Background Ketamine is widely used in paediatric practice because of its analgesic and sedative properties. Although cholangiopathy associated with chronic recreational ketamine use is well recognised, liver injury related to prolonged ketamine infusion in the intensive care setting has only recently been described in adults and has not yet been investigated in the paediatric population. Aim of the study To evaluate the association between prolonged ketamine infusion and the development of drug-induced liver injury (DILI) in critically ill paediatric patients. Secondary objectives were to characterise the biochemical pattern of hepatotoxicity, analyse the dose- and time-dependent relationship between ketamine exposure and liver injury, identify predictive cut-off values, and assess the long-term reversibility of hepatotoxicity. Materials and methods A retrospective observational study was conducted at the Paediatric Intensive Care Unit (PICU) and the Paediatric Gastroenterology Unit of the University Hospital of Padua. Paediatric patients aged <18 years admitted to the PICU between January 2017 and September 2023 were included if they required invasive mechanical ventilation for at least 5 days and received continuous ketamine infusion for more than 12 hours. Hepatotoxicity was classified according to the Common Terminology Criteria for Adverse Events (CTCAE) version 5.0. Clinical and laboratory follow-up was performed until May 2026. Results A total of 106 patients (40% male) were included, with a median age of 1.7 years (IQR 0.5–5.6). The median duration of ketamine infusion was 8.5 days (IQR 5–13.5), with a median cumulative dose of 321,000 mcg/kg. The overall incidence of hepatotoxicity was 81%. The predominant pattern was canalicular (75%), followed by hepatocellular (59%) and cholestatic (22%). Both infusion duration and cumulative dose were significantly correlated with peak gamma-glutamyl transferase (GGT) and alanine aminotransferase (ALT) levels (p<0.001). The median peak GGT level was 328.5 U/L, representing a 9.8-fold increase from baseline, and was reached after a median infusion duration of 11 days. Receiver operating characteristic (ROC) curve analysis identified predictive thresholds for hepatotoxicity of an infusion duration >4.9 days (AUC 0.814; sensitivity 84.9%; specificity 65%) and a cumulative dose >164,022.9 mcg/kg (AUC 0.801; sensitivity 76.7%; specificity 70%). In multivariable logistic regression analyses, ketamine exposure was an independent predictor of canalicular hepatotoxicity (aOR 1.68 for infusion duration and aOR 3.09 for cumulative dose; p<0.001 for both), and to a lesser extent of hepatocellular hepatotoxicity. At follow-up (median 4.8 months), 66% of patients had persistent abnormalities in liver function tests, and 23.5% had developed chronic liver disease. Radiological hepatic abnormalities increased from 29.2% at baseline to 63.2% following ketamine exposure, with de novo biliary dilatation or strictures observed in 29% of cases; 62.5% of these abnormalities persisted at the last available follow-up. Two patients (1.9%) developed secondary sclerosing cholangitis, and one patient (0.9%) required orthotopic liver transplantation. Conclusions Prolonged ketamine infusion is an independent, dose- and time-dependent determinant of hepatotoxicity, predominantly exhibiting a canalicular pattern, in critically ill paediatric patients. Thresholds of 4.9 days of infusion and a cumulative dose of 164,022.9 mcg/kg were identified as predictors of liver injury with good sensitivity and specificity. A substantial proportion of patients showed persistent liver abnormalities during long-term follow-up. These findings support the need for careful hepatic monitoring during prolonged ketamine infusions and for extended hepatological follow-up after treatment.
Introduzione La ketamina è ampiamente utilizzata in ambito pediatrico per le sue proprietà analgosedative. Sebbene la colangiopatia associata all’uso cronico ricreativo sia nota, il danno epatico correlato all’infusione prolungata in terapia intensiva è stato descritto solo recentemente negli adulti e non è stato finora studiato nella popolazione pediatrica. Scopo dello studio Valutare l’associazione tra infusione prolungata di ketamina e sviluppo di danno epatico farmaco-indotto (DILI) in pazienti pediatrici critici. Questo studio si propone di valutare l'associazione tra l'esposizione a infusione prolungata di ketamina e lo sviluppo di DILI in una coorte di pazienti pediatrici ricoverati in terapia intensiva. Gli obiettivi secondari includono la caratterizzazione del pattern biochimico di epatotossicità, l'analisi della relazione dose- e tempo-dipendente tra le variabili di infusione e il danno epatico, l'identificazione di cut-off predittivi e la valutazione della reversibilità dell’epatotossicità nel lungo termine. Materiali e metodi Studio osservazionale retrospettivo condotto presso la Terapia Intensiva Pediatrica e l’Unità di Gastroenterologia Pediatrica dell’Azienda Ospedaliera-Universitaria di Padova. Sono stati inclusi pazienti di età <18 anni ricoverati in Terapia Intensiva Pediatrica (TIP) tra gennaio 2017 e settembre 2023, sottoposti a ventilazione meccanica invasiva per almeno 5 giorni e trattati con infusione continua di ketamina per oltre 12 ore. L’epatotossicità è stata classificata secondo i criteri CTCAE v5.0. Il follow-up clinico e laboratoristico è proseguito fino a maggio 2026. Risultati Sono stati inclusi 106 pazienti (40% maschi), con età mediana di 1,7 anni (IQR 0,5–5,6). La durata mediana dell’infusione di ketamina è stata di 8,5 giorni (IQR 5–13,5), con dose cumulativa mediana di 321.000 mcg/kg. L’incidenza complessiva di epatotossicità è risultata pari all’81%. Il pattern prevalente è stato quello canalicolare (75%), seguito da quello epatocellulare (59%) e colestatico (22%). La durata dell’infusione e la dose cumulativa hanno mostrato una correlazione significativa con i picchi di GGT e ALT (p<0,001). Il picco mediano di GGT è stato di 328,5 U/L, corrispondente a un incremento di 9,8 volte rispetto al basale, raggiunto dopo una mediana di 11 giorni di infusione. L’analisi delle curve ROC ha identificato come soglie predittive di epatotossicità una durata di infusione >4,9 giorni (AUC 0,814; sensibilità 84,9%; specificità 65%) e una dose cumulativa >164.022,9 mcg/kg (AUC 0,801; sensibilità 76,7%; specificità 70%). Nei modelli di regressione logistica multivariata la ketamina è risultata un predittore indipendente di epatotossicità canalicolare (aOR 1,68 per durata; aOR 3,09 per dose; p<0,001 e, in misura minore, epatocellulare. Al follow-up (mediana 4,8 mesi), il 66% dei pazienti presentava alterazioni persistenti degli indici epatici e il 23,5% aveva sviluppato un’epatopatia cronica. Le anomalie radiologiche epatiche sono aumentate dal 29,2% al 63,2%, con dilatazioni o stenosi biliari de novo nel 29% dei casi; il 62,5% di tali lesioni persisteva all’ultimo controllo. Due pazienti (1,9%) hanno sviluppato colangite sclerosante secondaria e uno (0,9%) ha richiesto trapianto epatico. Conclusioni L’infusione prolungata di ketamina rappresenta un determinante indipendente, dose- e tempo-dipendente, di epatotossicità, prevalentemente canalicolare, nei pazienti pediatrici critici. Sono state identificate una soglia di 4,9 giorni di durata e di 164.022,9 mcg/kg di dose cumulativa, in grado di predire con elevata sensibilità e specificità l’insorgenza di danno epatico. Una quota non trascurabile di pazienti ha mostrato persistenza dell’epatopatia nel follow-up a lungo termine. I risultati supportano l’importanza del monitoraggio epatico durante infusioni prolungate di ketamina, nonché la necessità di un follow-up epatologico prolungato.
Dose and Duration Dependent Liver Toxicity Following Prolonged Ketamine Exposure in Critically Ill Children: Evidence Supporting Hepatic Monitoring.
FILIPPI, FRANCESCA
2025/2026
Abstract
Background Ketamine is widely used in paediatric practice because of its analgesic and sedative properties. Although cholangiopathy associated with chronic recreational ketamine use is well recognised, liver injury related to prolonged ketamine infusion in the intensive care setting has only recently been described in adults and has not yet been investigated in the paediatric population. Aim of the study To evaluate the association between prolonged ketamine infusion and the development of drug-induced liver injury (DILI) in critically ill paediatric patients. Secondary objectives were to characterise the biochemical pattern of hepatotoxicity, analyse the dose- and time-dependent relationship between ketamine exposure and liver injury, identify predictive cut-off values, and assess the long-term reversibility of hepatotoxicity. Materials and methods A retrospective observational study was conducted at the Paediatric Intensive Care Unit (PICU) and the Paediatric Gastroenterology Unit of the University Hospital of Padua. Paediatric patients aged <18 years admitted to the PICU between January 2017 and September 2023 were included if they required invasive mechanical ventilation for at least 5 days and received continuous ketamine infusion for more than 12 hours. Hepatotoxicity was classified according to the Common Terminology Criteria for Adverse Events (CTCAE) version 5.0. Clinical and laboratory follow-up was performed until May 2026. Results A total of 106 patients (40% male) were included, with a median age of 1.7 years (IQR 0.5–5.6). The median duration of ketamine infusion was 8.5 days (IQR 5–13.5), with a median cumulative dose of 321,000 mcg/kg. The overall incidence of hepatotoxicity was 81%. The predominant pattern was canalicular (75%), followed by hepatocellular (59%) and cholestatic (22%). Both infusion duration and cumulative dose were significantly correlated with peak gamma-glutamyl transferase (GGT) and alanine aminotransferase (ALT) levels (p<0.001). The median peak GGT level was 328.5 U/L, representing a 9.8-fold increase from baseline, and was reached after a median infusion duration of 11 days. Receiver operating characteristic (ROC) curve analysis identified predictive thresholds for hepatotoxicity of an infusion duration >4.9 days (AUC 0.814; sensitivity 84.9%; specificity 65%) and a cumulative dose >164,022.9 mcg/kg (AUC 0.801; sensitivity 76.7%; specificity 70%). In multivariable logistic regression analyses, ketamine exposure was an independent predictor of canalicular hepatotoxicity (aOR 1.68 for infusion duration and aOR 3.09 for cumulative dose; p<0.001 for both), and to a lesser extent of hepatocellular hepatotoxicity. At follow-up (median 4.8 months), 66% of patients had persistent abnormalities in liver function tests, and 23.5% had developed chronic liver disease. Radiological hepatic abnormalities increased from 29.2% at baseline to 63.2% following ketamine exposure, with de novo biliary dilatation or strictures observed in 29% of cases; 62.5% of these abnormalities persisted at the last available follow-up. Two patients (1.9%) developed secondary sclerosing cholangitis, and one patient (0.9%) required orthotopic liver transplantation. Conclusions Prolonged ketamine infusion is an independent, dose- and time-dependent determinant of hepatotoxicity, predominantly exhibiting a canalicular pattern, in critically ill paediatric patients. Thresholds of 4.9 days of infusion and a cumulative dose of 164,022.9 mcg/kg were identified as predictors of liver injury with good sensitivity and specificity. A substantial proportion of patients showed persistent liver abnormalities during long-term follow-up. These findings support the need for careful hepatic monitoring during prolonged ketamine infusions and for extended hepatological follow-up after treatment.| File | Dimensione | Formato | |
|---|---|---|---|
|
Francesca_Filippi.pdf
Accesso riservato
Dimensione
1.92 MB
Formato
Adobe PDF
|
1.92 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/109139