Aztreonam was compared with aminoglycoside antibiotics (tobramycin and amikacin) in a randomized, prospective, clinical trial in serious infections caused by gram-negative bacilli (GNB). A total of 43 evaluable patients with 47 infected sites were treated with aztreonam, and 41 evaluable patients were treated with aminoglycosides for 43 infections. Of patients treated with aztreonam, 17 were bacteremic, as were 12 of those treated with aminoglycosides. Clinical and microbiologic response rates were similar, except that only 5 of 11 patients with pneumonia were considered to be clinically cured with aminoglycoside therapy, while 5 of 6 patients with pneumonia treated with aztreonam were cured. Renal impairment was observed in 9 of 54 patients who received aminoglycoside antibiotics, but in only 2 of 53 patients treated with aztreonam. Hearing impairment developed in one patient treated with tobramycin. Transient elevations of serum transaminase levels occurred in 9 of 53 patients treated with aztreonam and in only 2 of 54 aminoglycoside-treated patients. Diarrhea and superinfection occurred with equal frequency in both groups. Serum concentrations of bactericidal activity could not be correlated with the outcome of therapy. Aztreonam appears to have comparable clinical efficacy with aminoglycoside antibiotics for the treatment of serious infections caused by aerobic and facultative GNB. Its use as a single agent for the treatment of serious lower respiratory infections caused by GNB warrants further evaluation.