Prevalence of clotting factor deficiencies in a large population with von Willebrand disease.

Ambrosi et al (1) recently reported levels of various endothelial markers in cardiac transplant recipients before and after being placed on the lipid-lowering agent fluvastatin. This intervention reduced levels of cholesterol and soluble thrombomodulin (both p <0.001) but, despite this, there was no difference in levels of von Willebrand factor. Our colleagues found this failure to respond puzzling, and suggest that von Willebrand factor may be more a marker of endothelial activation or inflammation than of injury. All the subjects in the study were taking the immunosuppressive drug cyclosporine, and 60% were taking the cytotoxic drug azathioprine. It has been shown that the cyclosporin vehicle, cremophor, enhances stimulated release of von Willebrand factor by human umbilical cord endothelial cells in vitro (2). Consequently, if the same effect is present in vivo, then this can simply explain the failure of von Willebrand factor to respond to the reducing levels of cholesterol. Licciardello et al (3) showed that von Willebrand factor increased when patients with squamous-cell carcinoma were placed on cisplatin-based chemotherapy. If the same phenomenon is true for azathioprine than this is a further reason for finding increased von Willebrand factor that does not respond to lipid-lowering therapy. It follows that if the endothelium is being damaged by cyclosporin, azathioprine and other agents, then changes in levels of soluble thrombomodulin must be responding to a different stimulus and are unresponsive to chemotherapy. This latter possibility is partially supported by our data from subjects undergoing bone marrow transplantation (4). Although both von Willebrand factor and soluble thrombomodulin were raised compared to controls in the 25 patients before transplantation, levels of von Willebrand factor increased further after transplantation (a likely response to the total body irradiation, cyclophosphamide and other agents including some cyclosporine) but there was no change in levels of soluble thrombomodulin. Thus our study, and that of Ambrosi et al, underline the importance of considering the possible deleterious effects of drugs on the endothelium, and that the factors involved in the release of von Willebrand factor are distinct from those controlling soluble thrombomodulin, As former is not a membrane constituent, being stored in intra-cellular organelles, and the latter is a membrane glycoprotein, these differences are entirely understandable.