Correlation of a Condensed Bleeding Score with New Diagnosis of a Congenital Bleeding Disorder in Patients Referred to a Tertiary Centre

Abstract 1226

The performance and utility of a condensed bleeding score (Bowman et al, J Thromb Haemost., 2008;6:2062) in relation to the diagnosis of a congenital bleeding disorder in new referrals to a regional haemostasis clinic over an 8 month period is presented. Between November 2010 and June 2011, 50 patients over the age of 16 (median age, 31 years; range, 16–79), including 32 females, were referred for investigation of a possible congenital bleeding disorder following detection of abnormal coagulation results and/or presentation with a bleeding history. A bleeding score was performed as part of their initial assessment. 12(24%) patients were from local referral and 38(76%) patients were referred from other hospitals in the region for further investigation of a suspected bleeding disorder. Basic coagulation tests (activated partial thromboplastin time (APTT), prothrombin time Clauss fibrinogen and platelet count) were normal in the referred patients from other centres. 50% (6/12) of the local referrals were for investigation of a prolonged APTT detected on routine coagulation screening prior to major surgery.

The median bleeding score was 6 with a range of −1 to 14 (Table 1). The presence of a congenital bleeding disorder was confirmed in 31 of the 50 patients (62%), including 19/31 (61%) of the female patients and 12/31(39%) of the males. Correlation of an abnormal bleeding score (score ≥ 4) with diagnosis of a congenital bleeding disorder was only seen for diagnosis of type 1 Von Willebrand Disease (VWD) (Table 2). Analysis of the cases with low scores and abnormal results identified two groups of patients; firstly, those who had not yet had a significant haemostatic challenge, and secondly, those in whom the abnormal coagulation results were explained by a non-haemostatically significant reduction in a coagulation factor level (e.g. FVII, 15%; dysfibrinogenaemia; F XII deficiency). These clinically insignificant laboratory abnormalities explain the discrepancy between the number of patients with abnormal laboratory tests (35) and the number of patients diagnosed with a congenital bleeding disorder (31).

Compared to previous reports the range of scores found with this assessment tool was narrow and could not exclude patients from further laboratory assessment. However the condensed bleeding score has only been validated prospectively for the diagnosis of type 1 VWD and all patients in this cohort who were diagnosed with type 1 VWD had an abnormal bleeding score (≥ 4). This observation supports the role of this scoring system in the assessment of patients for type 1 VWD. The use of the condensed bleeding score in assessing patients with suspected factor XI deficiency is difficult due to the lack of a phenotypic relationship between residual factor XI activity and a bleeding tendency. Furthermore, although factor XI deficiency is a rare congenital bleeding disorder in our cohort of patients 12/31(39%) were diagnosed with factor XI deficiency. This may explain the overall lack of correlation between bleeding score and diagnosis of a congenital bleeding disorder. Patients who have an abnormal bleeding score but normal laboratory tests need consideration of further investigations before concluding they are normal. The possibility of an acquired bleeding disorder should be considered. A thorough drug history is also important as one of the patients with a bleeding score of 14 was taking a non-steroidal anti-inflammatory drug.

The use of the condensed bleeding score in the detection of congenital bleeding disorders other than type 1 VWD requires further validation in a larger number of patients.