Methadone Dosing and Postoperative Pain Management in Cardiothoracic Surgery

Cardiac surgery is associated with significant postoperative pain. Methadone reduces postoperative pain and opioid consumption after cardiothoracic procedures, but the dosing of intraoperative methadone is unclear.  Our previous investigation of this retrospective cohort found that the use of intraoperative methadone was associated with lower pain scores and lower postoperative oral morphine equivalents after cardiac surgery compared to shorter acting opioids, but we did not examine dosing of methadone.  This secondary analysis of that previous study provides dose–response relationships for methadone administered intraoperatively to cardiac surgery patients for both analgesia and adverse effects.

Details of the study sample have been previously reported, but briefly, this retrospective cohort study (Mayo Clinic Institutional Review Board approval No. 22-007486) included 12,017 adults who underwent cardiac surgery with cardiopulmonary bypass from 2018 to 2023 at a large, multisite healthcare system.  The primary outcome of interest for this analysis was patient- and intensive care unit (ICU) nurse–reported daily mean pain score (0 to 10) until postoperative day 7. The secondary outcomes were daily total postoperative oral morphine equivalents until POD 7 and time until first postoperative opioid administration. Exploratory outcomes included opioid-related complications during the hospital admission: pruritis, urinary retention, time to extubation, ICU length of stay, hospital length of stay, postoperative nausea and vomiting, naloxone administration, reintubation, and delirium (Supplemental Digital Content 1 for data sources and definitions, https://links.lww.com/ALN/D685). The patients were categorized into four groups (data-driven) based on intraoperative administration of intravenous methadone: no methadone, methadone dosage of less than 0.25 mg/kg, methadone dosage of 0.25 to 0.30 mg/kg, and methadone dosage of greater than 0.30 mg/kg, using the total body weight and the methadone dose administered.

All aspects of perioperative management were at the discretion of the attending physicians (e.g., anesthesiologist, intensivist, surgeon), but management was based largely on institutional protocols, which strongly emphasize multimodal analgesia and enhanced patient recovery.  Covariates were included in the regression models based on their relationship with postoperative pain: age, body mass index, sex, year of surgery, Charlson Comorbidity Index, smoking status, history of chronic pain, diagnosis of fibromyalgia, use of home opioid within 90 days of surgery, administration of antiemetics and nonopioid analgesics preoperatively and intraoperatively, preoperative or intraoperative regional block, use of local anesthetic by the surgical team, length of surgery, and surgery type (table 1).  Differences in daily pain scores and oral morphine equivalents by methadone group were assessed with covariate adjusted mixed quantile regression models with random intercepts for patients. Differences within each day were assessed via group by day interaction terms. Differences in exploratory outcomes by methadone group were assessed with quantile regression, Cox proportional hazards, and logistic regression models, adjusting for covariates using Rv4.2.1. Sensitivity analyses using ideal body weight (Devine formula) were conducted. Alpha was set at 0.0167 (0.05/3) for the primary and secondary outcomes and 0.05 for exploratory outcomes. This article adheres to the guidelines set forth by the Strengthening the Reporting of Observational Studies in Epidemiology protocol.

The mean age was 65.0 yr (interquartile range [IQR], 55.0 to 72.0), 67.4% of patients were male, the median body mass index was 28.4 kg/m² (IQR, 25.0 to 32.7), and most patients underwent isolated valve surgery (61.3%). The median operating room time was 6.4 h (IQR, 5.3 to 7.8), and regional anesthesia was utilized in 20.0% of patients. The median methadone doses were 20.0, 25.0, and 30.0 mg for the groups receiving less than 0.25 mg/kg, 0.25 to 0.30 mg/kg, and greater than 0.30 mg/kg, respectively. The median intraoperative oral morphine equivalents by group was 150, 120, 144, and 150 for the groups given no methadone, less than 0.25 mg/kg, 0.25 to 0.30 mg/kg, and greater than 0.30 mg/kg, respectively. The median time to first intraoperative methadone administration was 62 min (IQR, 34 to 78), the median number of methadone doses was 3 (IQR, 2 to 4), and the time to first intraoperative opioid administration was 11 min (IQR, 7 to 17). The methadone groups had statistically significant lower postoperative pain scores compared to the no-methadone group until postoperative day 7, favoring lower pain scores with increasing methadone dose, for days 0 and 1 (table 1). The differences between the groups narrowed over time; by postoperative day 7, the median pain score for the greater than 0.3 mg/kg group was 0.1 lower (95% CI, −0.2 to 0.0), 0.2 lower in the 0.25 to 0.3 mg/kg group (95% CI, −0.3 to −0.1), and 0.2 lower in the less than 0.25 mg/kg group compared to the no-methadone group (95% CI, −0.3 to −0.1). There was a dose–response relationship with time to first postoperative opioid administration. Postoperative oral morphine equivalents followed a dose–response relationship for all days up until postoperative day 4, with statistically significant lower postoperative oral morphine equivalents requirement for the methadone groups compared to the no-methadone group except for the less than 0.25 mg/kg group on postoperative day 3 (fig. 1). There were no differences in pruritis, urinary retention, or reintubation by methadone group; i.e., patients receiving methadone were no more likely to experience these adverse events. However, time to extubation and hospital length of stay were longer in the higher-methadone-dose groups. There were also higher odds of PONV and delirium in the greater than 0.30 mg/kg group compared to the no-methadone group. There were higher odds of postoperative naloxone and delirium for the 0.25 to 0.30 mg/kg group compared to the no-methadone group. Sensitivity analyses using ideal body weight were similar (Supplemental Digital Content 2, https://links.lww.com/ALN/D686).

Although the difference in pain scores may not be clinically significant (0.1 to 0.9 lower), this finding of lower pain scores alongside a reduction in oral morphine equivalents is clinically relevant, especially in the context of marginal gains.  There were several important side effects including increased PONV and delirium with higher doses of methadone. These findings highlight the importance of clinical judgement in weighing risks versus benefits and the need for careful consideration and individualized approaches in methadone dosing to ensure patient safety.  It is essential to acknowledge the limitations, the retrospective design, and the lack of randomization, which can lead to selection bias. Although there will still be biases associated with retrospective study designs despite efforts to address biases, this study provides data on the use of intraoperative methadone for cardiac surgery among a large, pragmatic sample.