Cardiac complications in noncardiac surgery: Value of dobutamine stress echocardiography versus dipyridamole thallium imaging

Cardiac Complications in Noncardiac Surgery: Value of Dobutamine Stress Echocardiography Versus Dipyridamole Thallium Imaging Michael C. Kontos, MD, Kwame O. Akosah, MD, Lisa K. Brath, MD, John T. Funai, MD, and P.K. Mohanty, MD Objectives: The objective of this study was to determine the relative value of dobutamine stress echocardiography (DSE) and dlpyridamole thallium imaging (DT) in the preoperative assessment of cardiovascular risk before noncardiac surgery. Design: Prospectively DSE was performed in patients who had undergone DT as a part of their preoperative evaluation. Setting: A large urban veterans' affairs medical center. Participants: Thirty-seven patients undergoing major noncardiac surgery were assessed for complications during a 1-month follow-up period. Interventions: Both DSE and DT were performed before surgery. The medium interval between the two tests were 15.5 days. Measurements: Left ventricular wall motion was assessed at baseline and peak dobutamine dose in a standard fashion. Wall motion was scored and indexed using a 16-segment model. A positive DSE was defined as failure of augmentation, new or worsening of baseline wall motion abnormalities in two or more contiguous segments. Myocardial perfusion studies after DT were performed according to conventional method A positive DT was defined as a reversible perfusion defect, increased lung uptake, and/or transient left ventricular dilatation. Complications were defined as myocardial infarction or cardiac death occurring as a result of the operation, or need for revascularization before surgery.

Results: DSE was positive in 19 patients, whereas DT was positive in 25 patients. Fourteen patients had both an abnormal DSE and DT. Five patients had major postoperative cardiac complications: fatal myocardial infarction (1); fatal cardiac arrest (1); and severe coronary artery disease necessitating coronary artery bypass surgery (2) or percutaneous transluminal coronary angioplasty (1). DSE was positive in all 5 (100%), whereas DT was positive in 4 of 5 (80%) patients with complications. The sensitivity for each test was comparable: for DSE it was 100% (95% C.I. 56% to 100%) and for DT 80% (37% to 96%). Specificity for DSE (60%, 43%-74%) was somewhat higher than DT (38%, 24% to 54%), although this did not reach statistical significance

p

play in Identifying high-risk patients who are hkely to suffer cardiac complications related to noncardiac surgery.

A T I E N T S W I T H coronary artery disease (CAD) undergoing noncardmc surgery are at increased risk for developing perioperative cardmc complications, v4 A number of different nonlnvaswe tests have been used in an attempt to identify these patients preoperatively. 5-0 Because a significant n u m b e r of patients are unable to exercise secondary to coexisting peripheral vascular, orthopedac, or neurologic disease, a pharmacologic means of provoking myocardial lschemia is often used. Currently, the most commonly used test for further risk stratifying such pataents is dapyridamole thallium scintlgraphy. Although lmtlal studies have demonstrated high sensitivity of dlpyndamole thalhum imaging in predicting patients at risk for complications, 5.~° more recent data suggest ~mportant limitations in the ability of dipyridamole thalhum imaging to predict perioperative comphcations, lv15 Other major dasadvantages of dipyridamole thalhum include its contramdlcan o n in patients wath bronchospastac d~sease a n d / o r in pataents who are recewmg theophylllne treatment. Recently, dobutamine stress echocardaography has been shown not only to be a safe and effective noninvaswe imaging modahty m identifying patients with CAD, 16-1swith similar sensltwities and speclficitaes for predicting CAD as pharmacologic nuclear imaging I8 but is also able to predict patients at increased risk for developing complicataons after undergoing vascular surgery. ~922 Currently, there is limited information on the comparative role of these two tests in the preoperatwe assessment of cardiac risk in the same cohort of patients. Accordingly, the purpose of this study was to assess the comparative efficacy of dlpyridamole thallium scmhgraphy and dobutamine stress echocardlogra-

(p = 0.06). Conclusions: The ability of DSE to predict major cardiac complications related to noncardiac surgery appears to be similar to DT and may be used as an alternative to DT imaging in the preoperative risk assessment of patients undergoing noncardiac surgery.

Copyright © 1996 by W.B. Saunders Company

KEY WORDS: dobutamine stress, echocardiography, dipyrid a m o l e thallium, perioperative complications

METHODS Thirty-seven patients scheduled for 42 routine nonemergent, noncardlac major surgical procedures over a 9-month period comprised the study population All patients had undergone dlpyrldamole thalhum testing within 6 months (median 15.5 days) before their surgery and provided reformed consent to undergo dobutamlne stress echocardlography and participate m follow-up evaluation after surgery This protocol was approved by the Committee on the Conduct of Human Research at a university and the Research Committee of a veterans' affairs center The majority of the patients were examined by a consulting cardiologist or mtermst before their surgery Further recommendations, including the decision for coronary anglography, were based on that evaluation Invaswe monitoring during and after the surgery was performed as deemed necessary by the anesthesiologist, the patient's primary physician, and the consulting physician Postoperatively, all patients had dally creatme kmase (CK), CK-MB, and 12-lead electrocardiograms (ECG) for the first 4 postoperative days, and subsequently as deemed necessary by the patient's primary physlcmns During this period, all pahents were examined and had their charts and operatwe records reviewed to define cardiac complications by one of the lnveshgators (who was

From the Medtcal College of Vwgmta and McGutre VA Medwal Center, Rtchmond, VA Address reprmt requests to Kwame Akosah, MD, Cardtology Dtvtston (111J), McGutre VA Medtcal Center, 1201 Broad Rock Blvd, Rtchmond, VA 23249 Copyright © 1996 by W B Saunders Company 1053-0770/96/1003-000653 00/0

Journal of Cardlothoracm and Vascular Anesthesia, Vol 10, No 3 (April),1996' pp 329-335

329

330

bhnded to the dobutamine stress echocardlography results). Subsequent follow-up consisted of weekly clinical evaluation and chart review of medical records until time of discharge or 30 days. Patients were interviewed and examined by one of the investigators before surgery, with particular emphasis on the cardiac history and physical examination. Important variables such as previous history of myocardial infarction (MI), angina, congestive heart failure (CHF), cerebral vascular accident or transient ischemic accident, previous revascularlzatlon procedures, and relevant laboratory values were recorded. Angina was considered typical if it was primarily substernal, usually occurred with exertion, and was relieved with rest or nitroglycenn w~thln a few minutes; otherwise, it was labeled as atypical Patients were considered to have CHF if they had at least two of the followmg: effort dyspnea; orthopnea or paroxysmal nocturnal dyspnea; jugular vein distention; an $3 gallop; riles; or lower-extremity edema. Primary end points prospectively defined were MI or cardiac death occurring within 30 days of the operation and the need for revascularlzation before the scheduled noncardlac surgery Cardiac death was defined as death caused by MI, CHF, or primary cardiac arrest. MI was diagnosed if two of the following three criteria were present. ECG changes consistent with ischemla or infarction (ie, new Q waves; > 1 mm ST segment elevation or depression in two leads), angina or lschemic heart failure symptoms, and an elevation in both CK and CK-MB above normal values (CK > 250 U/L, MB > 5%) The dobutamine infusion protocol and dobutamxne stress echocardiographlc imaging have been reported previously in detail 23 from the authors' laboratory In brief, patients fasted the night before, and cardmc medications were withheld on the morning of the test. After baseline echocardlographlc images were obtained, the dobutamxne infusion was started at an lnmal dose of 5 ~g/kg/mln for 5 minutes. The infusion was then titrated to 10 izg/kg/mln, then increased by 10 tzg/kg/mln m 5-minute stages to a maximum dose of 40 ~g/kg/mln. End points for early discontanuanon of the test included reaching 85% of the patients' predicted maximum heart rate (PMHR, defined as 220--age), more than 2-ram additional ST segment depression compared with the baseline ECG, symptomatic decrease in systolic blood pressure, or precipitation of angina. Representative images of each view were digitized and stored in real time into cine loops at baseline, 5 ixg/kg/min, peak dobutamine infusion dose, and recovery Subsequently, the clne loops were transferred onto an optical disk controlled by a computer work station (ImageVue; Nova Mlcrosonics, Indianapolis, IN). Wall motion was scored by using the modified 16-myocardial segment model described previously from the authors' laboratory. 23 Stress echocardlograms were jointly analyzed by two experienced echocardiographers who had no knowledge of the preoperative chmcal findings or the thallium results. A normal response to dobutamme was defined as augmentation of baseline systolic wall motion. A positive result was defined as the development of a new or worsening of a previous wall motion abnormality in two or more contiguous segments, or failure to augment (xe, improvement in wall motion from baseline)18 Each segment was assigned a score of 1 to 4: normal, 1; hypoklnetlc, 2, aklnetic, 3; and dysklnetlc, 4 Disagreements were resolved by consensus Dipyridamole thalhum imaging was performed in a fasted state. Each patient received a controlled Infusion of dlpyrldamole, 0.56 mg/kg over 4 minutes, while undergoing constant blood pressure. electrocardiographic, and heart rate monitoring. Three minutes after completion of the dxpyridamole Infusion, the patient was given a tracer lnjectmn of 2 mCI of thallium-201 Planar images were then obtained in the anterior, 45-degree. and 70-degree left

KONTOS ET AL

anterior oblique projections Approximately 1 hour after the initial images were obtained, If there was evidence of a defect on the initial images, the patient was given a relnjectlon of 1 mCI of thalhum-201 as part of an immediate relnjeetion protocol. Approximately 3 hours after the initial tracer injection, repeat planar images of the heart were obtained, and the two sets of Images were compared. If there was a question whether a defect would later redistribute, the patient was brought back in 24 hours for repeat imaging. For each scintlgraphlc study, analog images, interpolated background-subtracted images, horizontal profiles, and washout rate analysis were used for analysis 24 The scans were divided into nine segments, anterior; anterior-lateral; septal, inferior; inferiorapical; inferior-basal; posterior-lateral; posterior-basal; and apical. All scans were analyzed by an experienced nuclear physician who had no knowledge of the protocol or the dobutamine echo results. Scans showing no thalhum perfuslon defects were considered normal; those with defects on initial images but not delayed scans were considered to have redistribution, and those with persistent defects on both immediate and delayed scans were considered to have fixed defects. Transient increase in left ventrlcular cavity size on the stress images as compared with the delayed images and those that showed an increased lung/heart thallium uptake ratio greater than 50% were also considered abnormal. 25.26Fixed defects alone did not classify the dipyrldamole thallium as abnormal Eight patients who had abnormal dipyrldamole thallium sclntlgraphy underwent coronary angiography. The maximal luminal diameter stenosis for each major coronary artery was visually estimated by an experienced angaographer who had no knowledge of the dobutamine stress echocardiography results. Fifty percent or greater luminal narrowing was considered hemodynamically significant. The decision to perform angiography and need for revascularizatlon was made by each patient's primary physician in collaboration with the consulting cardiologist. These physicians had access to the clinical and dipyrldamole thallium data, but not to the dobutamlne stress test results Sensitivity, specificity, and positive and negative predictive values were calculated m the usual fashion. Coronary artery distributions were divided into left and right systems, with the right considered as a combination of the left circumflex and right coronary arteries for determining the concordance between the two tests. Two-tailed Student's t test was used for comparisons. Differences were considered slgmficant at p < 0 05 Results are presented as mean _+ SEM (standard error of mean). Ninety-five percent confidence intervals for proportions were calculated as previously described 27 RESULTS

A total of 41 p a t i e n t s (all m e n ) u n d e r w e n t d o b u t a m i n e stress e c h o c a r d i o g r a p h y b e f o r e t h e i r s c h e d u l e d surgery. In four patients, t h e surgery was not d e e m e d to b e indicated or was canceled b e c a u s e of underlying n o n c a r d i a c medical conditions a n d were excluded from f u r t h e r analysis. T h e r e m a i n i n g 37 p a t i e n t s f o r m e d the study cohort. T h r e e p a t i e n t s r e q u i r e d coronary revascularlzatlon b e f o r e surgery. Two of these p a t i e n t s were s c h e d u l e d for vascular surgery, w h e r e a s o n e was s c h e d u l e d for o r t h o p e d i c surgery. Five of the 34 p a t i e n t s who h a d surgery u n d e r w e n t an additional p r o c e d u r e , for a total of 39 s e p a r a t e o p e r a t i o n s p e r f o r m e d (21 vascular; 7 orthopedic; 2 thoracic; 2 abdominal; 3 neurosurgical; and 4 general). T h e m e a n d u r a t i o n of surgery was 4.7 +- 0.5 hours. Forty-nine p e r c e n t u n d e r w e n t general anesthesia, 38% epidural, a n d 13% h a d both.

STRESS ECHO VERSUS THALLIUM IMAGING

331

Pulmonary artery catheterization and intra-arterial catheters were used for lntraoperative hemodynamic monitorlng, in 19% and 78% of patients, respectively. Intraoperative infusion of intravenous nitroglycerin was used for ischemic protection in 35% of patients. Preoperative characteristics of the patients are listed in Table 1. A significant number had underlying cardiovascular risk factors. Twenty-seven percent of the patients had typical angina; 32%, a history of a previous MI; 70%, hypertension; and 38%, diabetes mellitus. There were no significant differences in the baseline characteristics between patients with comphcations versus those without. Although there were a greater number of patients with complications who had CHF, this difference was not statistically significant (p = 0.062). Five patients had cardiac complications (Table 2). Two occurred postoperatively, one had a cardmc arrest and death, whereas the second sustained an MI comphcated by CHF, and subsequent death. Two additional patients required surgical revascularization for severe CAD, and one underwent percutaneous transluminal coronary angioplasty (PTCA) for myocardial revascularizatlon. Dipyrldamole thalhum results are presented in Table 3. Twenty-one of the 32 patients (66%) without complications had abnormal dipyrldamole thallium imaging compared with 4 of the 5 patients (80%) (p = 0.52) with comphcations. Although there was a trend towards more reversible (1 2 -+ 0.4 v 0.8 ± 0.1, p = 0 28) and more total defects (1.6 + 0.2 v 1.0 _+ 0.2;p = 0.19) in patients with complications compared with those without, it was not significant. There were a higher number of patients with increased lung uptake m the group with complications (p < 0.05). No patient had any significant complications resulting from dipyrldamole thallium imaging The dobutamlne stress echocardlographlc results are shown in Table 4. Nineteen of the 37 (51%) patients had a positive dobutamine stress echocardlography. Eight of 37 patients were taking beta-blockers (none of whom had postoperative complications). There were no major complications resulting from the dobutamme stress test. Eleven Table 1, Basehne Patmnt Characteristics

Age(yrs) Heart fadure Angina, typmal Angina, atypical Prewous PTCA Prewous CABG Previous MI H/O hypertension H/O diabetes melhtus H/O chronm renal fadure H/O CVA

No COMP (N = 32)

COMP (N = 5)

656-+ 16 9% 25% 22% 9% 19% 31% 72% 34% 9% 16%

614_+49 40% 40% 0% 0% 20% 40% 60% 60% 20% 0%

NOTE None ofthesewas different atP < =0 05 Abbrewatmns PTCA, percutaneous translummal coronary ang~oplasty, CABG, coronary artery bypass grafting, MI, myocardial mfarctmn, H/O, history of, CVA, cerebrovascular acmdent, COMP, comphcatlons

Table 2. Relative Findings in Dobutamine Stress Echocardiography and Dipyridamole Thalhum Scmtigraphy in Pattents With Malor Cardiac Events Comphcahon Cardiac arrest

MI/CHF/death

LM > CABG

LM/3 vessel Disease > CABG PTCA of an 80%-90% RCA lemon

DSE

DT

Multtple wall motmn Transtent left venabnormahtms at tncular ddatatlon, basehne, fadure to anterior reverstblhty augment wtth stress Posterior, inferior, One area reversible septal mductble defect, increased tschemm lung uptake Septal, anterior, mfe- Two areas of reversriot/posterior tbfl~ty, increased mductble tschemm lung uptake Posterior and septal Fixed posterior reducible tschemm defect Inferior reducible Inferior reverspble tschemta defect

Abbrewatmns' MI, myocardial mfarctmn; CHF, congestwe heart fadure, LM, left mare lesmn, CABG, coronary artery bypass graft, PTCA, percutaneous translummal coronary angtoplasty; RCA, right coronary artery, DSE, dobutamme stress echocardmgraphy, DT, dtpyrldamole thalhum scmtNraphy, >, led to

patients had the test terminated before reaching peak dobutamlne Infusion. Reasons included angina in 7, significant decrease in blood pressure in 1, hypertension in 1, significant dobutamine-lnduced wall motion abnormahties in 2, nonsustained ventrlcular tachycardia in 1, and attainmg greater than 85% of their P M H R in 2 patients (3 patients had more than one indication for stopping early). No patient developed diagnostic ST changes during the dobutamine Infusion, although 11 patients had ventricular ectopy, including one patient who had several runs of nonsustained ventrlcular tachycardla. Forty-four percent (14/32) of patients without complications had an abnormal dobutamlne stress echocardlogram, whereas it was abnormal in all patients with complications (p = 0.02). There were no significant differences in peak heart rate, baseline ejection fraction, rate pressure product, or peak blood pressure obtained with the dobutamine infusion between the patients who had complications and those who did not. There was a trend towards a higher baseline (18.8 _+ 0.9 v 21.8 + 3.4;p = 0.25) and peak wall motion scores (21.3 -+ 1.4 v 25.8 _+ 2.4, p = 0 23) In patients with complications compared with those without complications. Eight patients had coronary angaograms to define anatomy Table 3, Dtpyridamole Thallium Data

Mean no reversible defects Total no defects % Patients with transient LV dilatation % Patients with increased lung uptake % Patients with + DT

Without COMP (N=32)

With COMP (N=5)

0 78 _+ 0 14 1 O -+ 0 17 19 3 66

1 2 _+ 0 37 1 6 _+ 0 24 40 40* 80

Abbreviations COMP, complications, LV, left ventncular, DT, d~pyrJdamole thalhum *P < 0 05

332

KONTOS ET AL

Table 4. Dobutamine Stress Echocardiography Variables Without COMP (N = 32) Maximum dobutamlne dose, ~g/kg/min

With COMP (N = 5)

37 + 1

32 _+ 5

Peak heart rate, BPM

118 --. 4

120 ± 10

% PMHR Peak blood pressure

77 ± 3

76 +- 5

response, mmHg

148 -+ 5/64 -+ 3

149 + 10/66 -+ 11

17.0 + 0.7

8 0 _+ 1.9

Rate pressure product, mmHg x bpm x 1000 LVEF (baseline) Basehne wall mutton score Peak wall motion score

50 +- 3

40 _+ 5

188 ± 09

21 8 +- 3 4

21 3 ± 1 4

25 8 ± 2 4

44

1go*

% Patients with positive DSE

Abbreviattons: bpm, beats per minute; PMHR, predicted maximum heart rate; LVEF, left ventricular election fractton; DSE, dobutamlne stress echocardlography; and COMP, complications.

based on the preoperative workup. Two patients had no significant coronary artery disease; of these one had a dipyridamole thallium scan showing a fixed defect with transient left ventricular dilatation. However, this patient had a normal dobutamine stress echocardiogram. The second patient had one reversible and one fixed defect, whereas the dobutamine stress echocardiogram demonstrated multiple areas of inducible wall motion abnormalities. He had a resting left ventricular ejection fraction of 20%, consistent with a cardiomyopathy. The four patients with multivessel disease had positive dobutamine stress echocardiograms that demonstrated inducible wall motion abnormalities in at least two coronary artery territories. In three of these patients, the dipyridamole thallium scan was consistent with multivessel disease, whereas the fourth had a dlpyridamole thallium scan showing only a fixed defect despite significant three-vessel and left main disease. Two patients had one-vessel disease (one right coronary artery and one left circumflex), both having abnormal dobutamine stress echocardiograms and diyridamole thallium imaging with abnormalities in the right coronary artery distribution. The wall motion abnormality by dobutamine stress echocardiograms and perfusion defects by dipyridamole thallium tests were concordant in 67% of myocardial segments (50/74; 20 positive, 30 negatwe). In 14 segments, the dipyridamole thallium scan was positive, whereas the dobutamine stress echocardiogram was negative; and in 10 myocardial segments, the dobutamine stress echocardiography was positive, whereas the dlpyridamole thallium scan was negative. There was similar concordance between classification of coronary artery territories as ischemic or normal in both the left and right coronary distributions. Sensitivity and positive and negative predictive values of each test were comparable (Table 5). The sensitivity (100% [95% cardiac index [CI] 56% to 100%] versus 80% [CI, 37% to 96%]), positive (26% [CL 12% to 49%] versus 15% [CI, 6% to 35%]), and negative (100% [C!, 82% to 100%] versus 93% [CI, 64% to 99%]) predictive values were not significantly different between the dobutamine stress echocardio-

graphy and dipyrldamole thallium tests. There was a trend toward higher specificity of dobutamine stress echocardiography (60% [CI, 43% to 74%]) compared with dipyridamole thallium imaging (38% [CI, 24% to 54%]); (p = 0.06). The relative risk for cardiac complications of a positive dipyridamole thallium test was 2.1 (95%, CI 0.2 to 21). In contrast, dohutamine stress echocardiography demonstrated a relatwe risk index of 6.4 (CI, 0.7 to 63), which was higher than dipyridamole thallium tests. DISCUSSION

The major findings from this observational study are twofold: (1) a negative dobutamine stress echocardiographic study implies a low risk for cardiac complications, and (2) dobutamine stress echocardiography may be as effective as dipyrldamole thallium imaging in preoperative risk assessment of patients undergoing noncardiac surgery. Identification of patients who are at higher risk for cardiac complications resulting from noncardlac surgery is important because appropriate intervention might lead to improved patient outcome. Dipyrldamole thallium testing is the most frequently used pharmacologic stress test for this purpose and is generally well accepted. Recently, dobutamine stress echocardlography has emerged as a highly sensitive and specific tool for identifying patients with CADJ 6-18A number of studies addressing the capability of dobutamine stress echocardiography to predict patients at risk for cardiac complications associated with vascular surgery have found similarily high sensitivities and negative predictive values. 19-22To the best of the authors' knowledge, this study is the first that provides a head-tohead comparison between dobutamine stress echocardiography and dipyridamole thallium imaging in the same patients before noncardiac surgery In this study, comparable sensitivity and positive and negative predictive values were observed between dobutamine stress echocardiography and dipyrldamole thalhum imaging in predicting cardiac complications. Patients having either test positive were noted to be at increased risk for cardiac comphcations. Consistent with previous reports using either dipyridamole thallium imaging or dobutamine stress echocardiography to risk stratify patients, 1°,15,19,21,22 both tests had a low positive predictive value for the prediction of cardiac complications, especially the more severe events including MI or death. The specificity of Table 5. Sensitivity, Specificity, and Positive and Negative Predictive Value of Pthall and DSE

Sensitivity Specificity Positive predictive value Negative predictive value

DT

DSE

80 38 15 93

100 60 21 96

NOTE. DT--Had evidence of redistribution, increased lung uptake, or transient left ventrlcular dilatation DSE--Had worsening wall motion with stress (re, normal to hypokmettc or hypokinetlc to akmetlc) Values were computed for the total number of operations scheduled (n = 42) Abbreviations DT, dlpyridamole thallium; DSE, dobutamrne stress echocardlography, Pthall, Persantme thallium

STRESS ECHO VERSUS THALLIUM IMAGING

dobutamine stress echocardiography was higher than that of dlpyridamole thallium imaging, although this did not quite reach statistical significance probably because of small sample size. One possible explanation for higher sensitivity for the dipyridamole thallium test is that dipyrldamole is a vasodilator that causes flow heterogeneity, 28 and thus is capable of identifying coronary artery stenoses of limited magnitude, compromising specificity of the test. Josephson et al demonstrated that minor stenoses of 40% to 50% could cause thallium perfusion defects after dipyridamole vasodllation. 29 Lesions of this magnitude, although important for identification of atherosclerotic coronary artery disease, are highly unlikely to predispose to major cardiac postoperative complications. This study demonstrates that dobutamine stress echocardiography has a high sensmvlty and negative predictive value in predicting cardiac complications, similar to that reported by other investigators. 19,2°22 Eichelberger et a122 used dobutamine stress echocardlography to. evaluate 75 patients undergoing vascular surgery and found it to have a sensitivity of 100% and a specificity of 69% in predicting cardiac complications In contrast to the authors' study, there were few hard end points, with only two MIs and no deaths in the patient population reported by Eichelberger. 22 Lalka et a119found dobutamine stress echocardiography to have a sensitivity of 92% with a specificity of only 44% in predicting cardiac complications in 60 patients studied before aortic surgery. Poldermans et al 2° investigated the use of dobutamme stress echocardiography in predicting cardiac complications in 131 patients undergoing vascular surgery. All of the 15 patients with complications (5 patients with fatal MI; 9 with unstable angina; 1 with CHF) had inducible wall motion abnormalities compared with 20/116 (21%) of the patients without preoperative complications. Only one other study has specifically analyzed variables derived from the dobutamme stress test in order to determine whether there was any additional predictive value of this test over reducible wall motion abnormalities alone. Lalka et al I9 found that the patients who underwent aortic aneurysm repair and developed wall motion abnormalities at a heart rate less than 120 BPM were more likely to have comphcations. The dobutamine stress protocol used in this study differs from that of Lalka et aP 9 m that the authors did not specifically assess the relationship between heart rate and the time of onset of lschemic response. In contrast to the study of Lalka et al, 19 in which all the procedures were aortic surgery, this patient cohort was comprised of both vascular and nonvascular procedures and thus may account for a lower risk for complications The authors found increased lung uptake m thallium imaging studies to be significantly more common in patients with cardmc complications than those without. Previous studies have shown an increased rate of complications as the number of reversible defects increases25; there was a trend towards association between higher complication rate with increasing perfusion defects in the current study, but it was not statistically significant. Because it appears that similar prognostic information is

333

obtained from both tests, the advantages and disadvantages of both need to be considered when deciding on which test to obtain. Advantages of dobutamine stress echocardiography as compared with dipyridamole thallium imaging include no exposure to radiation, shorter total test time, and almost immediate availability of the results. Resting systolic function is readily available before and after dobutamine stress echocardiography as opposed to dlpyridamole thallium scintlgraphy. Disadvantages of dobutamine stress echocardiography include its subjective nonquantitative interpretation of the echo images and its limitations because of the lack of an adequate acoustic window in a minority of patients. 3° Advantages of dlpyridamole thallium imaging include its documented high sensitivity for the diagnosis of CAD, widespread experience in its use, and a considerable database that currently exists in predicting complications related to noncardiac surgery. However, despite its high sensitivity, more widespread use has led to inclusion of patient populations at low risk for cardiac complications. From a large pool of patients, recent studies have demonstrated a reduction m predictive value of dipyridamole thallium imaging for identifying patients who actually develop complicationsY -~5 There were no serious side effects related to either test in this study, which is consistent with previous reports. 22-25.31,32 In the largest series published thus far, Mertes et a131found no serious lschemic complicanons (MI, sustained venmcular tacbycardla, or syncope) related to high-dose dobutamine infusion m 1,118 patients. Yhas compares favorably with the safety profile of dipyridamole thallium testing in which major adverse effects (nonfatal MI or acute bronchospasm) occurred m only 0.26% of patients. 32 There are a number of differences in this study as compared with others. First, the authors used at least two segmental wall motion abnormalities rather than a singlesegment wall motion abnormality as an end point for early termination of the dobutamme stress test. This may have contributed to identification of patients with multivessel CAD. The identification of these patients is important because patients with multwessel disease are at a higher risk of operative cardiac comphcations. 33 The dobutamine infusion protocol used in this study is safe, despite the potential for greater ischemic burden. At this institution, the authors have performed more than 500 dobutamine stress echocardiograms using this protocol with no significant complications. Limitations of the study include the fact that not all patients who were tested underwent cardiac catheterization; therefore, a correlation with angiographlc coronary artery anatomy is not possible. However, in the small number of pattents with anglographlc data (N = 8), the dobutamine stress echocardiography results correlated more closely with the angiographic results than did dipyndamole thallium imaging. The specificity of dipyridamole thallium imaging was somewhat lower In this study compared with others. This may be partly because of differing definitions of end points. The fact that the physicians were not blinded to the

334

KONTOS ET AL

dipyridamole thallium results may have led to more aggressive hemodynamlc monitoring and anti-lschemlc measures, thus reducing the number of actual complications. In addition, a large number (60%) of the patients with an abnormal thallium result had only one area of reversibility, which may have placed them at a lower risk for cardiac comphcations.25 Although the positive predictive value of dipyridamote thallium imaging in the current study appears somewhat low compared with earher studies, more recent investigations have shown a low incidence of serious complications in patients with positive thallium scans. In a review of the use of diyndamole thallium scintigraphy from 16 studies with a total of 1,442 patients, the likelihood ratio was consistently In the range of 2 to 3 for the prediction of cardiac death and MI after vascular surgery in patients with a positive test. 34In a recent study, Baron et al found an odds ratio of only 1 i in a patient with thallium redistribution for predicting cardiac complications in 457 patients undergoing abdominal vascular surgery. Is Despite the small sample size of patients who underwent both tests, some insights into the relative value of the two widely used noninvasive tests are available from this study. The sensitivity, specificity, and positive and negative predictive values of these two tests an this study are consistent with prewous studies using either dobutamine stress echocardaography or dipyndamole thallium imaging. I°,~5J9,21,22 The concordance between the two tests in predicting ischemia in different coronary artery territories was 67%, which is similar to the results reported by Simek et al. 35 The authors defined their complications somewhat differ-

ently from previous studies. They included significant CAD requlrmg an intervention (either PTCA or coronary artery bypass graft) as an end point. Although the need for revascularization before noncardlac surgery is debatable, the standard of practice pattern suggests that patients with significant left mare or three-vessel disease undergo coronary artery revascularlzation before any major elective surgery. 36 These patients are at a higher risk of cardiac complications compared with patients with only one- or two-vessel disease. 33 Most previous studies have also used endpoints that can be considered somewhat soft, s 1015.19,21,22 such as unstable angina and ventricular or atrial arrhythmias, which, although resulting from underlying CAD, are rarely life threatening. The authors used quantitative planar thalhum scintigraphy rather than single photon emission computer tomography (SPECT). Although planar imaging has a lower sensitivity for predicting CAD compared with SPECT, the addition of quantification increases the sensitivity of planar imaging.37 Despite these limitations, this study demonstrates that the ability of dobutamlne stress echocardiography to predict major complications of noncardiac surgery is comparable with dipyridamole thallium imaging. Patients with a positive result from either test were at higher risk of complications, whereas a negatwe test predicted a low risk for an adverse cardiac event. Further prospective studies in larger cohorts of patients may be necessary to define the most effective cardiac noninvasive test for preoperative risk stratification.

REFERENCES

l v o n Knorrmg J' Postoperatwe myocardial infarction: A prospectwe study in a risk group of surgical patients Surgery 90'55-60, 1981 2. Jamieson WRE, Janusz MT, MlyaglshlmaRT, et al: Influence of Ischemlc heart disease on early and late mortality after surgery for peripheral occluswe vascular disease Circulation 66 192-197, 1983 (suppl 1) 3. Larsen SF, Olesen KH, Jacobsen E, et al. Prediction of cardiac risk an noncardlac surgery. Eur Heart J 8'179-185, 1987 4 Cooperman M. Pflug B, Martin EW Jr. et al: Cardiovascular risk factors In patients with peripheral vascular disease. Surgery 84'505-509, 1978 5 Boucher CA, Brewster DC, Darling RC, et al Determination of cardiac risk by dtpyridamole thalhum imagingbefore peripheral vascular surgery. N Engl J Med 312.389-394, 1985 6 Pasternack PF, Imparato AM, Riles TS, et al' The value of the radlonuchde angiogram m the prediction of perioperative myocardial infarction In patients undergoing lower extremity revascularizationprocedures. Circulation 72:II-13-II-17,1985 (suppl 2) 7. Gerson MC, Hurst JM, Hertzberg VS, et al' Cardmc prognosis in noncardlac germtric surgery Ann Int Med 103:832-837, 1985 8 Raby KE, Goldman L, Creager MA, et al Correlation between preoperative lschemIa and major cardiac events after peripheral vascular surgery N Engl J Med 321 1296-1300, 1989 9. McPhai1 N, Calvin JE, Shanatmader A, et al' The use of

preoperatwe exercise testing to predict cardiac comphcatlons after arterial reconstruction J Vasc Surg 7:60-68, 1988 10 Eagle KA, Singer DE, Brewster DC, et al' Dlpyrldamole thallium scanning in patients undergoing vascular surgery. JAMA 257:2185-2189, 1987 11 Mangano DT, London MJ, Tubau JF, et al: Dlpyridamole thalhum-201 sclntlgraphy as a preoperative screening test. A reexamination of it predictive potential. Circulation 84:493-502, 1991 12. MarwickTH, Underwood DA' Dipyrldamole thallium imaging may not be a reliable screening test for coronary artery disease in patients undergoing vascular surgery. Chn Cardiol 13 14-18, 1990 13. Flelsher LA, Rosenbaum SH, Hawes AD, et al The predictive value of preoperative sdent ischemia for postoperative Ischemic cardiac events in vascular and nonlnvaslve surgery panents. Anesthesiology 73 A75, 1990 14 Bertrand M, Coriat P, Baron JF, et al Dlpyridamole thallium scan is not accurate In detecting coronary stenoses in patients undergoing abdominal aortic surgery. Anesthesiology 73'A86, 1990 15 Baron JF, Mundler O, Bertrand M, et al' Dlpyndamole thallium scmtlgraphy and gated radionuclide angiography to assess cardiac risk before abdominal aortic surgery. N Engl J Med 330'663-669, 1994 16. Segar DS, Brown SE, Sawada SG, et al Dobutamlne stress echocardlography' Correlation with coronary lesson severity as

STRESS ECHO VERSUS THALLIUM IMAGING

determined by quantitative angiography J Am Coil Cardlo119:11971202, 1992 17. Marcovitz PA, Armstrong WF. Accuracy of dobutamine stress echocardlography in detecting coronary artery disease. Am J Cardlo169 1269-1273, 1992 18. Marwick T, Willemart B, D'Hondt AM, et al" Selection of the optimal nonexercxse stress for the evaluation of ~schemlc regional myocardial dysfunction and malperfuslon: Comparison of dobutamlne and adenosine using echocardIography and 99mTcMIBI single photon emission computed tomography. Circulation 87:345-354, 1993 19 Lalka SG, Sawada SG, Dalslng MC, et al' Dobutamine stress echocardlography as a predictor of cardiac events associated with aortic surgery. J Vasc Surg 15:831-842, 1992 20. Poldermans D, Floretti PM, Forster T, et al' Dobutamme stress echocard~ography for assessment of perioperatwe car&ac risk in patients undergoing major vascular surgery Circulataon 87:1506-1512, 1993 21. Dawla-Roman VG, Waggoner AD, Slcard GA, et al Dobutamme stress echocar&ography predicts surgical outcome in patients with an aortic aneurysm and peripheral vascular disease. J Am Coll Car&ol 21'957-963, 1993 22. Eichelberger JP, Schwartz KQ, Black ER, et al' Predictive value of dobutamme echocardlography just before noncar&ac surgery Am J Cardlol 72.602-607, 1993 23. Akosah KA, Porter TR, Simon R, et al: Ischemla-induced regional wall motion abnormahty is Improved after coronary angioplasty: demonstration by dobutamxne stress echocar&ography J Am Coil Cardiol 21 584-589, 1993 24. Berger BC, Watson DD, Taylor GJ, et al: Quantltatwe thallium-201 exercise scintlgraphy for detection of coronary artery disease. J Nucl Med 22"585-593, 1981 25 Lette J, Waters D, Bernler H, et al: Preoperatwe and long-term cardiac risk assessment: Predictwe value of 23 chmcal descriptors, 7 multwariate sconng systems, and quant~tatwe dipyrxdamole lmag4ng m 360 paUents Ann Surg 216:192-204, 1992 26 Kaul S, Finkelsteln DM, Homma S, et al: Superiority of quantitative exercise thalhum-201 variables m determining longterm prognosis in ambulatory patients with chest pain: A comparison with cardiac catheterization. J Am Coll Cardiol 12:25-34, 1988

335

27. Diamond GA. Limited assurances Am J Car&ol 63"99-100, 1989 28 Fung AY, Gallagher KP, Buda AJ: The physiologic basis of dobutamine as compared with dipyridamole stress interventions m the assessment of crit]cal coronary stenosis. Circulation 76:943951, 1987 29. Josephson MA, Brown BG, Hecht HS, et al" Noninvasive detection and localization of coronary stenoses in patients: Comparison of resting dipyrldamole and exercise thallium-201 myocardial perfusion imaging. Am Heart J 103:1008-1018, 1982 30. Plcano E, Lattanzl F, Orlandini A, et al: Stress echocardlography and the human factor: The importance of being expert. J Am Coil Cardlol 17-666-669, 1991 31 Mertes H, Sawada SG, Ryan T, et al: Symptoms, adverse effects, and complications associated with dobutamIne stress echocardiography: Experience in 1118 patients. Circulation 88:15-18, 1993 32 Ranhosky A, Kempthorne-Rawson J: The safety of intravenous &pyridamole thallium myocardial perfusion imaging. Circulation 81.1205-1209, 1990 33. Foster ED, Davis KB, Carpenter JA, et al. Risk of noncardiac operation in patients with defined coronary disease: The Coronary Surgery Study (CASS) registry experience Ann Thorac Surg 41:42-50, 1986 34. Wong T, Detsky AS' Preoperative cardiac risk assessment for patients having peripheral vascular surgery Ann Int Med 116:743-753. 1992 35. Slmek CL, Watson DD, Smith WH, et al: Dipyridamole thallium-201 imaging versus dobutamme echocar&ography for the evaluation of coronary artery disease in patients unable to exercise Am J Cardlol 72'1257-1262, 1993 36. Domanskl M, Ellis S, Eagle K: Does preoperatwe coronary revascularlzation before noncardlac surgery reduce the risk of coronary events in patients with known coronary artery disease? Am J Card 75:829-831, 1995 37. Beller GA. Myocardial perfuslon imaging with thallium-201. J Nucl Med 35'674-680, 1994