Associations between the PARIS ischaemic and bleeding risk scores and 1-year mortality in an all-comers Australian national PCI cohort

Abstract

Abstract Background Prediction scores can complement bedside clinical risk assessment in patients undergoing PCI. Purpose We sought to validate the PARIS ischaemic and bleeding risk scores in a national cohort of all-comer PCI patients in Australia, and to examine impact on 1-year mortality. Methods We retrospectively analysed the GenesisCare Outcomes multicentre Registry of prospectively collected data in patients undergoing PCI from 2008–2020. The ischaemic and bleeding scores were assessed in separate cohorts. Patients were included if all defining variables were available. The ischaemic integer score included diabetes, acute coronary syndrome, current smoking, GFR <60ml/min, prior PCI or CABG, and patients were grouped according as low (0 to 2), intermediate (3 or 4), and high (≥5) thrombotic risk. The bleeding integer score included age, body mass index, current smoking, GFR <60ml/min, anaemia and discharge on triple therapy after PCI, and patients were grouped as low (0 to 3), intermediate (4 to 7) and high (≥8) bleeding risk. One-year ischaemic and in-hospital bleeding endpoints were examined. Adjusted risk of 1-year mortality was assessed accounting for age, sex, family history, smoking, vascular disease, presentation status, hypertension, diabetes, obesity and dyslipidaemia. Results A total of 16,961 patients were included in the ischaemic risk cohort and 6431 patients in the bleeding risk cohort. The patient distributions by risk category for both scores are shown in the figure. Increasing ischaemic risk was associated with greater incidence of in-hospital (0.2% vs. 0.5% vs. 0.5%, p=0.01), and 1-year death (0.7% vs. 2.1% vs. 3.1%). Increasing bleeding risk was associated with greater in-hospital death (0.0% vs. 0.3% vs. 0.9%, p<0.01) and bleeding (0.6% vs. 0.9% vs. 3.2%, p<0.01), and greater 1-year death (0.3% vs. 1.4% vs. 2.8%, p<0.01). At 1-year compared to the low bleeding risk group, high risk patients had lower use of aspirin, P2Y12 inhibitors and statins. Similarly, compared to low ischaemic risk group, high-risk patients had lower use of aspirin and statins but higher use of P2Y12 inhibitors. With both risk scores, compared to low-risk patients, high-risk patients had significantly greater adjusted risk of 1-year mortality (See Figure). Conclusions The PARIS ischaemic and bleeding risk scores were associated with greater risk of in-hospital and medium-term adverse outcomes after PCI to 1-year. Prospective use of the scores can assist in guiding adherence to secondary prevention and optimal anti-thrombotic strategies to improve overall outcomes. Funding Acknowledgement Type of funding sources: None.