Spectrum Of The Acquired Von Willebrand Syndrome In a Large Cohort Of Patients Diagnosed In a Single Institution

The first description of an apparently acquired form of von Willebrand disease dates back to 1968 in patients with systemic lupus erythematosus. To differentiate between the congenital and acquired forms, the term acquired von Willebrand syndrome (aVWS) is now accepted for these patients. It turned out that there are three predominant underlying diseases that lead to an aVWS in a subset or even in almost all affected patients. More than 9.000 samples reach our laboratory yearly for diagnosis, confirming or (sub-) classification of a possible VWD. In about 15% of these the diagnosis VWD can be confirmed and 30% of these patients suffer from an apparently acquired form. Here we describe our experience in diagnosing aVWS patients over more than 9 years (January 2004 – February 2013).

The largest group are patients with cardiovascular diseases (483 = 40%). Since our first patients in 2005, the leading group are patients on non-pulsative left ventricular assist devices (LVADs) with still markedly increasing numbers. Aortic stenosis was the reason for aVWS in 139 patients, usually with less severe symptoms compared to patients on LVAD´s. In 67 patients the underlying diseases were congenital heart diseases. Miscellaneous cardiac defects were causative in 29 patients. Only nine patients were on an extracorporal membrane oxygenator. But all of them suffered from severe bleeding complications with a high mortality (five out of nine). The VWF:Ag was higher than normal (>160%) in all patient groups with the exception of the congenital cases. It seems to be influenced by the highth of the shear stress and the age of the patients. The mean VWF:CB was significantly lower compared to the VWF:Ag. Therefore the ratio between VWF:CB and VWF:Ag was low in all groups, but although our lower limit of a “normal” ratio is higher than usual (0.8), the sensitivity towards a loss of the large multimers is low with 60% in all groups. Thus without using multimer analysis as a first line diagnostic test, in a big part of patients with cardiovascular diseases, the aVWS will be overseen.

The acquired VWS in patients with thrombocythemia and the pathophysiology was described in 1984 and 1986. In our 362 patients (30% of the whole group), the large multimers nearly always show an absolute or relative reduction. An aVWS with a phenotype similar to inherited VWD1 is rarely seen. Whether a given patient suffers from thromboembolism or bleedings is strongly dependent from the platelet count. Most patients investigated were not actively bleeding but were tested to confirm the diagnosis or to plan for the best treatment in case of surgery or trauma. The ratio between VWF:CB and VWF:Ag was low in these patients. The sensitivity towards a loss of the large multimers is somewhat better than in cardiovascular cases, but does not exceed 80%.

Lymphoproliferative diseases as the underlying disorder associated with aVWS account for a significant portion of patients. We observed 243 patients between January 2004 and February 2013 (20% of the whole group). Typically in cases with a monoclonal IgG (201 patients) there is a relative decrease of the large multimers together with severely decreased or absent proteolytic bands. This is due to the preferential removal of the large multimers together with a too short time left for ADAMTS13 to cleave the molecule. In contrast, the majority of patients with monoclonal gammopathy of the IgM type had aVWS similar to inherited VWD type 1. They show an unmistakable multimeric pattern. This multimeric pattern can be explained by the giant VWF-IgM complexes which destroy the agarose during their passage through the gel and sometimes even leave holes. A brief inspection of the gels is thus sufficient for making correct diagnosis.

In 114 patiens (9% of the whole group), the mechanism leading to an enhanced clearance (type 1) or destruction of the VWF structure (type 2) is unknown.