The bleeding score: Useful in predicting spontaneous bleeding events in adults with bleeding of unknown cause?

Abstract

Despite extensive laboratory workup, no underlying coagulation de‐ fect is found in many patients with mucocutaneous bleeding (MCB) symptoms, and they are categorized as having bleeding of unknown cause (BUC). The prevalence of BUC in these patients ranges from 59% to 72%.1,2 The underlying pathophysiology has yet to be deter‐ mined, which creates difficulties for management. In general, the evaluation of bleeding symptoms is a recognized challenge due to the subjective nature of bleeding histories.3 Bleeding assessment tools (BATs) have been developed to better quantify the bleeding history.4 BATs translate bleeding symptoms into a summa‐ tive bleeding score (BS), alerting physicians to which patients have bleeding symptoms outside the range of normal.5 Many studies in different cohorts of patients support the use of BATs as a screening tool in the diagnosis of mild bleeding disorders.4,6,7 Currently, the ISTH‐BAT (International Society on Thrombosis and Haemostasis) is the tool endorsed by that society.8 The main utility of BATs to date has been as a screening tool for patients presenting with MCB. Studies have also used them to describe bleeding severity in patients with mild bleeding disorders. Recently, the utility of BATs has been extended to predict the risk of future bleeding in patients with von Willebrand disease (VWD). A prospective study of 796 Italian patients with all types of VWD showed that a BS >10 was the strongest predictor of bleeding events severe enough to require treatment with desmopressin (DDAVP), and/or von Willebrand factor (VWF)/FVIII concentrates.9 Therefore, the BS may be useful to inform treatment protocols and replacement therapies.9 No data currently exists regarding the predictive role of BS in other populations, including BUC. With this background, we aimed to determine whether the BS is predictive of future sponta‐ neous bleeding events in patients with BUC. We conducted a single‐centre retrospective chart review that aimed to determine whether the BS derived from the ISTH‐BAT pre‐ dicts risk of future spontaneous bleeding events in patients with BUC. We reviewed patients ≥18 years of age categorized as BUC after re‐ ferral for haemostatic evaluation at a Hematology Clinic at Kingston Health Sciences Centre (Ontario, Canada). Patients with a history of bleeding symptoms were classified as BUC after standard haemostatic testing revealed no diagnostic laboratory abnormality. The molecular and clinical markers for the diagnosis and management of VWD type 1 (MCMDM‐1) BS was generated at the time of initial haematological consultation.7 Bleeding scores from the MCMDM‐1 were then con‐ verted to an ISTH‐BS. A study by Elbatarny et al5 showed that there is a 90% overlap between BATs, providing evidence that merging bleeding scores derived from different bleeding questionnaires is ac‐ ceptable in this study. Number and site of future spontaneous bleed‐ ing events were collected from medical records. To be included in the study, patients must have had a positive BS at the time of diagnosis, defined as an ISTH‐BS ≥4 in adult males and ≥6 in adult females.5 Exclusion criteria included patients diagnosed with an inherited bleeding disorder, patients <18 years of age, active anti‐ coagulation and patients referred <6 months prior to data collection.5 The following laboratory tests were collected retrospectively from patient charts: (a) complete blood count, ABO blood type, coagulation profile (INR, PT, PTT, TT, fibrinogen), ferritin, iron, transferrin, trans‐ ferrin saturation; (b) VWF antigen level, VWF functional assay and factor VIII level; (c) platelet function studies; and (d) additional factor assays if performed. Data was imported into IBM SPSS and analysed using t tests, chi‐squared tests and logistic regression. A total of 90 patients were included in the study. The mean age was 46 years, and the sample was 98% female. There was a wide range of BSs among the study population (6‐20). The mean (SD) bleed‐ ing score among all BUC patients included in the study was 10 (3.22) (Figure 1). In this cohort, 57.8% of patients (52/90) had at least one spontaneous bleeding event following designation as BUC. Heavy menstrual bleeding (HMB) was the most common type of bleeding event, while muscle haematoma was the least common. There were