Acute kidney injury (AKI) requiring continuous renal replacement therapy (CRRT) is a common complication in critical illness and has a significant impact on pharmacokinetic factors determining drug exposure, including absorption, distribution, transport, metabolism, and clearance. In this review, we provide a practical guide to drug dosing considerations in critically ill patients undergoing CRRT, focusing on the most commonly used analgesic, anticonvulsant, and psychotropic medications in the clinical care of critically ill patients. A literature search was conducted to identify articles in which drug dosing was evaluated in adult patients receiving CRRT between the years 1980 and 2020. We included articles with pharmacokinetic/pharmacodynamic analyses and those that described medication clearance via CRRT. A summary of the data focused on practical pharmacokinetic and pharmacodynamic principles is presented, with recommendations for drug dosing of analgesics, anticonvulsants, and psychotropic medications. Pharmacokinetic and pharmacodynamic studies to guide drug dosing of analgesics, anticonvulsants, and psychotropic medications in critically ill patients receiving CRRT are sparse. Considering the widespread use of these medications, narrow therapeutic index of these drug classes, and risks of over- and underdosing, additional studies in patients receiving CRRT are needed to inform drug dosing.

Background

In acute kidney injury (AKI), medication dosing based on Cockcroft-Gault creatinine clearance (CrCl) or Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) estimated glomerular filtration rates (eGFR) are not valid when serum creatinine (SCr) is not in steady state. The aim of this study was to determine the impact of a kinetic estimating equation that incorporates fluctuations in SCrs on drug dosing in critically ill patients.

Methods

We used data from participants enrolled in the NIH Acute Respiratory Distress Syndrome Network Fluid and Catheters Treatment Trial to simulate drug dosing category changes with the application of the kinetic estimating equation developed by Chen. We evaluated whether kinetic estimation of renal function would change medication dosing categories (≥60, 30-59, 15-29, and <15mL/min) compared with the use of CrCl or CKD-EPI eGFR.

Results

The use of kinetic CrCl and CKD-EPI eGFR resulted in a large enough change in estimated renal function to require medication dosing recategorization in 19.3% [95 CI 16.8%-21.9%] and 23.4% [95% CI 20.7%-26.1%] of participants, respectively. As expected, recategorization occurred more frequently in those with AKI. When we examined individual days for those with AKI, dosing discordance was observed in 8.5% of total days using the CG CrCl and 10.2% of total days using the CKD-EPI equation compared with the kinetic counterparts.

Conclusion

In a critically ill population, use of kinetic estimates of renal function impacted medication dosing in a substantial proportion of AKI participants. Use of kinetic estimates in clinical practice should lower the incidence of medication toxicity as well as avoid subtherapeutic dosing during renal recovery.