The use of cefotaxime was monitored to evaluate the appropriateness and cost of its use. This was done by chart review and physician questionnaire. Of 57 patients reviewed, 19 were treated empirically for specific established infections; 19 for sepsis without an identified source; 3 for infections with a known pathogen; and 16 for surgical prophylaxis. In the majority of patients treated empirically for specific infection or sepsis, the use of cefotaxime was judged appropriate by infectious disease consultants, while for surgical prophylaxis, it usually was judged inappropriate. The average charge for cefotaxime was +396 per course vs. +318 for antibiotics commonly used before cefotaxime became available. When used alone, cefotaxime was less expensive to the patient than a combination of antibiotics in 22 of 29 cases. We concluded that cefotaxime was used appropriately for empiric therapy without increasing costs greatly.