Objectives: This study was designed to determine: 1) whether a routine invasive (RI) strategy reduces the long-term frequency of cardiovascular death or nonfatal myocardial infarction (MI) using a meta-analysis of individual patient data from all randomized studies with 5-year outcomes; and 2) whether the results are influenced by baseline risk.

Background: Pooled analyses of randomized trials show early benefit of routine intervention, but long-term results are inconsistent. The differences may reflect differing trial design, adjunctive therapies, and/or limited power. This meta-analysis (n = 5,467 patients) is designed to determine whether outcomes are improved despite trial differences.

Methods: Individual patient data, with 5-year outcomes, were obtained from FRISC-II (Fragmin and Fast Revascularization during Instability in Coronary Artery Disease), ICTUS (Invasive Versus Conservative Treatment in Unstable Coronary Syndromes), and RITA-3 (Randomized Trial of a Conservative Treatment Strategy Versus an Interventional Treatment Strategy in Patients with Unstable Angina) trials for a collaborative meta-analysis. A Cox regression analysis was used for a multivariable risk model, and a simplified integer model was derived.

Results: Over 5 years, 14.7% (389 of 2,721) of patients randomized to an RI strategy experienced cardiovascular death or nonfatal MI versus 17.9% (475 of 2,746) in the selective invasive (SI) strategy (hazard ratio [HR]: 0.81, 95% confidence interval [CI]: 0.71 to 0.93; p = 0.002). The most marked treatment effect was on MI (10.0% RI strategy vs. 12.9% SI strategy), and there were consistent trends for cardiovascular deaths (HR: 0.83, 95% CI: 0.68 to 1.01; p = 0.068) and all deaths (HR: 0.90, 95% CI: 0.77 to 1.05). There were 2.0% to 3.8% absolute reductions in cardiovascular death or MI in the low- and intermediate-risk groups and an 11.1% absolute risk reduction in highest-risk patients.

Conclusions: An RI strategy reduces long-term rates of cardiovascular death or MI and the largest absolute effect in seen in higher-risk patients.