Background: the OAT found that routine late (3-28 days post-myocardial infarction) percutaneous coronary intervention (PCI) of an occluded infarct-related artery did not reduce death, reinfarction, or heart failure relative to medical treatment (MED). Angina rates were lower in PCI early, but the advantage over MED was lost by 3 years.
Methods: angina and revascularization status were collected at 4 months, then annually. We assessed whether non-protocol revascularization procedures in MED accounted for loss of the early symptomatic advantage of PCI.
Results: seven per 100 more PCI patients were angina-free at 4 months (P < .001) and 5 per 100 at 12 months (P = .005) with the difference narrowing to 1 per 100 at 3 years (P = .34). Non-protocol revascularization was more frequent in MED (5-year rate 22% vs 19% PCI, P = .05). Indications for revascularization included acute coronary syndromes (39% PCI vs 38% MED), stable angina/inducible ischemia (39% in each group), and physician preference (17% PCI vs 15% MED). Revascularization rates among patients with angina at any time during follow-up (35% of cohort) did not differ by treatment group (5-year rates 26% PCI vs 28% MED). Most symptomatic patients were treated without revascularization during follow-up (77%).
Conclusions: in a large randomized clinical trial of stable post-myocardial infarction patients, the modest benefit on angina from PCI of an occluded infarct-related artery was lost by 3 years. Revascularization was slightly more common in MED during follow-up but was not driven by acute ischemia, and almost 1 in 5 procedures were attributed to physician preference alone.