Submitted August 8, 2007
Accepted December 21, 2007
Evidence-based diagnosis of type 1 von Willebrand disease: a Bayes theorem approach
Alberto Tosetto, Giancarlo Castaman, and Francesco Rodeghiero*
Hematology Department, San Bortolo Hospital, Vicenza, Italy
* Corresponding author; email: rodeghiero{at}hemato.ven.it.
The diagnosis of type 1 von Willebrand disease (VWD) is based on the presence of bleeding symptoms, reduced von Willebrand factor (VWF) levels and autosomal inheritance of the phenotype. To better appreciate the contribution of clinical and laboratory data to the final diagnosis of VWD, we computed the likelihoods of having VWD as a function of the bleeding score (LRscore), of VWF level (LRVWF) and of number of first-degree family members with reduced VWF levels (LRfamily). The three likelihoods were therefore combined using the Bayes' theorem, giving the final probability (odds) of having VWD. LRfamily and LRVWF were the two factors mostly influencing the final probability of having VWD. Data from the present study provide an evidence-based description of the minimal criteria for the diagnosis of type 1 VWD. As an example, presence of VWF levels below 40 IU/dL in at least two family members (including the proband) and a bleeding score of at least 1 were found to be required for a final odd of VWD above 2.0 (false positives rate less than 1/2). Validation of this approach and of its clinical utility is however required by analysis in other cohorts of well-characterized type 1 VWD patients.
