Routine Revascularization Versus Initial Medical Therapy for Stable Ischemic Heart Disease

BACKGROUND: Revascularization is often performed in patients with stable ischemic heart disease. However, whether revascularization reduces death and other cardiovascular outcomes is uncertain. METHODS: We conducted PUBMED/EMBASE/Cochrane Central Register of Controlled Trials searches for randomized trials comparing routine revascularization versus an initial conservative strategy in patients with stable ischemic heart disease. The primary outcome was death. Secondary outcomes were cardiovascular death, myocardial infarction (MI), heart failure, stroke, unstable angina, and freedom from angina. Trials were stratified by percent stent use and by percent statin use to evaluate outcomes in contemporary trials. RESULTS: Fourteen randomized clinical trials that enrolled 14 877 patients followed up for a weighted mean of 4.5 years with 64 678 patient-years of follow-up fulfilled our inclusion criteria. Most trials enrolled patients with preserved left ventricular systolic function and low symptom burden, and excluded patients with left main disease. Revascularization compared with medical therapy alone was not associated with a reduced risk of death (relative risk RR, 0.99 95% CI, 0.90-1.09). Trial sequential analysis showed that the cumulative z-curve crossed the futility boundary, indicating firm evidence for lack of a 10% or greater reduction in death. Revascularization was associated with a reduced nonprocedural MI (RR, 0.76 95% CI, 0.67-0.85) but also with increased procedural MI (RR, 2.48 95% CI, 1.86-3.31) with no difference in overall MI (RR, 0.93 95% CI, 0.83-1.03). A significant reduction in unstable angina (RR, 0.64 95% CI, 0.45-0.92) and increase in freedom from angina (RR, 1.10 95% CI, 1.05-1.15) was also observed with revascularization. There were no treatment-related differences in the risk of heart failure or stroke. CONCLUSIONS: In patients with stable ischemic heart disease, routine revascularization was not associated with improved survival but was associated with a lower risk of nonprocedural MI and unstable angina with greater freedom from angina at the expense of higher rates of procedural MI. Longer-term follow-up of trials is needed to assess whether reduction in these nonfatal spontaneous events improves long-term survival.