DOI: 10.1177/10600280261497379 ISSN: 1060-0280

Factors Affecting Tacrolimus Levels in Pediatric Liver Transplant Recipients During the Early Posttransplant Period

Eda Eyduran, Ayşen Durak Aslan, Merve Havan, Arzu Meltem Demir, Ceyda Tuna Kırsaçlıoğlu, Zarife Kuloğlu, Aydan Kansu, Elvan Onur Kırımker, Meltem Koloğlu, Erdinç Devrim, Tanıl Kendirli
Show PDF Cite

Background:

Tacrolimus is the cornerstone of immunosuppression in pediatric liver transplantation (LT) but exhibits high pharmacokinetic variability.

Objective:

This study aimed to identify clinical and laboratory determinants of tacrolimus trough levels (C0) during the initial 14 days posttransplant.

Methods:

This retrospective study analyzed pediatric LT recipients (January 2020-October 2025). We evaluated continuous variables for normality using the Shapiro–Wilk test and analyzed them with nonparametric bivariate tests, including Spearman’s rank correlation and Mann-Whitney U tests. To account for repeated measurements within individual patients, we fitted a linear mixed-effects model with a random intercept. Backward elimination with a conservative threshold of p > 0.50 was applied to remove nonsignificant variables, while retaining predefined clinical covariates. Model fit was evaluated using Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC), and the final parsimonious model was selected by minimizing BIC while excluding individual antimicrobial treatments to prevent overfitting. To quantify complex polypharmacy and avoid overfitting, we created a Pharmacological Interaction Score (PIS) by assigning weighted positive or negative values to co-administered medications based on their known CYP3A4/5 inhibitory or inducing effects on tacrolimus.

Results:

In 33 patients, hemoglobin (β = 0.78, P  < .001) and serum creatinine (β = 3.41, P  < .001) were positively associated with tacrolimus levels, while INR showed a negative association (β = −1.00, P  = .036). The model included albumin, a pharmacologic interaction score designed to simplify complex polypharmacy into a single numerical value, as well as platelet counts, which were retained but not statistically significant. Tacrolimus levels were not significantly influenced by sepsis, nutrition (enteral/parenteral), or extracorporeal therapies.

Conclusion and Relevance:

In this cohort, clinical complications such as sepsis, nutritional support methods, and extracorporeal therapies did not significantly affect drug trough levels. These results suggest that early postoperative fluctuations in whole-blood tacrolimus levels are mainly driven by patient hematological status, renal function, and graft recovery. Clinically, including hematological and biochemical parameters in routine drug monitoring could improve dose management and help prevent unnecessary empirical changes during critical care.

More from our Archive