ObjectiveTo determine the association of D-dimer to albumin ratio (DAR) with major adverse cardiovascular events (MACE) after percutaneous coronary intervention (PCI) in ischaemic heart failure patients with diabetes mellitus.DesignA retrospective observational cohort study.SettingSingle centre in Beijing, China, conducted at one of the largest cardiology centres in China.ParticipantsFrom June 2017 to June 2019, 3707 patients with heart failure and concomitant multiple vessel disease undergoing elective PCI were screened. A total 1021 of patients were enrolled after exclusion and the follow-up period was up to 36 months.Primary and secondary outcome measuresThe MACE was the primary measured outcome. The secondary outcomes were all-cause mortality, non-fatal myocardial infarction and any revascularisation.MethodsThese participants were grouped according to DAR tertiles. The cumulative incidence functions, Cox regression, restricted cubic spline and receiver operating characteristic curves were used to determine the association between DAR and outcomes. The subgroup analysis was also performed.ResultsAfter follow-up, MACE occurred in 404 (39.6%) participants. The cumulative hazards curve manifested significant differences in MACE, all-cause mortality and any revascularisation (log-rank test: all p<0.001). In adjusted models, DAR was an independent risk factor of MACE (tertile 2: HR 1.82, 95% CI 1.37 to 2.42; tertile 3: HR 1.74, 95% CI 1.28 to 2.36) and all-cause mortality (tertile 2: HR 2.04, 95% CI 1.35 to 3.11; tertile 3: HR 1.89, 95% CI 1.20 to 2.98). The optimal cut-off of DAR was 1.2. In the stratified analysis, sex, age, hypertension, hypercholesterolaemia, total revascularisation and any interfered vessel did not affect the independent predictive ability.ConclusionHigher DAR was independently associated with MACE and all-cause mortality after PCI in ischaemic heart failure patients with diabetes mellitus.