Comparison of Thrombin Generation Assay With Conventional Coagulation Tests in Evaluation of Bleeding Risk in Patients With Rare Bleeding Disorders

Based on the premise that the capacity of plasma to generate thrombin in vitro is a comprehensive and precise functional test of the clotting system, we designed a cross-sectional, single-center study involving 83 patients with rare bleeding disorders (RBDs) to compare the usefulness of the thrombin generation (TG) assay versus conventional tests including prothrombin time (PT) and activated partial thromboplastin time (aPTT) in predicting bleeding risk in patients with RBD in southern Iran. The TG parameters consisted of endogenous thrombin potential, lag time, peak, time to peak (ttPeak), and start tail. The area under the receiver-operating characteristic (ROC) curve showed statistically significant associations between bleeding risk and lag time, ttPeak, and start tail. We determined cutoff values for these 3 TG parameters and obtained a negative predictive value of 86% to 90% in patients with RBD who had a bleeding score (BS) ≤13. The ROC curves for the association of PT and aPTT with BS did not indicate any significant association. Correlation analysis supported the results of ROC curve analysis, only lag time, ttPeak, and start tail showed significant positive correlations with BS (P < .05). Disease severity based on plasma factor activity was significantly associated with prolonged lag time and ttPeak and with prolonged PT (P <.05). We suggest that TG assay is a potentially more useful tool for predicting the bleeding risk in patients with RBD. However, the small sample size in different RBD subgroups precluded subgroup analysis. Prospective multicenter studies with larger numbers of patients are therefore advisable.