Abstract: FR-OR66
Nuclear Magnetic Resonance-Based Glomerular Filtration Rate Estimation Is More Closely Associated with Serum Vancomycin Target Trough Achievement vs. Serum Creatinine-Based Equations
Session Information
- Innovative AKI Strategies and Drug Discoveries
October 25, 2024 | Location: Room 7, Convention Center
Abstract Time: 04:50 PM - 05:00 PM
Category: Pharmacology (PharmacoKinetics, -Dynamics, -Genomics)
- 2000 Pharmacology (PharmacoKinetics, -Dynamics, -Genomics)
Authors
- Barreto, Jason N., Mayo Clinic Minnesota, Rochester, Minnesota, United States
- Schwäble Santamaría, Amauri, Numares, Regensburg, Germany
- Grassi, Marcello, Numares, Regensburg, Germany
- Robertson, Andrew, Numares, Regensburg, Germany
- Lieske, John C., Mayo Clinic Minnesota, Rochester, Minnesota, United States
- Meeusen, Jeff W., Numares, Regensburg, Germany
- Schiffer, Eric, Numares, Regensburg, Germany
- Barreto, Erin F., Mayo Clinic Minnesota, Rochester, Minnesota, United States
Background
The clinical utility of glomerular filtrate rate (GFR) estimation based on a panel of analytes (creatinine, cystatin C, myoinositol and valine) measured by nuclear magnetic resonance spectroscopy (GFRNMR) as compared to standard eGFR equations based on serum creatinine for patients receiving vancomycin is unclear.
Methods
This prospective, single-center study included consecutive adults prescribed intravenous vancomycin at Mayo Clinic in Rochester, MN, from 2022 to 2024. Vancomycin indication, dose, frequency, and goal concentration range were determined by the clinical care team. Pearson’s product-moment correlation analysis and ordinary least squares regression techniques were used to compare GFR estimating equations as predictors of serum vancomycin trough concentration, accounting for relevant covariates.
Results
66 patients were enrolled. The median (IQR) age was 61 (50, 70) years, 45 (68%) were male, and 59 (89%) were non-Hispanic White individuals. Vancomycin was most often prescribed for skin, soft tissue, or musculoskeletal infections (n=24, 32%). Thirty-five (53%) patients had a steady state trough concentration within the goal range of 10-20 mg/L. Seven (11%) patients had troughs > 20 mg/L and six (9%) patients developed AKI within seven days of vancomycin initiation. Both GFRNMR and CKD-EPI 2021Cr were negatively correlated with vancomycin concentrations (GFRNMR: t64 = -6.81, p-value < 0.001, r = -0.65 [-0.77, -0.48]; CKD-EPI 2021Cr: t62 = -4.22, p-value < 0.001, r = -0.47 [-0.64, -0.26]). In multivariable analyses, the GFRNMR inclusive model for prediction of vancomycin trough concentrations had a higher adjusted R2 (mean [95%CI]) 0.51 [0.45, 0.6] than the CKD-EPI 2021Cr inclusive model at 0.34 [0.29, 0.4] (p-value = 0.008).
Conclusion
Vancomycin trough prediction models with GFRNMR were more predictive than models with CKD-EPI 2021Cr in hospitalized patients. Future research should evaluate the potential for GFRNMR to guide individualized vancomycin dosing.
Funding
- Commercial Support – Numares