Abstract

Oral trimethoprim-sulfamethoxazole (Bactrim) plus erythromycin (TMZ-E) was tested versus placebo (P) as prophylaxis for bacterial infection in a randomized, double-blind trial in adult cancer patients receiving cytotoxic chemotherapy expected to result in significant neutropenia. The incidence of adverse reactions attributable to TMZ and/or E was higher in drug-treated episodes (18 of 28 vs 3 of 29 for P, P less than 0.0005) resulting in poorer compliance. The incidence of fever was not significantly different between episodes treated with TMZ-E (18/27) and those treated with P (17/29), nor was there a significant difference in the median interval between the onset of neutropenia and the onset of fever. However, 14 of 18 fevers in TMZ-E recipients were without a documented infectious source compared with only 6 of 17 in P recipients (P less than 0.05). The same patterns were apparent even when episodes in which compliance with the regimen was either excellent or good were considered separately. There was no significant difference in the number of deaths from infection between TMZ-E and P recipients (3/27 vs 1/29). It is concluded that TMZ-E prophylaxis is of no practical benefit, may mask the cause of infection in febrile neutropenic cancer patients, and is associated with substantial toxicity.

View details for Web of Science ID A1984RZ06400025

View details for PubMedID 6418374