The discriminatory power of bleeding assessment tools in adult patients with a mild to moderate bleeding tendency

Abstract

Background: Bleeding assessment tools (BATs) have been developed to quantify bleeding severity. Their ability to predict for the diagnosis of a bleeding disorder has not been thoroughly investigated. Objectives: To evaluate the ability of the Vicenza BAT and the ISTH BAT to distinguish patients with an established bleeding disorder from those with bleeding of unknown cause (BUC). Patients/Methods: Three-hundred fifty-nine patients (228 with BUC, 64%) from the Vienna Bleeding Biobank were assessed in this study. Results: The bleeding scores were similar in patients with an established diagnosis of a bleeding disorder compared to patients with BUC. Both BATs had a low sensitivity and specificity for the diagnosis of a bleeding disorder with areas under the receiver operating characteristic (ROC) curves of 0.53 (95% confidence interval 0.47–0.60) for the Vicenza BAT and 0.52 (0.46–0.59) for the ISTH BAT. In terms of specific diagnoses, both scores were most accurate in diagnosing von Willebrand disease (VWD, areas under the ROC curve; Vicenza BAT 0.67 (0.45–0.90); ISTH BAT 0.70 (0.50–0.90)). A separate evaluation of different bleeding symptoms in patients who had undergone surgery and tooth extraction revealed that postpartum bleeding and bleeding from small wounds was predictive for diagnosing a MBD in multivariable analysis. Conclusions: The Vicenza- and the ISTH BAT have a low ability to distinguish patients with an established bleeding disorder from those with BUC.