The pharmacokinetics of ceftazidime were investigated during intermittent (II) and continuous (CI) infusion in eight healthy male volunteers in a crossover fashion. The total daily dose was 75 mg/kg of body weight per 24 h in both regimens, given in three doses of 25 mg/kg/8 h (II) or 60 mg/kg/24 h with 15 mg/kg as a loading dose (CI). After the third dose (II) and during CI, serum and blister fluid samples were taken. Seven new blisters were raised for each timed sample by a suction blister technique. Blisters took 90 min to form. Samples were then taken from four blisters (A samples) and 1 h later were taken from the remaining three (B samples). The concentration of ceftazidime was determined using a high-performance liquid chromatography method. After II, the concentrations in serum immediately after infusion (t = 30 min) and 8 h after the start of the infusion were 137.9 (standard deviation [SD], 27.5) and 4.0 (SD, 0.7) μg/ml, respectively. The half-life at α phase (t( 1/2 α)) was 9.6 min (SD, 4.6), t( 1/2 β) was 94.8 min (SD, 5.4), area under the concentration-time curve (AUC) was 285.4 μg · h/ml (SD, 22.7), total body clearance was 0.115 liter/h · kg (SD, 0.022), and volume of distribution at steady state was 0.178 liter/kg (SD, 0.023). The blister fluid (A) samples showed a decline in concentration parallel to that of the concentrations in serum during the elimination phase with a ratio of 1:1. The t( 1/2 ) of the A samples was 96.4 min (SD, 3.2). The concentration of ceftazidime in the B blister fluid samples was significantly higher (27%) than in the A samples over time. This shows that blister may behave as a separate compartment and establishes the need to raise new blisters for each timed sample. The mean AUC/h during continuous infusion was 21.3 μg · h/ml (SD, 3.0). The total body clearance was 0.113 liter/h · kg (SD, 0.018), the urinary clearance was 0.105 liter/h · kg (SD, 0.012), and the ceftazidime/creatinine clearance ratio was 0.885. The mean AUC of blister fluid per hour was 84.5% (18.0 μg · h/liter; SD, 3.6) compared with that of serum. The A samples did not differ significantly from the B samples. The implications of continuous infusion of beta-lactams for treatment of serious infections are discussed.
