Introduction
Postoperative pain management after major spine surgery, including multilevel lumbar fusion, presents a significant clinical challenge. Opioids remain a cornerstone of analgesia in this setting, but their use is associated with a well-documented profile of adverse effects, including respiratory depression, sedation, nausea, vomiting, and dependence risk [
1]. Multimodal analgesia, which combines different analgesic agents with various mechanisms of action, is now the standard of care, aiming to improve pain relief while reducing opioid consumption and related side effects [
2].
Nefopam is a centrally acting, non-opioid analgesic derived from benzoxazocine. Its mechanism of action is distinct from both opioids and non-steroidal anti-inflammatory drugs (NSAIDs), which involve serotonin, norepinephrine, and dopamine reuptake inhibition, as well as glutamatergic pathway modulation via sodium and calcium channels [
3]. Unlike opioids, nefopam does not cause respiratory depression or sedation at therapeutic doses and does not affect platelet function such as NSAIDs. However, it has side effects, including tachycardia, sweating, and nausea, attributed to its sympathomimetic-like actions [
4]. These characteristics make nefopam an attractive option as part of a multimodal analgesia regimen, which potentially provides opioid-sparing effects without the typical risks of opioids [
5].
Several studies have demonstrated the opioid-sparing efficacy of nefopam in various surgical contexts, including laparoscopic, gynecological, and orthopedic procedures [
6–
8]. Despite this evidence, the role of nefopam in spine surgery analgesia remains unclear. Spinal fusion, especially multilevel procedures, is related to intense postoperative pain and high opioid requirements. Eiamcharoenwit et al. [
9] revealed no significant reduction in 24-hour morphine consumption with a single-dose nefopam regimen in patients undergoing lumbar spine surgery. Similarly, Chalermkitpanit et al. [
10] in 2023 reported that a continuous nefopam infusion did not significantly reduce opioid consumption over the first postoperative days but decreased length of stay in open lumbar fusion. Similarly, continuous nefopam infusion, as part of a multimodal regimen, failed to confer additional analgesic or functional benefits in minimally invasive spine surgery [
11]. Potential explanations for these results include differences in dosing strategies, suboptimal timing or plasma concentrations, and robust baseline multimodal analgesia administration, which may obscure incremental benefits of nefopam.
Considering the limited and mixed evidence, further investigation is warranted, particularly in patients undergoing extensive spinal fusions who may have greater postoperative analgesic requirements. We hypothesized that a continuous infusion of nefopam could provide a significant analgesic benefit. Therefore, this study aimed to investigate the efficacy and safety of postoperative continuous intravenous (IV) nefopam for reducing morphine consumption in patients undergoing decompressive laminectomy and fusion for multilevel degenerative lumbar spinal stenosis.
Materials and Methods
Study design and ethical approval
This prospective, randomized, double-blind, placebo-controlled clinical trial was conducted at the largest university hospital in Thailand from January 2023 to February 2025. The Siriraj Institutional Review Board (SIRB) approved the study protocol (COA no., Si 772/2021), and all participants signed written informed consent before enrollment. The trial was registered at the Thai Clinical Trials Registry (TCTR20220615006).
Participants
Eligible participants included patients aged 18 years or older diagnosed with degenerative lumbar spinal stenosis who were scheduled to undergo elective decompressive laminectomy and spinal fusion with pedicular screw instrumentation at two or more levels. Exclusion criteria were a history of allergy or contraindication to nefopam, opioids, gabapentinoids, or ketorolac; severe renal impairment (estimated glomerular filtration rate of <30 mL/min/1.73 m2); a history of seizures; or a history of substance abuse.
Randomization and blinding
Patients were randomly allocated in a 1:1 ratio to either the intervention (nefopam) or the control (placebo) group. A computer program was used to generate the randomization sequence. The sealed envelope method was employed to maintain the allocation concealment. Patients, postoperative care providers, and outcome assessors were all blinded to the treatment allocation. The operating surgeon was not involved in postoperative assessment.
Intervention and surgical procedure
All patients underwent general anesthesia employing a standardized protocol. Five surgeons operated using the same surgical technique with standard open decompression, posterior pedicle screws instrumentation with posterolateral fusion or transforaminal interbody fusion. Immediately postoperatively, the intervention was initiated upon the start of skin closure. The intervention group received a continuous IV infusion of nefopam 80 mg (half-dose reduction in patients over 65 years old) diluted in 500 mL of normal saline, administered over 24 hours, whereas the control group received a continuous IV infusion of 500 mL of normal saline (placebo) administered over 24 hours.
Standardized perioperative care
Throughout the perioperative period, all patients received a comprehensive, standardized multimodal analgesic regimen.
Preoperative
On the morning of surgery, patients received a prophylactic oral dose of 1,000 mg of paracetamol (reduced to 500 mg for patients weighing <50 kg or with severe liver impairment).
Intraoperative
Anesthesia was standardized to exclude agents that could confound pain assessment, including ketamine, IV NSAIDs, dexmedetomidine, and IV lidocaine. Fentanyl and IV paracetamol were permitted. All patients received a local anesthetic wound infiltration, comprising 20 mL of 0.5% bupivacaine mixed with 30 mg of ketorolac before the final skin closure.
Postoperative
A multimodal regimen was initiated once patients were alert, comprising paracetamol (1,000 mg every 6 hours for 48 hours, then transitioning to a paracetamol/codeine combination) and gabapentin (300 mg every 12 hours). All patients were provided with a morphine IV patient-controlled analgesia (IV-PCA) device for breakthrough pain, managed by the hospital’s Acute Pain Service. Selective COX-2 inhibitors were withheld for the first 48 hours to ensure a clear assessment of the study intervention. Postoperative nausea and vomiting were treated with metoclopramide as needed.
Outcome measures
The primary outcome includes the total cumulative morphine consumption via IV-PCA at 24 hours postoperatively. Secondary outcomes include cumulative morphine consumption at 12, 36, and 48 hours; pain intensity at rest and during movement, measured using a 0–10 Verbal Analog Scale at 12, 24, 36, and 48 hours; and incidence of adverse effects, including nausea, vomiting, tachycardia (heart rate >100/min), hypotension, sedation, and respiratory depression.
Statistical analysis
The sample size was calculated using the data from Du Manoir et al. [
12], aiming for 80% power to detect a clinically significant difference in 24-hour morphine consumption with a two-sided alpha of 0.05. The initial target was 96 patients (48 per group). A group sequential design with three planned interim analyses (at 33%, 67%, and 100% of patient enrollment) was implemented using the O’Brien–Fleming alpha-spending function to maintain overall α at 0.05, with a stricter significance boundary (
p<0.0151) at interim, which may impact effect-size precision [
13]. This approach permits early termination of the trial for efficacy or futility while preserving the overall type I error rate. The study was terminated after the second planned interim analysis with 66 patients (33 per group, intention-to-treat population), as the pre-specified boundary for efficacy was crossed for the primary outcome. No safety concerns were observed to warrant stopping.
The Shapiro-Wilk test was used to first test continuous variables for normality. Normally distributed data are presented as mean±standard deviation (SD) and were compared between groups using the independent-samples t-test. Non-normally distributed data (including pain scores on Numeric Rating Scale and morphine consumption, which demonstrated skewed distribution) are summarized as median with interquartile range in addition to mean±SD and were compared using the Mann-Whitney U test. The chi-square test or Fisher’s exact test was used to compare categorical variables (e.g., incidence of side effects, gender, American Society of Anesthesiologists [ASA] class distributions) as appropriate. The interim-adjusted significance level of a p-value of <0.0151 was applied for the primary outcome (24 hours morphine) (as per O’Brien–Fleming correction). All p-values reported are two-tailed.
Discussion
This randomized double-blind trial revealed that continuous IV nefopam infusion, administered for 24 hours after multilevel lumbar spinal fusion, significantly produced a clear opioid-sparing effect without measurable differences in pain intensity, thereby supporting its role as an adjunct rather than a direct analgesic. Nefopam-treated patients required roughly half the amount of morphine in the first 24–48 hours while achieving the same pain control level as those receiving standard analgesia alone. To the best of our knowledge, this is one of the first studies to demonstrate a clear opioid-sparing benefit of nefopam in the context of extensive spine surgery. The results support our initial hypothesis that the incorporation of nefopam into a multimodal analgesic regimen improves postoperative analgesia (as evidenced by reduced opioid consumption) without compromising pain relief or safety.
Our results align with the previous evidence in other surgical populations that nefopam provides effective adjunct analgesia and reduces opioid needs [
5]. Nefopam’s mechanism of action likely underlies this opioid-sparing effect. Nefopam activates endogenous descending pain inhibition and prevents central sensitization by inhibiting monoamine reuptake and modulating NMDA (N-methyl-D-aspartate)-mediated pathways [
4]. This multimodal mechanism provides analgesia via a different pathway than opioids, thereby complementing opioid effects. In practical terms, nefopam-treated patients achieved adequate pain relief with fewer opioid doses. The PCA use data demonstrated that once the immediate postoperative period passed, nefopam-treated patients had less need to press the PCA button to maintain comfort. This indicates that nefopam contributed to pain relief in the late postoperative period. The similar pain scores between groups are interpreted as a positive outcome, whereas nefopam allowed equivalent analgesia at rest and with movement, despite significantly lower opioid consumption. This is particularly valuable because reducing opioids lessen opioid-related adverse effects and potentially facilitate faster recovery.
Notably, our trial focused on patients undergoing multilevel lumbar fusions, who generally experience severe postoperative pain and frequently require high doses of opioids. The benefits of nefopam may be more apparent in this setting of high-pain intensity. Our findings indicate that nefopam administration may be most pronounced in extensive spine surgeries, where pain intensity and opioid demand are higher. In patients with fusions that involve more than two levels, nefopam reduced morphine requirements by 6–12 mg within the first 24 hours and by over 20 mg within 48 hours, thereby representing a clinically meaningful opioid reduction. By contrast, in two-level fusions, opioid administration was lower overall, and nefopam did not provide additional sparing. This emphasizes the importance of tailoring multimodal analgesia strategies to surgical invasiveness. Further, nefopam may be particularly beneficial in cases with a higher expected pain burden. In contrast, previous studies in spine surgery, which have reported negative or null results, had some key differences. Eiamcharoenwit et al. [
9] investigated nefopam in a broader spine surgery context (including possibly less extensive surgeries) and used dosing limited to the intraoperative period (one dose before incision and/or one dose before closure). They revealed no significant reduction in 24-hour morphine consumption with nefopam, perhaps because the effect of those one or two doses may have worn off early in the postoperative course. In our study, by providing a continuous infusion throughout the first 24 hours, we ensured sustained plasma levels of nefopam, which may be required to observe a meaningful impact on opioid requirements over a full day. Our positive results somewhat contrast with the recent work of Chalermkitpanit et al. [
10], who administered a similar dosing regimen in lumbar fusion and reported no difference in 24-hour morphine administration (nefopam 10.4 mg vs. placebo 11.2 mg) and no improvement in pain beyond the immediate recovery room. They noted a benefit in terms of reduced length of hospital stay by approximately 1 day for nefopam-treated patients. Several factors could cause the discrepancy between their opioid consumption results and ours. First, their study included 100 patients with a robust multimodal analgesia protocol (including scheduled NSAIDs and pregabalin) that led to relatively low morphine usage even in the control group (mean 11.2±9.7 mg/24 hr). This “ceiling effect” may leave little room for nefopam to demonstrate further reductions. In our study, the control group’s morphine use was higher (20.49±16.60 mg/24 hr), thereby potentially revealing a larger margin for opioid-sparing by nefopam. Second, population differences may appear, particularly since our patients were slightly older (mean 70 years old vs. 63 years old) and possibly more opioid-naïve, which could amplify the relative benefit of adding a nonopioid analgesic. Further, differences in outcome definitions and study power could play a role. Chalermkitpanit et al. [
10] revealed that nefopam improved pain scores in the post-anesthesia care unit (immediate postoperative period) in their study, consistent with nefopam providing good early analgesia, but the difference dissipated by postoperative day 1. They hypothesized that the continuous infusion may not maintain analgesic efficacy once the initial bolus effect fades, and indicated intermittent bolus dosing of nefopam (20 mg every 6 hours) may be a better strategy. However, our data indicate that even with continuous infusion, opioid requirements continued to diverge beyond 12 hours, indicating ongoing analgesic contribution from nefopam through 24–36 hours. Differences in pain assessment (their patients had similar pain scores and thus morphine PCA may have been used less overall) or simply variability may explain the contrast. Importantly, our study’s positive result adds a new piece to the puzzle, indicating that nefopam’s effect in spine surgery may depend on factors, including surgery extent, baseline analgesic regimen, and perhaps patient-specific factors (e.g., pharmacogenomics affecting nefopam metabolism). Mobilization and length of stay were not predefined outcomes; however, clinical nursing records demonstrated similar early ambulation within 48 hours across both groups, indicating no delay in functional recovery.
The safety profile of nefopam in our study was consistent with previous reports and generally favorable. We found no increase in serious adverse event incidences with nefopam. In particular, concerns about respiratory depression or excess sedation were not observed, and the sedation case observed was mild and spontaneously resolved after a few hours. This reinforces the advantage of nefopam as an opioid alternative. It provides analgesia without the respiratory depressant effects inherent to opioids. Tachycardia and nausea were slightly more frequent in the nefopam group, which represents known side effects. However, these differences were not significant, and the absolute incidence was low. Other studies and a meta-analysis have revealed that nefopam may increase the risk of mild tachycardia and sweating, which may be related to its sympathomimetic properties (inhibition of dopamine reuptake peripherally or central activation). However, most studies reported no significant differences between placebo and nefopam administration [
6,
14,
15].
The incorporation of nefopam into multimodal analgesia for multilevel spine fusion appears beneficial in reducing opioid use, which is a key goal in the context of the opioid-sparing paradigm [
16]. By reducing opioid requirements, nefopam may indirectly decrease the incidence of opioid-related adverse effects, including nausea, constipation, urinary retention, and respiratory depression, thereby improving overall patient comfort and safety. In our study, the intervention group experienced excellent pain relief with approximately half the morphine dose of the control group, without an increase in pain or serious side effects. This indicates that anesthesiologists and acute pain teams can consider continuous nefopam infusion as an adjunct for major spine cases, especially in patients at risk from high opioid doses (older patients, those with obstructive sleep apnea or pulmonary issues, etc.). Further, since nefopam does not affect platelet function or bleeding risk, it is suitable for perioperative use in spine surgery where postoperative hematoma can be a concern (unlike NSAIDs, which some surgeons limit due to fusion healing or bleeding worries).
This study has some limitations. First, the sample size was relatively small, and the trial stopped early at the interim analysis, which may have overestimated the treatment effect despite α-adjustment. Second, this is a single-center study with standardized surgical techniques, which may limit generalizability. Third, outcomes were assessed only up to 48 hours; thus, longer-term pain control, functional recovery, and opioid use were not evaluated. Further, we did not specifically assess neuropathic pain symptoms. Considering the putative anti-neuropathic properties of nefopam, it could be interesting to see if patients with pre-existing neuropathic pain or high painDETECT scores benefit differently [
10]. Finally, while blinding was maintained, side effects, including tachycardia, could theoretically have unblinded some cases, although this is unlikely to have affected PCA-based outcomes.