Related Experiment Video
Updated: May 22, 2025
Bile Duct Ligation in Mice: Induction of Inflammatory Liver Injury and Fibrosis by Obstructive Cholestasis
Published on: February 10, 2015
The Albumin-Bilirubin-Fibrosis-4 Score for Predicting Teicoplanin-Induced Abnormal Liver Enzyme Levels in Patients
Hayahide Ooi1, Yuki Asai2, Yoshihiro Kondo3
1Pharmacy, National Hospital Organization Mie Chuo Medical Center, Tsu, Japan.
Background:
Although therapeutic drug monitoring (TDM) maintains serum teicoplanin (TEIC) concentration between 15 and 30 μg/mL, TEIC-induced liver injury may still occur. The albumin-bilirubin (ALBI)-fibrosis-4 (FIB-4) score may be useful for predicting TEIC-induced liver injury in patients undergoing TDM.
Objective:
This pilot study aimed to investigate whether the ALBI-FIB4 score can predict TEIC-induced abnormal liver enzyme levels in patients undergoing TDM.
Methods:
The multicenter retrospective cohort study included 140 patients undergoing TDM of TEIC at steady state. The primary outcome was TEIC-induced abnormal liver enzyme levels. Cut-off values for the alanine aminotransferase (ALT), ALBI score, FIB-4 index, and ALBI-FIB4 score were detected using receiver-operating characteristic curves. The cumulative risk of TEIC-induced abnormal liver enzyme levels was evaluated using Kaplan-Meier curves analyzed log-rank test. Subgroup analysis was performed to examine cumulative risk in patients with serum TEIC concentration of <30 μg/mL.
Results:
The incidence of TEIC-induced abnormal liver enzyme levels was 14.3% (20/140). Cut-off values were 24 IU/L for ALT (sensitivity: 0.800; specificity: 0.692; area under the curve [AUC]: 0.753), -1.33 for the ALBI score (sensitivity: 0.550; specificity: 0.617; AUC: 0.539), 2.73 for the FIB-4 index (sensitivity: 0.700; specificity: 0.475; AUC: 0.550), and -0.85 for the ALBI-FIB4 score (sensitivity: 0.800; specificity: 0.467; AUC: 0.572). The cumulative risk of TEIC-induced abnormal liver enzyme levels was significantly higher for patients with ALT ≥24 IU/L (P < 0.01) and ALBI-FIB4 score ≥-0.85 (P = 0.042). Patients with a serum TEIC concentration of <30 µg/mL exhibited a similar trend, with a higher cumulative risk for patients with ALBI-FIB4 score ≥-0.85 (P = 0.058).
Conclusion And Relevance:
An ALBI-FIB4 score ≥-0.85 may serve as a potential predictor for TEIC-induced abnormal liver enzyme levels in patients undergoing TDM. However, evidence supporting this threshold requires further statistical validation using a larger dataset.
Insights
The albumin-bilirubin (ALBI)-fibrosis-4 (FIB-4) score may predict teicoplanin (TEIC)-induced liver enzyme abnormalities in patients undergoing therapeutic drug monitoring (TDM). A score of -0.85 or higher indicates increased risk, warranting further validation.
Area of Science:
- Pharmacology
- Hepatology
- Clinical Chemistry
Background:
- Therapeutic drug monitoring (TDM) aims to keep serum teicoplanin (TEIC) concentrations between 15-30 μg/mL.
- TEIC-induced liver injury can still occur despite TDM.
- The albumin-bilirubin (ALBI)-fibrosis-4 (FIB-4) score shows potential for predicting TEIC-induced liver injury.
Purpose of the Study:
- To investigate the predictive value of the ALBI-FIB4 score for TEIC-induced abnormal liver enzyme levels.
- To assess the ALBI-FIB4 score in patients undergoing TDM for TEIC.
Main Methods:
- A multicenter retrospective cohort study of 140 patients undergoing TEIC TDM.
- Primary outcome: TEIC-induced abnormal liver enzyme levels.
- Receiver-operating characteristic (ROC) curves and Kaplan-Meier analysis were used to determine cut-off values and cumulative risk.
Main Results:
- The incidence of TEIC-induced abnormal liver enzyme levels was 14.3%.
- An ALBI-FIB4 score ≥-0.85 was associated with a significantly higher cumulative risk of abnormal liver enzymes (P=0.042).
- This trend was observed even in patients with serum TEIC <30 μg/mL (P=0.058).
Conclusions:
- An ALBI-FIB4 score of -0.85 or higher may predict TEIC-induced abnormal liver enzyme levels during TDM.
- Further statistical validation with larger datasets is needed to confirm this predictive threshold.
