Improved knowledge of the time course of antimicrobial activity should provide useful information for designing optimal dosage regimen. In contrast to $eta$-lactam, aminoglycosides tend to induce a prolonged postantibiotic effect against aerobic or facultative gram negative rods and clinical outcome was significantly correlated with achieving peak serum concentrations. The objective of this study was to compare the efficacy, safety of e same total daily dose of amikacin and gentamicin given either twice or thrice daily in the treatment of patients. Consecutive patients over 20 years old with a suspected or confirmed infection for which an aminoglycosides was indicated were eligible. Exclusion criteria were known allergy to aminoglycosides, renal impairment, granulocytopenia and pregnancy. Patients were treated with intravenous amikacin 15 mg/kg/day or gentamicin 4.5 mg/kg/day either in two devided or in three devided. Seventy-four patients with infection were enrolled in this study of amikacin twice daily (A2, n=29), gentamicin twice daily (G2, n=8) vs amikacin thrice daily (A3, n=30), gentamicin thrice daily (G3, n=7). Baseline characteristics were comparable in G2 and G3. The clinical cure rate (including partial improve) were $89.0\%;and;86.0\%$ in A2 group and A3 group respectively. The bacteriologic cure rate were $99.0\%;and;85.7\%$ in A2 group and in A3 group respectively. The clinical and bacteriologic effects were difficult to compare G2 with G3, because of the small numbers of patients. The serum creatinin rose in $3.44\%$ (1 in 29) of patients in the A2 group compared to $13.3\%$ (4 in 30) in e A3 group. Although audiometry was not performed, there was no clinical evidence of ototoxicity in any of the patients. In our opinion, twice-daily regimen of aminoglycosides is more effective and less nephrotoxic than thrice-daily regimen.