Purpose: We assessed the value of commonly used risk scores CHADS2 and CHA2DS2-VASc in predicting the outcome in patients with atrial fibrillation (AF) undergoing percutaneous coronary intervention (PCI). CHADS2 and CHA2DS2-VASc scores have been developed to estimate the thromboembolic risk in AF and to guide antithrombotic therapy, but their value in predicting the risk of thrombotic or bleeding complications of AF patients after PCI is unknown.

Methods: AFCAS is an observational, multicenter, prospective registry including patients (n=929) with AF referred for PCI. For analysis the risk scores were divided into low, intermediate and high; for CHADS 0-1, 2-3 and >3, and for CHAD2DS2-VASc 1-2, 3-4 and 5-9, respectively. Peri- and post-PCI antithrombotic treatment was administered at the operator's discretion. Primary study endpoints were 1) all cause mortality; 2) MACCE including all cause mortality, myocardial infarction, repeat revascularization, stent thrombosis, TIA, stroke, other arterial embolism; 3) bleeding complications; and 4) any harmful event (MACCE and bleeding complications) at 12 months follow-up.

Results: The patients were distributed as follows: CHADS2 low 29.5%, intermediate 55.2%, high 15.3%; CHA2DS2-VASc low 9.6%, intermediate 45.9%, high 44.5%. Patients with high CHA2DS2-VASc scores suffered more often MACCE compared to those with low scores (24.5% vs. 12.4%, p=0.014) and a high score was an independent predictor of MACCE (OR 1.52, 95% CI 1.08-2.12, p=0.015). A high CHADS2 score was not a significant predictor for MACCE or thromboembolism compared to those with low score (26.1% vs. 18.2%, p=ns). There was no difference in minor or major bleeding according to CHADS2 (11.7% vs. 12.6%, p=ns for minor and 10.6% vs. 12.0%, p=ns for major, respectively) or CHA2DS2-VASc score (13.5% vs. 12.4%, p=ns for minor and 7.9% vs. 12.3%, p=ns for major, respectively).

Conclusions: CHA2DS2-VASc score predicts adverse thrombotic outcomes after PCI in this high risk AF population and it provides better predictive value in comparison to CHADS2 score. No association with bleeding complications was seen with either score.