Comparative Evaluation of Five Different Assays for Ristocetin Cofactor Activity Determination,

Despite significant advances in the comprehension of the molecular pathogenesis of Von Willebrand Disease (VWD), the diagnosis of this condition remains one of the greatest challenges of daily hematology practice, especially in settings of low prevalence such as primary care services. Ristocetin cofactor activity (VWF:RCo) is one of the pillars of the diagnosis of VWD and is also essential for its management. However, reproducibility of tests available to measure VWF:RCo is still a major issue, as evidenced by coefficient of variations (CV) as high as 30%, 45% and 27% in the ECAT, NEQAS and PALQ external quality assessment program. Classical methods to measure VWF:RCo include light-transmission platelet agregometry (LPA) and visual agglutination with formaldehyde fixed human platelet (VA), but alternative methods continue to evolve. An automated method (BC) has been recently shown superior to LPA and VA. Other methods for which limited comparative precision studies are lacking include a new LIA-based automated assay (LIA) and flow citometry (FC), using fluorescence-marked platelets and ristocetin.

Here we evaluate five different VWF:RCo assays, including classical LPA and VA, two different commercially available automated coagulometer-based methods (BC and LIA), and FC. Reference ranges for the each method were established in twenty healthy adults. Plasma samples from thirty-three patients with VWD (type 1: n=26; type 2A or 2M: n=4; type 3: n=3) followed at our comprehensive care center (IHTC-Campinas, Brazil) were assayed for the comparative studies. Diagnosis of VWD was based on historical levels of VWF:Ag (assayed by ELISA) and VWF:RCo (assayed by both LTA and VA) obtained from medical records, and from bleeding history (evaluated by VWD bleeding score). Imprecision studies (intra-assay variability), were performed in different samples: commercial lyophilized normal and abnormal plasma, as well as plasma from two additional VWD patients with borderline results. All analyses were performed in the same samples, in an external quality assessment program-participating university laboratory.

Despite using similar conditions than those described in an original report, and repeated experiments, the flow cytometry-based assay (FC) was not reproducible in our hands, yielding extremely high CV that prevented its use in a clinical laboratory. Mean and reference ranges for VWF: RCo assays levels in healthy adults were: VA: 0.84 IU/ml (0.62–1.07); LTA: 1.13 IU/ml (0.92–1.35); BC: 0.93 IU/ml (0.75–1.12). The automated BC method yielded the lower intra-assay variability, in the range of 1.13% to 5.46% in different samples, compared to values as high as 29.4% for LTA. Correlation coefficient (r) ranged from 0.93 between BC reagent and LTA to 0.85 between VA and BC reagent. Mean VWF: RCo levels in patients ranged from 22.99 ± 17.80 with BC to 22.60 ± 18.50 with LTA and 20.11± 14.50 with VA. In twenty-six type I VWD patients, mean VWF:RCo/VW:Ag ratio ranged from was 0.7 (range 0.4–1.5) with the VA method to 0.8 (range 0.8–1.3) with the BC method. In four type 2 VWD patients, mean VWF:RCo/VW:Ag ratio was 0.2 (range 0.3–0.4) using VA method to 0.4 (range 0.4–0.5) with the BC method. Concordance, evaluated using the Bland-Altman plot, was higher between LTA and BC method.

In our study, LTA and VA performance was consistent with historical data, presenting very high CV, but good correlation and concordance with new automated methods. Of note, VA, a simple, inexpensive and equipment-independent method, presented better CV than LTA and should be regarded as a good alternative the diagnosis of VWD in areas with limited resources. Unfortunately, flow cytometry-based VWF:RCo methods apparently suffer from the same problems of LTA and VA regarding standardization. New automated assay present enhanced precision compared to VA and LTA, with the added benefits of automation in the setting of a Hemostasis laboratory. Ongoing improvements in protocols of these new assays, leading to even higher precision, could translate into more confident limits of normal and pathological VWF:RCo levels, enabling critical changes in the diagnosis, management and even in the definition of VWD.

No relevant conflicts of interest to declare.