We read with interest the commentary by Rodeghiero et al. [1] regarding the diagnosis of type 1 von Willebrand disease (VWD), which affects 80-90% of all patients with VWD. More than 40 years ago, Zimmerman and Edgington [2] reported that the molecules possessing factor VIII coagulant activity (FVIII:C) and von Willebrand factor (VWF) properties could be physically segregated and identified as different molecular entities. This pivotal discovery made it possible to diagnose type 1 VWD and the different VWD subtypes. The advent of mAbs, molecular biology approaches, cell adhesion under flow conditions, and transgenic animal models, among other developments, made it possible to understand the pathophysiology of VWD, the molecular diagnosis of the VWD subtypes, and the treatment alternatives for this disorder. Pathophysiologically, type 1 VWD is defined by partial quantitative deficiency of plasma VWF antigen level with a parallel decrease in VWF activity (usually measured as VWF ristocetin cofactor activity [VWF:RCo] and/or VWF collagen binding). Although this seems simple and straightforward, there is still no global consensus on the plasma VWF cut-off values for the laboratory diagnosis of this frequent subtype. A recent survey [3] showed that only 27% of North American specialized hemostasis laboratories adhered to National Heart, Lung and Blood Institute (NHLBI) expert panel guidelines [4].The objective of our study [5], referred to in the commentary by Rodeghiero et al., was to compare the various laboratory criteria recommended by authoritative groups for the diagnosis of type 1 VWD, which were based largely on expert opinion and little experimental evidence.We analyzed 4298 laboratory evaluations of patients referred to us during a 5-year period. Succinctly, the same data analyzed according to four proposed criteria led to an almost three-fold difference in the diagnostic rate of type 1 VWD, ranging from between 2.8% (NHLBI recommendation; similar to that recently adopted by the British Committee for Standards in Haematology [6]) to 8.3% (Zimmerman Program for the Molecular and Clinical Biology of VWD criterion) [7]. Therefore, adopting either choice leads to disparate outcomes. We think that these laboratory ranges are unacceptably broad and that their accurate demarcation should be a priority.Rodeghiero correctly criticizes the lack of bleeding score (BS) and family history in our patients; we share this concern, because, without these components, there is no disease to investigate. Diagnosing type 1 VWD on the basis of laboratory values only, without considering bleeding symptoms and inheritance, is as unreasonable as predicting VWD on the basis of the BS only. The authors are also correct in suspecting that an uncertain number of the patients included in our report could have had no abnormal bleeding had they had completed a comprehensive bleeding questionnaire. Nevertheless, this lack of information does not invalidate the central message: the wide variation in diagnotic yield obtained with the isolate...