The basic screen has three tests: von Willebrand factor antigen, activity measured as the RICOF (ristocetin cofactor assay), and factor VIII level, the last to exclude mild haemophilia. Most cases are type 1 with low antigen and low activity at a normal ratio, roughly 75% of von Willebrand disease.

The tests are unreliable, so the screen is done twice in a well-rested, unstressed patient. Stress releases extra von Willebrand factor, so babies under six months distressed by needles give unreliable results. Newborns are tested only if type 2B or type 3 is expected from family history, since management changes early.

Below 30 looks like type 1 and should be repeated. Between 30 and 50 depends on the bleeding history and is usually repeated. Above 50 is a negative result, and above 100, even on a single test, has a negative predictive value of 90 to 95%, so testing should stop.

Ristocetin, a failed antibiotic that caused thrombocytopenia, unfurls von Willebrand factor, mimicking shear stress. The assay measures only its ability to bind the platelet GP1B receptor. It used to use platelet aggregometry; now latex beads coated with recombinant GP1B act as fake platelets, avoiding variable lab platelets.

It separates haemophilia A from type 2N (Normandy). Patient von Willebrand factor is anchored to an ELISA well, calcium chloride strips off endogenous factor VIII, a fixed amount of factor VIII is added and bound factor VIII is measured. Homozygous 2N binds very little, heterozygous binds some.

RIPA is ristocetin-induced platelet agglutination using very low-dose ristocetin, normally too little to cause agglutination. Agglutination suggests type 2B or platelet-type pseudo von Willebrand disease. Repeating it with patient plasma plus reference platelets, then patient platelets plus reference von Willebrand factor from cryoprecipitate, shows where the defect lies.

By assessing high molecular weight multimers. Protein electrophoresis gels were labour intensive, so labs now use a collagen binding assay because high molecular weight multimers bind collagen best. Absent multimers and reduced collagen binding point to 2A; preserved multimers point to 2M, which the host calls mainly of academic interest.

Desmopressin is given IV or subcut and antigen, RICOF and factor VIII are measured before, at 30 minutes, one hour and four hours. In 1C levels rise initially then clear rapidly, sometimes back to baseline within two hours. It also shows whether desmopressin would be useful for bleeding.

Genetic testing is increasingly used to clarify type 2 subtypes, but two sets of antigen and activity screens must be done first. Options are the R121 single-gene analysis for von Willebrand disease or the broader R90 panel covering platelet and bleeding disorders.

In the case with RICOF 25, antigen 28 and factor VIII 27, factor VIII is low because antigen is low, pointing to type 1. Type 2N shows normal von Willebrand factor antigen with markedly reduced factor VIII, as in the 20-year-old woman with menorrhagia and normal multimers.