Amiodarone therapy is associated with several adverse effects, including hematologic ones such as pancytopenia, hemolytic anemia, and aplastic anemia. Very few cases of amiodarone-associated bone marrow granulomas have been reported. We report 2 cases of amiodarone-associated bone marrow granulomas. Patient 1 was an 81-year-old man who presented with leukopenia, thrombocytopenia, and hepatosplenomegaly after 2 years of amiodarone therapy. Patient 2 was an 80-year-old man who presented with pancytopenia 2 1/2 years after starting amiodarone treatment. Both patients had normal blood counts before amiodarone therapy. Bone marrow biopsies showed noncaseating granulomas in both patients. We reviewed the literature available on Medline for amiodarone-associated bone marrow granulomas and found 8 reported cases of amiodarone-associated bone marrow granulomas. One case also featured amiodarone-associated hepatic granulomas. Amiodarone therapy was stopped in 5 cases, with improvement of the granulomas occurring in 3 cases. We conclude that bone marrow granulomas, although rare, should be considered as a differential diagnosis for patients undergoing amiodarone therapy and presenting with cytopenias.

The Guillian Barré syndrome is an acute inflammatory disorder for which plasma exchange is effective treatment. Up to 10% relapse after plasma exchange suggesting that treatment sometimes finishes before disease activity has resolved. We studied whether plasma fibrinogen, an inflammatory marker, might be used to determine when to discontinue plasma exchange in patients with acute Guillain-Barré syndrome. We conducted a post-hoc analysis of apheresis database and hospital records of patients treated with plasma exchange for acute Guillain-Barré syndrome during 1999-2004. Data were analyzed from 28 patients who underwent a total of 29 courses of plasma exchange for acute Guillain-Barré syndrome. The mean (+/-SD) plasma fibrinogen concentration was 422.5 (+/-96.4) mg/dl at the time of presentation and, in 17 of the 29, it was above 400 mg/dl (reference range 200-400). Twenty of the 21 patients whose fibrinogen fell by more than 30% from baseline by the time of the final plasma exchange treatment had neurological improvement. There was improvement in only 3 of the 8 instances where fibrinogen decreased by less than 30% by the end of plasma exchange therapy. A > or =30% decrease in fibrinogen by the conclusion of plasma exchange was significantly associated with sustained neurological improvement (P = 0.0025). The plasma fibrinogen level appears to reflect disease activity in acute Guillain-Barré syndrome. A <30% fall in fibrinogen level despite plasma exchange may indicate the need to continue plasma exchange to maximize the benefit of treatment or minimize the risk of relapse. Therapeutic plasma exchange need not be extended when plasma fibrinogen remains > or =30% below its level at presentation by the time of the final planned plasma exchange procedure.

No abstract