Background: Little is known about the impact of bleeding site on mortality after percutaneous coronary intervention. The aim of this study was to assess the impact of access and non-access site bleeding within 30 days after percutaneous coronary intervention on mortality.
Methods and results: This study represents a pooled patient-level analysis of 14 180 patients recruited in 7 randomized trials. Access and non-access site bleeding were assessed using the Bleeding Academic Research Consortium criteria. The primary outcome was 1-year mortality. Follow-up was complete in 97.5% of the patients. There were 414 deaths within the first year after percutaneous coronary intervention: 44 deaths among patients with access site bleeding, 60 deaths among patients with non-access site bleeding, and 310 deaths among patients without bleeding (Kaplan-Meier estimates of mortality, 4.5%, 10.0%, and 2.5%, respectively; adjusted hazard ratio, 1.72 [95% confidence interval, 1.19-2.47] for access site bleeding versus no bleeding; hazard ratio, 2.78 [2.00-3.86] for non-access site versus no bleeding). The inclusion of non-access site bleeding (the absolute and relative integrated discrimination improvement, 0.005 and 8.9%; P=0.031) but not of access site bleeding (the absolute and relative integrated discrimination improvement, 0.0015 and 2.7%; P=0.084) was associated with an improvement of the discriminatory power of multivariable model for mortality prediction.
Conclusions: Both access and non-access site bleeding events occurring within 30 days of a percutaneous coronary intervention are independently associated with an increased risk of 1-year mortality. Non-access site bleeding is a stronger correlate of mortality than access site bleeding, and it improves the discriminatory power of models for mortality prediction.
Keywords: hemorrhage; mortality; percutaneous coronary intervention.