Clearance of amino acids, trace elements, and carnitine during CRRT: the carnitine dichotomy

Continuous renal replacement therapy (CRRT) provides vital organ support in acute kidney injury but is associated with the inadvertent extracorporeal clearance of essential low-molecular-weight nutrients. While the loss of amino acids and trace elements is clinically recognized, the specific dialytic filtration dynamics of free carnitine versus esterified acylcarnitine fractions remain poorly elucidated. This study aimed to quantitatively evaluate the 24-hour systemic plasma shifts and direct extracorporeal clearance of these metabolites in critically ill patients undergoing CRRT. This prospective observational study included 44 critically ill patients with stage 2 or 3 acute kidney injury, divided into a CRRT group ( n = 23; CVVHD mode, 30 mL/kg/h) and a non-CRRT control group ( n = 21). Blood samples were collected at baseline (h0) and 24 h (h24) to quantify trace elements, amino acids, free carnitine, and acylcarnitine subgroups [short- (SCAC), medium- (MCAC), and long-chain (LCAC)]. In the CRRT cohort, 24-hour effluent was continuously sampled to calculate actual dialytic mass removal and clearance kinetics. Statistical evaluations incorporated false discovery rate (FDR) corrections and advanced confounder-adjusted models (IPTW and PCA) to isolate the direct impact of extracorporeal clearance from baseline disease severity and systemic lactate. At baseline, specific acylcarnitines and hypercatabolism markers (1- and 3-methylhistidine) were significantly elevated in the CRRT cohort. Over 24 h, CRRT patients exhibited a statistically significant systemic depletion of free carnitine and SCACs. While substantial amounts of amino acids were removed via effluent, their systemic declines did not correlate with dialytic clearance. Conversely, the systemic reduction of specific SCACs (e.g., C2, C4-OH) strongly correlated with effluent clearance. Confounder-adjusted models (IPTW and PCA) confirmed that this active 24-hour dialytic depletion of SCACs remained completely independent of baseline disease severity and systemic lactate. Notably, all LCACs were completely undetectable in the effluent, demonstrating zero extracorporeal clearance. In this small, single-center, 24-hour observational study, CRRT was associated with reductions in free carnitine and short-chain acylcarnitines, whereas long-chain acylcarnitines were not detected in the effluent. These findings require further validation in larger studies incorporating different CRRT modalities and clinically relevant outcomes. The study was retrospectively registered with ClinicalTrials.gov (Identifier: NCT07467889; Date of registration: February 20, 2026).