Einfluss der Methode zur Bestimmung der von Willebrand-Faktor-Aktivität auf die klinische Diagnostik des von Willebrand-Syndroms : eine retrospektive Analyse

Abstract

Background: The diagnosis of VWD depends on clinical symptoms and laboratory findings. In this regard, a von Willebrand factor (VWF) activity assay is considered mandatory for the diagnostic work-up of patients with suspected inherited bleeding disorders. In September 2011, the VWF ristocetin cofactor assay (VWF:RCo) was replaced by the fully automated INNOVANCE® GPIb binding assay (VWF:Ac) at our institution. We retrospectively analyzed the effect of this methodological transition on the frequency and pattern of VWD diagnosis at our outpatient clinic.

Methods: All patients presenting with a presumed bleeding disorder during the year before (cohort VWF:RCo) and after introduction of the VWF:Ac assay (cohort VWF:Ac) were included in the analysis. Clinical data, including final diagnoses, and laboratory findings were extracted from electronic patient files.

Results: The VWF:RCo and VWF:Ac cohorts comprised 322 and 298 patients, respectively. In 10 patients, both assays were performed in parallel. Demographic patient characteristics were similar between the two groups (mean age, 50 ± 18 years; females, 70%). A decreased (i.e. below the normal reference range) VWF activity was more frequently detected in the VWF:RCo than in the VWF:Ac cohort (29% vs. 22%; P=0.04). Of the patients with decreased VWF activity, 22% (VWF:RCo) and 8% (VWF:Ac) had normal PFA-100® closure times (P=0.02), indicating regular VWF function under high shear conditions. VWF:Ac (r=0.92) correlated more closely with VWF collagen binding capacity (VWF:CB) than VWF:RCo (r=0.79). In the 10 patients with parallel measurements, results were significantly higher using the VWF:Ac (132 ± 50%) than the VWF:RCo assay (114 ± 41%; P=0.03). Importantly, in two of the ten patients with decreased VWF:RCo, but normal VWF:Ac, VWD was ruled out on the basis of repeated testing and clinical assessment. Likewise, in the VWF:Ac cohort, "possible VWD" was less frequently diagnosed (19% vs. 24%) and VWD was more often ruled out (65% vs. 59%) than in the VWF:RCo cohort. Consistently, the VWF:Ac assay had improved specificity (85% vs. 76%) and positive predictive value (44% vs. 33%) for the diagnosis of VWD than the VWF:RCo assay.

Conclusion: in this retrospective analysis of "real-world" patients presenting to an outpatient clinic for the diagnostic work-up of a presumed bleeding disorder, the VWF:Ac assay was associated with an improved diagnostic accuracy when compared to the VWF:RCo assay.