Use of trimethoprim-sulfamethoxazole to prevent bacterial infections in children with acute lymphoblastic leukemia.

We assessed the efficacy of prophylactic antibiotics in children receiving intensive chemotherapy for acute lymphoblastic leukemia. The patients were randomized to receive either trimethoprim-sulfamethoxazole (TMP-SMX) or placebo in a double-blind trial. Thirty patients were evaluated in each group. Children receiving TMP-SMX had fewer episodes of bacteremia (0 vs. 5) and otitis media (3 vs. 18). The geometric mean of the neutrophil nadir was 172 in the TMP-SMX group and 287 in controls. However, no increased delay or dose reduction of chemotherapy was observed in the TMP-SMX treated patients. Five patients who received TMP-SMX developed Gram-negative rods resistant to TMP-SMX on surveillance stool cultures. We conclude that TMP-SMX prophylaxis decreased certain bacterial infections in children with acute lymphoblastic leukemia without causing clinically significant toxicity. The emergence of Gram-negative rods resistant to TMP-SMX in treated patients suggests that TMP-SMX prophylaxis should be restricted to patients who are at high risk for developing a bacterial infection or Pneumocystis carinii pneumonia.