To the Editor:—
We read with interest the article by Lemberg et al. 1titled “Antinociception by Spinal and Systemic Oxycodone: Why Does the Route Make a Difference?” that was published in the October 2006 issue of Anesthesiology. In the following paragraphs, we highlight significant limitations confounding their conclusions.
Although there has been a large increase in the prescribing of the opioid analgesic oxycodone for clinical pain management in the past decade, the differences in the pharmacology of oxycodone relative to the prototypic μ-opioid agonist morphine are less well known. For example, although an analgesic potency ratio for oxycodone to morphine of 1.5 was reported in patients after intravenous administration for postoperative pain management,2the affinity of oxycodone for the μ-opioid receptor is much lower (approximately 40-fold) than that of morphine in radioligand binding studies performed with rat brain homogenate, as well as with membranes prepared from cultured CHO cells expressing the human μ-opioid receptor (hMOR1).3,4In addition, the potency of oxycodone compared with morphine is considerably lower after spinal administration in rats (approximately 14-fold) and humans (approximately 8-fold).5,6We have also shown that after intracerebroventricular administration in rats, the potency of oxycodone is approximately 44% that of morphine7and that intracerebroventricular oxycodone antinociception is blocked by the κ-selective opioid antagonist nor-binaltorphimine (nor-BNI).8
The recent article by Lemberg et al. 1reported that in addition to producing only weak naloxone-reversible antinociception after intrathecal administration to Sprague-Dawley rats, G-protein activation induced by oxycodone in the dorsal horn of the spinal cord was lower compared with that of morphine and oxycodone’s O-demethylated metabolite, oxymorphone. In addition, the authors present data purporting to show that oxycodone’s antinociceptive effects after subcutaneous administration were blocked by naloxone, but not by nor-BNI. They conclude that the reduced potency of oxycodone after intrathecal administration in rats is related to its low efficacy and potency for G protein–mediated μ-opioid receptor activation, and that an active metabolite such as oxymorphone that is not formed in the central nervous system may account for the differences in oxycodone’s antinociceptive effects after subcutaneous and intrathecal administration.
However, there are a number of problems with the authors’ interpretation of their data, and their results1do not justify the conclusions drawn. In previous human studies, circulating oxymorphone concentrations remained very low (< 1.2 ng/ml) after systemic and oral modes of administration.9–12Recent studies of oxycodone’s pharmacokinetics and metabolism after oral administration in humans found that oxycodone was metabolized predominantly to noroxycodone, with area under the curve ratios for oxycodone:noroxycodone and oxymorphone:oxycodone of 1.19 and 0.04, respectively.4Similarly, we have shown that circulating oxymorphone levels also remain very low (< 2.1 ng/ml) in rats after subcutaneous administration of oxycodone.13In addition, studies in human CYP2D6-extensive metabolizers showed that oxycodone’s pharmacodynamic effects were not significantly altered after blockade of CYP2D6-mediated O-demethylation of oxycodone to oxymorphone by quidinine,12further discounting the possible contribution of metabolically derived oxymorphone to the analgesic effects of systemically administered oxycodone.
Furthermore, there are serious methodologic problems with the study involving nor-BNI in the article by Lemberg et al. 1that confound their data interpretation. In previously published dose–response studies, suppression of the antinociceptive effects of the κ-selective opioid agonist, U69,593, by nor-BNI was shown to be maximal after intracerebroventricular administration in rodents, when nor-BNI was administered 24 h before U69,593.14Likewise, in other studies, subcutaneous nor-BNI at 5 and 20 mg/kg blocked the antinociceptive effects of the κ-selective opioid agonist U-50,488H in rodents only when nor-BNI was administered at least 2 h before U-50,488H.15Furthermore, in the same study, Endoh et al. 15showed that at 30 min after subcutaneous administration, nor-BNI behaves as a μ-antagonist rather than a κ-antagonist. Because nor-BNI was administered 30 min before oxycodone in the study by Lemberg et al. 1and a positive control such as U69,593 or U-50,488H was not included in their study design, the authors’ conclusion that nor-BNI did not block the antinociceptive effects of oxycodone is not justified. Moreover, the assertion by Lemberg et al. that oxycodone’s antinociceptive effects are mediated by μ- rather than κ-opioid receptors is not correct.
Although the reason for oxycodone’s low potency after spinal routes of administration remains unclear, it is highly unlikely that oxymorphone or some other active metabolite can account for this, as the authors speculate. Rather, it is our view that the findings by Lemberg et al. 1further demonstrate that the antinociceptive effects of oxycodone and morphine are mediated by distinctly different mechanisms and that oxycodone is not a μ-opioid agonist.
*School of Pharmacy, University of Queensland, Brisbane, Australia. m.smith@pharmacy.uq.edu.au
