Morphine and hydromorphone pharmacodynamics in human volunteers: population-based modelling of interindividual response variability and utility.

Abstract

Morphine and hydromorphone differ in onset, magnitude, and duration of effects, all which exhibit considerable interindividual variability, but do not differ in interindividual pharmacokinetic variability. This secondary analysis of a previous study modelled morphine and hydromorphone pharmacodynamics (concentration-effect relationships) to compare their pharmacodynamic variability. In a balanced crossover, 51 healthy volunteers received a 2-h infusion of 0.05 mg kg^-1 morphine or 0.2 mg kg^-1 hydromorphone. Measurements included analgesic response to thermal stimulus (maximally tolerated temperature, verbal pain scores at specific temperatures), pupil diameter, respiratory rate, and end-expired CO₂ concentration. Morphine and hydromorphone pharmacodynamic effects were analysed together in a single model, for each clinical endpoint, with drug as a covariate, to allow assessment of interindividual and drug differences. Pharmacodynamic parameters were centred at their average, whereas a factor determined the difference from average. Models were fitted in NONMEM. Utility functions were constructed using effect-site concentrations. Analysis focused on potency parameters, blood-effect-site equilibration half-lives (t½k_e0) and interindividual variabilities. Limit temperature modelling afforded separate morphine (46.9 ng ml^-1) and hydromorphone (4.4 ng ml^-1) potencies, whereas t½k_e0 was not significantly different between morphine (0.71 h) and hydromorphone (0.75 h), although it was generally slower for morphine for other outcomes. Interindividual variabilities (% coefficient of variation) in potency and t½k_e0 for various clinical effects were large, and varied between different effect measures, but none of the interindividual variabilities were significantly different between opioids. There was considerable interindividual variability in pharmacodynamic effect parameters for both morphine and hydromorphone, but no major variability differences between the opioids. Pharmacodynamic potencies for the various endpoints were different between opioids, but of the same order of magnitude for each opioid. The utility function was more favourable for hydromorphone than for morphine. These results might influence patient- and procedure-specific opioid selection.