Since the bactericidal effects of beta-lactam antibiotics are time dependent, the optimum strategy for their administration could be continuous infusion. In this prospective, randomised controlled trial to evaluate the clinical efficacy of continuous infusion therapy, we evaluated the outcomes for 40 septic critically ill patients who received piperacillin either continuously (2 g intravenously (i.v.) over 0.5 h as a loading dose followed by 8 g i.v. daily over 24 h (n=20)) or as an intermittent infusion (3 g i.v. every 6h over 0.5 h (n=20)). Results from our study demonstrated that the clinical efficacy of piperacillin as a continuous infusion is superior to intermittent administration in critically ill patients. Change in APACHE II scores from baseline at the end of the second, third and fourth days, respectively, were 4.1, 5.1 and 5.2 for continuous infusion and 2.0, 2.6 and 2.8 for intermittent infusion (P< or =0.04). Considering minimum inhibitory concentrations (MICs) of 16 microg/mL and 32 microg/mL, the percentage of time for which piperacillin plasma concentrations were higher than the MIC (%T>MIC) was calculated for each patient in the two groups. For MICs of 16 microg/mL and 32 microg/mL, %T>MIC in the continuous infusion group was 100% and 65% of the dosing interval, respectively; in the intermittent infusion group, %T>MIC was only 62% and 39% of the dosing interval. There was a significant relationship between clinical results and laboratory data. It was shown that the superiority of the clinical efficacy of continuous infusion could be related to piperacillin pharmacodynamics. Continuous infusion significantly reduced the severity of illness as demonstrated by APACHE II scores during therapy.