Cyclosporin for dermatomyositis.

SIR, We completely agree with Dr E N Harris that carrying out clotting assays to detect lupus anticoagulant is more difficult than an enzyme linked immunosorbent assay (ELISA) for anticardiolipin antibodies. Specialised laboratories are necessary, coagulation tests are time consuming, and improper preparation and handling of plasma samples strongly influence the results of lupus anticoagulant assays. The goal of our article was not to propose the determination of lupus anticoagulant only and to sweep aside the anticardiolipin antibody determination but to emphasise that testing anticardiolipin antibodies alone is not enough to identify patients with increased risk of thrombotic complications. From the values we found for sensitivity and specificity with lupus anticoagulant and anticardiolipin assays we conclude that in patients with systemic lupus erythematosus (SLE) the lupus anticoagulant assays correlate better with the presence of a history of thrombotic complications. We did not claim that the anticardiolipin antibody ELISA is not a useful predictor for thrombosis, fetal loss, or thrombocytopenia, but that lupus anticoagulant tests are better predictors. Confirmation of our results by studies in other laboratories is necessary, however. Therefore we endorse the last sentence of Dr Harris's letter completely that 'with the knowledge we have today both tests have to be performed'. As mentioned by Dr Harris our studies differ with respect to the percentage of lupus patients included and the selection of patients. We evaluated 111 consecutively seen lupus patients who were unselected apart from the fact that they were seen at a university hospital. Dr Harris selected on the availability of clinical notes 121 patients (76% with SLE) from a total of 300 patients with various autoimmune diseases (60% with SLE) who had been screened for IgG anticardiolipin antibody levels. In our paper we incorrectly used the term exchange of sera when we referred to the freeze dried samples which were kindly provided by Dr Harris for participation in the International Anticardiolipin Standardisation Workshop (April 1986). Although we introduced some modifications, our assay appeared to be valid. The limits we gave in units for low, medium, and high levels were obtained by transformation of the known anticardiolipin antibody concentrations (in tg/ml) in the provided samples. We agree that apart from standardisation of the anticardiolipin antibody ELISA an international consensus with respect to the definition of the lupus anticoagulant is also urgently needed. Unless we can be sure that only results that are obtained with valid assays performed on proper samples are presented, methodological differences remain a possible cause for conflicting reports. Therefore, we welcome and strongly support the tremendous efforts made by Dr Harris and coworkers to reach international uniformity.