Antimicrobial practice. A prospective laboratory-based audit of gentamicin use and therapeutic monitoring

We report a study investigating the proportion of patients in whom therapeutic serum concentrations of gentamicin are achieved in the early phase of treatment and to determine the underlying reasons for sub-optimal therapy. A laboratory based prospective study of 83 courses of gentamicin was performed (excluding patients receiving renal replacement therapy or with bacterial endocarditis) in a London teaching hospital. Of 83 monitored courses 74 had paired levels tested. Initial trough concentrations were > 2 mg/L in nine (12%) and were < 1 mg/L in 51 (69%) indicating levels were seldom in the toxic range. The first monitored peaks were sub-therapeutic in 58 (78%) courses using a cut-off of 5 mg/L and only seven (9%) were greater than 6 mg/L. Of these seven patients, six had troughs levels of greater than 2 mg/L. Of those with initial levels below 5 mg/L, 33 had further serum levels tested, of which 26 (79%) remained below 5 mg/L. Of those for whom dosing information was available 73% received 80 mg t.d.s. and the mean dose for those for which body weight was known was 3.3 mg/kg/day (S.D. = 0.7). Most patients continued to receive 8-hourly dosing with 80 mg of gentamicin, leading to subtherapeutic peak levels in the majority of cases. In those patients with satisfactory peaks, such dosing frequently leads to elevated troughs. This suggests that such dosing practices should be abandoned and replaced with dosing based on body weight and divided into no more than two daily doses.