Purpose: An increased incidence of thrombosis has been reported in patients with a prolonged activated partial thromboplastin time (APTT) due to a lupus anticoagulant (LA), which is an antibody to negatively charged phospholipids. The antiphospholipid antibodies can be quantitated in an enzyme-linked immunoabsorbent assay (ELISA) that utilizes cardiolipin as the antigen. With the development of the ELISA, two major areas of controversy have arisen. First, the correlation between assay results for LA and for the ELISA has varied widely among laboratories. Second, some investigators have described a correlation between high levels of anticardiolipin antibodies (ACA) and thrombotic disorders, whereas others have found no association between ACA levels and thrombosis in a general population of medical patients. To explore these issues further, the present study determined the sensitivity and specificity of an LA assay for detecting ACA in medical patients with a prolonged APTT. The association between the isotype and titer of ACA and thrombosis was examined in those patients positive for LA.
Patients and methods: Plasma samples from 70 medical patients with a prolonged APTT by routine screening studies were tested for the presence of LA by dilution of phospholipid in an APTT system and for IgM and IgG ACA according to a standardized ELISA. Clinical records were reviewed for a history of thrombotic events by an investigator who had no knowledge of the laboratory results.
Results: The ACA assay gave positive results in 47 patients, 44 of whom also tested positive for LA. Thus, the sensitivity for the LA assay for detecting ACA was 94% (confidence interval, 82% to 99%). The result of the LA assay was negative in 20 of 23 patients who were ACA-negative. The specificity of the LA assay was 87% (confidence interval, 67% to 98%). Twelve of the 47 patients (26%) had a history of venous or arterial thrombosis. Of these patients, 75% tested in the high-positive range for IgG or IgM ACA, or both. Of the 35 patients without thrombosis, only 14% were in this range. Patients with thrombosis had either underlying systemic lupus erythematosus, lymphoma, or no apparent etiology for LA. There was no history of thrombosis in patients with LA associated with infection or medication.
Conclusion: A test for LA in medical patients with a prolonged APTT can be sensitive and specific for ACA. Determination of ACA levels in patients who have LA that is not induced by medication or infection may define those patients at increased risk for thrombosis.