Variations in Practice to Therapeutic Monitoring of Tacrolimus following Primary Adult Liver Transplantation

Abstract

Background: There is limited clinical evidence evaluating the correlation between immunosuppressant monitoring practice and transplant outcomes.

Objective: To assess current practice of tacrolimus trough monitoring in early post-operative period following liver transplantation (LT), and its impact on outcomes.

Methods: The duration to trough levels (DTT) were calculated in patients undergoing primary LT. The impact of variability in DTT on graft rejection episodes, serum tacrolimus level and renal function was assessed. These results were converted into a drug level estimation tool, which was validated in a prospective cohort of patients.

Results: 2946 events in 274 patients were evaluated. The median DTT was 7:19 hrs (range: 27 min to 19:38 hrs). In 72% (2140 events) of the occasions, DTT was <8 hrs. There was a significant (p=0.022) correlation between DTT and tacrolimus level. Despite clinical decisions were taken to modify the dose of tacrolimus based on trough level, neither did DTT affect the average creatinine levels (p=0.923), nor the variability in DTT did affect acute rejection (p=0.914, and 0.712, respectively). A dose estimation tool was developed and applied to validation cohort (n=612), and returned a moderate R(2) value of 0.50.

Conclusion: There is a significant variation in the "real world" monitoring of tacrolimus with DTT in majority of measurements falling below recommendations; reassuringly, this did not lead to adverse transplant sequelae.

Keywords: Drug monitoring; Immunosuppressive agents [Pharmacological action]; Outcome assessment (Health care); Patient outcome assessment; Tacrolimus; Transplantation.

Figure 1

Percentage of tacrolimus monitoring events vs. duration to trough (DTT

Figure 2

Variation in the mean tacrolimus levels (A) and mean creatinine levels (B) vs. DTT, from the model presented in Table 2

Figure 3

Predicted tacrolimus levels vs. actual tacrolimus levels in the modelling cohort (A), and validation cohort (B), based on the dose estimation tool. The solid line indicates the target for equivalence; the broken likes are the 95% prediction limits