. TRULYSILENTISCHEMIA

JACC Vol. 23, No. 2 F&run9 19%:3@-46

Tahk 5. Comparison of Amount of Ischemic Myocardium

Abnormal Segments/R or V

All Pts No b41

Average ScorelSegment

All Pts

Pts Group I 4.6 + 3.8 5.1 + 3.5 1.1 c 0.8 17.3 z!z 9.4 17.2 f 9.2 Group II 4.5 2 4.6 5.2 + 4.4 I.0 + 0.9 1.0 f 0.6 15.2 f 8.9 5.7 t 8.6 Group 111 2.0 t 4.4* 2.0 t 4.2* 0.5 2 0.42 0.6 r 0.6 8.7 x9.1: 9.3 -L 9.1”

.AlI vessels Group I 3.6 2 1.9 3.9 c 1.9 I.1 t 0.6 1.1 = 0.6 7.6 t 6.2 7.9 I 6.2 Group II 3.7 + 1.8 4.3 d 1.7 I.0 2 0.5 1.0 9 0.6 7.9 + 6.0 8.3 !? 6.3 Group HI 1.7 f 1.7* 2.9 2 1.61 0.5 r 0.611 0.6 2 0.6 4.3 ? 4.7* 5.4 -c 5.05

LAD Group I 4.9 * 1.7 5.5 r 1.2 1.8 -c 0.8 1.8 r 0.6 14.0 + 6.1 14.8 -c 4. Group II 4.2 k 2.2 4.6 +s 2.2 1.4 + 0.7 1.4 2 0.S 10.9 + 5.7 Il.3 9 6.1 Group III 2.2 f 2.4S 2.8 1 2.2# 0.8 + 0.7” 0.9 + 0.78 6.4 2 5.7”” 7.3 c 5.66

RCA Group 1 2.3 z 2.1 2.5 ?L 2.1 9 9 0.6 1.0 + 0.6 4.4 2 3.2 5.1 ” 3.2 Group II 1.9 + 2.2 2.6 zz 2.18 1.0 r I.1 5.2 r 5.6 6.1 + 6.2 Group 111 0.Y I l.6W 1.8 2 I.5 0.5 + 0.78 2.6 r 3.49 3.4 + 4.2

LCX Group I I.8 * 1.2 1.6 + 1.3 1.3 4 u.9 I.2 d 0.9 3.9 + 2.6 3.6 1. 2.8

1.4 a 0.9 3.5 + 2.7 4.3 -c 2.7 4.1 2 3.4

definitions as in Tables I and 2.

significantly greater in Groups I and II compared with 111 for almost every cornpa~s~~ with the exceptio circumflex coronary artery. There was significa ischemic myocardium in patients exercisin

.___ ~. .~ Abnormal SegmentsPt or V

All Pts No Ml

- Sum of ScoresEt or V

All Pts No MI -

Pts Group I: 7.9 2 3.8 7.8 2 4.1 3.2 + 0.1 3.1 + 0.8 50.4 + 11.1 49.2 h 11.6 Group II 7.5 z!z 3.8 7.3 2 3.5 3.2 + 1.2 3.1 t 0.7 50.4 5 11.9 49.3 2 9.4 Group 111 6.2 2 4.3 5.1 2 4.2* 2.9 f 0.9 2.7 2 0.7 47.3 2 12.lP 42.9 2 9.9t

All vessels Group I 4.i + 2.1 4.1 f 2.3 3.2 + 0.7 3.1 r 0.7 19.4 2 9.7 19.2 t 9.8 Group II 4.4 ” 2.2 4.6 f 2.1 3.1 9 0.7 3.1 z 0.6 20.6 2 9.6 21.1 f 10.4 Group 111 3.3 + 1.9 3.5 + 1.9 2.9 t 0.8 2.7 2: 0.6 17.8 f 8.5 18.7 + 5.6

LAD

Group I 5.6 + I.: 5.8 t 1.0 3.9 + 0.7 3.9 + 0.7 31.1 + 5.3 31.2 4 5.2 Group If 5.0 + 1.8 4.9 f 1.9 3.6 2 0.8 3.5 2 0.7 28.5 ? 6.0 28.1 r 5.9 Group III 4.1 c l.9$ 3.8 t 2.1-l 2.4 2 0.80 3.4 + 0.88 27.0 + 6.89 25.? + 6.28

RCA

Group I 2.9 + 2.1 2.7 + 2.2 3.3 f 0.8 3.1 k 0.7 16.3 + 4.2 15.3 ?: 3.4 Group II 2.8 + 2.u 3.0 + 2.0 3.4 + 1.3 3.4 5 1.3 17.2 4 6.6 17.7 zt 6.5 Group 111 2.5 + 1.7 2.1 a I.5 3.4 f 1.1 3.9 2 0.9 16.7 r 5.3 !4.8 L 4.4

LCX

Group I 2.1 k 1.2 1.8 2 1.3 3.5 + 1.0 3.3 2 0.9 13.5 f 2.9 9.8 2 2.9 Group II 2.1 + 1.3 2.2 2 1.3 3.4 t 1.2 3.4 + 0.9 10.3 + 3.8 10.1 f 2.8 Group III 2.3 r I.2 2.0 + 1.3 3.9 2 1.1 3.6 + 1.2 11.6 t 3.2 10.9 f 3.5

*p < 0.01 versus Group 1 and p < 0.05 versus Group 11. tp < 0.01 versus Groups 1 and II. #p < 0.001 versus Group I and p C 0.05 versus Group 11. BP < 0.01 versus Group II. Values presented are mean value + SD. Abbreviations and definitions as in Tables 1,2 and 5.

JACC Vd. 23, No. 2 RbrUiphy P994:369-76

Avesage ScorelSegmena

0.2 1- 0.5 0.4 9 I.1 0.3 Itr I.1

0.1 ” 0.4 0.0 4 0.0

2.1 r 0.4

2.2 t 0.9

2.6 1: 0.9

2.2 + 0.4

2.2 r I.tJ

2.1 IO.2 17.6 -e 3.3

?.I 2 0.3 17.8 ix 4.2 2.3 2 0.6 20.4 + 7.4

2.1 ‘- 0.5

2.2 + 0.8

2.2 It 0.6

2.1 c 0.2 2.0 ” 0.2

Il.9 c 3.7

11.7 c 3.9

14.1 + 5.2

6.6 2 1.2 6.9 + 2.9

16.4 + 1.5 16.8 f 12.7 18.0 "- ,I.!3

Group 111 0.9 -t 1.7" 0.3 I 0.6 2.8 z!t 1.II 2.2 I 0.7 8.3 z 3.28 6.8 + ii

‘Q < 8.05 versus GBOUQS II and II. Tp c WI versus Groups I and II. tQ < 0. 1 versus Group 1. $Q < 0.01 versus Group 1, and p < 0.05 versus &-oup 11.

Values presented are mean value + SD. Abbseviations and ~e~~~t~~rns as in TabPes 1, 2 and 5.

e average score per seg

peak exercise (Table 6),

&cant differences were noted for the ing coronary artery in Groups I and II compared wit I11 for the number of abnormal segments and in Group HI compared with Group 111 for average score per segment and summed wall motion score. The ~erna~~der of the compari-

right coronary arteries (p < 0.05) and the average score per segment for the total vessels and left circumflex coronary artery (p < 0.95). When patients with a prior myocardial

ed, there were no si

en Croups P and II at peak exercise or rest or for is

wever, it clearjy dem-

imaging in 112 patients with coronary artery disease d strating partially or totally reversible thallium defects.

374 HECHTETA!. JACC Vol. 23, No. 2 TRULYSILENT ~~~~~~~~A February 19!l4:369-76

were no differences in the number of myocardial segments demonstrating partial or total redistribution between patients exercising with and without chest pain or ST segment abnormalities. Casperetti et al. (2) evaluated 103 patients with planar thallium imaging and tbun either fixed or reversible thallium abn exercising with and without chest pain. in 219 patients with coronary artery disease, noted no differences in tomographic thallium enormities, either total or reversible, between patients exercising with and without chest pain. Pooling all thallium imaging studies with cardiac catheterization data (l-3,9) yields an incidence of “truly silent ischemia” of 4i% in 489 patienos. However, because thallium perfusi alit& may be associated with, but are not synony , isc~cmia, the use of the term silclnt j~~~~~~~~ in

cially in the subset with truly silent iscbem~~ who h

exercise, is an excellent tool for the evaluation of co

coronary artery disease, with a sensitivity, s

treadmill exercise (6). The sitivities in the present stu this tool to evaluate ischemia.

The measures of ischemia include the as assessed by the number of abnormal merits; the severity, as determined by the ave

obal index, the summed wall both extent and severity. These variables

were obtained at rest and at peak exercise and, most important, for the change from rest to peak exercise, which represents ischemia. To further remove the influence of rest left ventricular dysfunction resulting from prior myocardial infarction, the data were also analyzed excluding with prior infarction. The myocardial

silent ischemia (l-3) have evaluated only the ic extent of abnormal perfusion, not the severity. In because 24-h or reinjection images, which may

increase the amount of reversible perfusion abno~ality in uP to 50% of patients were not obtained, the value of these reports is in further question because they have undoubtedly underestimated the amount of ischemia.

Nome We. The term “silent ischemia” has been used 1) to label patients with coronary artery disease who exer-

inaction and history of

). The shorter duration of exercise in pain (Group I) is similar to several

previous reports ( 1.9, IO). as is the lower peak maximal heart rate in patients with chest pain (3,18). Despite the shorter duration of exercise and lower peak maximal heart rate in

ati~nts expc~e~ci chest pain, there were no differ- ecces in work load or rate-pr re pro cantly higher percent of abnor stress exercising with versus without chest pain (83% vs. 38%, p <

1) is similar to data in previous thulium imaging studies, t et al. (1) reported 70% versus 32%, Gasperetti et al.

(2) 64% versus 41% and et al. (3) 73% versus 35%, respectively.

Amount of abnormaMy contracting mywwdium. The vir- tually identical amount of ischemic myocardium on a patient

JAW. Vol. 23, No. 2 February 1994:369-‘76

AECHT ET AL. 375

nt acquits of ischemia, of whi

e factors nor does it n of chest pain to isc~em~a

silent ischemia Rave failed to provide conclusive evidence that patients who exercise with or without chest pain have

because patients wit

376 HECHTETAL. TRULYSlLENTlSCHEMlA

by computerized edge detection programs would be prefer- able to the semiquantitative approach used, but no reliable quantitative analysis is available at the present time.

CO This study clearly demonst~tes that pa- tients who exekise with ST segment depression have more ischemia than do those who exercise without ST nt depression and that the presence or absence of e- induced chest ‘II bears no relation to the amount of ischemia. It su ts a reclassification of silent ischemia that focuses on both symptomatic and ECG silence and provides a framework for understanding the silent is~bemia data.

I. H&I HS, Shaw RE, B ilent ischemia: evaluation by exercise and redist~buti ittin- my~itr~iai ima J Am COB Cardiol 1989

2. Gasperetti CM, Butwell LR, Belier GE. Prevalence of and variables on exercise thallium-201 stress

3.

4.

minal coronary angioplasty. Am J Cardiol 1993:71:293-6. 5. Hecht 2%. DeBord L, Shaw R. et al. Digital supine bicycle stress

echocardiography: a new technique for evaluating coronary artery dis- ease. J Am Coil Cardiol 1993;21:950-6.

6. Hecht HS, DeBord L, Shaw R, et al. Supine bicycle StiCSS echocardiog-

titation of the left venlricle by two,dimensiona~ ech~nrdiogra~hy. Amer- ican Society of ~~h~ardio~~by Committee on St~dards, Sa~ommit-

mensional ~c~a~iu~aphy. J Am Sot

drix GW. Carabello BA, Usher JF. ind

infarction in patients Am J Cardiol 198?:59:

rcise

6, de&vi S, Poma E. et al. Clinical significance of exercise- silent myocardial ischemia in patients with coronary artery

. 3 Am Coil Catdiol 1987:9:29S-9. II. , Bacharach SL. Green MV. LaFreniere RL. Epstein SE.

implications of symptomatic versus asymptomatic Isiknt) &hernia induced by exercise in mildly symptomatic and in

patients with angiographiially documented coronary artery Cardiol 1987;601778-$3.

12. Weiner DA, Ryan TJ, McCabe CH. et al. Significance of silent myocardial xercise lesting in patients with coronary artery disease.

P. Shapiro EP. Chattdra MC. Goltlieb SH. Chew PK. Gottlieb hit and functional co son of patients with silent eadmill ischemia early r myocardial infarction. Am

e K. tntarachot V. Josephson MA. Rieders D, Mody FV. Siisgh stic signilicance of silent myocardial ischemia ia patients with

na. J Am Coil Cardiol 1987;lO: i-9.

31. Semtys PW. Wijns W. vandettBrand M. et al. Left ventricular perfor- mance regional blood flow. wall motion, and lactate metabolism during transluminal a~oplasty. Circulation 1984:70:25-36.

32. Hauser AM,Gangadhamn V. Ramos RG. Gordon S. Tiuuttis GS. Sequence of mechanical. electrocardiographic and clinical effects of repeated coronary artery occlusion in human beings: echocardiograpbic observations during comnary at@oplasty. J Am Colt Catdioll985;5:1937.

33. Wohlgelemter D. Cleman M, Highman HA, et al. Regional myocardial dysfunction during coronary angioplasty: evaluation by two-dimensional echocardiography and 12 lead electrocardiography. J Am Coil Cardiol 1986;73 1245-54.

34. Woh~ge~e~ter D. Jaffe CC, Cabin S, Yeatmau LA, CIeman M. Silent ischemia during coronary occlusion produced by balloon inflation: rela- tion to regional myocardial dysfunction. J Am Cdl Cardioll987;10:491-8.

35. Campbell S, Barry J. Rocco MB, et al. Features of the exercise lest that reflect the activity of ischemic heart disease out of hospital. Circulation 1986;74:72-80.

IS. Go&b SO. Weisfeldt ML, Guyang P, Mellits D, Gerstenblith G. Silent ischemia predicts infarction and death during 2 year follow-up of unstable

36. Mulcahy D. Keegan J, Sparrow J, Park A, Wright C. Fox K. lschemia in

angina. J Am Coil Cardiol1987;10:756-60. the ambulatory setting-the total ischemic burden: relation to exercise

16. Stem S. Gavish A, Wcisz G, Benhorin J, Keren A. Tzivoni D. Charac- testing and investigative and therapeutic implications. J Am Coil Cardiol 1989;14:1166-72.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29,

30.

JACC Vol. 23, No. 2 February 1~:3~9-~6

t&tics of silent and symptomatic myocardial ischemia during daily activities. Am J Cardiol 1988;61: 3223-8.

erm prognosis. J Am Coil Cdrdi

Rligonul V. Vandormael

exercise after coronary angio~last~. Am J Cardiol 1989$4:1-S.

Guyang P, Cbaa~ra hC. Gottlieb SO. F~e~Mea~y and irn~o~~~~~~ of silent rny~~dia~ iscbemia ide~ti~ed with i~rnbulat~~y electroc;tirdiographic

~eva~eace and prognostic rdial ischemia detected by y in asy~i~t~~ati~ vohm-

Assoc~tioo of periop

et al. Prevalence and prognostic

y J, Qras J. Bouassin BA. Raby KE. Selwyn AP. Effeec’: of asymptomatic ischemia ott long-term nosis in chronic stable coronary disease. Ci~cu~atioa l~~:83:1598 Hemandez RA. Macaya C, Jniguez A, et al. em outcome of patients with asymptomatic restenosis after coronary ba~l~a angioplasty. J Am Coil Cardiol $9: 1402-9. Weiner DA. abe CH, Ryan TJ. Identification of patients with left main and three vessel coronary disease witb clinical and exercise test variables. Am J Cardiol 1980;46:21-7.