Abstract
Study Design
Retrospective Propensity-Matched Analysis.
Objectives
Optimal control of postoperative pain while minimizing opioid consumption is paramount in spine surgery. While gabapentinoids have robust utilization for the treatment of neuropathic pain and neuroprotective effects, their interplay in multimodal analgesia following ACDF is unclear. The objective of the present study was to investigate the association between postoperative gabapentinoid use and postoperative opioid use following ACDF.
Methods
We conducted a retrospective cohort study using the TriNetX Research network, identifying adult patients who underwent ACDF between 2003-2023. Patients with chronic opioid use were excluded. After 1:1 propensity score matching, cohorts were constructed based on receipt of postoperative gabapentinoids vs acetaminophen monotherapy. Outcomes included opioid utilization and surgical/systemic complications across standardized follow-up intervals (30 days, 90 days, 6 months, 1 year, 2 years, and 5 years). Risk ratios with 95% confidence intervals were calculated, and Kaplan-Meier analyses assessed time-to-event outcomes.
Results
After matching, 32 455 patients were included in each group. Gabapentinoid use was associated with higher opioid consumption at every interval, persisting through 5 years (2.29% vs 0.51% at 5 years, RR 4.61, P < 0.001). Gabapentinoid recipients had a greater risk of pneumonia and respiratory failure across multiple timepoints. Kaplan-Meier curves demonstrated durable separation between groups for pneumonia, respiratory failure, and opioid use (all log-rank P < 0.001).
Conclusion
Gabapentinoids use following ACDF was associated with increased opioid utilization and higher complication rates, challenging their presumed benefit in this setting. These findings suggest that gabapentinoids may not be an effective adjunct in multimodal pain regimens for ACDF.
Introduction
Effective control of pain following spine surgery, while minimizing unnecessary narcotic consumption, is central to optimizing clinical outcomes. Patients with uncontrolled pain following spine surgery are more likely to experience sleep disturbances, diminished functional outcomes, and increased incidence of postoperative mental health pathologies, thus significantly interfering with postoperative rehabilitation trajectories. 1
Opioid consumption in the treatment of spine conditions has been linked with the unfortunate ongoing public health burden of chronic opioid misuse, abuse, and fatal overdoses.2–5 Previous literature has demonstrated that decreasing opioid utilization following spine surgery is linked to improved patient satisfaction, outcome scores, and reduced prolonged opioid dependence. 6 In efforts to combat opioid misuse, the spine community has sought to decrease postoperative narcotic consumption through the incorporation of multimodal analgesic regimens, utilizing various enhanced recovery after surgery (ERAS) pathways.3,7 These ERAS pathways incorporate several synergistic pain medications to combat the multifactorial nature of postoperative pain, including inflammatory, nociceptive, and neuropathic targets for pharmacologic pain control. 7
One such target of multimodal pain medication regimens is the gabapentinoid class, which includes gabapentin and pregabalin. 8 These medications are designed to target neuropathic pain through inhibitory effects upon presynaptic calcium channels, which are concentrated within the dorsal root ganglion and spinal cord, thus preventing the release of excitatory neurotransmitters from afferent sensory neurons and subsequent activation of nociceptive pathways. 9 Gabapentinoids have demonstrated neuroprotective and antinociceptive properties in preoperative settings, with evidence supporting their use for attenuating central sensitization and reducing neuropathic pain when administered before surgery.10–12 Despite their important role in pre-operative settings, there remains evidence to suggest that concomitant use of gabapentinoids and opioids in the postoperative setting may increase the risk of opioid-associated complications such as respiratory depression and overdose.13,14 Currently, equipoise exists within the spine literature, as Karnati et al found that gabapentinoid use reduced long-term opioid use following long-segment lumbar fusion, whilst Tao et al found that opioid use actually increased in those patients taking gabapentinoids following one to three level anterior lumbar interbody fusion (ALIF).15,16
Even though anterior cervical discectomy and fusion (ACDF) remains one of the most commonly performed spine surgeries in the United States, with more than 137 000 annual procedures, no literature to date has investigated the interplay of postoperative gabapentinoids and their role in opioid consumption.17,18 Therefore, the purpose of the present study is to clarify the utility of gabapentinoids following ACDF, and their relationship to postoperative opioid requirements, thereby helping surgeons elucidate the safest and most effective methods for postoperative pain control.
Methods
Study Design and Data Source
This retrospective cohort study was conducted using the TriNetX Research Network, a global federated database that compiles de-identified electronic health records from more than 133 academic medical centers and community healthcare organizations. The platform enables HIPAA-compliant access to patient-level data, including clinical diagnoses, medication prescriptions, laboratory results, procedures, and longitudinal follow-up encounters. All contributing institutions maintain data use agreements to ensure patient confidentiality and data integrity. As the dataset consists exclusively of de-identified information, the study was classified as non-human subjects research and was exempt from institutional review board oversight. Study methodology adhered to the guidelines outlined by the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement.
Patient Selection and Cohort Construction
Cohort Characteristics
Exposure and Outcome Measures
The primary exposure was the receipt of a postoperative gabapentin prescription within 30 days following ACDF, compared to acetaminophen monotherapy as described in previous methods.16,20 Baseline clinical and demographic characteristics were evaluated before and after matching to confirm balance between groups including nonsteroidal anti-inflammatory drug(NSAID) use. Primary outcomes included opioid use and respiratory complications assessed at multiple standardized timepoints: 30 days, 90 days, 6 months, 1 year, 2 years, and 5 years postoperatively. Outcomes were identified using structured queries of ICD-10 diagnosis codes, CPT procedure codes, and pharmacy records within the TriNetX database.
Statistical Analysis
All statistical analyses were performed within the TriNetX analytics platform, with supplementary validation in R (v4.3.0, Vienna, Austria). Descriptive statistics were used to summarize cohort characteristics, with comparisons made using chi-square tests for categorical variables and t-tests for continuous variables. Risk ratios (RR) with 95% confidence intervals (CI) were calculated to assess differences in outcome incidence between groups across each time point. Cumulative incidence curves were generated for time-to-event outcomes such as opioid use and respiratory complications, with between-group comparisons assessed using log-rank testing. Cohort-level data were analyzed to quantify the association between opioid use and respiratory complications (pneumonia and respiratory failure) across multiple follow-up intervals (3 months, 6 months, 1 year, 2 years, and 5 years). First, Pearson correlation coefficients were calculated between opioid use rates and the incidence rates of each respiratory outcome to assess the direction and strength of the association. Subsequently, separate simple linear regression models were fitted for each outcome, using the incidence rate of pneumonia or respiratory failure as the dependent variable and opioid use rate as the independent variable. Coefficients (β) and 95% CIs were reported to describe the change in outcome rate per unit change in opioid rate. To model event counts and account for varying cohort sizes, Poisson regression with a log link and offset term for total cohort size was performed. Model overdispersion was assessed using the deviance-to-degrees-of-freedom ratio; when overdispersion was detected, negative binomial models were used. Exponentiated coefficients were reported as incidence rate ratios (IRRs) with corresponding 95% CIs. Model performance was assessed using the coefficient of determination (R2) for linear models and Akaike Information Criterion (AIC) for count-based models. Statistical significance was set at a two-sided P-value <0.05 for all primary analyses.
Results
Pre- and Post-matching Demographics
Following 1:1 propensity score matching, the gabapentin and acetaminophen cohorts were each composed of 32 455 patients (Table 1). The mean age at time of surgery was 55.8 years and 55.7 years for the gabapentin and acetaminophen cohorts, respectively (P = 0.78). Single-level fusions constituted 62.06% of the gabapentin group and 62.03% of the acetaminophen group, while 37.94% of the gabapentin and 37.97% of the acetaminophen cohorts received multi-level fusions. The proportion of single- and multi-level fusions was well balanced between cohorts (P = 0.89 and P = 0.85, respectively). No significant differences were noted concerning diagnoses of myelopathy and radiculopathy amongst cohorts (P > 0.05). Following matching, no significant differences were found in NSAID use (P > 0.05). The distribution of males and females was well balanced among groups, with 51.7% males in the gabapentin cohort compared to 51.8% in the acetaminophen cohort (P = 0.64), and 45.4% vs 44.7% females, respectively (P = 0.072).
After matching, the BMI was 30.4 kg/m2 in the gabapentin group and 29.6 kg/m2 in the acetaminophen group (P < 0.001), though this difference is likely not clinically significant. Type 2 diabetes were present in 12.0% of gabapentin and 11.8% of acetaminophen patients (P = 0.36), while type 1 diabetes was more significantly common in gabapentin patients (0.36% vs 0.46%, P = 0.037). Hypertension was reported in 31.4% of gabapentin and 31.6% of acetaminophen patients (P = 0.49). COPD (3.1% vs 3.3%, P = 0.17) and chronic kidney disease (2.3% vs 2.4%, P = 0.29) were not significantly different between groups. Coagulation defects (1.2% vs 0.9%, P < 0.001) and peripheral vascular disease (0.69% vs 0.57%, P = 0.047) were also more common among gabapentin patients. Rates of heart failure (1.2% vs 1.3%, P = 0.52), malignancy (1.4% vs 1.3%, P = 0.49), smoking (11.1% vs 11.4%, P = 0.300), liver disease (0.24% vs 0.20%, P = 0.24), benign neoplasm (0.51% vs 0.64%, P = 0.25), and dementia (0.17% vs 0.19%, P = 0.41) were not significantly different (Table 1).
3-Month to 2-Year Outcomes
Cohort Comparison Analysis, Opioid Utilization, Respiratory Complication
Association Between Opioid Exposure and Respiratory Complications in the Gabapentin Cohort Across Linear and Negative Binomial Models
Cumulative incidence curves for opioid use, respiratory failure, and pneumonia between Gabapentin and acetaminophen cohorts
5-Year Outcomes and Cumulative Incidence
At 5 years, gabapentin use remained associated with significantly greater opioid utilization (2.29% vs 0.51%, RR 4.61, 95% CI 3.90-5.45, P < 0.001) (Table 2). Respiratory complications were also higher, including pneumonia (4.52% vs 2.26%, RR 2.00, 95% CI 1.84-2.19, P < 0.001) and respiratory failure (3.92% vs 2.10%, RR 1.87, 95% CI 1.70-2.04, P < 0.001). Cumulative incidence curves demonstrated sustained separation between the gabapentin and acetaminophen cohorts for opioid use, pneumonia, and respiratory failure over the 5-year follow-up. At each time point, cumulative incidence was higher among gabapentin patients across all three endpoints, with log-rank tests confirming significant between-group differences (P < 0.001 for all). (Tables 2 and 3, Figure 1).
Opioid and Respiratory Complication Correlation Analysis
In cohort-level regression models evaluating the overall relationship between opioid exposure and respiratory outcomes, linear regression demonstrated significant associations between higher opioid use and both pneumonia (β = 1.73, 95% CI 1.29-2.16, P < 0.001, R2 = 0.91) and respiratory failure (β = 1.41, 95% CI 1.09-1.74, P < 0.001, R2 = 0.93) (Table 3). Negative binomial models using event counts showed increased opioid exposure corresponding to higher incidence of pneumonia (IRR = 1.003, 95% CI 1.001-1.006, P = 0.002, AIC = 132.8) and respiratory failure (IRR = 1.003, 95% CI 1.001-1.004, P = 0.003, AIC = 130.0). (Table 3).
Discussion
The present investigation of more than thirty thousand patients undergoing ACDF demonstrates that gabapentinoid use is associated with a significantly increased utilization of opioids in the immediate postoperative period, a trend that continues to be significant at 5 years following surgery. Beyond this, patients receiving gabapentinoids were at consistently higher risk of complications across nearly every postoperative interval. These complications included a greater risk of post-operative pneumonia and respiratory failure. Kaplan-Meier curves demonstrated durable separation between cohorts over a 5-year follow-up for opioid utilization, suggesting that gabapentinoid utilization may not serve as a beneficial adjunct in a postoperative multimodal pain regimen for patients undergoing ACDF.
Gabapentin and pregabalin, collectively termed gabapentinoids, have been widely adopted as adjuncts for multimodal perioperative analgesia in patients undergoing spine surgery. 7 These medications modulate neuropathic pain pathways in the central nervous system and were postulated to provide opioid-mitigating effects after surgery. 21 To our knowledge, no studies have specifically evaluated the safety and efficacy of gabapentinoids after ACDF. Prior studies have yielded conflicting results about the effectiveness and safety of gabapentinoids after spine surgery. Tao et al demonstrated a modest 2.9% reduction in daily opioid use with concomitant use of gabapentinoids following long-segment lumbar fusion; however, the gabapentinoid cohort had 37% longer duration of opioid prescriptions, 42% total higher opioid consumption post-discharge, a higher rate of pain-related readmissions, and more diagnoses of postoperative drug abuse. 15 Alongside this, Cetik et al examined chronic preoperative gabapentinoid use in patients undergoing spinal fusion surgery and discovered that the concomitant use of gabapentinoids and opioids was associated with worse pain and functional outcomes, as well as more opioid use, suggesting certain patient populations may not benefit from gabapentinoids as part of their multimodal pain regimen. 20 While these aforementioned findings are important for spinal surgeons to be cognizant of, they may not be directly applicable to anterior cervical procedures where there exists a significant difference in “surgical intensity,” compared to multilevel lumbar fusions. The present study builds on this prior work by demonstrating that gabapentinoid use after ACDF is, in this dataset, associated with sustained increases in opioid consumption at multiple time points. Being mindful of these novel findings is critical for spinal surgeons to simultaneously provide appropriate postoperative analgesia while limiting the need for increased narcotic usage. Understanding the interplay of gabapentinoid medications and postoperative opioid consumption is paramount, given the robust utilization of these neuromodulatory and neuroprotective medications in the preoperative setting.
Beyond questions of efficacy, the findings from the present investigation serve to add to the growing body of evidence that gabapentinoids can pose additional risks to post-operative patients, particularly when combined with other CNS depressants. Bykov et al found that adding a gabapentinoid to opioid-based analgesia strategies was associated with nearly double the risk of opioid overdose and higher risks of respiratory complications. 14 Similarly, Park et al conducted a propensity-matched retrospective cohort study of nearly 1 million surgical patients over age 65, noting that perioperative gabapentin use was associated with a greater risk of pneumonia. 22 In 2019, the Food and Drug Administration (FDA) released a warning about the use of gabapentinoids in patients at risk of respiratory depression, including those taking concomitant opioids. 23 The setting of anterior cervical surgery necessitates a particular focus on alleviating risk factors for postoperative respiratory complications, as these can confer increased mortality, readmission rates, and decreased patient-reported outcomes.24–27 The present investigation is the first of its kind to demonstrate that gabapentinoid use was associated with an increased risk of postoperative pneumonia and respiratory failure following ACDF, data that is integral to keep in mind in the perioperative arena.
These findings have important implications for perioperative pain management in patients undergoing ACDF. The primary goal of incorporating gabapentinoids into multimodal analgesia regimens is to reduce opioid consumption, especially in light of the opioid crisis, yet our data suggest they may have the opposite effect after ACDF, suggesting increased opioid use. This increase in utilization is particularly concerning, especially given the clearly established link between prolonged postoperative opioid use and higher rates of respiratory complications, readmissions, and worse functional outcomes.28–32 Gabapentinoid use in the postoperative period may therefore compound these established risks. At a population level, even small increases in consumption could contribute to meaningful long-term exposure to opioids and their downstream sequelae. Further, our data demonstrated increased risk of other potentially devastating complications, including respiratory complications and readmissions. Surgeons must therefore weigh the theoretical benefit of gabapentinoids with the demonstrated risks of gabapentinoid use after ACDF. Our results also highlight the importance of tailoring enhanced recovery after surgery (ERAS) pathways to the specific pain profile of ACDF rather than extrapolating from data from lumbar or other spine procedures. A one-size-fits-all ERAS pathway may not be appropriate for anterior cervical surgery. Developing individualized, procedure-specific protocols optimizes pain control while minimizing potential complications and should be a target of future research. 33
Any discussion of gabapentinoid medications necessitates the consideration of potential abuse and represents a growing public health concern, as rates of reported abuse have been estimated as high as 1.6% in the general population. 34 These rates are substantially higher among individuals diagnosed with concomitant opioid use disorder, as recent reports have demonstrated increased opioid-related overdose mortality and hospital utilization with concomitant gabapentinoid use. 35 Most reported misuse instances are cited for achieving euphoric effects, self-medication for pain control, or withdrawal symptoms which can involve significant psychiatric symptoms.36,37 Keeping this in mind, surgeons should exercise appropriate caution when administering gabapetinoids, and monitor for signs of abuse, which can include early refills and reluctance to taper. 37
The present investigation is not without limitations. First, as this investigation was conducted using a large, national database, its limitations lie in the lack of detail inherent to ICD and CPT coding systems, as well as the retrospective nature of the study. While matching was performed using previously validated propensity systems, it is possible that comorbidities and additional confounding variables may not have been adequately accounted for. With this, adverse events such as pulmonary complications years following gabapentin exposure may have been associated with alternative etiologies, such as additional procedures or illnesses that were not captured by the database. Because the dataset includes only data from academic medical centers, differences in patient populations at community medical facilities may limit the generalizability of our findings. Alongside this, the present investigation was not able to garner patient-reported outcomes (PROMS) such as Visual Analog Scale neck or arm scores secondary to the limitations of the database, which would have provided granularity regarding potential differences in the pain and severity of pre- and postoperative symptoms experienced by patients in the gabapentinoid cohort when compared to our control cohort. Additionally, the database from which the study was conducted does not provide data on dosing regimens or medication compliance. Additionally, as combination analgesic regimens are frequently used postoperatively, our findings may not fully capture real-world multimodal pain management patterns, and future studies should evaluate outcomes associated with combination therapy. With this, if patients were exposed to over the counter (OTC) medications such as acetaminophen outside of prescription, this would not have been captured by the present study. Finally, as the database utilized does not contain quantifiable data concerning prior opioid use, this limits the ability to conclude that gabapentinoid use may directly correlate to increased opioid use.
Conclusion
In this large propensity-matched cohort of more than thirty thousand ACDF patients, perioperative gabapentinoid use was associated with increased opioid consumption and higher rates of respiratory complications across short- and long-term follow-up. These findings challenge the routine incorporation of gabapentinoids into multimodal pain regimens for ACDF, suggesting that their theoretical benefits in neuropathic pain modulation may not translate to anterior cervical procedures. Future prospective studies are needed to clarify causal mechanisms, identify subgroups of patients who may still benefit from gabapentinoids, and guide the development of procedure-specific ERAS pathways for ACDF.
Footnotes
Funding
The authors did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Consent Declaration
The use of entirely de-identified records qualified this study as non-human research and is therefore exempt from institutional review board oversight.