Complete versus culprit lesion-only revascularisation for acute myocardial infarction (Complete Revascularisation Trialists' Collaboration): an individual patient data meta-analysis of randomised trials

Background: In patients presenting with acute coronary syndromes and multivessel coronary artery disease, the question of whether to undertake a strategy of complete revascularisation in cases in which percutaneous coronary intervention (PCI) is performed routinely on non-culprit lesions (in addition to the culprit lesion) or whether to restrict PCI only to the culprit lesion is a common dilemma. The Complete Revascularisation Trialists' Collaboration aimed to determine, based on the totality of data from randomised trials, the effect of a complete revascularisation strategy on major cardiovascular events and whether it reduces cardiovascular death. Methods: In this individual patient data meta-analysis, trials were included if they enrolled at least 250 patients, compared a complete revascularisation strategy (with PCI) to a culprit lesion-only PCI strategy, and enrolled patients presenting with acute ST-segment elevation myocardial infarction or non-ST-segment elevation myocardial infarction. To ensure that no trials were overlooked, we searched Ovid MEDLINE, Embase, and the Cochrane Central Register of Controlled Trials (CENTRAL) for randomised controlled trials published between 1996 and Sept 15, 2025. The primary outcomes were the composite of cardiovascular death or new myocardial infarction and cardiovascular death alone. Hierarchical testing of cardiovascular death alone was planned contingent on reduction in cardiovascular death or new myocardial infarction based on the prespecified alpha level of 0·04. A one-stage individual patient data meta-analysis was performed using a Cox frailty model. All-cause death was the secondary outcome; non-cardiovascular death and new myocardial infarction were additional outcomes. All analyses included all randomly assigned patients. The meta-analysis was registered in PROSPERO, CRD420251124098. Findings: Six randomised controlled trials involving 8836 individuals were included. The median age was 65·8 years (IQR 57·0-76·0), and 2088 (23·6%) patients were female and 6748 (76·4%) were male. Overall, 7768 (87·9%) patients presented with ST-segment elevation myocardial infarction and 1068 (12·1%) with non-ST-segment elevation myocardial infarction. At a median follow-up of 36·0 months (IQR 30·6-48·0), cardiovascular death or new myocardial infarction occurred in 382 (9·0%) of 4259 patients in the complete revascularisation group compared with 528 (11·5%) of 4577 patients in the culprit lesion-only group (hazard ratio [HR] 0·76 [95% CI 0·67-0·87], p<0·0001). There were 155 (3·6%) cardiovascular deaths in the complete revascularisation group compared with 209 (4·6%) in the culprit lesion-only group (HR 0·76 [95% CI 0·62-0·93], p=0·0091). All-cause death occurred in 308 (7·2%) patients in the complete revascularisation group compared with 370 (8·1%) patients in the culprit lesion-only group (HR 0·85 [95% CI 0·73-0·99], p=0·039). Non-cardiovascular death was similar between the groups (153 [3·6%] in the complete revascularisation group vs 161 [3·5%] in the culprit lesion-only group; HR 0·98 [95% CI 0·78-1·22], p=0·85). Complete revascularisation reduced new myocardial infarctions compared with culprit lesion-only PCI (255 [6·0%] vs 357 [7·8%]; HR 0·76 [95% CI 0·65-0·90], p=0·0011). Interpretation: In patients presenting with acute myocardial infarction and multivessel disease, complete revascularisation reduced the composite of cardiovascular death or new myocardial infarction as well as cardiovascular death alone compared with a culprit lesion-only PCI strategy. In addition, all-cause death was lower with complete revascularisation. These data provide the strongest and most robust evidence to date that complete revascularisation improves important cardiovascular clinical outcomes. Funding: None.