To the Editor
We read with interest the recent study by Katzi et al evaluating the influences of a single dose of epidural morphine after vaginal delivery on opioid consumption, pain control, and quality of recovery. Other than the limitations described by them in the discussion, we have several questions about their methods and results and wish to get their reply.
First, as a primary outcome of this study, opioid consumption was converted into intravenous (IV) morphine milligram equivalent (MME) for the between-group comparison. However, the MME conversion factors for oral and IV hydromorphone are questionable. Based on the newest guideline of the Centers for Disease Control and Prevention for prescribing opioids for pain—United States, 2022, the oral MME conversion factor of oral hydromorphone is 5 rather than 4 reported in this study. In the current pain management practices, moreover, the recommended oral MME conversion factor for IV hydromorphone is 18. Given that 1 mg IV morphine is equivalent to 3 mg oral morphine, as described by Katzi et al, 1 mg IV hydromorphone should be converted to 6 mg IV morphine. That is, the MME conversion factor of IV hydromorphone to IV morphine should be 6, rather than 3 reported in this study. The lower MME conversion factors for oral and IV hydromorphone used in this study may not only influence the results of between-group comparison regarding 24-hour opioid consumption, but also they can result in difficulty in comparing the results of this study with the findings of other works.
Second, the choice of a 2.5-MME difference in 24-hour opioid consumption as the effector to estimate the sample size was not based on literature evidence or guidelines. In the available literature, the recommended minimal clinically important difference of total opioid consumption for postoperative analgesia in a randomized controlled trial is 10 mg IV morphine in 24 hours. The MME consumed in the 24 hours was statistically decreased in the interventional group, but the median treatment effect was 0, which does not achieve their assumed effector of 2.5 MMEs. Most importantly, opioid consumption is not a patient-centered outcome and cannot imply any patient benefit unless it is matched with improved pain control or decreased side effects such as respiratory depression, nausea and vomiting, dizziness, and delayed gastrointestinal function recovery. Indeed, median pain score at 24 hours was statistically reduced in the interventional group and the between-group difference of median pain scores at 24 hours achieved the minimal clinically important difference of 1 point recommended in available literature. However, median pain scores at 24 hours in the 2 groups were 3 or less, indicating that most patients only experienced a mild pain, that is, a clinically acceptable level of pain control.5 In addition, this study did not assess patient satisfaction with postdelivery pain control. Considering the facts that epidural morphine after vaginal delivery resulted in more side effects associated with neuraxial opioid use and did not improve any outcome of postdelivery recovery, we cannot determine if decreased opioid consumption and improved pain control by this intervention are only statistically significant, but clinically insignificant. At least, the clinical importance of using this intervention to enhance recovery after vaginal delivery is questionable.
Finally, in the methods, the authors described that secondary end points included total opioid consumption until discharge and total opioid consumption through week 1 of discharge. However, their results did not report and compare these variables. They reported the rates of opioid use within 24 hours of delivery, but not the time to first opioid need and the time from the completion of epidural morphine to discharge from the hospital. In fact, these outcome variables are useful for the complete evaluation of the analgesic efficacy, duration, and potential benefits of epidural morphine after vaginal delivery.