Filtered plasma as a potential cause of clinical misdiagnosis: inappropriate testing in a haematology laboratory.

We have identified situations in which filtered plasma is provided to the laboratory for combined assessment of lupus anticoagulant, as well as other diagnostic haemostasis laboratory procedures. On the basis of laboratory investigation, however, we report that significant errors in test result interpretation can arise should this occur and depending on the specimen processing procedure, and whether filtered plasma is used for multiple assays. Citrated plasma samples were obtained from 12 normal individuals and a portion of each sample filtered using a standard 0.22 micron filter. Small volumes were frozen for later analysis. We could detect a statistically significant difference (P < 0.01) between filtered plasma and non-filtered plasma in most test results. Most notable were changes in the APTT, fibrinogen, FVIII and vWF, where results often fell outside the normal reference range of the assay. Such assay results closely mimic those obtained with type 2 von Willebrand's disease (vWD) patients, thus should filtered plasma unknowingly be tested for vWF, a wrong conclusion of type 2 vWD could easily arise. Assay results for other parameters may similarly draw incorrect conclusions, such as mild haemophilia A and dysfibrinogenaemia. In summary, we report that following filtration not only are a variety of coagulation test results altered, but reduced plasma levels of some coagulation factors could potentially lead clinicians incorrectly to diagnose congenital or acquired deficiencies or defects. We recommend that laboratories note that (other than for lupus anticoagulant) filtered plasma is not appropriate for the coagulation laboratory, and that any abnormal haemostasis test result following such combined testing be considered a potential artifact and independently repeated to confirm any putative abnormality prior to drawing any clinical conclusions.