INTRAVENOUS IMMUNOGLOBULINS IN SEPTIC SHO CK THROMBOCYTOPENIA

OBJECTIVES: Intravenous immunoglobulins have been shown to be effectiv e in the treatment of immunologically mediated thrombocytopenia, Sever al articles have been published on the positive effect of immunoglobul ins in sepsis-related death. We retrospectively studied the effect of intravenous immunoglobulins used during septic shock thrombocytopenia over a 5 year period in a polyvalent intensive care unit PATIENTS AND METHODS: Inclusion criteria were development of acute thrombocytopenia under 75 G/I during. septic shock, excluding all cases of disseminate d intravascular coagulation. Thirty-five patients were included in the study; 18 were given polyvalent intravenous immunoglobulins (group Ig IV) and 17 were not (controls). The two groups were comparable for SAP S and APACHE II gravity scores at admission and at day 0 (first day of septic shock with platelet count under 75 G/I), age, sex. platelet co unt at admission, OSF score, type of referral unit, McCabe score. and the presence of 4 parameters which might affect platelet count: hemofi ltration, ARDS, surgery, Swan-fanz catheter. RESULTS: Platelet counts increased on day 8 in the treatment, group (63.5 G/I, range 25-453 ver sus 105.7 G/I, range 38-355; p = 0.0505). The number of days with thro mbocytopenia was the same in both groups. Overall mortality was high ( 60%) but there was a difference between the two groups in favor of the treated group (74.7%) versus 44.4%; p = 0.053). The number of red cel l units (214 vs. 164) and plasma units (175 vs. 54) transfused was hig her in the control group. Platelet transfusion was equivalent in the t wo groups. DISCUSSION: Although we were unable to demonstrate a signif icant difference in the effects of immunoglobulins on platelet level a nd mortality the trend during this evaluation was comparable with that reported in the literature I For transfusion, and although the result s. were not significant, a notion of reduced risk was evident Prospect ive trials are needed to confirm these observations.