Changes in blood coagulation proteins following plasmapheresis were reported as early as 1975 [I]. In the recent letter from Pearlman et al. [2] they reported a 60% reduction in fibrinogen and antithrombin following therapeutic plasmapheresis and commented that this balanced reduction was not accompanied by evidence of haemorrhage or thrombosis. Such changes have been reported previously for a variety of coagulation factors 13-51, the extent of reduction being dependent on the volume of plasma exchange and the rate of recovery after plasma exchange [4]. Our own studies would suggest Pearlman's data relate to exchange of somewhat less than one plasma volume, since we found 70% depletion of fibrinogen and 60% of antithrombin following single plasma volume exchange. Recovery of fibrinogen after exchange is somewhat slower than for antithrombin, but in common with Pearlman we found no evidence of thrombosis 141 and only rare cases of haemorrhage, usually ascribable to thrombocytopenia or inadequate heparin neutralisation [3]. Reviewing the literature available in 1983 yielded generally similar conclusions [ 5 ] .
The changes reported by Pearlman in fibrin-split products and D-dimer are of interest in suggesting fibrinolytic activation occurs in a proportion of patients. However, as the results are only expressed as the proportion of patients with abnormal results, it is difficult to determine whether such effects are of major significance or are more probably, since they only occur in a few patients, borderline events.