Determinants of complications with recombinant factor VIIa for refractory blood loss in cardiac surgery

Abstract

Recombinant factor VIIa (rFVIIa) is being used for refractory, excessive blood loss (EBL) after cardiac surgery, but its safety for this indication is not known. The unadjusted and risk-adjusted adverse event (AE) rates were compared between 114 consecutive cardiac surgical patients who received rFVIIa for refractory EBL and 541 concurrent patients who developed EBL but did not receive rFVIIa. Similarly, timing of rFVIIa therapy was assessed by dichotomizing rFVIIa patients based on median number of red blood cell (RBC) units received before therapy. The measured AE was a composite of death, stroke, renal failure, myocardial infarction, and major vein thrombosis. For risk adjustment, logistic regression models for this outcome were constructed using known predictors of AEs. The median RBC units transfused before rFVIIa therapy was eight. The AE rates in the untreated, early (< or = 8 U), and late (> 8 U) treated patients were 24% (129/541), 30% (20/66), and 60% (29/48). The risk-adjustment model included total RBC units, pump time, weaning difficulty, gender, weight, and age. The unadjusted and adjusted AE odds ratios (OR) in the treated vs untreated groups were 2.41 [confidence interval (CI) 1.58-3.67; P < 0.0001] and 1.04 (CI 0.60-1.81; P = 0.9). In the rFVIIa group, the adjusted AE OR was lower in the early treated group (OR 0.41; CI 0.18-0.92; P = 0.03). In cardiac surgical patients with refractory hemorrhage, rFVIIa therapy is not associated with increased risk of AEs, and early treatment may be associated with better outcomes.