Predicting Mortality and Periprocedural Adverse Events in Patients Undergoing Complex and Higher Risk Percutaneous Coronary Intervention.

Background: The estimation of periprocedural risk in patients undergoing complex and higher risk percutaneous coronary intervention (CHIP-PCI) is challenging. In this study we aimed to compare the performance of 3 different risk-scoring systems---National Cardiovascular Data Registry (NCDR) CathPCI, British Cardiovascular Intervention Society (BCIS)-CHIP, and Blue Cross Blue Shield of Michigan Cardiovascular Consortium 2 (BMC2)---in predicting mortality and major adverse cardiac and cerebrovascular events (MACCEs) in a population of patients undergoing CHIP-PCI at a specialized centre.

Methods: The primary endpoints of this study were in-hospital mortality and MACCE. Score performances were evaluated based on discrimination ability (area under the curve [AUC] method) and calibration.

Results: The study included 4287 patients, with rates of in-hospital mortality and MACCE of 2.5% and 3.3%, respectively. The BMC2 score (AUC = 0.93, 95% confidence interval [CI] 0.91-0.95; and AUC = 0.87, 95% CI 0.84-0.90) outperformed both the NCDR CathPCI (AUC = 0.89, 95% CI 0.85-0.92, P = 0.002; and AUC = 0.83, 95% CI 0.80-0.87, P = 0.009) and BCIS-CHIP (AUC = 0.81, 95% CI 0.76-0.85, P < 0.001; and AUC = 0.78, 95% CI 0.73-0.82, P < 0.001) scores for the prediction of both mortality and MACCE. Accuracy metrics showed a similar pattern. In subanalyses, the superior performance of the BMC2 model was even more pronounced in chronic total occlusion and multivessel PCI. Calibration analysis revealed good predictive accuracy but highlighted a trend toward overestimation in low-risk patients for the BMC2 score.

Conclusions: The BMC2 risk score demonstrated superior performance in predicting in-hospital mortality and MACCEs in a large cohort of patients undergoing CHIP-PCI.