To our knowledge, this retrospective study is the first to evaluate the effectiveness of perioperative gabapentin combined with intrathecal morphine for pain control after PSF for AIS. We previously showed that intrathecal morphine combined with oral analgesics provides safe and effective pain control after PSF for AIS.10,12 When our institution initiated our intrathecal morphine for PSF for AIS protocol in 2014, anesthesiologists administered 12 µg/kg of intrathecal morphine (maximum 1000 µg).10 Ninety percent of patients experienced postoperative nausea/vomiting and 40% of patients had pruritus. The protocol was modified by decreasing the dose of intrathecal morphine to 8–10 µg/kg (maximum 800 µg) and implementing more consistent dosing of adjuvant medications (acetaminophen, ketorolac, diazepam).12 While patients had lower and more consistent pain scores after implementation of the updated protocol, nausea/vomiting and pruritus were still experienced by 89% and 64% of patients, respectively. Although the rates of postoperative nausea/vomiting and pruritus after decreasing the dose of intrathecal morphine were similar to our historical control group who received a hydromorphone epidural infusion10, we strove to further improve our protocol to reduce opioid-related side effects while providing effective analgesia.
In 2018, our institution initiated oral gabapentin administration in the perioperative period. Due to concerns for potential oversedation, standard of care was to decrease the dose of intrathecal morphine administered. The results of this retrospective study show that while addition of gabapentin as adjuvant therapy did not seem to have an effect on pain scores in the early postoperative period, it did result in reduced oral opioid consumption and more consistent postoperative pain scores throughout the hospitalization after PSF for AIS. This postoperative opioid-sparing effect was seen even despite patients in this group receiving a lower dose of intrathecal morphine in the operating room. The patients who received intrathecal morphine alone had significantly higher pain scores in the afternoon and evening of POD 1 compared to earlier postoperative time periods. This may be secondary to increased pain after mobilization with physical therapy. The patients who received gabapentin would have had one to two postoperative doses by the afternoon of POD 1 and this may explain the more consistent pain scores in that group.
Previous studies have assessed the effect of perioperative gabapentin combined with intravenous PCA after PSF for AIS, and showed a single dose of preoperative gabapentin did not result in a difference in opioid use or pain scores compared to placebo.17 A randomized, double-blind, controlled trial conducted by Rusy et al. demonstrated lower pain scores in PACU and the morning after surgery in patients who received a preoperative dose of gabapentin and postoperative gabapentin for 5 days.3 Patients who received perioperative gabapentin also had reduced postoperative morphine consumption during the first two postoperative days. As such, the authors recommended continuing gabapentin for only the first two days after surgery. Similarly, Trzcinski et al. retrospectively found that patients who received perioperative gabapentin had improved pain scores and decreased opioid use for 48 to 72 hours after surgery.5 Thomas et al. demonstrated a decrease in time to complete physical therapy goals when patients received perioperative gabapentin but there was no difference in length of hospital stay.4
Choudhry et al. retrospectively showed that PSF patients who received perioperative gabapentin combined with intravenous PCA had decreased total PCA doses, decreased morphine use on POD 1, and shorter time to transition to orals compared to patients who received PCA alone.1 However, postoperative pain scores, time to ambulation, and length of hospital stay were similar. These authors found that addition of a clonidine transdermal patch for 7 days postoperatively to perioperative gabapentin and intravenous PCA resulted in the shortest time to transition to orals, shortest time to ambulation, and shortest length of hospital stay, but there was no difference in total PCA doses and morphine use compared to patients who received gabapentin and intravenous PCA.
Although we did not observe a shorter time to ambulation, physical therapy discharge, or length of hospital stay with the addition of perioperative gabapentin to intrathecal morphine, PSF patients who are managed according to our intrathecal morphine for PSF for AIS protocol appear to achieve these goals earlier than what has been reported by other authors. The mean time to ambulation for both groups in our study was 19 to 21 hours, mean time to physical therapy discharge was 2.0 to 2.3 days, and mean length of hospital stay was 2.4 to 2.6 days. While Thomas et al. found a decrease in time to complete physical therapy goals when patients received perioperative gabapentin combined with intravenous PCA, only 28% of those patients ambulated on POD 1, 52% of those patients completed all physical therapy goals on POD 2, and mean length of hospital stay was 3 days.4 Similarly, in Choudhry et al.’s study, the gabapentin and gabapentin plus clonidine groups had a mean time to ambulation of 36 hours and 27 hours, and a mean length of hospital stay of 87 hours and 77 hours, respectively.1
Prior studies have shown no difference in the rate of adverse events, including postoperative nausea/vomiting, pruritus, and oversedation, with the addition of perioperative gabapentin to intravenous PCA after PSF for AIS.1,3,5 Nausea/vomiting is a common adverse event after PSF. Trzcinski et al. reported nausea/vomiting in 75% of their entire retrospective cohort.5 The patients who received perioperative gabapentin in our study experienced a significantly lower rate of postoperative nausea/vomiting and pruritus, possibly secondary to receiving a lower dose of intrathecal morphine. However, this difference may also be directly related to antiemetic effects of gabapentin at central nervous system sites18 as some prior studies have shown that it appears to have antiemetic effects in at least some patient populations,19 including spinal surgery.20 In our ITM + GABA group, 52% still either received an antiemetic medication or had documented nausea/vomiting and 44% either received antipruritic medication or had documented pruritus. It can be difficult to retrospectively determine the rate of these minor adverse events. In addition to documentation of an adverse event in the medical record, we also considered administration of an antiemetic or antipruritic as evidence of nausea/vomiting or pruritus. However, it can be difficult to know whether these medications were administered for prophylaxis or treatment so the rate of these adverse events may have been over-reported. Also, at our institution, ondansetron is ordered as treatment for both nausea/vomiting and pruritus, making it difficult to tell retrospectively for which indication it was given.
Another limitation of this study is inconsistent perioperative gabapentin dosing. Per institutional protocol, gabapentin was to be administered at 10 mg/kg preoperatively followed by 5 mg/kg three times daily for 2 days beginning on POD 1. In reality, there was a wide range of gabapentin dosing due to many factors. One was the lack of prompt availability of liquid gabapentin in our preoperative area for patients who were not able to swallow pills. Additionally, the pediatric anesthesiology group at our institution is large (> 35 anesthesiologists) with variable preferences for gabapentin dosing and uncertainty about the impact that the addition of gabapentin might have on their preferred anesthetic technique for these cases. We found wide variability in gabapentin dosing in published reports on perioperative gabapentin combined with intravenous PCA in PSF patients with AIS.1, 3–5,17 We hope that the retrospective data we have collected will allow us to implement a standard perioperative gabapentin protocol for all patients undergoing PSF for AIS at our institution and we have since worked with our operating room pharmacy to increase the availability of liquid gabapentin in the preoperative area. Another limitation is the small sample size. Our study may have been underpowered to detect differences in some of our outcomes. Lastly, this was a retrospective study so data collection was dependent on accurate documentation in the medical record.