Suboptimal Aminoglycoside Dosing in Critically Ill Patients : Therapeutic Drug Monitoring (original) (raw)

Original Article

Rea, Rhonda S PharmD*; Capitano, Blair PharmD*; Bies, Robert PharmD, PhD†; Bigos, Kristin L PhD‡; Smith, Randall PhD‡; Lee, Howard MD, PhD§

From the *Department of Pharmacy and Therapeutics, School of Pharmacy; †Department of Pharmaceutical Sciences and Psychiatry, Schools of Pharmacy and Medicine; ‡Department of Pharmaceutical Sciences, School of Pharmacy, University of Pittsburgh, Pittsburgh, Pennsylvania; and §Center for Drug Development Science, Department of Biopharmaceutical Sciences, School of Pharmacy, University of California, Washington Center, Washington, DC.

Received for publication February 12, 2008; accepted August 20, 2008.

Current address: Rhonda S. Rea, Metabolism, US Medical Affairs, Sanofi-Aventis Pittsburgh, Pennsylvania; Blair Capitano, Pfizer Pittsburgh, Pennsylvania; Kristin L. Bigos, Clinical Brain Disorders Branch, National Institute of Mental Health, NIH Bethesda, Maryland.

Correspondence: Howard Lee, MD, PhD, Center for Drug Development Science, Department of Biopharmaceutical Sciences, School of Pharmacy, University of California, UC Washington Center, 1608 Rhode Island Avenue, NW, Washington, DC 20036 (e-mail: [email protected]).

K.L.B. received support from NIH F31MH076420. R.B. received support from NIBIB P41 EB-001975.

Abstract

Maximal aminoglycoside (AG) killing requires that the ratio of peak serum concentrations (_C_max) to the minimum inhibitory concentration (MIC) of the pathogen exceeds by ≥10. This has been shown to hasten resolution of infection in the general patient population. It was postulated that critically ill patients, likely to have larger intravascular volumes, are underdosed. The primary aim was to determine _C_max to MIC target attainment rate in medical intensive care unit (MICU) patients. A retrospective review of MICU patients who received at least 1 intravenous dose and serum concentration of either gentamicin or tobramycin was performed. A population pharmacokinetic model was developed, and MIC distributions for AG were used in determining the _C_max/MIC and in calculating the probability of attaining the pharmacodynamic (PD) target. One hundred two unique patients with 211 AG concentrations were analyzed to determine population pharmacokinetic parameters. Mean maximum clearance (CL) was 3.14L/h (95% confidence interval: 1.26-4.54 L/h), and mean volume of distribution (V) was 53 L (95% confidence interval: 38-66.8 L/h). Glomerular filtration rate and standardized body weight were identified as significant covariates for clearance in the final model. Standardized body weight also significantly affected V. There was only a 20% and 40% probability that patients receiving 7 mg/kg of gentamicin and tobramycin, respectively, will achieve PD target over the range of MIC distributions. Based on these data, the majority of critically ill patients would not be predicted to achieve the PD target under current dosing regimens. This may be a result of intensive care unit patients having a larger volume of distribution than reported in the literature. Future recommendations for treating gram-negative infections in the MICU population include using initial doses of 7 mg/kg of either gentamicin or tobramycin, measuring _C_max after the first dose, and determining MIC for the pathogen(s) with adjustment of subsequent doses to achieve the PD target.