Clinical applicability of cystatin C-based renal function assessment and HPLC-based therapeutic drug monitoring for predicting cefepime overexposure

Abstract

Abstract Background Cefepime (CFPM)-associated neurotoxicity is linked to elevated plasma concentrations, particularly in patients with impaired renal function. Although cystatin C-based estimated glomerular filtration rate (eGFRcys) has been proposed as a potentially superior marker of renal function for dose optimization, its clinical applicability for predicting CFPM exposure remains insufficiently established. We developed a clinically applicable high-performance liquid chromatography (HPLC) method for plasma CFPM quantification and evaluated the clinical utility of eGFRcys, with particular focus on its discordance with creatinine clearance (Ccr). Methods This prospective study enrolled 22 hospitalized patients receiving CFPM. Steady-state trough concentrations were measured using the developed HPLC method. Overexposure was defined as ≥ 16 µg/mL, based on a previously reported threshold associated with an increased risk of cefepime neurotoxicity. Associations between trough concentrations and renal function indices were evaluated using Spearman’s correlation, and the impact of discordance between Ccr and body surface area-adjusted eGFRcys [eGFRcys(BSA)] on CFPM exposure was investigated. Results The HPLC assay demonstrated excellent analytical performance, including high precision and robust chromatographic separation under polypharmacy conditions, enabling reliable quantification in routine clinical practice. Despite Ccr-guided dosing, 55% of patients had trough concentrations ≥ 16 µg/mL. Renal function-normalized dose correlated with trough concentrations, with the strongest association observed for eGFRcys(BSA) (ρ = 0.86, P < 0.001). Notably, all patients with preserved renal function according to Ccr (≥ 60 mL/min) but impaired renal function according to eGFRcys(BSA) (< 60 mL/min) exhibited trough concentrations ≥ 16 µg/mL, suggesting that Ccr-based assessment alone may underestimate the risk of CFPM overexposure. Furthermore, all patients with impaired renal function (Ccr < 60 mL/min) also exhibited trough concentrations ≥ 16 µg/mL. Conclusions This validated HPLC method provides a practical approach for routine CFPM therapeutic drug monitoring using conventional laboratory infrastructure. eGFRcys(BSA) showed the strongest correlation with CFPM exposure among the renal function indices evaluated and may improve identification of patients at risk of overexposure. However, renal function assessment alone was insufficient to explain exposure variability in patients with impaired renal function. These findings support the clinical applicability of eGFRcys(BSA) for CFPM dose optimization while highlighting the complementary role of HPLC-based therapeutic drug monitoring for individualized minimize the risk of overexposure.