EARLY IDENTIFICATION OF NEUTROPENIC PATIENTS AT RISK OF GRAM-POSITIVEBACTEREMIA AND THE IMPACT OF EMPIRICAL ADMINISTRATION OF VANCOMYCIN

The aim of this multicentre randomised trial was to determine whether it was possible to predict gram-positive bacteraemia, and whether the empirical use of vancomycin would lead to reduced morbidity and mortal ity. 35 of 113 patients (31%; confidence interval, CI 8.5), who presen ted with a skin or soft tissue infection and had received empirical va ncomycin in addition to either ceftazidime or piperacillin-tobramycin, had initial bacteraemia with a single gram-positive bacterium compare d with 135 of the 784 (17%; CI 2.6), who presented with another infect ion and who had been givers ceftazidime or piperacillin-tobramycin wit hout vancomycin (P<0.001). Empirical vancoanycin resulted in a higher rate of eradication (P=0.033, relative risk 1.2), but not a better cli nical outcome and was associated with more toxicity (P=0.042, relative risk 1.6). Irrespective of the initial treatment regimen, fever laste d an average of 8 days, the empirical regimen was modified in more tha n 50% of cases and mortality attributed to gram-positive infection was less than 2%. Incorporating vancomycin in the initial empirical antib iotic regimen for febrile neutropenic patients does not appear necessa ry, even for skin and soft tissue infections associated with gram-posi tive bacteraemia. Copyright (C) 1996 Elsevier Science Ltd