Low cefpirome levels during twice daily dosing in critically ill septic patients: pharmacokinetic modelling calls for more frequent dosing

Objectives: To measure plasma levels and pharmacokinetics of cefpirome in c ritically ill septic patients with normal renal function. To use the pharma cokinetic model to simulate alternate dosing regimens and identify those th at predict sustained levels. Design and setting: A prospective, open-label, descriptive study in a 22-be d, multidisciplinary, adult ICU in a university-affiliated, tertiary referr al hospital. Patients: Twelve adults with normal renal function on enrolment and with su spected or documented sepsis in whom cefpirome was judged to be the appropr iate therapy by the managing clinician. Interventions: Cefpirome 2 g was infused intravenously over 3 min every 12 h. Timed blood samples were taken prior to and during two dosing intervals. Urine was collected for creatinine clearance determination. Measurements and results: Two patients were non-evaluable due to renal dysf unction post-enrolment. The median cefpirome trough level was 1.1 mg/l (ran ge 0.5-8.1 mg/l) after the initial dose and 1.4 mg/l (range < 0.5-15.9 mg/l ) at 'steady state'. The volumes of distribution and elimination half-lives were greater and more variable than in healthy volunteers. Pharmacokinetic simulation predicted that more frequent bolus dosing, increased doses and continuous infusions would result in concentrations greater than 4 mg/l for over 60 % of the dosing interval for all patients. Conclusions: Cefpirome 2 g twice daily produced low plasma troughs in a num ber of patients. Our data suggest that this drug regimen may be inadequate for successful treatment of difficult-to-treat infections in critically ill patients with normal renal function.