External Validation of a Recently Developed Population Pharmacokinetic Model for Hydromorphone During Postoperative Pain Therapy.

BACKGROUND AND OBJECTIVE: We recently developed a new population pharmacokinetic model for hydromorphone in patients including age and bodyweight as covariates. The aim of the present study was to evaluate prospectively the predictive performance of this new model during postoperative pain therapy. METHODS: This was a prospective, single-blinded, randomized, single-center study with two parallel arms. Fifty patients aged 40-85 years undergoing cardiac surgery involving thoracotomy were enrolled. Hydromorphone was administered postoperatively on the intensive care unit as target controlled infusion (TCI) for patient controlled analgesia (TCI-PCA) using the new pharmacokinetic model, or as conventional patient controlled analgesia (PCA). Arterial blood samples were taken for measurement of the hydromorphone plasma concentration. The predictive performance of the pharmacokinetic model was assessed by the median performance error (MDPE), the median absolute performance error (MDAPE), wobble and divergence. For comparison, the performance indices were also determined for three older models from the literature. RESULTS: 903 plasma concentrations of 41 patients were analyzed. The mean values (95 % CI) of MDPE, MDAPE, wobble and divergence for the new pharmacokinetic model were 11.2 % (3.9 to 18.7 %), 28.5 % (23.9 to 33.0 %), 21.4 % (18.0 to 24.9 %) and -1.6 %/h (-2.3 to -0.8 %/h). When compared with older models from the literature, performance was better with less overshoot after bolus doses. CONCLUSION: The new pharmacokinetic model of hydromorphone showed a good precision and a better performance than older models. It is therefore suitable for TCI with hydromorphone during postoperative pain therapy. TRIAL REGISTRATION: EudraCT 2013-002875-16, Clinical Trials NCT02035709.