Administration of Preoperative Gabapentin to Patients Undergoing Laparoscopy: A Double-Blinded,Placebo-Controlled Randomized Trial

Introduction

Minimally invasive procedures have been shown to decrease complications and hospital length of stay (LOS). ¹ However, immediate postoperative pain still causes issues for patients in the recovery area. Persistent postoperative pain can contribute to delay in patient discharge after surgery, slower return to routine activities, and overall increased healthcare costs.

Many studies have explored different techniques preoperatively, intraoperatively, and postoperatively to decrease issues with postoperative pain. Treatment of postoperative pain mainly involves the use of opioids and anti-inflammatory medications. These medications, especially opioids, can have considerable side-effects, including dizziness, fatigue, constipation, and the potential for addiction.

One medication that could prove to be beneficial is gabapentin, a third-generation antiseizure drug. It binds to the voltage-gated calcium channels at the α-2-δ subunit and inhibits neurotransmitter release. ² This agent induces central and peripheral antinociceptive activity. Several studies have evaluated gabapentin's use in perioperative pain management.^3,4

A meta-analysis evaluated the use of preoperative administration of gabapentin prior to abdominal hysterectomy. ⁴ The dose and timing of administration has varied among the studies that were evaluated. Dosing ranged from a one-time preoperative dose of 300-mg dose to multiple doses, given both pre- and postoperatively, totaling up to 1200 mg. Laparoscopically, the lowest dose shown to be effective was 300 mg. This decreased postoperative pain and rescue analgesic use following laparoscopic cholecystectomy. ⁵ Taking these doses into consideration, and that this study would utilize laparoscopy, a dose of 300 mg preoperatively was chosen.

This was a double-blinded, placebo controlled randomized trial to assess the effect of a single immediate preoperative dose of gabapentin (300 mg) on immediate postoperative pain in patients undergoing laparoscopy for benign gynecologic indications.

Materials and Methods

Prior to starting this study, approval was obtained from the Penn State Hershey institutional review board. The study was registered with Clinicaltrials.gov on February 9, 2015, with the name “Administration of Pre-Operative Gabapentin to Patients Undergoing Laparoscopy (Gabapentin),* as trial number NCT 02359110.

Patients were identified preoperatively in the division of minimally invasive gynecologic surgery clinic. Inclusion criteria included female patients, ages 18–80, who were scheduled to undergo a laparoscopic or robotic-assisted procedure for benign gynecologic indications.

Exclusion criteria included pregnant patients as well as those having laparotomy, vaginal, or hysteroscopic procedures only or in conjunction with a laparoscopic procedure. Exclusions also excluded patients who were allergic to gabapentin, those who used the drug chronically, and those with a histories of renal insufficiency (as gabapentin is excreted renally). However, patients on chronic opioids were not excluded, secondary to the uncertain analgesic efficacy of opioids in the treatment of neuropathic pain. ⁶

Patients were identified and offered participation during their preoperative visits by utilizing their medical records. If patients wished to proceed and met all of the inclusion criteria, they were consented at their preoperative appointments, in the clinic setting, prior to the day of surgery. Each patient's characteristics, including age, body mass index (BMI), gravidity, parity, and prior surgical history—such as prior cesarean section, laparoscopy, and laparotomy—was recorded from her medical record.

The biggest potential confounder for this randomized trial was a history of chronic pelvic pain. Chronic pelvic pain was defined as a history of pelvic pain >6 months self-reported by the patient. History of chronic pelvic pain was used as a randomization stratification factor to ensure balance between the treatment and placebo groups within each stratum (i.e., no chronic pelvic pain or chronic pelvic pain). The stratified randomization scheme used a random number generator to create variable size-permuted blocks of sizes 2, 4, and 6 patients to ensure that the number of participants in each group was balanced after each set of “B” randomized participants (wherein B is the block size) within each stratum. Randomization occurred prior to the day of surgery and patients were randomized into either the gabapentin or placebo groups using a 1:1 allocation.

Randomization prior to the day of surgery allowed the medications to be ready to dispense to the preoperative area on the day of surgery with inclusion of first-start cases. Investigators and study participants were blinded to the assigned groups. The medication and placebo were identical in appearance. The capsules were administered 30 minutes prior to leaving the preoperative area to allow for transport time to the operating room and setup so that the dose was given ∼1 hour before surgery start time.

All patients received standard general anesthesia for induction and intraoperatively. In addition, the surgical technique was standardized. Three to four ports (5–12 mm) were used in each case; no local or intraperitoneal anesthesia was used; irrigation was always kept to a minimum; and no intraperitoneal humidifiers were used. Postoperatively, patients were given a standard as-needed pain regimen. The amount of medication used during the postoperative hospital stay was tallied and converted into morphine-equivalent dosing in mg. In the recovery area, pain was assessed by the postoperative nursing staff. Scores were obtained from the patients' charts at 2, 4, 6 and 8 hours using a numeric rating scale (NRS) pain rating scale. The NRS ranging from 0 to 10, implemented with adults. ⁷ Having no pain is rated as a 0 and the worst pain the patient could tolerate is rated as a 10. At 2 and 6 hours, pain was also assessed using a visual analogue scale (VAS) pain score. The VAS scoring uses a straight line that corresponds to the values of 1 mm to 100 mm. Total morphine-equivalent dosing was calculated over the entire postoperative hospital stay and at time intervals 0–2, 2–4, 4–6, and 6–8 hours to correspond to the 2-hour increments at which the pain scales were assessed.

Postoperative LOS was defined from the time the patient left the operating room until the time of discharge and was recorded in minutes.

After being discharged to go home, patients recorded their average daily pain using the NRS scale on postoperative days 1 through 7 on a log that they were given prior to discharge. Patients were not asked to track postoperative medication use, in the hope that fewer postoperative tasks would enable with more compliance on returned results. However this did not allow for correlation of pain scores with medication usage.

The primary outcome for this study was the NRS pain score. A 2-point difference in NRS pain scores between the treatment groups was decided to be clinically relevant, and the standard deviation (SD) was assumed be 3.5 points. Additionally, a very conservative approach was taken to sample size estimation and no correlation was assumed among the repeated NRS pain scores over time. It is a conservative estimate because the assumption of 0 correlation among the repeated measurements required a larger sample size to achieve at least 80% power, compared to a sample-size calculation that incorporates some non-0 correlation among the repeated measurements.

Given these assumptions, a sample size of 50 subjects per group provided 81% statistical power to detect a difference of 2 points on the NRS between the 2 treatment groups at any given timepoint, using a two-sided test having a significance level of 0.05. As this study was longitudinal, a 10% subject dropout was anticipated and, therefore, it was planned to recruit a total of 112 subjects.

Pain scores were analyzed using linear mixed-effects models to assess differences between the treatment groups over time. The linear mixed-effects model is an extension of linear regression that accounts for the correlation of measurements over time within a subject. These models were adjusted for the stratification-factor history of chronic pelvic pain as well as morphine-equivalent unit dosing levels. All other continuous data were reported as either mean (SD) with a two-sample t-test for normally distributed data or median (25th, 75th percentile) with a Wilcoxon rank-sum test for skewed distributions. Categorical data were reported as N (%) with either a Mantel–Haenszel χ ² test for ordinal data or Pearson's χ ² test for binary data. All hypotheses tests were two-sided and all analyses were performed, using SAS software, version 9.4 (SAS Institute Inc., Cary, NC).

Results

Of 125 patients enrolled in this study, 109 completed the protocol. The 16 excluded patients did not undergo surgery, rescheduled their procedures, or did not receive the study drug prior to going into the operating room. There were 55 patients in the gabapentin group and 54 patients in the placebo group. A total of 64 patients (32 per treatment group) had histories of chronic pelvic pain (Fig. 1). Groups were similar in patient characteristics including age, BMI, gravidity, parity, and past surgical histories as shown in Table 1.

OPEN IN VIEWER

FIG. 1.

Patient enrollment.

Sixty-two patients underwent hysterectomy; of these, 35 received gabapentin and 27 received placebo. Forty-seven patients underwent operative laparoscopy; of these, 20 received gabapentin and 27 received placebo. Operative laparoscopies included excision/ablation of endometriosis, adnexal surgery, and salpingectomies. The percentage of hysterectomies were not significantly different between gabapentin (63.6%) and placebo (50.0%; p = 0.15).

Postoperative pain was assessed using the NRS and VAS scales and adjusted for morphine doses received. There was no significant difference in pain scores between the treatment and placebo groups at any of the immediate postoperative hours (Table 2). The secondary analysis based on procedure type of hysterectomy or operative laparoscopy also showed no significant difference in pain scores (Table 3).

Pain was also assessed on postoperative days 1 through 7 using the NRS. There was no significant difference in pain scores on any postoperative day in the groups as a whole (Table 4) or with subgroup analysis of hysterectomy and operative laparoscopy (Table 5).

Total opioid consumption was obtained by converting narcotic use during the postoperative hospital stay to morphine-equivalent doses in mg. Total opioid consumption was 14.3 (12.6) mg for the gabapentin group and 14.7 (12.3) mg for the placebo group (p = 0.88). The LOS was similar between the 2 groups, with a median (25th, 75th percentile) of 884.0 (352.0, 1243.0) minutes for gabapentin versus 902.5 (317.0, 1191.0) minutes for placebo (p = 0.67). The LOS for hysterectomy was 968.0 (389.0, 1286.0) minutes for gabapentin versus 1068.0 (407.0, 1325.0) minutes for placebo (p = 0.66), and for operative laparoscopy was 566.5 (255.5, 1189.5) minutes for gabapentin versus 440.0 (273.0, 1022.0) minutes for placebo (p = 0.56).

Discussion

No significant decrease in postoperative pain was identified with preoperative administration of gabapentin. This is in contrast to the finding in a meta-analysis showing that preoperative gabapentin decreased postoperative pain in abdominal hysterectomies. ⁴ One theory that might explain this was proposed by Kissin. ⁸ Laparoscopy has been shown to have decreased postoperative pain compared to laparotomy. The surgeries in that study—all laparoscopic—would have low-intensity noxious stimuli and fallen into this decreased pain group. This postoperative pain might only be nociceptive, not neuropathic; thus, gabapentin would have no analgesic benefit. This is in comparison to the possibility that laparotomy produces both nociceptive and neuropathic pain, which would benefit more from gabapentin.

A similar finding was reported by Paech et al, showing that a single preoperative dose of pregabalin did not reduce acute pain or improve recovery after minor surgery involving the uterus. ⁹ Pregabalin is a structural analogue of γ-aminobytyric acid, binding to the presynaptic α-2-δ subunit of voltage-dependent calcium channels, similar to gabapentin's action but noted to have more than 90% bioavailability. However, pregabalin was given only for minor gynecologic surgeries and included dilation and curettage. No cases had incisions of skin or mucosa.

An interesting finding was that, on postoperative day 1, hysterectomy patients' adjusted mean pain score was 4.2 versus operative laparoscopy patients' adjusted mean pain score of 5.9. This finding is most likely the result of the current authors' patient population. The majority of their laparoscopies are performed for pelvic pain and endometriosis, whereas hysterectomies are performed for many indications, including pelvic pain, endometriosis, abnormal uterine bleeding, fibroids, and risk-reducing surgeries.

Another reason a difference in pain scores were not found is the possibility that the dose of gabapentin was subtherapeutic. Studies show dosing ranging from 300 mg to 1200 mg and from single- to multidose regimens. A single pre-operative dose of 300 mg was chosen for the current study. This was potentially the lowest effective dose and that has shown to be beneficial in laparoscopic procedures. ⁵ Although it was effective for laparoscopic cholecystectomies, the anatomical location of the operation might have made a difference in the outcome; this could be verified with further studies. A multidose regimen was not administered, as a meta-analysis demonstrated that multiple postoperative doses had no significant effect on VAS scores at 24 hours, and, thus, a single dose was favored. ⁴ Subtherapeutic dosing is especially a consideration in patients with chronic pelvic pain, as they might have preexisting neuropathic pain and sensitization.

In addition to the dose, when initially powering and stratifying patients, the procedure being performed was not considered.. Strength of this study would be improved if the patients were stratified and the study powered by both chronic and no chronic pain as well as operative laparoscopy and hysterectomy. Pain scores were only separated based on procedure for the subgroup analysis after the main study was completed.

One of the main strengths of this study was its randomized placebo-controlled design. In addition, there was a single surgical team using consistent techniques throughout the duration of the study. The surgical team included a single minimally invasive gynecologic surgeon, minimally invasive gynecologic surgery fellows, and obstetrics and gynecology residents as well as a dedicated gynecologic operating room team. In addition, patients received standardized preoperative, intraoperative, and postoperative care.

The single gynecology-trained surgical team, while in many ways a strength, might also have been a limitation of this study as this kind of team might not be applicable to other centers with limited resources. Another weakness of this study is that adverse events were not addressed separately. Gabapentin is associated with adverse events such as dizziness, drowsiness, ataxia, and fatigue. Preoperatively, the patients were given gabapentin and then remained sitting or lying on stretchers until they were transported to surgery. There were no occurrences of such events. However, there was a short time interval of 30 minutes between administration and transport to the operating room, and with limited activity, these adverse events would be difficult to assess. There were no documentations of nausea or vomiting in the patients' preoperative nursing records; there was no a formal process for collecting this data. side effects were did not assessed postoperatively, as it would not have been possible to distinguish if these were the result of preoperative gabapentin, intraoperative anesthesia, or postoperative opioid medications.

In the future, consideration could be given to repeating this study with hysterectomies only and excluding operative laparoscopies, with the idea that gabapentin might prove more beneficial for surgeries that have been reported to cause more pain. Although there was no difference in the subgroup analysis, this study was not powered to detect this, and a more-complete evaluation could be beneficial. It might also be beneficial to consider repeating this study with a higher or multidose regimen.

Conclusions

A single dose of 300 mg of preoperative gabapentin did not decrease postoperative pain significantly in gynecologic patients undergoing laparoscopy for benign indications.