Introduction
Cesarean section (CS) is the world’s most common major surgical procedure (Pfuntner, W. L. M., and Stocks 2013). It accounts for approximately 32% of births in the United States and is predicted to trend upward through 2030 (Betran et al. 2021; Martin et al. 2015). CS involves a large abdominal incision, making the effective delivery of postoperative analgesia of primary importance. Indeed, patients rank perioperative avoidance of pain as their highest priority regarding anesthesia outcomes (Carvalho et al. 2005). Inadequate pain control contributes to post-partum depression, impaired bonding with infants, and is a risk factor for developing chronic pain syndromes (Eisenach et al. 2008; Apfelbaum et al. 2016; Toledo, Miller, and Wisner 2018; Sutton and Carvalho 2017).
Regional anesthesia for CS, consisting typically of an intrathecal injection of a local anesthetic in addition to a lipophilic opioid, provides analgesia post-operatively for up to 36 hours while also reducing post-operative opioid requirements (Sutton and Carvalho 2017). However, many opioid naïve patients will require persistent opioid use postoperatively (Bateman et al. 2016). Considering the rising number of individuals with opioid use disorder, alternative analgesic agents and strategies have been explored to limit opioid consumption after CS (Lyden and Binswanger 2019).
Gabapentin is an anticonvulsant medication first discovered in the 1970, receiving FDA approval in 1993 for use in the United States (Lumsden et al. 2019). Initially used as a muscle relaxer and an anti-spasmodic, it has now been approved by the FDA for multiple conditions such as epilepsy and neuropathic pain conditions (Yasaei, Katta, and Saadabadi 2022; Felder et al. 2019a). It works by binding voltage-gated calcium channels inhibiting the release of excitatory neurotransmitters (Yasaei, Katta, and Saadabadi 2022). Gabapentin has been used as preemptive analgesia or as a supplement for postoperative (i.e. abdominal and vaginal hysterectomies) analgesia, resulting in decreased opioid-associated vomiting and pruritus (Ho, Gan, and Habib 2006). Gabapentin also demonstrated reduction in postoperative pain scores and opioid requirements 24 hours following hysterectomy when administered perioperatively (Alayed et al. 2014). There has been an increased interest in using gabapentin as part of pain-reducing strategies for patients undergoing CS.
The objective of this study, a systematic review (SR), was to characterize perioperative gabapentin use and outcomes in patients undergoing CS. The related goal of this study was to determine if using gabapentin in the perioperative timeframe reduced opioid requirements and postoperative pain in patients undergoing a CS.
Materials and methods
We conducted a systematic review of the literature for studies utilizing gabapentin as part of the perioperative pain management strategy for patients undergoing a CD. This is a PROSPERO trial (ID: CRD42024538992), submitted on April 22nd, 2024, and registered on May 5th, 2024.
The search strategy was composed and conducted by a medical librarian in Ovid MEDLINE, Ovid Embase, and Cochrane. The final literature search was run on May 6th, 2024, around the main concepts for “Cesarean Section,” “Gabapentin,” and “Pregabalin” and adapted for each database as necessary, from their inception until May 2024. The search strategy for Ovid MEDLINE included: (“Cesarean Section” OR “Cesarea*” OR “Caesarea*” OR “c section*” OR “abdominal deliver*” OR “ex utero intrapartum treatment procedure*”) AND (“Gabapentin” OR “Pregabalin” OR “gabapentin*” OR “Neurontin*” OR “pregabalin*” OR “ci1008*” OR “ci 1008” OR “Lyrica*”). Additional reference lists of identified original articles or reviews were searched manually. Randomized controlled trials (RCTs) and observational studies were included. No restrictions regarding parity, age, ethnicity, maternal comorbidities, or indication for CS were applied. Studies were included if they used gabapentin or pregabalin pre- and/or postoperatively at the time of CS. Studies were excluded if there was no comparison group in which a placebo or intervention other than gabapentinoids was given. Observational studies were included if it was clear which participants received the gabapentin or pregabalin and which did not.
Retrieved articles were assessed independently by two reviewers (HF and KO) to verify their eligibility. Any discrepancies in reviewers’ opinions on inclusion or exclusion of any article were discussed with a third reviewer (DH) to reach a consensus. Titles and abstracts were initially assessed and ones that appeared relevant had full text manuscripts retrieved, when possible, for full evaluation.
A search of PROSPERO May 2024 using the terms “gabapentin” and “cesarean” revealed no hits for other systematic reviews on the site.
Data extraction was focused on design (authors, study design), participants (number of participants in intervention and control groups), interventions (different regimens, frequency, timing, control type, con-interventions, analgesics used during CS), and outcomes (Mixed Morphine Equivalents (MME), Pain Scores (VAS, NRS, or POD), Patient satisfaction). Data from studies was also extracted by 2 of the authors and any discrepancies resolved by consensus.
We conducted risk of bias assessments using the Cochrane tools for randomized control trials (RoB-2 tool) (Sterne et al. 2019), the Risk of Bias Assessment tool for non-randomized studies of Interventions (ROBINS-I) tool for the observational studies (Sterne et al. 2016), and reported it according to the PRISMA Statement 2020 (Page et al. 2021).
Two reviewers (HF and KO) independently assessed the risk of bias in each study using the Risk of Bias Assessment tool for non-randomized studies of Interventions (ROBINS-I) tool for the observational studies (Sterne et al. 2016). RoB 2 tool was used to assess the risk of bias of the RCTs (Sterne et al. 2019). Any discrepancies were resolved by a third author (DH).
Summary descriptive information from each included study were extracted. The primary outcome was total cumulative opioid pain medication required after cesarean as morphine milligram equivalents (MME), typically measured during the first 48 hours of the hospital stay. Secondary outcomes included maximal pain scores measured with movement at 24 hours after the procedure (NRS pain scores), and participant satisfaction with pain. Outcomes of pain at activity at 24 hours was estimated from the VAS, NRS, or POD found in the articles, by converting all to the NRS (Supplemental material). Similarly, satisfaction with pain was measured on different scales and converted to 0 to 10. Appropriate statistical methods were used to convert different summary statistics provided to the mean and standard deviations for use in the network meta-analysis (Hozo, Djulbegovic, and Hozo 2005; Wan et al. 2014; Cai, Zhou, and Pan 2021). Only descriptive statistics were provided for adverse events due to the heterogeneity in assessment.
A network meta-analysis (NMA) was conducted for each outcome using a frequentist approach which adjusts for the correlation between multiple comparisons within multi-arm studies (Franchini et al. 2012; Rücker and Schwarzer 2014). Analysis was conducted with the netmeta package in R software (Balduzzi et al. 2023). Network meta-analysis allows us to incorporate the most evidence and can provide direct and indirect evidence on contrasts of interest such as pairwise difference between two means. Mean differences in outcomes and associated 95% confidence intervals are reported in league tables. For each outcome, a random effects NMA was fit including all studies and also for the subset of randomized controlled trials only. Random effects models were conducted due to the heterogeneity in studies and populations.
Results
The study selection process is described in Figure 1. Our initial search yielded 726 titles, of which 18 were deemed relevant and full texts requested. We included 13 studies in the final analysis, 10 RCTs and 3 observational studies (Moore et al. 2011; Short et al. 2011; Nofal, Mahmoud, and Al Alim 2014; Memari et al. 2015; Monks et al. 2015; Hafez et al. 2017; Randolf et al. 2019; Najafi Anaraki and Mirzaei 2014; Karami, Hoshyar, and Jafari 2021; El Kenany and El Tahan 2016; Grasch et al. 2023; MacGregor et al. 2021; Ende et al. 2022). Risk of bias was reported to be overall low in both observational studies and randomized control trials, except for the Memari study that had some concerns in biases arising from randomization and outcome measurement (fig.2-3) (Memari et al. 2015). All the RCTs were placebo-controlled studies with blinding of investigators and participants, with an exception in the Anaraki trial that used intrathecal fentanyl in the control group as opposed to pre-operative oral gabapentin (table 1) (Najafi Anaraki and Mirzaei 2014). Most trials included healthy pregnant patients with a singleton gestation, undergoing scheduled cesarean delivery at term. Study location, sample sizes, inclusion and exclusion criteria are described in Table 1.
A total of 597 patients were in the intervention groups and 498 in the control group. VAS pain on movement at 24 hours post-op was the most reported outcome (n=7), followed by MMEs and NRS pain scores (Table 1). Five studies reported patient satisfaction outcomes. Two observational studies incorporated gabapentin as part of CS enhanced recovery protocols (ERAS). Both observational studies noted benefits. Seven of the eight trials using only preoperative gabapentin showed some benefit. The typical mean gestational age at randomization was about 38 weeks. The maternal age, gestational age, and BMI describing the patient populations are found in Table 2. Administered doses of gabapentin ranged from 200mg to 900mg, and from 150-300mg for pregabalin. Timing of doses was up to 1 hour pre-op for most studies, except for two studies where gabapentin was continued post-operatively for 48 hours (Grasch et al. 2023; MacGregor et al. 2021).
From a random effects NMA model, treatment with 600 mg of gabapentin preoperatively, compared to no treatment or placebo, was associated with a decrease in cumulative opioid requirements post-operatively in MME (-20.72, 95% [CI -33.52, -7.93]) when including all study designs (Figure 4). When limiting the analysis to randomized trials, treatment with 300 mg of gabapentin or pregabalin preoperatively also led to reduced opioid use (MME: -4.30, 95% CI [-8.25, -0.36]; -6.90, 95% CI [-8.51, -5.29], respectively) (Figure 5, Table 3). Using the same model for RCT studies only for the outcome of pain, 600 mg and 900 mg of gabapentin were associated with lower pain scores (-1.01, 95% CI [-1.98, -0.04] and -2.50, 95% CI [-4.28, -0.71], respectively) (Figure 6, Table 4). Results are similar when including all study designs for pain on movement (Supplemental Figure 1). From the limited studies that report satisfaction (n= 4; all using gabapentin), satisfaction was higher for those in treatment groups (200 mg/both, 300 mg, and 600 mg), but only the 300 mg dose of gabapentin provided significantly higher satisfaction (1.35, 95% CI [0.45, 2.24]) than the control group (Figure 7).
Discussion
Principal findings
Administration of 300-600mg of gabapentin and 300mg of pregabalin led to a significant decrease in MMEs in term-pregnant patients undergoing CS. Even small reductions in opioid requirements after CS can be important for patient recovery. Regarding pain scores, 600-900mg of gabapentin led to lower scores overall. Additionally, 300mg of gabapentin drove satisfaction with pain control scores significantly higher.
Results
The findings demonstrated are in accordance with the results of a previous SR-MA by Felder et. al. evaluating the relation between perioperative gabapentin and post-CS pain control (Felder et al. 2019b). This study differs from the Felder SR-MA due to the inclusion of 5 additional studies, 3 RCTs (Karami, El Kenany, and Anaraki), and 2 observational (MacGregor and Grasch) assessing the implementation of an ERAS protocol containing gabapentin, in addition to evaluating the effect of pregabalin (Najafi Anaraki and Mirzaei 2014; Karami, Hoshyar, and Jafari 2021; El Kenany and El Tahan 2016; Grasch et al. 2023; MacGregor et al. 2021). These additional studies allowed the demonstration of an effect on MME at 300mg doses of gabapentin and pregabalin, findings not present in the previously mentioned SR-MA that showed efficacy using 600mg of gabapentin (Felder et al. 2019b).
Clinical Implications
Clinical utilization of gabapentinoids has long been a controversial topic in the literature. Despite demonstrating lower MMEs, pain scores, and nausea when used pre-operatively in abdominal hysterectomies (Alayed et al. 2014), multiple publications have been published opposing their use, citing statistically but not clinically relevant effects on postoperative pain, subacute pain and no effects on chronic postoperative pain in both gabapentin and pregabalin, regardless of administration timing. Most of these reviews did not support the routine use of gabapentinoids in the management of postoperative pain in adults (Verret et al. 2020; Eipe et al. 2015; Mishriky, Waldron, and Habib 2015; Fabritius et al. 2017b, 2017a; Goodman and Brett 2019).
ERAS pathways were developed to optimize patient outcomes by introducing interventions that were data supported (Kalogera and Dowdy 2016). ERAS implementation has shown promising results after for gynecologic oncology procedures, delivering improved clinical outcomes such as length of stay, complications, readmissions, and cost (Bisch and Nelson 2022). Going further, Meyer et. al. found a 72% reduction in mean opioid consumption among gynecologic oncology patients after the implementation of the ERAS protocol, in addition to 16% of patients achieving an opioid-free hospital stay without significant differences in pain scores (Meyer et al. 2018). Meyer et. al. also reported that the ERAS group had an earlier resolution of impairment in activities and return to walking (Meyer et al. 2018). Decreases in opioid consumption and pain scores were similarly observed in ERAS protocols that included gabapentin, in both pre-CS or pre- and post-CS administration (Grasch et al. 2023; MacGregor et al. 2021).
Implementation of ERAS protocols containing gabapentinoids would then, hypothetically, lead to safer pain control methods compared to opioids. One of the benefits of gabapentin is its mild side-effect profile when compared to opioids (Yasaei, Katta, and Saadabadi 2022). Additionally, gabapentin does not concentrate in breast milk, leading to low serum levels and no observed side effects in breastfed infants (Ohman, Vitols, and Tomson 2005; Kristensen et al. 2006). Pregabalin has also been described as moderately safe during breastfeeding (Hale and Rowe 2016). Side effects mentioned in studies analyzed included sedation, nausea, vomiting, dizziness, and blurred vision (Nofal, Mahmoud, and Al Alim 2014; Monks et al. 2015; Hafez et al. 2017; Karami, Hoshyar, and Jafari 2021; El Kenany and El Tahan 2016). In light of these findings, an RCT designed by Memari et. al. evaluated the effect of 600mg of Gabapentin pre-op on post-CS nausea and vomiting (Memari et al. 2015). The double-blind, placebo-controlled trial demonstrated significantly lower nausea severity at 1hr postop (p= 0.041), with lower incidence of vomiting (p=0.048) in the Gabapentin group (Memari et al. 2015).
Research Implications
Whether gabapentin or pregabalin is the optimal choice for CS ERAS regimens has not been well examined. The included RCTs by El Kenany and Karami both compared pregabalin to placebo (Karami, Hoshyar, and Jafari 2021; El Kenany and El Tahan 2016). However, we were unable to locate any trials comparing gabapentin to pregabalin directly. The use of a network approach to the meta-analysis allowed for indirect comparisons of gabapentin and pregabalin. In our network analysis limited to RCTs, 300mg doses of pregabalin and gabapentin demonstrated similar reductions in consumed MMEs. A trial directly comparing the two drugs may be warranted.
Strength and Limitations
A strength of our study is that the methodology followed was compliant with the Cochrane Handbook for Systematic Reviews of Interventions (Higgins et al. 2019). Limitations of this study include different dosing and scheduling regimens for gabapentinoids used in the studies (pre-op vs both pre- and post-op), which may have limited the ability to demonstrate some significant differences and to synthesize results from different studies. There was also some inconsistency in the dose findings, likely due to smaller numbers in different dosing groups. Additional limitations include the limited sample sizes of studies included, non-uniformity such as in inclusion and exclusion criteria amongst studies evaluated, and lack of information regarding duration of surgery, or presence/absence of adhesions. A clear research priority should be harmonization of gabapentin regimens for CS ERAS protocols, as well as measurement standardization for outcomes, including participant satisfaction. While the inclusion of observational studies removes some ability to discuss causation, these represent some “real world” implementation using the drug and can help add to the evidence base. Additionally, outside of the observational studies, only the small Randolf trial used gabapentin continuing into the postoperative period (Randolf et al. 2019).
Conclusion
In conclusion, implementation of ERAS protocols including gabapentin or pregabalin may decrease total MME requirements in patients undergoing CS. Reducing MME, even in small increments, can be important in postoperative care. Gabapentinoids may also safely provide lower pain scores and improve satisfaction rates among patients. Given these findings, a large, multicenter trial is likely warranted to find the optimal ERAS gabapentin regimen to reduce opioid use after CS.