The need for MAC prophylaxis in advanced HIV disease is recognized. Prophylactic therapy is complicated, however, due to drug toxicity, potential drug interactions, patient noncompliance with multiagent regimens, microbial drug resistance, and high treatment costs. This study assesses the feasibility of deferring MAC prophylaxis in patients whose CD4+ cell counts rebound in response to antiretroviral therapy.

In this double-blind, placebo-controlled trial, patients are randomized to receive azithromycin weekly or matching placebo. Patients are switched to open-label azithromycin if they have an average CD4+ cell count less than 50 cells/mm3 on 2 consecutive counts obtained at least 48 hours apart (preferably no more than 60 days).

[AS PER AMENDMENT 10/23/97: SUBSTUDY CPCRA 054: A subset of participants of CPCRA 048 have oropharyngeal swabs taken at baseline and 4 months after randomization. Antibiotic susceptibility tests are then performed on isolates of S. pneumoniae.] [AS PER AMENDMENT 9/3/99: Participants in protocol version 3.0 are followed for data collection purposes every 4 months for 18 months until January 31, 2001. Data collection includes CD4 cell counts, viral RNA measurements, macrolide susceptibility in MAC isolates, and antibiotic susceptibility of isolates from bacterial pneumonia episodes.]