Trimethoprim-sulphamethoxazole appears more effective than aerosolized pentamidine as secondary prophylaxis against Pneumocystis carinii pneumonia in patients with AIDS.

OBJECTIVE

We compared the efficacies of low-dose trimethoprim-sulphamethoxazole (TMP-SMX; one tablet: TMP, 160 mg, SMX, 800 mg, twice daily, twice a week) and aerosolized pentamidine (300 mg every 4 weeks) as secondary prophylaxis against Pneumocystis carinii pneumonia (PCP) in patients with HIV infection.

DESIGN

A retrospective controlled study.

SETTING

The study was performed at St Vincent's Hospital, Sydney, Australia, which is a tertiary referral university hospital.

PATIENTS, PARTICIPANTS: Over a 4-year period, following primary episodes of PCP, 60 patients received TMP-SMX and 73 aerosolized pentamidine. Thirty-eight patients who received no secondary prophylaxis served as historical controls.

MAIN OUTCOME MEASURES

The rate of and time to PCP relapse was recorded for patients receiving low-dose TMP-SMX, aerosolized pentamidine, or no prophylaxis.

RESULTS

Only one (1.7%) patient in the TMP-SMX-treated group relapsed, compared with 31 (42.5%) of those in the aerosolized pentamidine group and 21 (55.1%) of those in the control group (P less than 0.0001). Median PCP-free survival times were greater than 1153 days in the TMP-SMX group, 496 days in the pentamidine group, and 265 days in the control group (P less than 0.0001 between all groups). The rate of or time to relapse was not influenced by CD4+ lymphocyte count at the start of prophylaxis, primary therapy of PCP, history of allergy to TMP-SMX, or zidovudine therapy during the period of secondary prophylaxis in any patient group. Both therapies were well tolerated, with three (5%) of those receiving TMP-SMX and four (5%) of those receiving pentamidine discontinuing therapy as a result of side-effects.

CONCLUSIONS

Low-dose TMP-SMX appears to be more effective compared with aerosolized pentamidine as secondary prophylaxis against PCP in HIV-infected patients. Zidovudine was well tolerated by both groups, but did not influence the rate of or time to relapse.