Lack of Benefit for Routine Functional Testing Early after Coronary Artery Bypass Graft Surgery: Results from the ROSETTA-CABG R

More than 500,000 coronary artery bypass graft (CABG) procedures are performed each year in North America.¹ Although CABG is a durable procedure, the development of graft closure can occur within the first year following CABG leading to angina, myocardial infarction (MI) and death. In order to identify early graft closure, some physicians employ routine functional testing in all their post-CABG patients. This approach may lead to the early identification and treatment of graft closure with either intensified medical therapy or revascularization. Either of these treatments may lead to a reduction in clinical events. In contrast, other physicians employ a selective or clinically-driven strategy in which only patients who develop symptoms undergo functional testing. However, selective functional testing will not identify patients with asymptomatic graft closure, and this approach may be associated with a higher rate of follow-up clinical events. Guidelines issued by the American College of Cardiology and the American Heart Association suggest that functional testing should not be performed routinely in all patients following percutaneous coronary intervention (PCI).² However, there are no parallel guidelines regarding the use of functional testing post-CABG. For this reason, we initiated the Routine versus Selective Exercise Treadmill Test after Coronary Artery Bypass Graft (ROSETTA-CABG) Registry. The purpose of this registry was to examine the patterns of use and outcomes of routine and selective functional testing strategies. The primary objective of this report is to examine the effects of routine post-CABG functional testing on follow-up cardiac procedures and clinical events. Methods Study centers and patient population. During the period between May 30, 1999 and May 30, 2002, a total of 408 patients were enrolled in the Routine Versus Selective Exercise Treadmill Testing after Coronary Artery Bypass Graft Surgery (ROSETTA-CABG) Registry (Appendix A). Twelve-month follow up was complete by May 2003, with 10 patients (2.5%) lost to follow up. Because of early death after CABG, an additional 3 patients could not be classified into the routine or selective functional testing groups. Therefore, data from a total of 395 patients are included in this analysis. In each institution involved in the study, the Research and Ethics Committee approved the study, and written informed consent was obtained from each patient. Patients were enrolled in the registry immediately after their CABG, but before they were discharged from the hospital. Patients were included if they had a successful CABG defined as one in which all ischemic areas were thought to be revascularized and the patient had no major in-hospital events (e.g., perioperative MI, stroke, need to return to the operating room). Inclusion criteria included an isolated CABG surgery. Patients who had valve surgery, aortic repair, etc. were not eligible. Exclusion criteria were as follows: 1) participation in conflicting clinical studies; 2) contraindications to repeat cardiac procedures (cardiac catheterization, percutaneous coronary intervention, repeat CABG); 3) contraindications or inability to undergo follow-up functional testing; 4) future revascularization procedure planned; 5) pregnancy or likelihood of becoming pregnant; 6) medical condition with a prognosis Baseline clinical and procedural characteristics. Clinical and procedural data were collected at baseline. This information included demographic and clinical characteristics such as comorbid conditions, prior procedures and medical therapy. Data recording the primary indication for the CABG as well as the numbers and types of bypass grafts were also collected. Twelve-month follow up. Clinical follow-up data were obtained 12 months after CABG. The research nurse at the clinical center conducted a telephone interview with each patient to determine if they had undergone any post-CABG functional testing and to document if any endpoints occurred. In addition to the telephone interview, the nurse contacted the patient’s cardiologist (and other physicians if necessary) for 12-month follow-up information. For every functional test during the 12-month follow up period, the nurse collected the documentation of the type and results of the testing. Similarly, the nurse also obtained documentation regarding secondary endpoints (clinical events and cardiac procedures). If tests or events occurred at other institutions, that institution was contacted to obtain the appropriate endpoint documentation. A functional test was defined as one of the following: exercise treadmill testing, stress echocardiography, stress nuclear perfusion imaging, and others such as positron emission tomographic (PET) imaging. The functional testing strategy was defined by the reason for the first functional test after the CABG. The patient was considered to have had routine functional testing if the first functional test was performed as a routine follow up. In contrast, the patient was categorized in the selective group if the first functional test was performed for a clinical indication, or if they underwent no functional testing. Statistical analysis. Continuous data are presented as the mean ± standard deviation, and dichotomous data are presented as percentages. The initial data analysis examined the clinical characteristics and outcomes of the patients undergoing routine and selective functional testing strategies. We then performed univariate and multivariate analyses to determine potential predictors of the prespecified composite endpoints at 12 months after CABG. Dichotomous and continuous variables were examined using logistic regression modeling. We previously found that the use of post-CABG functional testing is largely unrelated to baseline clinical and procedural characteristics, and is chiefly determined by the clinical center at which the patient underwent the CABG.³ Therefore, we included the clinical center at which the patient had the CABG in the multivariate analysis. If a particular variable was associated with follow-up clinical events by univariate analysis with a p-value ? 0.10, it was entered into a multivariate logistic regression model. Potential interaction between variables was also examined. Finally, we performed analyses to examine antianginal medication use (beta blockers, calcium channel blockers and nitrates) at discharge and at 12 months post-CABG with respect to functional testing results among the patients who underwent functional testing during the 12 months. All statistical tests were two-tailed, and a p-value ? 0.05 was considered to be statistically significant. Results Clinical and procedural characteristics. Although patients were not randomized to the routine and selective functional testing strategies, there were few differences in baseline clinical and procedural characteristics between patients in the two groups (Table 1). Patients in the selective functional testing strategy group were more likely to have undergone prior PCI. In contrast, patients in the routine functional testing strategy group were more likely to be men and to have radial artery and saphenous vein bypass grafts. In both the routine and selective functional testing groups, the vast majority of patients had multivessel CABG with the use of left internal mammary artery (LIMA) grafts and with a mean of 3.5 grafts per patient. Functional testing. Among the 395 patients, 111 (28%) were observed to undergo a routine functional testing strategy. In contrast, 284 patients (72%) underwent a selective functional testing strategy in which they either had no functional testing during the 12-month follow up (88%), or they underwent one or more functional tests in which the first was performed for a clinical indication (12%). Patients in the routine functional testing group were more likely to undergo two or more functional tests than patients in the selective group (21.6% vs. 1.4%, respectively; p Relationship between functional testing and clinical outcomes. To control for clinical and procedural differences between the routine and the selective functional testing groups, regression modeling was performed. We first performed univariate analyses, which demonstrated an association between a number of clinical and procedural variables and subsequent clinical events (unstable angina, MI and death) (Table 2). We then performed a multivariate logistic regression analysis to control for clinical and procedural characteristics. We found that routine functional testing had a persistent independent association with a reduction in the composite clinical event rate (odds ratio = 0.10; 95% confidence interval: 0.01–0.78; p = 0.028). We also found an increase in the occurrence of clinical events depending on the clinical center at which the patient had the CABG, the patient’s history of cerebrovascular disease and the pre-admission need for insulin. We then examined functional test results as well as procedural and medical therapy in patients undergoing both functional testing strategies. We wanted to determine whether an increase in procedures or an increase in medications was responsible for the association between routine functional testing and a reduction in the composite clinical event rate. Functional test results were found to be associated with subsequent clinical and procedural events. Patients who had a positive functional test were more likely to experience a clinical event and to undergo a procedure during the year following CABG than those who had an indeterminate or negative test. In contrast, patients who had a negative functional test during the 12-month follow up were much less likely to experience any events or procedures. For patients who had an indeterminate functional test, we found no significant association between the results of the functional test and subsequent clinical and procedural events. We then performed similar analyses, separating the patients who underwent functional testing into the routine and selective functional testing groups with respect to functional testing results (Table 3). The majority of clinical and procedural events occurred in the selective group who had a positive functional test (clinical events = 33%; procedural events = 40%). Event rates were more common in the selective group regardless of the results of functional testing or even if patients did not undergo functional testing. In contrast, events and procedures were very rare among the 111 patients in the routine functional testing group. Patients with a positive or indeterminate test had no events and underwent no procedures (0%), while only 1 patient with a negative test had an event (6.3%) and underwent a procedure (6.3%). These data suggest that procedures are not responsible for the protective effect. Routine functional testing results appear to have had little impact on treatment management. We then examined the timing of the procedures and events with respect to functional testing results. The majority of events occurred before functional testing (70%), whereas the majority of procedures were performed after functional testing (60%). In the routine functional testing group, the only event occurred after a negative functional test, and all of the procedures were also performed after a negative test. In the selective functional testing group, 4 of 5 (80%) events occurred before or the same day as a positive functional test, and 6 of 11 (55%) procedures were performed before or the same day as a positive test. Furthermore, for patients in the selective group who had an indeterminate functional test, 2 of 3 (67%) events occurred before or the same day as the test, and all procedures were performed after functional testing. Finally, for patients in the selective group who had a negative functional test, 1 of 2 (50%) had unstable angina before testing. Finally, we found that functional test results had little impact on prescription patterns of antianginal medication (beta blockers, calcium channel blockers and nitrates). Both beta blocker and calcium channel blocker use did not increase from discharge to 12 months following a positive test (N = 25; 84% vs. 80%; p = NS; 16% vs. 16%; p = NS), while nitrate use increased seven-fold (4% vs. 28%; p = 0.03). However, following a negative test (N = 100), beta blocker use decreased modestly (85% vs. 70%; p = 0.01), while both calcium channel blocker and nitrate use were unchanged (30% vs. 20%; p = NS; 4% vs. 6%; p = NS, respectively). Finally, following an indeterminate test (N = 21), antianginal medication use was unchanged (data not shown). Discussion The objective of the ROSETTA-CABG Registry was to examine the effects of post-CABG functional testing strategies on the use of follow-up cardiac procedures and clinical events. We found a nine-fold difference in the intensity of functional testing during the 12-month follow up period between patients who underwent the routine and selective testing strategies. We also found that routine functional testing was associated with an increase in catheterization rates, but not with an increase in the use of cardiac revascularization procedures. In addition, routine testing was associated with a substantial reduction in cardiac events. However, functional testing results had little association with subsequent events and procedures. These results suggest that the protective effect associated with routine testing is likely due to physicians sending lower-risk patients for routine functional testing and not to the early identification and treatment of patients with impending or complete graft closure. Because event rates were extremely low in the routine group, routine functional testing appeared to be of limited value. Very few prospective studies have examined the clinical utility of routine functional testing after CABG. Thus, the results from our study are significant, since there is little consensus on the appropriate use of functional testing early after CABG. As demonstrated in this study, a routine functional testing strategy is employed by many physicians. Other physicians employ a more conservative approach, employing functional testing only if symptoms develop. To our knowledge, this is the first study to examine this issue prospectively. Previous studies. Although it has been shown that routine functional testing after PCI does not lead to a reduction in clinical events,4 we found no studies that examined the strategies of routine and selective functional testing among post-CABG patients. Previous studies mainly assessed the utility of functional testing as a prognostic and diagnostic method; however, the strategies in which functional testing is used have not been evaluated. For example, several studies examined it as a method of risk stratification following CABG^5–21 while other studies assessed functional testing as a means of identifying postoperative graft stenosis.²² Many earlier studies examined the ability of functional tests to accurately risk-stratify asymptomatic patients following CABG.^5–11 Functional testing early after CABG was found to be a good predictor of future cardiac events and an effective tool for the risk stratification of patients. Arruda et al., for example, suggested that exercise echocardiography had incremental value in identifying high-risk patients for clinical events (p = 0.02) and cardiac death (p 5–7 Similarly, Miller et al. found that exercise thallium-201 SPECT imaging could be used to stratify their 411 post-CABG patients into low- and high-risk subgroups within 2 years after CABG.⁸ Similarly, Lauer et al. concluded that significant numbers of asymptomatic post-CABG patients have positive functional tests and are at high risk for future clinical events (11% vs. 4%; p = 0.0002).⁹ In contrast, Elhendy et al. suggested that routine functional testing in patients with a low pretest probability does not appear to be cost-effective and thus should not be recommended.¹⁰ A number of earlier studies also investigated the utility of post-CABG functional testing to noninvasively identify graft stenosis or progression of arterial disease in native vessels. We previously conducted a meta-analysis in which we found that exercise treadmill testing is a poor diagnostic test at identifying the presence of graft stenosis (sensitivity of 45% and specificity of 82%).²² In contrast, the use of stress myocardial perfusion imaging and stress echocardiography had substantially improved sensitivities and specificities (sensitivity of 68% and 86%; specificity of 84% and 90%, respectively). In summary, none of these studies have found functional testing to be useful routinely for protective purposes and have determined whether a routine functional testing strategy significantly impacts on subsequent procedural and clinical events, functional status and quality of life. Study limitations. Several potential limitations of our study should be noted. First, determining the primary reason for functional testing can often be complex. Thus, some misclassification of patients into the routine and selective functional testing groups may have occurred. However, misclassification should lead to the finding of no differences between groups, while we found substantial differences between patients undergoing routine and selective functional testing. A second potential limitation is the observational design of our study: patients were not randomized to the routine or the selective groups. In addition, different strategies were used at different centers with respect to the types of functional tests used, as well as the timing of functional tests. Importantly, the majority of functional tests performed were exercise treadmill tests with no ventricular imaging. It is possible that the use of ventricular imaging would have increased the early detection of graft closure and thus reduced the occurrence of a greater number of subsequent clinical events. Conclusions The objective of the ROSETTA-CABG Registry was to examine the effects of post-CABG functional testing on the use of follow-up cardiac procedures and clinical events. Our results showed a nine-fold difference in the use of functional testing between the routine and selective groups. This difference was associated with an increase in the use of cardiac catheterization, but not with an increase in the rate of cardiac revascularization procedures. However, we also found a strong association between routine functional testing and a low frequency of follow-up clinical events. This association is likely to be due to physicians sending lower risk-patients for routine functional testing. This association is not attributable to an increase in medication and/or procedure rates. Because event rates are low in patients undergoing routine functional testing after CABG, this strategy does not appear to be warranted.

1. American Heart Association. Heart Disease and Stroke Statistics – 2003 Update. Dallas, Texas: American Heart Association, 2002. 2. Gibbons RJ, Balady GJ, Bricker JT, et al. ACC/AHA 2002 guideline update for exercise testing: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on Exerciser Testing). 2002. American College of Cardiology web site. Available at: www.acc.org/clinical/guidelines/exercise/dirIndex.htm. 3. Eisenberg MJ, Okrainec K, Wou K, et al. Utility of routine functional testing after coronary artery bypass graft surgery: Results from the ROSETTA-CABG Study. J Am Coll Cardiol 2004;(Suppl A)43;273A. 4. Eisenberg MJ, Blankenship JC, Huynh T, et al. Evaluation of routine functional testing after percutaneous coronary intervention. Am J Cardiol 2004;93:744–747. 5. Arruda AM, McCully RB, Oh JK, et al. Prognostic value of exercise echocardiography in patients after coronary artery bypass surgery. Am J Cardiol 2001;87:1069–1073. 6. Arruda AM, Das MK, Roger VL, et al. Prognostic value of exercise echocardiography in 2,632 patients > or = 65 years of age. J Am Coll Cardiol 2001;37:1036–1041. 7. Arruda-Olson AM, Juracan EM, Mahoney DW, et al. Prognostic value of exercise echocardiography in 5,798 patients: Is there a gender difference? J Am Coll Cardiol 2002;39:625–631. 8. Miller TD, Christian TF, Hodge DO, et al. Prognostic value of exercise thallium-201 imaging performed within 2 years of coronary artery bypass graft surgery. J Am Coll Cardiol 1998;31:848–854. 9. Lauer MS, Lytle B, Pashkow F, et al. Prediction of death and myocardial infarction by screening with exercise thallium testing after coronary artery bypass grafting. Lancet 1998:615–622. 10. Elhendy A, Shub C, McCully RB, et al. Exercise echocardiography for the prognostic stratification of patients with low pretest probability of coronary artery disease. Am J Med 2001;111:18–23. 11. Elhendy A, Arruda AM, Mahoney DW, Pellikka PA. Prognostic stratification of diabetic patients by exercise echocardiography. J Am Coll Cardiol 2001;37:1551–1557. 12. Weiner DA, Ryan TJ, Parsons L, et al. Prevalence and prognostic significance of silent and symptomatic ischemia after coronary artery bypass surgery: A report from the Coronary Artery Surgery Study (CASS) randomized population. J Am Coll Cardiol 1991;18:343–348. 13. Yli-Mayry S, Huikuri HV, Airaksinen KEJ, et al. Usefulness of a postoperative exercise test for predicting cardiac events after coronary artery bypass grafting. Am J Cardiol 1992;70:56–59. 14. Wallis JB, Supino PG, Borer JS. Prognostic value of left ventricular ejection fraction response to exercise during long-term follow-up after coronary artery bypass graft surgery. Circulation 1993;88(Part 2):99–109. 15. Nallamothu N, Johnson JH, Bagheri B, et al. Utility of stress single photon emission computed tomography (SPECT) perfusion imaging in predicting outcome after coronary artery bypass grafting. Am J Cardiol 1997;80:1517–1521. 16. Cottin Y, Rezaizadeh K, Touzery C, et al. Long-term prognostic value of 201Tl single-photon emission computed tomographic myocardial perfusion imaging after coronary stenting. Am Heart J 2001;141:999–1006. 17. Vanzetto G, Ormezzano O, Fagret D, et al. Long-term additive prognostic value of thallium-201 myocardial perfusion imaging over clinical and exercise stress test in low to intermediate risk patients: Study in 1,137 patients with 6-year follow-up. Circulation 1999;100:1521–1527. 18. Dubach P, Froelicher V, Klein J, Detrano R. Use of the exercise test to predict prognosis after coronary artery bypass grafting. Am J Cardiol 1989;63:530–533. 19. Borges-Neoto S, Shaw IJ, Kesler K, et al. Usefulness of serial radionuclide angiography in predicting cardiac death after coronary artery bypass grafting and comparison with clinical and cardiac catheterization data. Am J Cardiol 1997; 9:851–855. 20. Palmas W, Bingham S, Diamond GA, et al. Incremental prognostic value of exercise thallium-201 myocardial single-photon emission computed tomography late after coronary artery bypass surgery. J Am Coll Cardiol 1995;25:403–409. 21. Ritchie JL, Bateman TM, Bonow RO, et al. Guidelines for clinical use of cardiac radionuclide imaging: Report of the American College of Cardiology/American Heart Association Task Force on Assessment of Diagnostic and Therapeutic Cardiovascular Procedures (Committee on Radionuclide Imaging), developed in collaboration with the American Society of Nuclear Cardiology. J Am Coll Cardiol 1995;25:521–547. 22. Chin AS, Goldman LE, Eisenberg MJ. Functional testing after coronary artery bypass graft surgery: A meta-analysis. Can J Cardiol 2003;19:802–808.