A Meta-analysis of Randomized Controlled Trials Comparing Percutaneous Coronary Intervention With Medical Therapy in Stable Angina Pectoris

• Published in: Canadian journal of cardiology 2013-04, Vol.29 (4), p.472-482
• Main Authors: Thomas, Sabu, MD, FRCPC, FACC, Gokhale, Rohit, MD, Boden, William E., MD, FACC, FAHA, Devereaux, P.J., MD, PhD, FRCPC
• Format: Article
• Language: English
• Subjects: Angina, Stable - drug therapy; Angina, Stable - surgery; Angina, Stable - therapy; Angioplasty, Balloon, Coronary; Bias; Cardiovascular; Confounding Factors (Epidemiology); Evidence-Based Medicine; Humans; Odds Ratio; Percutaneous Coronary Intervention; Randomized Controlled Trials as Topic; Research Report - standards; Risk Assessment; Vasodilator Agents - therapeutic use
• ISSN: 0828-282X; 1916-7075
• DOI: 10.1016/j.cjca.2012.07.010

Abstract There continues to remain uncertainty regarding the effect of percutaneous coronary intervention (PCI) vs medical therapy in patients with stable angina. We therefore performed a systematic review and study-level meta-analysis of randomized controlled trials of patients with stable angina comparing PCI vs medical therapy for each of the following individual outcomes: all-cause mortality, cardiovascular (CV) mortality, myocardial infarction (MI), and angina relief. We used 8 strategies to identify eligible trials including bibliographic database searches of MEDLINE, PubMed, EMBASE, and the Cochrane Controlled Trials Registry until November 2011. Two independent reviewers undertook decisions about study eligibility and data abstraction. Data were pooled using a random effects model. Ten prospective randomized controlled trials fulfilled our eligibility criteria and they included a total of 6752 patients. We did not detect differences between PCI vs medical therapy for all-cause mortality (663 events; relative risk [RR], 0.97 [confidence interval (CI), 0.84-1.12]; I2 = 0%), CV mortality (214 events; RR, 0.91 [CI, 0.70-1.17]; I2 = 0%), MI (472 events; RR, 1.09 [CI, 0.92-1.29]; I2 = 0%), or angina relief at the end of follow-up (2016 events; RR, 1.10 [CI, 0.97-1.26]; I2 =85%). PCI was not associated with reductions in all-cause or CV mortality, MI, or angina relief. Considering the cost implication and the lack of clear clinical benefit, these findings continue to support existing clinical practice guidelines that medical therapy be considered the most appropriate initial clinical management for patients with stable angina.