Population pharmacokinetics and exposure-response analysis of tigecycline in patients with hospital-acquired pneumonia.

ABSTRACT

BACKGROUND: Tigecycline has been widely used to treat hospital-acquired pneumonia (HAP) off-label since it is effective against a wide range of multidrug-resistant bacteria. However, no recommended dosage for this indication has been evaluated, resulting in possible inadequate treatment. AIMS: The aims of this study are to establish the population pharmacokinetic (PPK) model of tigecycline in Chinese patients with HAP, as well as to evaluate the exposure-response relationship for the treatment of HAP with multidrug-resistant gram-negative bacteria. METHODS: A PPK analysis of tigecycline was conducted on pooled data from 328 blood samples obtained from 89 patients with HAP. Tigecycline plasma concentrations were measured by a two-dimensional liquid chromatographic system and the data were analysed using Phoenix NLMETM software. Exposure-response analyses for efficacy were performed based on the data from 79 HAP patients with multidrug-resistant gram-negative infections. Classification and regression tree and logistic regression analyses were employed to identify which pharmacokinetic-pharmacodynamic (PK-PD) indices and magnitudes were the significant predictors of tigecycline efficacy. RESULTS: A two-compartment model with zero-order absorption and first-order elimination adequately described the data. A larger body weight was associated with increased central volume of distribution and clearance (P < .005), and increased age, baseline creatinine concentration and aspertate aminotransferase were associated with decreased clearance (P < .005). The AUC0-12h  × V/MIC ratio, APACHEII score and combined Pseudomonas aeruginosa infection are the strong predictors for tigecycline clinical response. Classification and regression tree analyses indicated that the combination of APACHEII score < 24 and AUC0-12h  × V/MIC ratio ≥ 100 was associated with clinical success. CONCLUSIONS: The proposed PPK model may serve as the basis for estimating tigecycline exposure for PK-PD analyses, and the PK-PD index and magnitude found in this study could be used for designing proper dosage regimens of tigecycline.