Haemophilia commentary: The utility of <scp>BAT</scp>s

In contrast to the questionable accuracy of haemostatic tests, personal and family history of bleeding have long been accepted as important indicators of an underlying inherited bleeding disorder.1-5 However, obtaining an efficient and comprehensive bleeding history requires skill that is not universally available. Bleeding histories are subjective, which can both lead to over- or under-emphasis by the patient and physician. Thus in 1995, Sramek et al6 examined the discriminatory elements of a bleeding questionnaire when administered to patients with a bleeding disorder vs normal controls. They identified a positive family history of bleeding and postsurgical bleeding as the most distinctive elements of the questionnaire. In 2005, the International Society on Thrombosis and Haemostasis (ISTH) Scientific and Standardization Committee (SCC) on von Willebrand factor (VWF) proposed a provisional collection of criteria with a set threshold for frequency and severity of mucocutaneous bleeding to support the diagnosis of type 1 VWD.7 Since the development of these criteria, quantitative bleeding assessment tools (BATs) have been evolving to standardize the bleeding history in order to act as an effective screening tool for patients being investigated for the first time for an inherited bleeding disorder, prevent unnecessary laboratory testing, prevent false-positive diagnoses resulting from initial screening tests (e.g low VWF without bleeding symptoms) and simplify communication about bleeding with a summative score.1-5 Initial iterations of BATs were developed by a group of investigators from Vicenza, Italy, and thus were named Vicenza-based BATs.3 Rodeghiero and his group devised and validated the first summative BAT for adult patients with type 1 VWD. A total score of three for men and five for women had a specificity of 98% and sensitivity of 69% for the diagnosis of type 1 VWD.3 A subsequent BAT, the European Molecular and Clinical Markers for the Diagnosis and Management of type 1 VWD (MCMDM-1 VWD) BAT, was modified to enhance sensitivity as higher scores were awarded for bleeding episodes requiring higher intensity medical therapy (e.g transfusions) or surgical therapy (e.g hysterectomy).5 However, these initial BATs were time-consuming (more than 40 minutes) and cumbersome to administer. Therefore, a Condensed version of the MCMDM-1 VWD BAT was developed to maximize speed (administered in 5-10 minutes) without compromising test accuracy. The Condensed MCMDM-1 VWD BAT had a sensitivity of 100%, specificity of 87%, positive predictive value (PPV) of 0.20 and negative predictive value (NPV) of 1.0 for the diagnosis of VWD in the primary care setting.1 This Condensed BAT was later studied in a group of women with heavy menstrual bleeding, and it could discern women with the presence or absence of a bleeding disorder with a sensitivity of 85%, specificity of 90%, PPV of 0.89 and NPV of 0.86.8 The Condensed BAT bleeding score correlated with the severity of VWD, as individuals with type 3 VWD had the highest scores.8 The Vicenza-based BATs have been mainly concentrated on VWD. However, the Condensed MCMDM-1 VWD BAT has also been studied as a screening tool for the diagnosis of platelet function disorders (sensitivity 86%, specificity 65%, PPV 0.50 and NPV 0.92).9 The Condensed BAT was also studied for the detection of mild bleeding disorders, and a normal bleeding score had a NPV of 0.99 assuming a 1% prevalence of mild bleeding disorders in the general population.10 The ISTH established a Working Party in 2010 to generate and sanction the use of a single consensus BAT for adults and children.4 Multiple studies have demonstrated that BATs can differentiate normal from abnormal bleeding and are effective screening tools for mild bleeding disorders, in particular VWD, which is the most prevalent bleeding disorder. More recently, a study by James et al11 reported that haemophilia carriers had a higher mean bleeding score using the ISTH-BAT compared to age-matched controls (5.7 vs 1.43, P < .0001). While BATs have a documented track record as effective screening tools for bleeding disorders, most require expert (e.g nurse or physician) administration. More recently, a patient self-administered bleeding assessment tool (Self-BAT) has been validated as an effective screening tool for type 1 VWD and haemophilia carriership.12, 13 As implied, the utility of current BATs lies in their suggestion of a mild inherited bleeding disorder as they cannot effectively describe bleeding in severe disorders due to score saturation. Mild bleeding disorders include VWD, platelet function disorders and mild coagulation factor deficiencies (e.g haemophilia carriership). Importantly, the test characteristics (sensitivity, specificity, PPV and NPV) of BATs are predicated on the presence of a gold standard laboratory test to confirm the diagnosis of a mild bleeding disorder. Moenen et al set out to critically appraise, for the first time, the existing literature that evaluates the diagnostic accuracy of BATs for the identification of mild inherited bleeding disorders in the secondary (i.e haematology) care setting. They found a high specificity (>90%) but a relatively low sensitivity (59%-85%) and as such concluded that BATs are of limited utility in patients referred to a specialist for a bleeding evaluation.14 It is important, however, to keep in mind the intuitive intention of a BAT. A BAT provides structure and order to the bleeding evaluation. A BAT therefore remains a helpful tool for the healthcare provider who is not well versed in bleeding disorders. A BAT can help inform the need for specialist referral and should therefore be incorporated in the initial assessment. If we accept that BATs are simply a way to standardize the bleeding history, then the poor aggregate sensitivity of BATs found by Moenen et al is concerning as it is calls into question the clinical relevance of a positive laboratory test in the absence of a personal history of bleeding (i.e negative BAT). Therefore, the poor sensitivity instead of arguing against the utility of BATs may instead argue against the relevance of poorly selected coagulation laboratory testing. Moenen et al do acknowledge that some mild bleeding disorders fail to have a true laboratory diagnostic standard, however, this cannot be overly emphasized as over 50% of the studies included in their appraisal had patients belonging to this category. Also, some patients in these studies may not have not had sufficient haemostatic challenges (e.g childbirth, surgery), which limits the opportunity for BAT point accrual.14 The poor aggregate sensitivity certainly would have been affected by the inclusion of both such patients in the appraised studies. The experienced hematologist, however, will incorporate the BAT score along with other important data points such as the family history or multiple unusual bleeding episodes, both of which may prompt additional laboratory testing. Therefore, a better assessment of the accuracy of BATs in the haematology care setting would be to compare the predictive value of BATs vs the expert-administered unstructured bleeding history. Also, the poor aggregate specificity of BATs described by Moenen et al is unlikely to be problematic from a clinical point of view. It is well recognized that clinical history cannot distinguish one mild bleeding disorder from another; therefore, a seasoned hematologist will not have the expectation that a BAT can do this any better. It is true that haematologists accustomed to evaluating patients with bleeding symptoms are less likely to require a questionnaire to guide their assessment and are also more aware of the laboratory test pitfalls that further interfere with making the diagnosis. Many haematologists are comfortable acquiring and interpreting a bleeding history, although in reality some may be more comfortable than others. Also, a BAT will have additional value in academic haematology clinics with medical trainees as a BAT provides explicit guidance on how to obtain a good bleeding history. Finally, the findings of the systematic review of Moenen et al must be cautiously interpreted particularly as they included studies that lack a diagnostic gold standard and because the majority of studies were of “low quality.” As such, I believe that BATs continue to play a valid role in the clinical assessment of a patient with a possible bleeding disorder whether in the primary or secondary care setting. The authors stated that they had no interests which might be perceived as posing a conflict or bias.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.