Bleeding complications in patients undergoing percutaneous coronary intervention in coronary artery bypass grafts: the impact of clinical presentation, access site and selection bias

Abstract

Abstract Background Percutaneous coronary intervention (PCI) in patients with prior coronary artery bypass graft (CABG) surgery is commonly done via the femoral approach. Relatively few cases are done via the radial approach. Purpose To assess bleeding complications in patients with either radial/brachial or femoral access when undergoing graft PCI, saphenous vein graft or internal mammary artery (IMA) graft. Method Data was collected for quality control on all patients that underwent PCI between 2010 and 2018 at 6 tertiary care hospitals in our system. Outcomes were classified and reported in accordance with the Bleeding Academic Research Consortium (BARC) criteria. BARC Type 4 (CABG-related) and BARC Type 5 (fatal) bleeds were excluded. Major bleeding was defined as BARC Type ≥3. Major adverse cardiac and cerebrovascular events (MACCE) was the combined endpoint of hospital death; post procedural myocardial infarction; cerebrovascular events (ischemic and hemorrhagic) and major bleeding complications. Due to the high intraclass correlation of procedures within patients, only the first procedure within the study period was selected from each patient for analysis. The two-sample t-test, Wilcoxon rank sum test, chi-square or Fisher's exact test were used as appropriate. Results A total of 1153 of 1196 patients who underwent graft intervention had BARC outcome data. Of these only 81 (7%) was via radial/brachial access. Baseline clinical characteristics are shown in the Table. A higher percentage of BARC ≥3 patients presented with NSTEMI/STEMI or had IMA graft intervention. Access site was not associated with bleeding complications. However, MACCE, post procedural heart failure (HF) and post procedural myocardial infarction (MI) rates were higher in patients who had femoral access (Figure). Conclusions In a large contemporary dataset of post CABG patients undergoing graft interventions: 1) the majority of patients (93%) underwent graft PCI via femoral access; 2) clinical presentation and acuity was associated with more severe bleeding complications; 3) access site did not seem to be associated with bleeding complications (possible selection bias); and 4) there was a trend towards higher MACCE rates in patients who had a femoral access. Additional analysis will be done to further investigate our findings. Funding Acknowledgement Type of funding sources: None.