Comparative prognostic value of clinical risk indexes, resting two-dimensional echocardiography, and dipyridamole stress thallium-201 myocardial imaging for perioperative cardiac events in major nonvascular surgery patients

Comparative prognostic value of clinical risk indexes, resting two-dimensional echocardiography, and dipyridamole stress thallium-201 myocardial imaging for perioperative cardiac events in major nonvascular surgery patients The relative prognostic value of widely accessible resting two-dimensional echocardiographic ventricular function data has not been compared with recognized clinical and scintigraphic risk markers ‘in patients who are unable to exercise before major nonvascular surgery. To this end, 53 consecutive patients aged 67 + 13 years undergoing preoperative evaluation (intraabdominal, 23%; orthopedic, 30%; thoracic, 9%; other, 36%) for known or suspected coronary artery disease were followed up to evaluate the prognostic value of these studies for the perioperative cardiac events (cardiac death [n = 41, myocardial infarction [n = 21, unstable angina [n = 31, and pulmonary edema [n = 61) that occurred in 13 of the 53 patients (25%). Dipyridamole thallium-291 myocardial redistribution defects occurred in 15 (26%) patients. Resting echocardiographic left ventricular dysfunction was present in 21 (40%) patients. Multivariate analysis of clinical, echocardiographic, and scintigraphic risk predictors revealed that cardiac events were not predicted by clinical variables, including Goldman class or score. Cardiac events were independently predicted only by the presence of significant left ventricular dysfunction on resting two-dimensional echocardiography (p < 0.042) and dipyridamole thallium-201 defect redistribution (p < 9.026). A dipyridamole-induced reversible thallium-201 perfusion defect was predictive of subsequent cardiac death or myocardial infarction (p < 0.02), whereas left ventricular dysfunction on resting echocardiography was predictive of perioperative pulmonary edema (p < 0.023). We conclude that stress thallium-201 perfusion imaging and resting two-dimensional echocardiography provide independent prognostic information in patients undergoing major nonvascular surgery who are at significant risk for ischemic cardiac events and who are unable to perform standard exercise stress tests. Clinical risk indexes were not predictive of cardiac events in these patients when analyzed in combination with noninvasive risk assessment. (AM HEART J 1993;126:1099-106)

Bonpei Takase, MD, Liwa T. Younis, MD, PhD, Sheila L. Byers, RN, Leslee J. Shaw, MA, Arthur J. Labovitz, MD, Bernard R. Chaitman, MD, and D. Douglas Miller, MD St. Louis, MO.

High-risk vascu1a.r surgery patients who are identified by clinical or noninvasive preoperative cardiac risk assessment may benefit from preoperative coroFrom Louis Supported Received

the Department of Internal University Medical Center. in part

Medicine,

by the Lichtenstein

for publication

Feb.

12, 1993;

Division

by Mosby-Year Book, + .lO 4/l/48896

St.

Foundation. accepted

March

Reprint requests: D. Douglas Miller, MD, Department Louis University Medical Center, Division of Cardiology-14th Vista Avenue at Grand Blvd., P.O. Box 15250, St. Louis, Copyright Q 1993 0002.8703/93/$1.00

of Cardiology,

Inc.

26, 1993. of Medicine, St. floor, 3635 MO 63110-0250.

nary revascularization or special perioperative therapies and monitoring to reduce their risk of cardiac events.lM3 Certain clinical variables, such as hypertension, prior myocardial infarction or congestive heart failure, diabetes mellitus, and advanced age, have been correlated with ‘postoperative cardiovascular morbidity and mortality.4-s Goldman risk factor analysis is the most widely applied clinical method for preoperative surgical cardiac risk assessment.g Dipyridamole stress thallium-201 myocardial imaging further discriminates which patients to undergo major vascular surgery are at greater risk of periop-

1099

1100

Takase et al.

Table

I. Characteristics of the study population (N = 53) Characteristic

Age (yrs) Age >70 years Male History of MI History of angina History of heart failure Diabetes mellitus Hypertension Hypercholesterolemia Smoking Family history of ischemic heart disease History of revascularization (PTCA-CABG) Se gallop or jugular vein distention NYHA class ~2 Q wave on ECG Any abnormality on ECG Goldman index Goldman class >_2

American

(%I 67 + 13 26(49) 28(53) 17(32) 19(36) 5 (9) 15(28) 37(70) 8(15) 18(34) ll(21) 12(23) 2 (4) 19(36) ll(21) 23(43) 5.4 f 4.9 12(34)

PTCA, Percutaneous transluminal coronary angioplasty; CABG, coronary artery bypass surgery; NYHA, New York Heart Association.

erative cardiac events in a population in which underlying ischemic heart disease is prevalent.lO-r7 Resting radionuclide left ventricular function studies have also demonstrated predictive value for postoperative events. 15s161ls Rowever, the preoperative prognostic utility of resting two-dimensional echocardiography-a widely available and relatively inexpensive noninvasive assessment of global and regional ventricular function-has not been established and has not been related to clinical risk indices and pharmacologic stress perfusion imaging data in the same population. Therefore the purpose of this study was to compare the relative prognostic values of resting two-dimensional echocardiographic left ventricular function assessment, intravenous dipyridamole myocardial perfusion thallium-201 imaging, and clinical risk variables (including Goldman risk factor analysis) in patients with suspected or known ischemic heart disease undergoing major nonvascular surgery. METHODS Patient population.

Preoperative resting two-dimensional echocardiography and intravenous dipyridamole thallium-201 myocardial imaging were performed and analyzed retrospectively in 53 consecutive patients with suspected or known coronary artery disease who were to undergo major noncardiovascular surgery and Who had been referred to St. Louis University Medical Center -for preoperative

November 79(Y3 Heart Journal

evaluation. None of the patients was able to perform adequate exercise, Characteristics of the study population are summarized in Table I. Sixteen (30%) patients underwent orthopedic surgery; 12 (23 % ) patients had intraabdominal surgery; and 25 (47%) patients had other operations (intrathoracic, urologic, gynecologic, renal transplantation, otopharyngeal, and reconstructive) under general anesthesia. Preoperative

risk assessment

Clinical assessment. Goldman

risk factor analysis was performed by retrospective medical record review without knowledge of the subsequent postoperative outcome. Of 36 clinical variables evaluated, 15 were necessary for computation of the Goldman risk index. Other clinical variables are given in Table I. Two-dimensional echocardiography. Two-dimensional echocardiography was performed in the standard parasternal long- and short-axis, apical fourchamber and two-chamber, and subcostal four-chamber and short-axis views. Doppler color flow imaging was also used to detect and evaluate suspected valvular abnormalities. All echocardiographic studies were technically adequate and were reviewed and interpreted in real time by two expert echocardiographers blinded to the clinical variables and study protocol. The degree of global ventricular dysfunction and segmental systolic wall motion abnormality was qualitatively evaluated as being mild, moderate, or severe hypokinesis. If present, coexistent valvular regurgitation was quantified as mild, moderate, or severe. Intravenous dipyridamole thallium-201 myocardial imaging. Dipyridamole thallium-201 scintigra-

phy was performed according to a previously reported method.ig Dipyridamole was infused intravenously at a rate of 0.56 mg/kg body weight over 4 minutes, and this was followed by active leg swinging. Then, 2 mCi of thallium-201 was injected 4 minutes after dipyridamole injection, and planar myocardial imaging was initiated within 5 minutes of isotope injection. Symptoms and a 12-lead ECG were monitored throughout the study. ST-segment depression of more than 1.0 mm at 80 ms after J-point (horizontal or downsloping) was considered abnormal. Planar thallium-201 myocardial images were acquired using a small-field gamma camera equipped with a low-energy, high-resolution, parallel-hole collimator in the anterior, 45-degree left anterior oblique, and left lateral views. Data were acquired from the 80 KeV thallium-201 photo-peak with a 20% window and stored on a 256 X 256 byte computer matrix. Approximately 500,000 counts were acquired during a lo-minute period in each view. Four hours after termination of infusion, delayed images were

Volume 126, Number 5 American Heart Journal

obtained in identical views. Images were interpreted by consensus of two expert observers without knowledge of the patient’s clinical or echocardiographic information. As previously described,lg the left ventricle was divided into five segments per view, which were each graded as being normal or as having either fixed or reversible thallium-201 myocardial perfusion abnormalities. Follow-up for cardiac events. Medical records of all patients were reviewed to ascertain the presence of 36 relevant risk variables available to the clinicians making the decision for preoperative medical clearance. Recommendations for further medical management or diagnostic testing were made by this physician, who had access to both the clinical and noninvasive test data. Postoperative cardiovascular events were prospectively defined as: (1) cardiac death, (2) postoperative myocardial infarction-electrocardiographic (Xl new Minnesota code Q wave) or serial cardiac enzyme changes (peak CPKMB > 5 %), (3) unstable angina (22 episodes of chest pain with documented ischemic electrocardiographic changes), (4) pubmonary edema documented by clinical record evidence of pulmonary rales and chest radiographic findings. Statistical analysis. Categorical variables were compared using either chi-square analysis or the Fisher exact test. Continuous variables were compared using the Student t test. Significant univariate predictors were entered thereafter into multivariate regression analysis. Univariate analyses were performed on 36 clinical, echocardiographic, and dipyridamole thallium-201 image variables to determine their predictive value for postoperative cardiac events. Relative risk of postoperative cardiac events was calculated for each variable. A probability (p) value <0.05 was considered significant for all statistical analyses. RESULTS Preoperative

risk assessment. Dipyridamole thallium-201 imaging revealed that 15 (28%) patients had ~1 reversible defect and that 6 (21% ) patients had z 1 fixed thallium-201 defects. Twenty-one patients (40 % > had abnormal resting echocardiographic left ventricular function. Four and two patients had mild and moderate global hypokinesis, respectively. Fifteen patients had segmental wall motion abnormalities, which were classified as mild (n = S), moderate (n = 4), severe hypokinesis (n = 2), or akinesis (n = 3). Seven (13%) patients underwent cardiac catheterization before surgery based on the treating physician’s interpretation of the results of preoperative cardiac evaluation (six had a reversible thallium-

Takase et al.

1101

201 defect and four had abnormal left ventricular function). Of four patients who underwent a coronary revascularization procedure (PTCA) before elective surgery, none had postoperative cardiac events. Postoperative outcomes. Thirteen patients (25 % ) had perioperative cardiac events. These were cardiac death (n = 4), m:yocardial infarction (n = 2), unstable angina episodes (n = 3), and acute episodes of pulmonary edema (n = 8) (Table II). The cardiac events were not counted as mutually exclusive in the endpoint analysis. The inCidences of postoperative cardiac events were slightly higher in Goldman classes ~2; however, this was not significantly different from class 1. Reversible thallium-201 myocardial perfusion defects and abnormal echocardiographic left ventricular systolic function study results were both correlated with cardiac events. Abnormal echocardiographic left ventricular function was highly correlated with the occurrence of perioperative acute pulmonary edema (p = 0.023). Higher incidence of both left ventricular global and original wall abnormalities were observed, but no differences were noted in the frequency of use of perioperative invasive monitoring. The influence of Goldman classification and dipyridamole thallium-201 imaging results on the occurrence of any type of cardiac event is illustrated in Fig. 1. Patients with Goldman class 1 and normal thallium-201 imaging results had no cardiac events, whereas 5 of 15 (33%) patients with Goldman class 1 and a abnormal thallium-201 perfusion defect had cardiac events. Four of seven (57%) patients with Goldman class ~2 and abnormal dipyridamole thallium scan had an event, whereas only 3 of 12 (25%) patients with normal scan had events (JJ < 0.05). The relation of Goldman class to echocardiographic left ventricular function assessment for prediction of perioperative pulmonary edema is shown in Figure 2. Only 1 (4%) of 24 patients in Goldman class 1 with normal left ventricular function developed pulmonary edema, whereas 4 (36 % ) of 11 patients in Goldman class 1 and abnormal left ventricular function had pulmonary edema. A similar pattern existed for patients with Goldman class 22. Table III compares the 13 patients who had postoperative cardiac events to the 40 patients who did not. Only echocardiographic left ventricular systolic dysfunction and reversible thallium-201 myocardial perfusion defects were significantly more frequent in patients with perioperative cardiac events. A history of myocardial infarction and the use of nitrates were insignificantly greater in patients with cardiac events. Other clinical variables, including Goldman index

1102

Takase et al.

American

November 1993 Heart Journal

Number of Patients

25

i8 CE (-1

m GE(+)

0 Normal Fixed ‘Reversible GOLDMAN CLASS > 2 (rk18)

Normal Fixed -Reversible GOLDMAN CLASS 1 (rk35)

Fig. 1. Incidence of postoperative cardiac events in 53 patients, stratified by Goldman classand dipyridamole thallium scintigraphy. (CE, Cardiac events.) Table II. Incidence of postoperative cardiac events when stratified by the Goldman class,dipyridamole-thalliu scintigraphy (DPT), and LV systolic function by two-dimensional echocardiography (ECHO) Cardiac death

Myocardial

infarction

(%I

Goldman class Class 1 (n = 35) Class 22 (n = 18) DPT Normal (n = 32) Redistribution (n = 15) Fixed defect (n = 6) LV systolic function on ECHO Normal (n = 32) Abnormal (n = 21) *Number pNumber

f%)

angina

Unstable

Pulmonary

edema

Any

cardiac events”

(%)

(%)

(%I

5 (14) 3 (16)

7 (20) 6 (33)

2 Wt 2 (11)

l(3)

l(3)

1 (6)

2 (11)

0 2 (13) 2 (33)

0 2 (13) 0

2 (6)

l(3)

1 (7) 0

5 (33) 2 (33)

3 (9) 6 (53)f 2 (33)

2 (6)

0

2 (10)

2 (10)

l(3) 2 (10)

1 (3) 7 (33) $

4 (13) 9 (43)

of patients with any cardiac events. in parentheses represents percentage.

$P < 0.05 YSnormal.

and class, were not significantly correlated with adverse outcome. Univariate and multivariate predictors of cardiac events are listed in Table IV. Patients who died or had myocardial infarction had a significantly higher incidence of reversible thallium-201 perfusion defects (67% vs 23%; p < 0.02). Only reversible thallium-201 perfusion defects and echocardiographic left

ventricular systolic dysfunction were predictive of any postoperative cardiac event by multivariate analysis. Occurrence of a cardiac death or a myocardial infarction was correlated with reversible and fixed thallium-201 perfusion defects by multivariate analysis. Echocardiographic left ventricular systolic dysfunction was the only multivariate predictor of perioperative pulmonary edema (p = 0.023).

Volume 126, Number 5 American Heart Journal

Number

Takase et al.

1103

of Patients

El CE (-) n

Normal LVF Abnormal LVF GOLDMAN CLASS 1 (n=35)1

Normal LVF Abnormal LVF GOLDMAN CLASS 12 (nd8)

Fig. 2. Incidence of postoperative pulmonary edema in 53 patients, stratified by Goldman echocardiographic left ventricuIar function. (CE, Cardiac events.)

The comparative prognostic value of clinical, dipyridamole thallium-201, and echocardiographic variables for any perioperative cardiac event was assessed by comparison of derived relative risk ratios. Goldman class 22, the presence of reversible thallium-201 defects, and abnormal echocardiographic left ventricular systolic dysfunction had relative risk ratios of 1.7, 3.3, and 2.2, respectively. DISCUSSION

The global risk of perioperative cardiac events in patients with ischemic heart disease is increased approximately lo- to 50-fold as compared with patients without ischemic heart disease.20, ZJ Goldman’s clinical risk criteria are validated indexes of clinical risk assessment that may be applied for preoperative screening of a general population. Among patients with intermediate to high risk of perioperative cardiac events (i.e., vascular surgery patients), dipyridamole thallium-201 myocardial scintigraphy and exercise radionuclide left ventricular function analysis have both been reported to be powerful independent predictors for postoperative cardiac events.16V1s*22 In our study, the power of dipyridamole thallium-201

CE (+)

class and

myocardial scintigraphy and resting two-dimensional echocardiographic assessment for predicting perioperative cardiac events in patients who were unable to exercise before major nonvascular surgery superceded the prognostic value of previously recognized clinical risk variables. A third of this population gave a medical history of myocardial infarction or angina pectoris. Half (49 % ) the patients studied were at increased risk because of their age (170 years). Dipyridamole thallium-201 abnormalities (49 % ) and echocardiographic left ventricular systolic dysfunction (39 % ) were therefore quite prevalent. These characteristics confirm that the patient population studied was at considerable risk for perioperative cardiac events.23> 24 Among the ischemic risk variables tested ,in the multivaiiate analysis, only reversible dipyridamole thallium-201 perfusion defects and echocardiographic left ventricular systolic dysfunction were significantly predictive of cardiac events. More specifically, reversible thallium-201 defects predicted cardiac death or myocardial infarction, whereas echocardiographic left ventricular dysfunction predicted acute perioperative heart failure. When the

November

1304

Table 111.Comparison of patients with and without cardiac

events after nonvascular surgery Variables

Age > 70 years Male History of MI History of angina Diabetes mellitus Hypertension Hypercholesterolemia Smoking Family history of ischemic heart disease History of revascularization (PTCA-CABG) Sa gallop or jugular vein distention NYHA class 22 Q wave on ECG Any abnormality on ECG Goldman index Goldman class 22 LV systolic dysfunction Regional wall motion abnormality Global LV dysfunction Redistribution of thallium Fixed thallium defect Perioperative pulmonary pressure monitoring

1943

AmericanHeartJournal

Takaseet al.

Events (%) No events (%) (n = 40) (n = J3)

8 (62) 6 7 5 5 7 1 5 3

(46) (54) (38) (381 (54)

(8) (38) (23)

p Value

Table IV. Significant predictors by univariate and multivariate regression analysis for postoperative cardiac events

(A), death and myocardial edema (6).

infarction Any

lS(45) 22 (55) lO(25) 14(35)

8 (20)

,38 .24 .I2 .43 .15 .15 .51 .58 .52

(24

.79

10 (25) 30 (75) 7 (18) 13 (33)

(B), pulmonary cardiac events

Univariate Predictors

p ualue

Multivariate p value

A ECHO LV systolic dysfunction Redistribution of thallium History of MI Usage of nitrate

0.007 0.002 0.054 0.072

0.042 0.026 0.292 0.459

Redistribution of thallium Thallium fixed defects Usage of nitrate

0.026 0.074 0.067

0.018 0.023 0.369

Echo LV systolid dysfunction Redistribution of thallium Fixed thallium defect Usage of nitrate

0.003 0.004 0.055 0.017

0.023 0.076 0.316 0.093

B 3 (23)

9

0 (0)

2 (5)

.57

C

6 (46) 4(31) 4 (31) 5.6 t 4.6 6 (46) 9 (69)

6 (46) 3 (23)

8 (62) '

4 (31) 2 (16)

12 (33) 7 (18)

19 (48) 5.4 It 5.0 12 (30) 12 (30) 9 (24 3 (8) 7 (18) 7 (18) 5 (20)

.29 .31 .46 .97 .86 ,007

<.Ol <.09 <.OOl .31 .18

relative risk of cardiac events was calculated, dipyridamole thallium-201 hypoperfusion and echocardiographic dysfunction had higher associated risk than did Goldman’s clinical risk criteria. Two-dimensional echocardiographic left ventricular systolic function was qualitatively assessed by a consensus of expert echocardiographers. Computer calculations of left ventricular ejection fraction from echocardiography using off-line analysis are timeconsuming, and echocardiographic ejection fractions are not routinely calculated in clinical practice. Subjective estimations of left ventricular ejection fraction by expert echocardiographers are reported to be as reliable as computer-assisted quantitative measurements of left ventricular ejection fraction.25-27 The prognostic value of the current qualitative assessment supports the utility of this clinical method. In our study the incidence of global and regional wall motion abnormality was higher in patients with cardiac events. Previous studies. Pasternack et al. have reported that resting radionuclide ventriculographic function predicted perioperative cardiac events in peripheral vascular surgery and abdominal aortic aneurysm re-

section patients.15 Other reports have previously substantiated the prognostic value of dipyridamole thallium-201 myocardial scintigraphy compared with clinical variables in high-risk noncardiac surgery patients.7> r3pl7 One study of an aortic aneurysm surgery population concluded that a reversible dipyridamole thallium-201 scintigraphy defect could predict perioperative cardiac events, whereas clinical variables (including Goldman criteria) as well as left ventricular ejection fraction (obtained from either radionuelide angiogram or echocardiography) could not.22 The influence of left ventricular function on perioperative heart failure was not evaluated in these studies. Exercise testing is a better predictor for postoperative cardiac events than clinical variables alone.28 Noninvasive cardiac evaluation was useful in geriatric major noncardiovascular surgical candidates at significant risk of coronary artery disease, but not in the routine preoperative evaluation of low-risk patients.2g>30 Our findings support and extend these results, substantiating the prognostic role of pharmacologic stress testing in an intermediate to highrisk population that was unable to perform exercise stress. The “hard” cardiac event rate in this study (6 of 53 = 11%) is comparable to the intermediate-highrisk patient group reported by Eagle et alI4 The incidence of postoperative heart failure (13 % ) was relatively ~greater but was comparable to other studies in which postoperative heart failure was reported.30 A history of myocardial infarction, heart failure, and

Volume American

126, Number 5 Heart Journal

Takase et al.

diabetes mellitus have been identified as significant predictors for perioperative pulmonary edema. The relatively high incidence of heart failure in the early postoperative p:hase in our population could explain why only echocardiographic left ventricular systolic function predicted these postoperative cardiac events, despite ,the low (9 % ) incidence of preoperative heart failure symptoms. Of 100 patients evaluated for noncardiovascular surgery, Coley et al. 31 have reported that clinical variables such as congestive heart failure and age 170 years iden.tify patients for whom subsequent noninvasive testing is beneficial. In this earlier study, eight of the nine patients with cardiac events had reversible thallium-201 defects. Eight other patients with significant reversible defects were excluded from the study because of cancellations of surgery after the result of dipyridamole was known, potentially influencing the total number of events in the group with an abnormal scan result.31 Conclusions. Our multivariate analysis of clinical assessment, dipyridamole thallium-201 imaging, and resting echocardiographic variables indicates that noninvasive cardiac evaluation is superior to clinical risk assessment alone. Dipyridamole thallium-201 scintigraphy was highly predictive of cardiac death or myocardial infarction. Rest two-dimensional echocardiographic assessment-a study more widely available than stress dipyridamole thallium-201 scintigraphy-may also provide significant prognostic data regarding the risk of perioperative pulmonary edema. The combination of these noninvasive studies predicted acute ischemic events and left ventricular decompensation during or immediately following major nonvascular surgery in these patients with a significant likelihood of coronary artery disease. However, these data do not preclude the application of clinical evaluation in the decision to perform noninvasive testing in a population at risk and do not absolve clinicians from carefully integrating available noninvasive and clinical risk parameters in the preoperative evaluation of major nonvascular surgical risk. We thank Lori Gallini Meeker the preparation of the manuscript.

for her secretarial

assistance

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

in

REFERENCES

1. Goldman L. Assessment of the patient with known or suspected ischemic heart disease for non-cardiac surgery. Br J Anesth 1988;61:38-43. 2. Abraham SA, Coles A, Coley CM, Strauss HW, Boucher CA, Eagle KA. Coronary risk of noncardiac surgery. Prog Cardiovast Dis 1991;342205-34. 3. Eagle KA, Boucher CA. Cardiac risk of noncardiac surgery. N Engl J Med 1989;321:1330-2. 4. Knorring JV. Postoperative myocardial infarction: a prospec-

20.

21.

22.

1105

tive study in a risk group of surgical patients. Surgery 1981; 90:55-60. Calvin JE, Kieser TM, Walley VM, McPhail NV, Barber GG, Scobie TK. Cardiac mortality and morbidity after vascular surgery. Can J Surg 1986;29:93-7. Cooperman M, Pflug B, Martin EW Jr, Evans WE. Cardiovascular risk factors in patients with peripheral vascular disease. Surgery 1987;84:505-9. Eagle KA, Singer DE, Brewester DC, Darling RC, Mulley AG, Boucher CA. Dipyridamole-thallium scanning in patients undergoing vascular surgery. JAMA 1987;257:2185-21. Detsky AS, Abrams HB, McLaughlin JR, Drucker DJ, Sasson Z, Johnston N, Scott JG, Forbath N, Hilliard JR. Predicting cardiac complications in patients undergoing non-cardiac surgery. J Gen Intern Me61986;1:211-19. Goldman L. Caldera DL. Nussbaum SR. Southwick FS. Krogstad D,‘Murray B, Burke DS, O’Malley TA, Goroll AH; Caplan CH, Nolan J, Carabello B, Slater EE. Multifactorial index of cardiac risk in noncardiac surgical procedures. N Engl J Med 1977;297:845-50. Hendel RC, Layden JJ, Leppo JA. Prognostic value of dipyridamole thallium scintigraphy for evaluation of ischemic heart disease. J Am Co11 Cardiol 1990;15:109-16. Boucher CA. Brewester DC. Darline C. Okada RD. Strass HW. Pohost GM.’ Determination of cardiac risk by dipyridamole: thallium imaging before peripheral vascular surgery. N Engl J Med 1985;312:389-94. Lette J, Waters D, Lassonde J, Rene P, Picard M, Laurendeau F, Levy R, Cerino M, Nattel S. Multivariate clinical models and quantitative dipyridamole-thallium imaging to predict cardiac morbidity and death after vascular reconstruction. J Vast Surg 1991;14:160-9. Leppo J, Plaja J, Gionet M, Tumolo J, Paraskos JA, Cutler BS. Noninvasive evaluation of cardiac risk before elective vascular surgery. J Am Coil Cardiol 1987;9:269-76. Eagle KA, Coley CM, Newell JB, Brewester DC, Darling C, Strass HW, Guiney TE, Boucher CA. Combining clinical and thallium data optimize preoperative assessment of cardiac risk before major vascular surgery. Ann Intern Med 1989;110:85966. Pasternack PF, Imparato AM, Bear G, Riles TS, Baumann FG, Benjamin D, Sanger J, Kramer E, Wood P. The value of radionuclide angiography as a predictor of perioperative myocardial infarction in patients undergoing abdominal aortic aneurysm resection. J Vast Surg 1984;1:320-5. Brown KA, Rimmer J, Haisch C. Noninvasive cardiac risk stratification of diabetic and noncardiac uremic renal allograft candidates using dipyridamole-thallium-201 imaging and radionuclide ventriculography. Am J Cardiol 1989;64:1017-21. Younis LT, Aguirre F, Byers S, Dowel1 S, Barth G, Walker H, Carrachi B, Peterson G, Chaitman BR. Perioperative and long-term prognostic value of intravenous dipyridamole thallium scintigraphy in patients with peripheral vascular disease. AM HEART J 1990;119:1287-92. Laser L, Russell JC, DaSilva J, Radford M. Use of the multiple uptake gated acquisition scan for the preoperative assessment of cardiac risk. Surg Gynecol Obstet 1988;1678:234-8. Lam JT, Chaitman BR, Glaezner M, Byers S, Fite J, Shah Y, Goodgold H, Samuels L. Safety and diagnostic accuracy of dipyridamole-thallium imaging in the elderly. J Am Co11 Cardiol 198&11:585-g. Tarhan S, Moffitt EA, Taylor WF, Giuliani ER. Myocardial infarction after general anesthesia. JAMA 1972;220:14511454. Stogoff S, Keats AS. Does perioperative myocardial ischemia lead to postoperative myocardial infarction. Anesthesiology 1983;62:107-14. McEnroe CS, O’Donnell TF Jr, Yeager A, Konstam M, Mackev WC. Comoarison of eiection fraction and Goldman risk factor analysis to dipyridamole-thallium-201 studies in the evaluation of cardiac morbidity after aortic aneurysm surgery. J Vast Surg 1990;11:497-504.

November

1993

Newman et al. 23. Shaw L, Chaitman BR, Hilton TC, Stocke K, Younis LT, Caralis DG, Kong BA, Miller DD. Prognostic value of dipyridamole thallium-201 imaging in elderly patients. J Am Co11 Cardiol 1992;19:1390-8. 24. Camp AD, Garvin PJ, Hoff J, Marsh J, Byers SL, Chaitman BR. Prognostic value of intravenous dipyridamole thallium imaging in patients with diabetes mellitus considered for renal transplantation. Am J Cardiol 1990;65:1459-63. 25. Amico AF, Lichtenberg GS, Reisner SA, Stone CK, Schwartz RG, Meltzer RS. Superiority of visual versus computerized echocardiographic estimation of radionuclide left ventricular ejection fraction. AM HEART J 1989;118:1259-65. 26. Kan G, Visser CA, Koolen JJ, Dunning AJ. Short and long-term predictive value of admission wall motion score in acute myocardial infarction: a cross-sectional echocardiographic study of 345 patients. Br Heart J 1986;56:422-7. 27. Mueller X, Stauffer JC, Jaussi A, Goy JJ, Kappenberger L. Subjective visual echocardiographic estimation of left ven-

28.

29.

30.

31.

tricular ejection fraction as an alternative to conventional echocardiographic methods: comparison with contrast angiography. Clin Cardiol 1991;14:898-907. Carliner NH, Fisher ML, Plotnick GD, Garbart H, Rapoport A, Kelemen MH, Moran GW, Gadacz T, Peters RW. Routine peroperative exercise testing in patients undergoing major noncardiac surgery. Am J Cardiol 1985;56:51-8. Gerson MC, Hurst JM, Hertzberg VS, Doogan PA, Cochran MB, Lim SP, McCall N, Adolph RJ. Cardiac prognosis in noncardiac geriatric surgery. Ann Intern Med 1985;103:832-‘7. Charlson ME, MacKenzie CR, Gold JP, Ales KL, Topkins M, Shires GT. Risk for postoperative congestive heart failure. Surgery Gynecol Obstet 1991;172:95-104. Coley CM, Field TS, Abraham SA, Boucher CA, Eagle KA. Usefulness of dipyridamole-thallium scanning for peroperative evaluation of cardiac risk for nonvascular surgery. Am J Cardiol 1992;69:1280-5.

Effects of canbe m.yocardia~l infarction on sympathetic efferent neuronal funct.i-on: Scintigraphic and electrophysiologic correlates We studied the effects of nondecentralized left stellate ganglion stimulation on regianal epicardial monophasic action potential duration at 50% (APDso) and 90% (APD~,J) repolarization from 104 sites in 10 surviving dogs with a chronic myocardial infarction model. These effects were correlated with thathum-201 and iodine-123 metiodobenzylguanidine (MIBG) imaging to identify areas of viable but denervated myocardium. Mean infarct size was 5.2% & 0.8% total heart weight, and the planimetered areas of denervatiorl were always larger (18% r 4% total heart area). During constant ventricular pacing, stellate stimulation tended to shorten the APDgo only in normally innervated areas (364 + 5 to 358 rt 5 msec) and to increase in denervated areas (358 f 5 to 362 + 5 msec), (p value not significant (NS) for prestellate and poststellate stimulation; p < 0.05 for difference between denervated vs innervated). The APDso significantly shortened in innervated areas from 287 t 5 to 270 + 3 msec (p < 0.05) compared with denervated areas (283 i: 4 to 274 k 5 msec, p = NS). We conclude that MIBG imaging demonstration of denervation identifies areas with impaired shortening of the epicardial APDso in response to stellate stimulation and that nontransmural myocardial infarction produces areas of denervation larger than areas of necrosis. (AM HEART J 1993;126:1106-12.)

David Newman, MD, Luisa Munoz, MD, Michael Chin, BSc, Michael Franz, MD, PhD, John Herre, MD, Melvin M. Scheinman, MD, Elias Botvinick, MD, and Michael Dae, MD San Francisco, Calif. From the Department of Medical and Cardiovascular Research Institute, University of California, San Francisco. Supported in part by grants from the National Institutes of Health (HL-38105 and HL-25847) and the Fannie Rippel Foundation and clinical grant CG-89007 from DuPont, Bellerica, MA. Dr. Newman was supported by a fellowship from the Medical Research Council of Canada. Received for publication Feb. 17, 1993; accepted April 1, 1993. Reprint requests: David Newman, MD, St. Michael’s Hospital, Division of Cardiology, 30 Bond St., Toronto, Ontario, M5B lW8. Copyright @ 1993 by Mosby-Year Book, Inc. 0002-8703/93/$1.00 + .lO 4/l/48877

1106

The sympathetic nervous system has an important role in postmyocardial infarction arrhythmias.l This role may be related to the acquired imbalance and heterogeneity of sympathetic innervation of the heart produced by myocardial infarction. Both animal and limited human data have demonstrated that myocardial infarction can produce areas of denervation around the area of infarction and distal to the area that is infarcted.2p 3 In canine d&a, denervation dis-