Abstract
Optimal pain relief in day-case surgery is imperative to patient comfort and timely discharge from hospital. Short-acting opioids are commonly used for analgesia in modern anaesthesia, allowing rapid recovery after surgery. Plasma concentration fluctuations from repeated dosing of short-acting opioids can cause patients to oscillate between analgesia with potential adverse effects, and inadequate analgesia requiring rescue dosing. Methadone’s unique pharmacology may offer effective and sustained analgesia with less opioid consumption, potentially reducing adverse effects. Using a double-blind, randomised controlled trial, we compared post-anaesthesia care unit opioid consumption between day-case gynaecological laparoscopy patients who received either intravenous methadone (10 mg), or short-acting opioids intraoperatively. The primary outcome was post-anaesthesia care unit opioid consumption in oral morphine equivalents. Secondary outcomes included total opioid consumption, discharge opioid consumption, pain scores (0–10) until discharge, adverse effects (respiratory depression, postoperative nausea and vomiting, excess sedation), and rate of admission. Seventy patients were randomly assigned. Patients who received methadone consumed on average 9.44 mg fewer oral morphine equivalents in the post-anaesthesia care unit than the short-acting group (18.02 mg vs 27.46 mg, respectively, 95% confidence interval 0.003 to 18.88, P = 0.050) and experienced lower postoperative pain scores at every time point, although absolute differences were small. There was no evidence of lower hospital or discharge opioid consumption. No significant differences between the methadone and short-acting groups in other outcomes were identified: respiratory depression 41.2% versus 31.4%, Padjusted >0.99; postoperative nausea and vomiting 29.4% versus 42.9%, Padjusted >0.99; overnight admission 17.7% versus 11.4%, Padjusted >0.99; excess sedation 8.82% versus 8.57%, Padjusted >0.99. This study provides evidence that, although modestly, methadone can reduce post-anaesthesia care unit opioid consumption and postoperative pain scores after day-case gynaecological laparoscopy. There were no significant differences in any secondary outcomes.
Keywords
Introduction
Postoperative pain remains a primary therapeutic challenge for clinicians, with approximately 80% of patients reporting inadequate postoperative analgesia.1,2 This may cause an array of problems, including delayed recovery and increased hospital stay, prolonged opioid use, and increased risk of cardiac and pulmonary complications. 2 Notably, poorly controlled pain is an important risk factor for developing chronic post-surgical pain, 3 present in 10–50% of patients.3,4 Despite innovations in perioperative care, postoperative pain management has failed to improve over the past two decades,5,6 a period in which short and ultra-short-acting opioids have become the mainstay of perioperative analgesia. 7 Such a trend may have consequences relating to how employing these opioids, both intraoperatively and in postoperative analgesia, causes wide fluctuations in plasma opioid concentrations, resulting in patients oscillating between periods of sedation with potential respiratory depression, and inadequate analgesia when rescue dosing is required.
For women undergoing gynaecological laparoscopy, a minimally invasive, moderately painful procedure, managing postoperative pain can be notoriously difficult, and there is a vast array of possible permutations to a routine anaesthetic in the absence of an ideal regimen. 8 Often conducted as a day-case procedure, analgesia is of great importance. However, laparoscopic procedures often have both under-appreciated and undertreated postoperative pain. 9
Methadone is a unique opioid employed in chronic pain and palliative care, and as a way of managing withdrawal symptoms in people addicted to opioids. 10 In addition to mu-receptor activity, methadone is an N-methyl-D-aspartate receptor antagonist and inhibitor of serotonin and noradrenaline reuptake, 11 offering several attractive properties for use in moderately painful surgical procedures. Furthermore, it has a rapid onset of effect and the longest elimination half-life of any clinically used opioid, providing sustained analgesia and reducing the need for rescue dosing with short-acting opioids. 12 Numerous studies have shown that either an intraoperative dose of methadone or its use in the post-anaesthesia care unit (PACU) can reduce both pain and opioid consumption postoperatively in complex spine,10,13 cardiac,14 –16 abdominal17,18 and paediatric 19 surgery.
The aim of this study was to compare PACU opioid consumption in patients who underwent laparoscopic gynaecological procedures who received either intraoperative methadone or short-acting opioids. This study hypothesised that intraoperative methadone use would reduce PACU opioid consumption. We also hypothesised that opioids consumed after discharge, at one week, would be reduced.
Materials and methods
Trial design
This was a single-centre, double-blind, randomised controlled trial with two parallel groups: those who received a single intraoperative dose of methadone and those who received conventional, short-acting opioids (fentanyl, morphine, or oxycodone), allocated at a ratio of 1:1.
Patient recruitment
This investigation ran from January 2021 to May 2022. Patients were identified from Wodonga Hospital operating theatre lists and were contacted for information to be provided regarding participation prior to their admission. Wodonga Hospital is one of two Albury Wodonga Health campuses that provide services to more than 250,000 people across north-east Victoria and southern New South Wales, Australia. Wodonga Hospital is the regional referral centre for public obstetrics and gynaecology.
At admission, patients were screened for eligibility and exclusion criteria. Prior to randomisation, informed oral and written consent was obtained by a single assessor who also collected all data throughout each participant’s involvement in the trial.
Inclusion criteria
Adults undergoing gynaecological laparoscopy at Wodonga Hospital; Planned same-day discharge; Able to speak and read English.
Exclusion criteria
Body mass index (BMI) >35 kg/m2; Obstructive sleep apnoea; Significant renal or liver dysfunction; Pregnant or breastfeeding women; Preoperative opioid dependence or chronic opioid use; Known or suspected allergy to methadone; Anaesthetist preference to avoid methadone administration for any reason; Emergency laparoscopic operations (e.g. ectopic pregnancy); Long QT syndrome, or long QT interval (defined as >450 ms) measured on a preoperative electrocardiogram (ECG), performed as a safety precaution in the context of the trial; Subjects already taking a selective serotonin reuptake inhibitor (SSRI) or a monoamine oxidase inhibitor.
Approval and registration
This investigation was approved by the Albury Wodonga Health Human Research Ethics Committee (ID: 66418). The study was registered with the Australian New Zealand Clinical Trials Registry (ACTRN12620001131943) and approval for the off-label use of methadone was granted by the Therapeutic Goods Administration.
Randomisation and blinding
A randomisation schedule (at a ratio of 1:1) was prepared by the study statistician using randomised permuted blocks with block sizes of 2, 4 and 6, and transferred into sequentially numbered, opaque, sealed envelopes by a research team member who was not involved in patient recruitment or data collection. These envelopes were opened by the anaesthetist for each consented patient to reveal group allocation. The patient and single assessor remained blinded.
Procedure
Premedication, anaesthesia induction and maintenance were at the attending anaesthetist’s discretion, although in the methadone group, the only opioid allowed was methadone. The methadone dose to be administered was 10 mg intravenously at induction, 18 although lower doses were allowed at the anaesthetist’s discretion (e.g. in smaller patients) at 0.15 mg/kg, providing this was less than 10 mg. In the short-acting opioid group, the opioid used was fentanyl, morphine or oxycodone, chosen at the anaesthetist’s discretion. Use of any other analgesic adjuncts was permitted, as was local anaesthetic infiltration.
Data collection
After randomisation, demographic data, including age, height, weight, BMI and smoking status were recorded.
At the completion of surgery, patients were transferred to the PACU in the usual way, and pain was managed as per protocol. Pain protocols for fentanyl, morphine and oxycodone were available for use. The assessor, a member of the research team who remained blinded to group allocation, established pain scores on arrival at the PACU (verbal numerical rating scale (VNRS), 0–10), then every 15 min for the first hour, then hourly for 4 h, and then prior to discharge. If the patient was admitted to hospital for any reason, this was recorded. Opioid dose (in morphine equivalents) consumed postoperatively until discharge was recorded. Opioid consumption at one week after discharge was recorded by a scheduled telephone call.
Following the completion of surgery, surgical notes and the operative assistant were consulted to obtain surgical data. After the completion of patient recruitment, the anaesthesia and medication charts were reviewed to reveal intraoperative medications, any adverse events, and further medications administered.
Episodes of respiratory depression, defined as a respiratory rate of less than 10 breaths per minute for 10 min, oxygen desaturation episodes (<90%) or where intervention was required (e.g. naloxone administration) were recorded. Discharge analgesia was prescribed according to the anaesthetist’s usual practice.
Outcomes
Primary outcome
Opioid dose, in oral morphine equivalents 20 (OMEs) administered in the PACU.
Secondary outcomes
OMEs administered prior to discharge from hospital; OMEs consumed after discharge, at one week; Pain scores at time points detailed above; Postoperative nausea and vomiting (PONV) at any time point (moderate or severe); Pasero opioid-induced sedation scale (POSS) sedation scores of 3 or 4 at any time point; Rate of admission to hospital; Respiratory depression in the PACU.
Sample size estimation
We conducted a small retrospective audit of patients who had undergone gynaecological laparoscopy at Wodonga Hospital to inform our sample size estimation. PACU opioid consumption (in intravenous morphine equivalents) was used. With a mean of 5.6 mg, a standard deviation of 5.8, and an anticipated decrease to 1 mg, at least 31 subjects were required in each group (with 80% power and an alpha of 5%). To ensure we reached this target, 70 patients were recruited and randomly assigned. This sample size was consistent with similar previous investigations.10,16,18
Statistical analysis
Statistical analysis was performed using SPSS (IBM SPSS Statistics, version 26; IBM Corp., Armonk, NY, USA). All opioid consumption was converted to OMEs 20 for comparison. If a single, pre-intubation bolus of either remifentanil or alfentanil was administered, this was not included. Methadone 10 mg and morphine 10 mg were considered equivalent. 21
The primary outcome and those secondary outcomes comparing mean opioid consumption and mean highest PACU pain score were analysed with an unpaired two-sided Student’s t-test or Mann−Whitney U-test, as appropriate.
Secondary outcomes comparing categorical variables were analysed with the χ2 test or Fisher’s exact test. Differences in group pain score trajectories over time were compared using a linear mixed model. Assumptions of parametric models were checked by visually inspecting residual plots.
Statistical significance was set at 0.05 and secondary outcome type 1 error from multiple testing was controlled for using the Holm method. Adjusted P values are reported throughout the text.
Midway through our investigation (in September 2021), the primary investigator was unblinded to group allocation after 42 participants had completed their involvement in the study. This analysis served only to permit university thesis requirements and in no way altered recruitment progression, outcomes, or statistical analysis of the completed sample, regardless of any information revealed. The statistics used in the formation of this report did not differ in any way from the study protocol, results were not shared with those involved in patient management, and the analysis was not triggered by any concern for patient safety. Blinding continued as described for all participants recruited into the trial after this.
Results
The recruitment flowchart is presented in Figure 1. One hundred and ninety-four patients undergoing gynaecological laparoscopy were assessed for eligibility. Preoperative use of a serotonergic agent and having a BMI greater than 35 were the most common reasons for exclusion. Seventy patients were randomly allocated either to receive short-acting opioids or methadone intraoperatively and 63 participants completed follow-up to quantify discharge opioid consumption.
Recruitment proceedings. Seventy participants were randomly allocated into two groups of 35 participants. Six subjects in the short-acting group did not participate in the follow-up telephone call to quantify discharge opioid consumption. BMI: body mass index; SSRI: selective serotonin reuptake inhibitor; SNRI: serotonin noradrenaline reuptake inhibitor.
The study data from one participant allocated to the methadone group was mishandled, leaving 35 and 34 patients in the short-acting and methadone groups, respectively, for analysis. The missing participant’s datasheet is believed to have been misplaced after recording inpatient data, before being transferred to an electronic record.
Baseline characteristics and operative data (Tables 1 and 2) did not meaningfully differ between groups.
Baseline characteristics.
Values are mean (standard deviation) or numbers (percentage).
BMI: body mass index.
Operation details.
Values are numbers (percentage).
ASA: American Society of Anesthesiologists physical status score.
As presented in Figure 2, those subjects who were randomly assigned to receive methadone consumed on average 18.02 mg OME in the PACU, 9.44 mg less (95% confidence interval (CI) 0.003 to 18.88, P = 0.050) than those randomly assigned to receive a short-acting opioid (average 27.46 mg); however, both hospital (8.71 mg vs 9.63 mg, Padjusted >0.99) and discharge opioid consumption after one week (56.96 mg vs 54.49 mg, Padjusted >0.99) did not differ (Table 9).
Post-anaesthesia care unit opioid consumption with 95% confidence intervals. Consumption is represented in oral morphine equivalents (in mg).
Statistical analysis.
Values are mean (standard deviation), numbers (percentage), mean difference with 95% CI, or relative risk with 95% CI.
MD: mean difference; CI: 95% confidence interval; PACU: post-anaesthesia care unit; PONV: postoperative nausea or vomiting; POSS: Pasero opioid-induced sedation scale; RR: relative risk.
Hospital opioid consumption was counted from discharge from the PACU to discharge from hospital.
Given gynaecological laparoscopy encompasses an array of surgical interventions, details of the procedures performed were recorded. Surgical entry technique and surgical procedures performed are reported in Table 2. More subjects in the short-acting group had either peritoneal biopsies taken or underwent surgical resection of an ovary or Fallopian tube.
Anaesthesia details for both groups are presented in Tables 3 –5. The anaesthetic comparison is relatively homogenous and no important differences in anaesthetic techniques, medications utilised, or other adjuncts are observable. Overall, the postoperative care, including prescription of discharge opioids (Table 6), is relatively similar between groups. Of note is an appreciable difference in mean intraoperative opioid consumption, in OMEs, between the methadone (32.54 mg) and short-acting (46.94 mg) group. This is likely to be attributable to the ability to utilise multiple agents and doses in the short-acting group.
Discharge opioid prescription.
Values are number (percentage) or mean (standard deviation).
GP: general practitioner; OMEs: oral morphine equivalents (in mg).
Anaesthetic comparison.
Values are numbers (percentage).
TIVA: total intravenous anaesthesia.
Intraoperative opioid information.
Values are numbers (percentage) or mean OMEs in mg.
OMEs: oral morphine equivalents.
Anaesthetic adjuncts.
Values are numbers (percentage).
No patients received naloxone for respiratory depression. However, one patient in the methadone cohort was administered naloxone in the PACU for pruritis. A single patient in the methadone group was subjected to a medical emergency team (MET) call for hypotension with bradycardia. Similar rates of intraoperative hypotension were seen in both groups, with 10 patients from each receiving vasopressors or inotropes.
There was one subject in the methadone group who received fentanyl (100 µg) in addition to methadone, due to the attending anaesthetist misunderstanding the study protocol. This participant was included in the methadone group’s analysis.
There were no significant differences in side effects or clinically significant complications between groups. There were six and seven participants in the short-acting and methadone groups, respectively, who presented to hospital or their general practitioner (GP) after discharge from hospital regarding pain relief. Five and seven of these participants, respectively, received an additional prescription of opioids (Table 7). If consumed, these subsequent opioids were included in the totals for post-discharge opioid consumption and total remaining opioids, described in Table 6, but were not included in the reporting of hospital discharge prescriptions to highlight any difference in anaesthetist prescribing practices. The GP in this case would not have been aware of group allocation, nor study involvement, without the patient offering the latter.
Progression.
Values are mean (standard deviation) or numbers (percentage).
GP: general practitioner.
The rate of respiratory depression in the PACU did not differ meaningfully between the short-acting (31.4%) and methadone (41.2%) groups (relative risk (RR) 1.31 (0.69–2.44) χ2 (1) 0.709, Padjusted >0.99). There was no evidence that suggested those patients who received short-acting opioids compared with methadone experienced different rates of PONV (15 (42.9%) vs 10 (29.4%), Padjusted >0.99), with the use of intraoperative prophylactic antiemetics being similar in each group (Table 5). The same number of patients experienced excess sedation postoperatively as measured by the POSS (RR 1.02 (0.44–2.34), Padjusted >0.99).
On average, the procedure duration was slightly longer in the short-acting group (Table 7). The prescription of pain protocols for the PACU, reported in Table 8, was similar for both groups.
Pain protocol prescribed.
Values are numbers (percentage).
Six subjects in the methadone group were admitted (uncontrolled pain and sedation (one), unable to mobilise safely (two), bradycardia and hypotension (MET call) (one), urinary retention (one), uncontrolled pain and vomiting (one)) compared with four in the short-acting group (unable to mobilise safely (one), uncontrolled pain and urinary retention (one), nausea, urinary retention and low oxygen saturations (one) and nausea (one)). No patients remained in hospital for more than one night.
All outcomes are reported in Table 9.
Pain scores
Pain score data, presented in Figure 3, were analysed over time using a linear mixed model. We found strong evidence (F(1,583) 21.30, Padjusted = 0.00003) that methadone decreased pain by 0.76 (95% CI 0.44 to 1.09) on the VNRS scale relative to the short-acting group. We found no evidence (F(8,583) 1.26, P = 0.262) that the trend over time differed for the two groups. Pain scores were lower in the methadone group at all time points.
Estimated mean and 95% confidence interval of postoperative pain scores. The methadone cohort had lower postoperative pain scores; however, the trajectory of pain did not differ between short-acting opioids and methadone.
Discussion
Reducing postoperative opioid consumption yields benefits for both the hospital system and individual patients. The current reappraisal of methadone as an adjunct to anaesthesia in an array of settings is worthy of attention. However, our investigation failed to reproduce the impressive reductions in postoperative opioid requirements that other studies have shown when methadone has been employed over short-acting opioids.
Methadone is not a new drug, but is one whose appealing properties may be overlooked by the current fashion towards short-acting and opioid-free regimens. Investigations have repeatedly demonstrated comparable or superior efficacy. In our trial, methadone provided adequate and sustained analgesia for women undergoing gynaecological laparoscopy compared with the commonly used short-acting opioids fentanyl, oxycodone and morphine, albeit with a small reduction in the average PACU OME dose. Importantly, this did not come with any appreciable difference in adverse effects, with PONV, hospital admission and respiratory depression being comparable among groups, although our study was insufficiently powered to come to definitive conclusions about these outcomes.
Despite receiving lower intraoperative OME exposure, alongside a modest decrease in postoperative consumption in the PACU, those who received methadone experienced significantly less pain throughout their hospital stay, although not at a clinically significant level. This likely results from methadone’s rapid central nervous system equilibration with plasma concentrations compared with other opioids (effect-site equilibration half-time 24 s, compared with fentanyl 5 min, and morphine 2–4 h). 21 The relatively small improvement in analgesia revealed in our study may not be clinically significant, contrasting with prior studies which have demonstrated a greater measure of effect.13,15,17,21,22 This discrepancy could result from innate differences in population groups or surgical focus between studies, with our randomised controlled trial investigating only gynaecology patients, or it could be due to surgical differences between our two groups. Regardless of this, it does affirm methadone as an effective analgesic in this population when compared with conventionally used opioids.
In our investigation, discharge prescribing was according to the attending anaesthetist’s usual practice and was mostly homogenous. Those patients who received methadone consumed slightly fewer opioids in the first 7 days after their surgery; however, this is not statistically or clinically significant and this investigation was not powered to identify a difference in this outcome.
The incidence of postoperative respiratory depression was notably high in our population compared with other investigations. For example, the incidence of respiratory depression was 31.2% among joint arthroscopy patients who underwent a general anaesthetic, 23 12.8% in patients who underwent a hysterectomy, 24 and 15.3% in patients undergoing laparoscopic surgery. 25 Our rates may arise from how we defined an episode of respiratory depression, be inflated by the small sample size effect or be due to closer monitoring than in other investigations, without someone at the bedside throughout a patient’s time in the PACU. Further, we note that standardising the administration of supplemental oxygen to participants in the PACU could have eliminated the confounding effects of supplemental oxygen on all three parameters we used to define respiratory depression.
Our investigation has several strengths. By focusing on a narrow population who often have undermanaged postoperative pain, our findings relate strongly to their care. Further, while a multicentre design may yield external validity and aid in applying findings to the wider population, the single-centre design reduces the disparity in intra and postoperative care. Likewise, having a single data collector throughout the trial eliminates validity concerns relating to the interpretation and use of scales for data collection.
Limitations
While measuring PACU opioid consumption through a patient-controlled analgesia device may provide more accurate representations of analgesic requirements, while also limiting the involvement of unblinded nursing staff administering analgesia, this does not reflect clinical practice for day-case surgery. We also acknowledge that not standardising anaesthetic management or PACU oxygen therapy allows other variables to contribute, although randomisation of a reasonable sample should account for this. Furthermore, varied discharge opioid prescribing from different anaesthetists compounds this outcome and should be explored more vigorously in future studies. We also acknowledge that the consumption of discharge opioids is impacted by admission.
When designing this protocol, approval was not granted for the preparation of methadone and short-acting opioids by the pharmacy into unlabelled syringes due to associated costs and lack of funding. As such, although preferable, blinding of the attending anaesthetist was not possible in our investigation. This may have introduced bias surrounding perioperative care, pharmacological management and discharge opioid prescriptions.
As described, there was a small number of protocol breaches; however, the nature of these breaches and their relative presence in both groups is unlikely to impact our findings. Our protocol stated that those who were randomly assigned to receive intraoperative methadone were not to receive any other intraoperative opioids; however, one patient in the methadone group received fentanyl and three patients in each group received intraoperative tramadol in addition to their allocated opioids, despite those in the short-acting group only being technically able to receive fentanyl, morphine or oxycodone. In addition, as with any opioid conversion, there is always potential for error when using dose equivalents. For parenteral methadone in particular, there is somewhat limited evidence on the conversion ratio to OMEs. We considered methadone to be equipotent to intravenous morphine 10 mg when a single dose was given, as has been used in previous research. 21
Developing perioperative strategies to decrease the reliance on opioid medications and enhance patient outcomes is a clear focus of postoperative pain research and enhanced recovery after surgery protocols. Methadone’s unique mechanistic profile, rapid onset and long duration of action make it a suitable analgesic and an appealing alternative to current strategies, with patients benefiting from lower opioid consumption and enhanced analgesia. If postoperative opioid requirements can be reduced, not only will patients benefit in terms of greater comfort and fewer dose-related opioid adverse effects, but concerns around overprescribing discharge opioids may be alleviated. This randomised controlled trial confirms methadone as an effective alternative to conventionally used opioids for women undergoing gynaecological laparoscopy. Subsequent analysis and further investigations regarding those concerns holding methadone back from clinical uptake need to be addressed before patients may benefit from its incorporation into routine care.
Moving forward, it is perhaps time to move away from exploring the opioid-sparing effects and superior analgesia achieved by using methadone, and instead address those concerns that are holding it back from routine clinical use. Debate around methadone’s potential for use as part of enhanced recovery after surgery protocols continues to present itself in the literature, most recently when concern about a paucity of evidence for safety or its utilisation in ambulatory surgery was raised once again, while others call for its consideration. 26 The literature reports conflicting and limited evidence on methadone’s propensity to cause postoperative respiratory depression compared with conventionally used opioids. Further concern stems from methadone’s ability to affect cardiac electrical activity in much higher or sustained dosing; however, little evidence suggests this cardiotoxicity occurs after a single intravenous bolus used for anaesthesia. Prolongation of the QT interval puts patients at risk of torsades de pointes, a potentially fatal arrhythmia, although no such arrhythmia was observed in our sample. It would be reasonable to perform a preoperative ECG on all patients in whom the use of intraoperative methadone is being considered. Establishing a cost–benefit analysis and day-case admission rates would be beneficial, as would determining whether there are any long-term benefits to chronic pain and opioid use as a result of methadone utilisation.
A final question posed pertains to methadone’s suitability in day-case anaesthesia, in which patients are discharged under the prolonged therapeutic effects of opioids in addition to their discharge prescriptions. While inpatient procedures would benefit from those findings discussed, the same value would not be afforded to day-case patients if consequences result from sustained plasma opioid concentrations after discharge. Additional investigation around discharge and post-discharge patient satisfaction may also address such concerns.
Conclusion
Although our study provides evidence of a small reduction in PACU opioid consumption in day-case gynaecological laparoscopy patients receiving methadone over short-acting opioids, and a similar slight reduction in pain scores, these reductions are minimally clinically significant and are smaller than previous studies have suggested. With previous robust investigations directly comparing methadone with individual opioids in a standardised anaesthetic, the pragmatic design of this study might lead to the previously touted benefits of methadone being viewed more modestly.
Data sharing
Due to consent constraints, individual participant data will not be available.
Footnotes
Author Contribution(s)
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no specific grant from any funding agency in the public, commercial or not-for-profit sectors. All drugs were supplied by Albury Wodonga Health.
