Validation of prediction tools for GI bleeding in patients on dual anti-platelet therapy after percutaneous coronary intervention

Abstract

Background and aims: The management of dual anti-platelet therapy after percutaneous coronary intervention (PCI) and GI bleeding (GIB) remains a clinical dilemma. We sought to identify predictors of GIB and recurrent bleeding and to determine whether recurrent bleeding increases the risk of major adverse cardiovascular events (MACEs).

Methods: In this single-center retrospective study, patients undergoing PCI were identified. The primary and secondary endpoints were GIB at 180 days and recurrent bleeding or MACE at 365 days. Logistic regression was used to identify predictors of GIB and recurrent bleeding. Cox proportional hazards modeling was used to determine whether recurrent bleeding can predict a MACE.

Results: Five hundred thirty-six patients were included. On multivariable analysis, PCI for acute coronary syndrome was associated with a 95% increased odds of GIB (P < .001). The P2Y₁₂ inhibitor was continued in >90% of patients, which trended toward significance for recurrent bleeding (P < .10). The HAS-BLED score (Hypertension, Abnormal renal and liver function, Stroke, Bleeding tendency or predisposition, Labile INRs, Elderly, Drugs), including a labile international normalized ratio and prior major bleeding, was strongly associated with recurrent bleeding (P ≤ .009). Recurrent bleeding was associated with a 115% increased risk of MACEs (P = .02). We derived a novel risk score, named the SIGE score ([S]TEMI at PCI, having a labile [I]NR at PCI, index [G]IB within 180 days of PCI, and previous precatheterization [E]ndoscopy within 6 months), to predict recurrent bleeding at 365 days with a high predictive accuracy (area under the curve, .773; 95% confidence interval, .702-.845).

Conclusions: The SIGE score may help to predict recurrent bleeding, which was shown to be associated with an increased risk of MACEs. Further external validation is needed.