ORIGINAL RESEARCH I Comparative Effectiveness of CABG and PCI this study lack important clinical details about Potential Conflicts of Interest: Disclosures can be viewed at www patients, such as measures of left ventricular fund myocardial ischemia, and the extent of coronary Reproducible Research Statement: Study Protocol: The protocol is fully This study used nonrandomized data to compare the described in the manuscript. Statistical code. Available on request. Data effectiveness of CABG and PCl. Comparing treatments by set: Medicare data sets are available to qualified researchers but cannot be using observational data has been controversial because of released by the investigators. potential biases related to treatment selection and the lack Requests for Single Mark A. Hltaky, MD, Stanford Univer- of data on important clinical factors that might affect out sity School of Medici Redwood Building, 259 Campus Drive, We used 2 mitigate these obstacles. Fi Stanford, CA 9430 we restricted the population to patients with multivessel PCI or multivessel isolated CABG, which produced a pop- annals. org Current author addresses and author contributions are available at www ulation that more closely resembled patients eligible for randomized trials of these procedures. Second, we used propensity score matching to further control for potential References treatment selection bias, which closely balanced observed 1. BARI Investigators. The final 10-year follow-up results from the BARI ran- patient characteristics Table) that differed greatly in the 2. King SB 3rd, Kosinski AS, Guyton RA, Lembo N), Weintraub ws. Eight initial study population. This yielded an estimate of the year mortality in the Emory Angioplasty versus Surgery Trial(EAST). JAm Coll CABG-PCI HR(0.92)that was similar to the one Cardiol. 2000: 35: 1116-21[PMID: 10758949 derived from pooling randomized trial data(0.91). Our 3. Serruys Pw, Ong AT, van Herwerden LA, Sousa JE, Jatene A, Bonnier J) finding that several baseline clinical characteristics modi treatment of multivessel disease: the final analysis of the Arterial revascularization fied the comparative treatment effectiveness of CABG and Therapies Study (ARTS)randomized trial. JAm Coll Cardiol. 2005: 46-575-81 PCI is also consistent with the results of the earlier pool analysis(8). The broad agreement between the present 4. Booth J, Clayton T, Pepper J, Nugara F, Flather M, Sigwart U, et al; Sos study and the results of the pooled randomized trials sug- sus percutancous coronary intervention in patients with multivessel coronary ar. gests that carefully conducted analyses of obser rational tery disease: six-year follow-up from the Stent or Surgery Trial(SoS).Circulation. data can provide reliable estimates of real-world treatment 2008; 118 381-8.[PMID: 186069191 effectiveness 5. Hueb W, Lopes N, Gersh BJ, Soares PR, Ribeiro EE, Pereira AC, et al. Ten-year follow-up survival of the Medicine, AngiopLasty, or Surgery Study The major limitation of this study is that it is based on (MASS I): a randomized controlled dinical trial of 3 therapeutic strategies for claims data and lacks detail on many key baseline clinical multivessel coronary artery disease. Circulation. 2010: 122: 949-57. [PMID: characteristics and information on medications and behav- 207331021 6. Rodriguez AE, Baldi J, Fernandez Pereira C, Navia J, Rodriguez Alemparte G s during subsequent follow-up. These limitations may M, Delacasa A, et al: ERACI II Investigators. Five- year follow-up of the Argen- be particularly relevant to our analyses of tobacco use and heart failure, which seem to modify the CABG-PCi treat- surgery in patients with multiple vessel disease(ERACI ID). J Am Coll Cardiol. ment effect but may be correlated with other unobserved 2005;46:582-8PMID:16098419 clinical characteristics. In addition, the claims data used 7. Serruys PW, Morice MC, Kappetein AP, Colombo A, Holmes DR, Mack 1, et al; SYNTAX Investigators. Percutaneous coronary intervention versus in this study do not contain information on symptom severity or on the functional status or quality of life of 2009, 360: 961-72.(PMID: 19228612 coronary artery disease. N Engl JMed. Patients. 8. Hlatky MA, Boothroyd DB, Bravata DM, Boersma E, Booth J, Brooks This study provides strong evidence that clinical char- interventions for multivessel disease: a collaborative analysis of individual patient acteristics modify the comparative effectiveness of CABG data from ten randomised trials. Lancet. 2009: 373: 1190-7. [PMID: 19303634 and PCI on mortality, especially for diabetes, which had 9. Parsons LS. Reducing bias in a propensity score matched-pair sample using been previously shown (8). Our study suggests that addi- greedy matching techniques. Presented at 26th Annual SAS Users Group Inter- ational Conference, 22-25 April 2001, Long Beach, California. ease,and heart failure, also modify the CABG-PCI treat- ED, Kolm P, et al. Comparative effectiveness of revascularization strategie ment effect. These variations in comparative effectiveness N Engl Med.2012;366:1467-76. [PMID:22452338 underscore the need to per 11. Farkouh ME, Domanski M, Sleeper LA, Siami FS, Dangas G, Mack M, dations for coronary revascularization among patients with et al; FREEDOM Trial l treatment recommen for multivessel revascularization in patients with diabetes. N Engl JMed. 2012: 367: 2375-84. [PMID: 231213231 coronary disease. 12. van Domburg RT, Meeter K, van Berkel DE, Veldkamp RF, van Herw- erden LA, Bogers AJ. Smoking cessation reduces mortality after coronary artery From Stanford University School of Medicine, Stanford; Kaiser Per bypass surgery: a 20-year follow-up study. J Am Coll Cardiol. 2000: 36: 878-83 nente Northern California, Oakland; and University of California, San [PMID:10987614] Francisco, San Francisco, California 13. Hasdai D, Garratt KN, Grill DE, Lerman A, Holmes DR Jr. Effect of smoki on the lon vascularization. N Engl J Med. 1997: 336: 755-61. [PMID: 90526531 pport: By grant HL099872 from the National Heart, Crouse JR 3rd, Hagaman AP. Smoking cessation in relation to cardiac procedures. Am) Epidemiol. 1991: 134: 699-703. [PMID: 1951275] 21 May 2013 Annals of Internal Medicine Volume 158. Number 10 www.annals.orgthis study lack important clinical details about these patients, such as measures of left ventricular function, myocardial ischemia, and the extent of coronary artery disease. This study used nonrandomized data to compare the effectiveness of CABG and PCI. Comparing treatments by using observational data has been controversial because of potential biases related to treatment selection and the lack of data on important clinical factors that might affect out￾comes. We used 2 steps to mitigate these obstacles. First, we restricted the population to patients with multivessel PCI or multivessel isolated CABG, which produced a pop￾ulation that more closely resembled patients eligible for randomized trials of these procedures. Second, we used propensity score matching to further control for potential treatment selection bias, which closely balanced observed patient characteristics (Table) that differed greatly in the initial study population. This yielded an estimate of the overall CABG–PCI HR (0.92) that was similar to the one derived from pooling randomized trial data (0.91). Our finding that several baseline clinical characteristics modi- fied the comparative treatment effectiveness of CABG and PCI is also consistent with the results of the earlier pooled analysis (8). The broad agreement between the present study and the results of the pooled randomized trials sug￾gests that carefully conducted analyses of observational data can provide reliable estimates of real-world treatment effectiveness. The major limitation of this study is that it is based on claims data and lacks detail on many key baseline clinical characteristics and information on medications and behav￾iors during subsequent follow-up. These limitations may be particularly relevant to our analyses of tobacco use and heart failure, which seem to modify the CABG–PCI treat￾ment effect but may be correlated with other unobserved clinical characteristics. In addition, the claims data used in this study do not contain information on symptom severity or on the functional status or quality of life of patients. This study provides strong evidence that clinical char￾acteristics modify the comparative effectiveness of CABG and PCI on mortality, especially for diabetes, which had been previously shown (8). Our study suggests that addi￾tional factors, particularly smoking, peripheral arterial dis￾ease, and heart failure, also modify the CABG–PCI treat￾ment effect. These variations in comparative effectiveness underscore the need to personalize treatment recommen￾dations for coronary revascularization among patients with multivessel coronary disease. From Stanford University School of Medicine, Stanford; Kaiser Perma￾nente Northern California, Oakland; and University of California, San Francisco, San Francisco, California. Grant Support: By grant HL099872 from the National Heart, Lung, and Blood Institute. Potential Conflicts of Interest: Disclosures can be viewed at www .acponline.org/authors/icmje/ConflictOfInterestForms.do?msNumM12 -1564. Reproducible Research Statement: Study protocol: The protocol is fully described in the manuscript. Statistical code: Available on request. Data set: Medicare data sets are available to qualified researchers but cannot be released by the investigators. Requests for Single Reprints: Mark A. Hltaky, MD, Stanford Univer￾sity School of Medicine, HRP Redwood Building, 259 Campus Drive, Stanford, CA 94305-5405; e-mail, hlatky@stanford.edu. Current author addresses and author contributions are available at www .annals.org. References 1. BARI Investigators. The final 10-year follow-up results from the BARI ran￾domized trial. J Am Coll Cardiol. 2007;49:1600-6. [PMID: 17433949] 2. King SB 3rd, Kosinski AS, Guyton RA, Lembo NJ, Weintraub WS. Eight￾year mortality in the Emory Angioplasty versus Surgery Trial (EAST). J Am Coll Cardiol. 2000;35:1116-21. [PMID: 10758949] 3. Serruys PW, Ong AT, van Herwerden LA, Sousa JE, Jatene A, Bonnier JJ, et al. Five-year outcomes after coronary stenting versus bypass surgery for the treatment of multivessel disease: the final analysis of the Arterial Revascularization Therapies Study (ARTS) randomized trial. J Am Coll Cardiol. 2005;46:575-81. [PMID: 16098418] 4. Booth J, Clayton T, Pepper J, Nugara F, Flather M, Sigwart U, et al; SoS Investigators. Randomized, controlled trial of coronary artery bypass surgery ver￾sus percutaneous coronary intervention in patients with multivessel coronary ar￾tery disease: six-year follow-up from the Stent or Surgery Trial (SoS). Circulation. 2008;118:381-8. [PMID: 18606919] 5. Hueb W, Lopes N, Gersh BJ, Soares PR, Ribeiro EE, Pereira AC, et al. Ten-year follow-up survival of the Medicine, Angioplasty, or Surgery Study (MASS II): a randomized controlled clinical trial of 3 therapeutic strategies for multivessel coronary artery disease. Circulation. 2010;122:949-57. [PMID: 20733102] 6. Rodriguez AE, Baldi J, Ferna´ndez Pereira C, Navia J, Rodriguez Alemparte M, Delacasa A, et al; ERACI II Investigators. Five-year follow-up of the Argen￾tine randomized trial of coronary angioplasty with stenting versus coronary bypass surgery in patients with multiple vessel disease (ERACI II). J Am Coll Cardiol. 2005;46:582-8. [PMID: 16098419] 7. Serruys PW, Morice MC, Kappetein AP, Colombo A, Holmes DR, Mack MJ, et al; SYNTAX Investigators. Percutaneous coronary intervention versus coronary-artery bypass grafting for severe coronary artery disease. N Engl J Med. 2009;360:961-72. [PMID: 19228612] 8. Hlatky MA, Boothroyd DB, Bravata DM, Boersma E, Booth J, Brooks MM, et al. Coronary artery bypass surgery compared with percutaneous coronary interventions for multivessel disease: a collaborative analysis of individual patient data from ten randomised trials. Lancet. 2009;373:1190-7. [PMID: 19303634] 9. Parsons LS. Reducing bias in a propensity score matched-pair sample using greedy matching techniques. Presented at 26th Annual SAS Users Group Inter￾national Conference, 22–25 April 2001, Long Beach, California. 10. Weintraub WS, Grau-Sepulveda MV, Weiss JM, O’Brien SM, Peterson ED, Kolm P, et al. Comparative effectiveness of revascularization strategies. N Engl J Med. 2012;366:1467-76. [PMID: 22452338] 11. Farkouh ME, Domanski M, Sleeper LA, Siami FS, Dangas G, Mack M, et al; FREEDOM Trial Investigators. Strategies for multivessel revascularization in patients with diabetes. N Engl J Med. 2012;367:2375-84. [PMID: 23121323] 12. van Domburg RT, Meeter K, van Berkel DF, Veldkamp RF, van Herw￾erden LA, Bogers AJ. Smoking cessation reduces mortality after coronary artery bypass surgery: a 20-year follow-up study. J Am Coll Cardiol. 2000;36:878-83. [PMID: 10987614] 13. Hasdai D, Garratt KN, Grill DE, Lerman A, Holmes DR Jr. Effect of smoking status on the long-term outcome after successful percutaneous coronary revascularization. N Engl J Med. 1997;336:755-61. [PMID: 9052653] 14. Crouse JR 3rd, Hagaman AP. Smoking cessation in relation to cardiac procedures. Am J Epidemiol. 1991;134:699-703. [PMID: 1951275] Original Research Comparative Effectiveness of CABG and PCI 734 21 May 2013 Annals of Internal Medicine Volume 158 • Number 10 www.annals.org