In order to assess the efficacy and toxicity of ceftazidime as a substitute for aminoglycosides in the treatment of intra-abdominal sepsis, a prospective randomized trial was conducted. Ninety-four patients (49% trauma) were randomized to receive ceftazidime/clindamycin (CAZ/C) (n = 47) or tobramycin/clindamycin (T/C) (n = 47). CAZ (2.0 gm) and C (0.9 gm) were administered intravenously every 8 hours while T dosage was adjusted to maintain peak (5-8 mg/L) and trough (less than 2 mg/L) concentrations. Age, sex, baseline serum creatinine, and etiology of infection were comparable in the two groups. Clinical cure was similar in culture-positive and culture-negative patients who received CAZ/C (94% vs 88%). The clinical cure rate however was significantly lower in the T/C culture positive (73%) than in the culture negative patients (100%) (P = 0.016). Pathogenic organisms were eradicated in 100% (30/30) and 76% (13/17) of CAZ/C and T/C patients, respectively (P = 0.0006). Nephrotoxicity Nephrotoxicity or ototoxicity was observed in none of the CAZ/C patients and in one and two T/C patients, respectively. CAZ/C more effectively eradicated the bacteria isolated from these patients and no significant difference in clinical response was observed in culture-positive patients. These findings plus the lack of toxicity suggest that CAZ/C is an effective alternative for treatment of IAI.