The predictive performance of a remifentanil pharmacokinetic data set during Target Controlled infusion

M.J. Mertens, J. Vuyk, A.G.L. Burm, J.G. Bovill. Dept of Anaesthesiology, Leiden University Medical Center, Leiden, The Netherlands.

Background & goal of the study

The pharmacokinetic properties of the new synthetic opioid remifentanil (REM) make it a suitable agent for target controlled infusion (TCI). We studied the predictive performance of a recently determined pharmacokinetic data set for REM ¹ during TCI.

Material and methods

With ethics commitee approval and informed consent, 6 ASA 1‑2 female patients undergoing lower abdominal surgery, received a TCI of REM in combination with a TCI of propofol. Arterial blood was sampled at several REM target concentrations for determination of the plasma REM concentration. The performance error² was calculated as ((C_m‑C_p)/C_p)*100, where C_m and C_p are the measured and predicted REM concentrations, respectively. Subsequently, the bias and inaccuracy of the data set was assessed by determination of the median (MDPE) and the median absolute (MDAPE) performance error, and the coresponding 95% confidence intervals.

Results and discussion

A total of 91 blood samples were analysed. The MDPE and the MDAPE of the REM pharmacokinetic data set were ‑22% and 23%.

Conclusion

Since the 95% confidence interval of the MDPE (-30.4% to -21.5%) included zero, it was concluded that a significant bias had occurred.

Figure: Measured versus predicted remifentanil serum concentrations

References

1. Anesthesiology. 1997 Jan; 86(1): 10‑23