A retrospective analysis of pharmacokinetic/pharmacodynamic indices as indicators of the clinical efficacy of ciprofloxacin.

A retrospective analysis of the relationship between estimated pharmacokinetic/pharmacodynamic indices and the reported efficacy of ciprofloxacin has been carried out using different correlation models. f1.gif" BORDER="0">, T(ss) > MIC, f2.gif" BORDER="0"> and AUIC(ss) were calculated for each clinical case included in the study, from simulated plasma level curves corresponding to the dosage regimen administered. A univariate correlation analysis was performed considering efficacy (%) as the dependent variable and indices as the independent variables according to linear and non-linear pharmacokinetic-pharmacodynamic models (PK-PD models). The results prove that log-transformation of the independent variable improves the data fitting to linear model. The four estimated indices show a log-linear relationship with outcome, T(ss) > MIC and AUIC(ss) being the parameters best correlated with percentage efficacy. The E(max) model with intrinsic response is an additional correlation strategy for T(ss) > MIC, leading to estimated values of E(max) and E(0) of 100.34 +/- 25.09% and 24.40 +/- 11.7%, respectively. The wide range of bacteria responsible for the infections considered, including Gram-positive pathogens such as staphylococci, might explain the good correlation between T(ss) > MIC and percentage efficacy found for ciprofloxacin in this study.