Pharmacokinetic/pharmacodynamic-guided gentamicin dosing in critically ill patients: a revisit of the Hartford nomogram

The appropriateness of the Hartford nomogram based on 7 mg/kg gentamicin with administration interval adjustment is questioned in critically ill patients. This study aimed to perform a pharmacokinetic/pharmacodynamic (PK/PD) evaluation of the Hartford nomogram and to assess the influence of PK/PD indices on gentamicin dosing. Gentamicin data were extracted from a critical care database to construct the population PK model. Simulations were performed to evaluate the probability of target attainment (PTA) and risk of toxicity for the gentamicin Hartford nomogram. C_max/MIC ≥ 10 and AUC/MIC ≥ 100 were the PK/PD targets considered, and the non-toxicity targets included concentration < 0.5 mg/L for at least 4 h within a dosing interval and trough concentration < 1 mg/L. A one-compartment model was optimal to describe gentamicin PKs, and creatinine clearance (CL_Cr) was included as a time-varying covariate on gentamicin clearance. The PTA of C_max/MIC ≥ 10 (MIC = 1 mg/L) for the Hartford nomogram was suboptimal after the first dose but was desirable (near or greater than 90%) at steady-state, and > 90% PTA based on AUC/MIC ≥ 100 was readily achieved after the first dose in patients with CL_Cr < 60 mL/min. Significant PTA differences between the PK/PD targets were observed at an MIC of 2 mg/L, but the PTAs were all low. The predicted risk of toxicity was high regardless of the applied toxicity targets. The Hartford nomogram provided adequate gentamicin exposure in critically ill patients with an MIC ≤ 1 mg/L by considering the combined PK/PD indices.