Author + information
- Received January 22, 1997
- Revision received July 1, 1997
- Accepted July 10, 1997
- Published online November 1, 1997.
- Alfred F. Parisi, MD, FACCA,*,
- Pamela M. Hartigan, PhDB,
- Edward D. Folland, MD, FACCC,
- for the ACME Investigators1
- ↵*Dr. Alfred F. Parisi, Chief, Cardiology, The Miriam Hospital, 164 Summit Avenue, Providence, Rhode Island 02906.
Objectives. We sought to evaluate the prognostic ability of cardiac exercise stress tests in predicting cardiac mortality and morbidity in a low risk group of patients with established coronary artery disease (CAD).
Background. Although previous studies have demonstrated the superior value of stress nuclear cardiac scintigraphy in the prognosis of patients with CAD, none of these studies have focused on patients with a proven angiographic low risk profile (i.e., single- and double-vessel CAD).
Methods. Three hundred twenty-eight patients with documented single- and double-vessel disease were treated by random assignment to percutaneous transluminal coronary angioplasty or medical therapy in the Angioplasty Compared to Medicine (ACME) trial. Six months after randomization, maximal symptom-limited exercise tests were performed with electrocardiography (n = 300) and thallium scintigraphy (n = 270). Patients were followed up for a minimum of 5 years thereafter.
Results. A reversible thallium perfusion deficit documented after 6 months of either therapy was associated with an adverse mortality outcome (18% mortality rate with a reversible thallium perfusion defect and 8% mortality rate with no reversible thallium perfusion deficit, p = 0.02). Moreover, an important mortality gradient was demonstrated in relation to the number of reperfusing defects (0 = 7%, 1 to 2 = 15%, >3 = 20%, p = 0.04). Exercise electrocardiography did not predict this mortality outcome.
Conclusions. A reversible thallium perfusion deficit demonstrated 6 months after medical therapy or coronary angioplasty is a valuable prognostic marker in patients with angiographically documented single- and double-vessel disease and is superior to exercise electrocardiography in this regard.
A number of reports have indicated that planar thallium scintigraphic imaging can predict survival and adverse cardiac events in patients being evaluated for chest pain [1–20]. These studies have been done in patient groups with a broad spectrum of risk profiles. In some of these studies, the patients did not have coronary angiography. In other studies in which angiography was performed, patients proved to have anywhere from no to three-vessel or even left main coronary artery disease (CAD) [5, 9, 10]. This analysis was undertaken because the Veterans Affairs–sponsored Angioplasty Compared to Medicine (ACME) trial presented a unique opportunity to evaluate outcomes of a highly defined group of prospectively followed patients previously characterized by exercise testing with thallium scintigraphy—all of whom had chronic stable angina and a low risk profile on angiography.
All patients in ACME were required to have single- or double-vessel disease with ejection fractions >30% at cardiac catheterization. The initial treatment strategy proceeded along two well defined lines to render patients angina-free with either percutaneous transluminal coronary angioplasty (PTCA) or optimized medical management with triple-drug therapy. All patients were asked to return 6 months after their random assigned treatment for follow-up exercise electrocardiography with thallium scintigraphy; all patients were then followed for a minimum of 5 years thereafter. The specific intents of this report were to evaluate whether observations with thallium scintigraphy performed after 6 months of these predefined treatment strategies have prognostic import in this selected low risk group and to compare the thallium study results with those of exercise electrocardiography.
1.1 ACME Protocol
The protocol and methods involved in ACME have been presented elsewhere in detail . Pertaining to this report, the methods are described briefly below.
The ACME study was designed to evaluate whether PTCA was superior to optimized medical therapy in relieving angina in patients with ejection fractions >30% who had single- and double-vessel disease. A total of 328 patients were randomized to one or the other of these therapies. More specifically, the study entailed a clinical trial of 212 patients with subtotally obstructive (70% to 99% stenosis) single-vessel disease and a pilot evaluation of 101 patients with double-vessel disease , as well as 15 patients with totally occlusive single-vessel disease randomized according to the identical protocol.
The highlights of the study protocol are outlined in Fig. 1. After randomization and implementation of PTCA or medical management, patients were seen monthly. Recurrent angina was considered an indication to either repeat PTCA in PTCA-assigned patients or to intensify medical management in medically assigned patients until all three classes of antianginal drugs (nitrates, beta-blockers and calcium antagonists) were used in optimal doses to provide relief of angina. At the end of 6 months of therapy, all patients returned for a final exercise test with thallium scintigraphy. This 6-month exercise test was performed on the optimized medical regimen for the patients who were assigned to medical treatment; in contrast, because the goal of ACME was to assess the effects of angioplasty per se as a treatment strategy, no antianginal drugs were allowed in the preceding 24 h for those patients assigned to PTCA therapy. The outcomes of the patients who received this 6-month exercise study form the basis for this report.
1.2 Exercise Electrocardiography and Thallium Scintigraphy
All exercise electrocardiograms were performed according to the modified Bruce protocol (initial stage: 5% elevation, 1.7 mph) and proceeded as a symptom-limited maximal test. Approximately 1 min before the end of exercise, 2.5 to 3.0 mCi of intravenous thallium-201 chloride was injected; patients were imaged in three planes (anterior, left anterior oblique and lateral) immediately after the conclusion of exercise as well as 3 h thereafter. Planar thallium scintigrams were qualitatively scored (grades 0 to 2, where 0 = absent activity; 1 = definitely reduced but not absent; and 2 = normal) in nine segments (three segments in each of three views) according to a well validated protocol at a central core laboratory. This was done by individuals who had no knowledge of the patients’ characteristics or the nature of the treatments they received. A response was considered positive for a reversible defect when an exercise-induced area of hypoperfusion in any segment improved by one or more grades in the delayed rest study.
Exercise electrocardiographic (ECG) responses were evaluated by computer algorithm, where a positive response was considered 1-mm horizontal or downsloping ST segment depression 0.08 s after the J point persisting for at least 15 s, which reverted to baseline after exercise. After the 6-month exercise study, local investigators were allowed to undertake patient treatment according to their best judgment, which included their own interpretation of the test results, but they had no knowledge of the central core laboratory evaluations.
1.3 Follow-Up Procedures
All patients were followed for a minimum of 60 months after randomization, initially (first 36 months) by the study assistants at each of the eight participating Veterans Affairs Medical Centers and later (all remaining months) through a centralized telephone follow-up interview conducted under the auspices of the West Haven Veterans Affairs Coordinating Center.
Survival outcomes that were recorded included all-cause mortality and cardiovascular mortality (as best as could be determined from medical records, autopsy and death certificates and information reported by the decedents’ family). Two of the authors (A.F.P. and E.D.F.), who had no knowledge of the exercise study results, made these determinations independently; any disagreements were resolved by mutual consensus. In addition, nonfatal myocardial infarctions, subsequent hospital admissions for unstable angina and subsequent cardiac revascularization procedures were ascertained and verified through hospital records.
1.4 Statistical Analyses
Categoric variables were compared using the chi-square test or the Fisher exact test, and continuous variables were compared using the Student ttest . Survival analyses were based on time to failure methods, including Kaplan-Meier estimation, log-rank testing and proportional hazard regression models . These last models were used to estimate relative risks. All reported p values herein are two-sided.
2.1 Stress Test Outcomes and Survival
The mean study follow-up time was 59.1 and 61.1 months for PTCA and medical therapy, respectively; the median times (with 95% confidence interval [CI]) were 64.7 (57.8 to 68.4) and 65.8 (60.1 to 67.6) months, respectively (p = NS). Of the 328 randomized patients, 300 returned after 6 months for an exercise ECG; in three tests, the ECG recordings were not adequate for interpretation. Of these remaining 297 tests, 270 were performed with thallium scintigraphy. There were a total of 43 deaths occurring after the exercise tolerance test (ETT), 36 of which occurred in patients whose 6-month ETT involved thallium scintigraphy. Of the 297 exercise tests with interpretable ECG studies, 153 were considered positive and 24 (16%) of these 153 patients died. One hundred forty-four of the exercise ECG responses were negative, and of these, 19 patients (13%) died. Of the 270 thallium studies, 205 were interpreted as abnormal, having either fixed (n = 59) or reversible (n = 146) defects (Table 1). There were 32 (16%) deaths in those with abnormal tests (n = 205) compared with four deaths in the 65 patients (6%) with normal tests (p = 0.06). Six (10%) of the deaths occurred in the 59 patients with fixed defects; the majority of the deaths (26 of 32) occurred in patients with reversible defects, which showed the highest mortality (26 [18%] of 146) of any test result by either modality (p = 0.05).
A further analysis of those with reversible thallium defects as opposed to any other result on thallium scintigraphy is shown in Table 2. Of the thallium studies, 146 patients showed reversible defects, 26 (18%) of whom died as compared with 124 patients who did not have reversible defects, 10 (8%) of whom died (p = 0.02) (Table 2top). Kaplan-Meier survival plots of these patient groups segregated by exercise electrocardiography and thallium scintigraphy are shown in Figs. 2 and 3. ⇓⇓Of particular note in Fig. 2is the continuously widening separation of survival curves associated with negative and positive thallium studies. At 3 years the proportion of surviving patients who were thallium negative was 97% and thallium positive was 94%; by 5 years these values were 93% and 82%, respectively. When mortality was tracked by the patient’s initial random assignment to medical therapy or PTCA, a strong trend remained for reversible defects detected by thallium scintigraphy to be associated with mortality (Table 2bottom).
A mortality gradient was also demonstrable in relation to the number of reversible defects. There was a 7% (10 of 124) mortality rate in patients with no reversible defects; mortality increased to 15% (11 of 72) in those patients with one or two reversible defects and was highest at 20% (15 of 74) in those with three or more reversible defects (p = 0.04).
We attempted to define the exact cause of death (as described in the Methods section) of those 36 patients who had thallium exercise scintigraphy at 6 months. In two instances, information was too sketchy to define an exact cause. Of the remaining 34 deaths, 13 were deemed to be noncardiovascular. Of the remaining 21 cardiovascular deaths, 15 occurred in patients with reversible defects and 6 in patients without reversible defects.
2.2 Nonfatal Cardiac Events
Table 3lists the number of nonfatal myocardial infarctions, hospital admissions for unstable angina and late (after the 6-month ETT) revascularizations in patients with positive and negative exercise ECG and thallium scintigraphic studies. There was no significant difference in the distribution of these nonfatal events between patients with positive studies and those with negative studies, irrespective of the specific type of stress test. This also remained true whether the initial patient assignment was to PTCA or medical therapy.
2.3 Comparability of Patients With and Without Reversible Perfusion Defects
The baseline characteristics of the patients with and without reversible defects are listed in Table 4. The clinical risk profile of these patients is identical in terms of age, underlying disease, current smoking status, lipid profile and more. By 6 months, cholesterol and low density lipoprotein cholesterol had decreased in all patients, but were not different between those with and those without reversible defects (Table 4). However, there was a greater proportion of revascularized patients with no reperfusion defects; this latter group had a greater duration on treadmill exercise than the group with reversible defects. In contrast, a greater proportion of medically assigned patients had reversible defects (Table 4).
2.4 Predictors of Survival
Lifetable regression analyses evaluating risk factors pertaining to survival in this patient group are shown in Table 5. Univariate regression analyses showed that age, ejection fraction, diabetes, reversible thallium defect and current smoking had the strongest individual relations to subsequent survival. When these five variables were combined in a multivariate model, the highest relative risks were observed in patients with diabetes, a reversible thallium defect and current smoking.
3.1 Findings in Relation to Characteristics of the ACME Study Group
The findings reported herein support and extend multiple previous observations that stress thallium scintigraphy has important prognostic value in patients with established CAD. The patients in ACME were randomized to two distinct initial treatment strategies. All the patients had a low risk profile in comparison to the previously evaluated group of patients with CAD. None had an acute coronary syndrome; all completed a maximal treadmill exercise test while off of all medications in order to enter the study. None had left main or three-vessel disease and none had a previous revascularization procedure at randomization. Our thallium studies were performed after all of the patients had received standardized treatment along two acceptable lines (i.e., either maximal medical therapy to relieve angina in the medical cohort or PTCA, as well as repeat PTCA if needed), with the goal of achieving complete freedom from angina in the PTCA cohort. Exercise thallium scintigraphy proved superior to exercise electrocardiography in predicting subsequent mortality—the majority of which was cardiovascular. When evaluated concomitantly with clinical risk factors, only diabetes and current smoking emerged as predictors of comparable magnitude.
3.2 Previous Studies of Exercise Thallium Scintigraphy and Patient Outcomes
The published reports focusing primarily on planar thallium imaging in patients with known or suspected CAD were reviewed by Brown in 1991 and more recently by Heller and Brown in 1994 and an American Heart Association/American College of Cardiology Task Force . Multiple independent studies involving more than 6,000 patients have shown that a perfusion deficit, in particular, a reversible deficit on planar exercise thallium scintigraphy, is the dominant predictor of subsequent cardiac events, including cardiac death, nonfatal myocardial infarction and late coronary revascularization. However, of the more than 6,000 patients represented in these reviews, <25% had had a previous cardiac catheterization. Only two of these studies involved follow-up periods of more than 5 years [9, 11].
Kaul et al. reported on 299 ambulatory patients with chest pain followed for a mean duration of 4.6 years, studied at the University of Virginia. At cardiac catheterization, 93 of these patients had no identifiable CAD, 77 had three-vessel disease and 5 left main disease. The remaining 124 had single- and double-vessel disease. The best predictors of subsequent outcome were the number of diseased vessels and the number of reversible defects on exercise thallium scintigraphy. Another large study involving cardiac catheterization data of 204 patients studied for a mean of 5.3 years was also performed by Kaul et al. at the Massachusetts General Hospital. Again, the patient group included individuals with triple-vessel disease. Exercise thallium scintigraphy had a slightly better predictive ability than cardiac catheterization. Of the 20 deaths reported, 15 were in patients with reversible defects.
In 1993, Brown and Rowen reported on 75 patients with negative thallium studies, 60 of whom (80%) had single- and double-vessel disease. At a mean 2-year follow-up, there were no deaths and only one nonfatal myocardial infarction in this cohort. Our survival analysis of 124 patients with no redistributing defects (Fig. 2) shows similarly excellent survival at 2 years, by which time only one of these patients had died. Previously, Iskandrian et al. reported on a series of 781 patients with suspected or known CAD and found that the total number of perfusion defects was the single most powerful predictor of adverse events (cardiac death and nonfatal myocardial infarction). Our finding of a significant increase in mortality in relation to the total number of reversible defects is consistent with this finding.
No previous study has exclusively focused on patients with proven CAD who had a low risk profile clinically and angiographically and then were prospectively followed. Our findings thus extend previous observations supporting the value of thallium scintigraphy as a predictor of subsequent mortality in such low risk patients.
3.3 Study Limitations
Our conclusions are applicable to men >60 years of age and who have the clinical and angiographic risk profile defined earlier. The ACME protocol required use of antianginal medication for medically assigned patients on the 6-month exercise study; such medications were proscribed for the 24 h preceding this test in the PTCA-assigned patients. Nevertheless, treatment assignment did not confer a survival advantage or disadvantage when evaluated by regression analysis (Table 5). Single-photon emission computed tomographic thallium scintigraphy has now replaced planar thallium imaging in most centers. This modality has comparable, if not superior, predictive power to that of planar thallium imaging for determining prognosis in patients with CAD [10, 30–36], particularly when a reversible defect is elicited [20, 30, 32, 35]. Finally, our study reports all-cause mortality. A number of patients in the 60-year-old age group, when followed for 5 or more years, will die of cancer. Nevertheless, the majority of deaths (21 of 34) in this group were directly attributable to cardiovascular disease; of these 21 deaths, there were over two times (15 vs. 6) more cardiovascular deaths in patients with reversible defects.
Exercise thallium scintigraphy is a valuable predictor of subsequent mortality in men with a low angiographic risk profile who present with stable coronary syndromes. In the ACME study, the prognostic power of exercise thallium scintigraphy was superior to that of exercise ECG findings and comparable to diabetes and current smoking as an adverse marker for subsequent survival.
↵fn1 This study was supported by the Cooperative Studies Program of the Department of Veterans Affairs, Washington, D.C. and was presented in part at the 69th Scientific Sessions of the American Heart Association, New Orleans, Louisiana, November 1996.
- Angioplasty Compared to Medicine (trial)
- coronary artery disease
- electrocardiogram, electrocardiographic
- exercise tolerance test
- percutaneous transluminal coronary angioplasty
- Received January 22, 1997.
- Revision received July 1, 1997.
- Accepted July 10, 1997.
- The American College of Cardiology
- Brown KA,
- Boucher CA,
- Okada RD,
- et al.
- Ladenheim ML,
- Pollack BH,
- Rozanski A,
- et al.
- Kaul S,
- Lilly DR,
- Gasho JA,
- et al.
- Staniloff HM,
- Forrester JS,
- Berman DS,
- Swan HJC
- Iskandrian AS,
- Chae SC,
- Heo J,
- Stanberry CD,
- Wasserleben V,
- Cave V
- Gill JB,
- Ruddy TD,
- Newell JB,
- Finkelstein DM,
- Strauss HW,
- Boucher CA
- Fleg JL,
- Girstenblith G,
- Zonderman AB,
- et al.
- Wackers FJ,
- Russo DJ,
- Russo D,
- Clements JP
- Heo J,
- Thompson WO,
- Iskandrian AS
- Brown KA
- Pollock SW,
- Abbott RD,
- Boucher CA,
- et al.
- Folland ED,
- Parisi AF,
- Hartigan P
- Snedecor GW,
- Cochran WG
- Cox DR,
- Oakes D
- AHA/ACC Task Force on Assessment of Diagnostic and Therapeutic Cardiovascular Procedures, Committee on Radionuclide Imaging, developed in collaboration with the American Society of Nuclear Cardiology
- Johnson J,
- Pancholy S,
- Wasserleben V,
- et al.
- Berman DS,
- Hachamovitch R,
- Kiat H,
- et al.
- Hachamovitch R,
- Berman DS,
- Kiat H,
- et al.
- Marie P-Y,
- Danchin N,
- Durand JF,
- et al.
- Machecourt J,
- Longere P,
- Fagret D,
- et al.