Empiric antimicrobial therapy with aztreonam or ceftazidime in gram-negative septicemia.

: In an open, comparative study, 225 patients with severe underlying diseases and suspected gram-negative bacillary septicemia were randomly assigned to receive aztreonam or ceftazidime empirically, 2 g intravenously three times daily. Twenty-five patients in the aztreonam group and 22 in the ceftazidime group had blood cultures that grew aerobic gram-negative bacilli and were evaluable for response to therapy. All pathogenic strains were sensitive to treatment. In the aztreonam group, 22 (88 percent) patients had cures, three (12 percent) had failures, and seven (28 percent) had development of superinfections (five were caused by gram-positive cocci and two by fungi). In the ceftazidime group, 18 (82 percent) patients had cures, one had improvement, three (14 percent) had failures, and three had superinfections. The median peak serum bactericidal activity was 1:2,048 after aztreonam administration and 1:512 after ceftazidime administration. Failures were not related to resistant strains or to low serum bactericidal activity.