Performance of the HAS-BLED high bleeding-risk category, compared to ATRIA and HEMORR2HAGES in patients with atrial fibrillation: a systematic review and meta-analysis

Atrial fibrillation (AF) patients' major bleeding risk should be evaluated through risk scores such as HAS-BLED, HEMORR2HAGES or ATRIA. These scores were validated in independent studies with different methods. Therefore, we aimed to review and estimate the value added by ATRIA and HEMORR2HAGES compared to HAS-BLED. Medline and Cochrane Library (July 2013) were searched, as well as reviews and references of obtained articles. We looked for studies reporting data for diagnostic accuracy of HAS-BLED with any of HEMORR2HAGES or ATRIA scores, concerning Major Bleeding events. We determined the sensitivity, specificity, and diagnostic odds ratio (DOR) of ATRIA and HEMORR2HAGES compared to HAS-BLED within the same studies. Random effects meta-analysis was performed in order to derive diagnostic accuracy estimates. Heterogeneity was assessed through I (2) test. Six studies fulfilled inclusion criteria. Five studies evaluated simultaneously HAS-BLED and HEMORR2HAGES. Sensitivity, specificity, and DOR were respectively 0.53 (0.52-0.54), 0.65 (0.65-0.65) and 2.11 (1.91-2.35) for HAS-BLED, and 0.27 (0.26-0.27), 0.89 (0.89-0.89) and 2.90 (2.77-3.04) for HEMORR2HAGES. Four studies compared HAS-BLED with ATRIA. Sensitivity, specificity, and DOR were respectively 0.41 (0.35-0.48), 0.78 (0.76-0.79) and 2.22 (1.08-4.55) for HAS-BLED, and 0.23 (0.17-0.29), 0.91 (0.90-0.91) and 1.98 (1.29-3.03) for ATRIA. The 'high-risk' categories of the evaluated major bleeding-risk scores are not sensitive. HAS-BLED, due to its sensitivity (compared to other scores) and ease to apply, is recommended for the assessment of AF patients' major bleeding risk.