Comparative efficacy and toxicity of amikacin/carbenicillin versus gentamicin/carbenicillin in leukopenic patients. A randomized prospective trail

Between July 1974 and August 1976, 157 leukopenic patients who had a neoplastic disease or bone marrow failure syndrome were assigned at random to receive either amikacin (7.5 mg/kg every 8 hours)/carbenicillin (100 mg/kg every 4 hours) or gentamicin (2 mg/kg loading, 1.5 mg/kg every 8 hours)/ carbenicillin on admission to the hospital or on completion of a previous course of antibiotic therapy. Fever and evidence suggesting gram-negative rod infection prompted initiation of therapy, and its duration was determined by the patient's clinical course and culture data. A total of 155 courses of amikacin therapy and 140 courses of gentamicin therapy were given. The over-all clinical response rate was 75 per cent for both antibiotic regimens. Cure of urinary tract or soft tissue infections was observed in over 90 per cent of the cases. In gram-negative bacillemias only 15 of 23 patients (65 per cent) treated with amikacin and 10 of 17 patients (59 per cent) given gentamicin survived. Two bacteremic patients with gentamicin-resistant (but amikacin-sensitive) rods were assigned at random to receive gentamicin; they died. Improved survival in both treatment groups was associated with the use of combinations of drugs that interacted synergistically in vitro. There was no significant difference in the incidence of ototoxicity and nephrotoxicity. For patients with infections due to an organism susceptible to both amikacin and gentamicin, a regimen which incorporates carbenicillin and either drug is equally effective, but empiric use of amikacin is indicated when there is a reasonable possibility of infection due to gentamicin-resistant organisms.