The purpose of this study is to evaluate the patient experience when using gabapentin with other pain control medications after posterior spinal fusion surgery for scoliosis in adolescents. These results will be compared to patients who underwent the same procedure during the study period and received the same standardized pain control regimen excluding gabapentin. Effects on pain level, opioid use, and satisfaction will be measured. Opioid side effects including nausea, sedation and urinary retention (inability to empty one's bladder) will also be recorded.The null hypotheses are as follows:

1. There is no significant difference in pain control when adding gabapentin to a multimodal pain management protocol in pediatric post-operative posterior spinal fusion patients.
2. There is no significant difference in the amount of opioid medication required for pain control in pediatric post-operative posterior spinal fusion patients.

Patients aged 10-19 years with idiopathic scoliosis, and classified as American Society of Anesthesiology (ASA) class I to III who intended to undergo posterior spinal fusion for deformity correction were enrolled. Prior to surgery, subjects were randomized into either the experimental or control group by the OHSU research pharmacy using an online randomization tool, which utilized block randomization upon patient enrollment to result in equal sized groups at study completion. Patients, caretakers and providers remained blinded to the group assignments.

Patients filled out the Scoliosis Research Society, SRS-22 standardized form at a pre-operative clinic appointment. Following hospital admission, patients recorded initial pain level with the Visual Analog Scale (VAS) prior to receiving standardized pre-operative medications. The VAS scale used in this study was a 10 cm line with anchors of "no pain" at the left and "worst pain imaginable" at the right; each point was measured to the nearest millimeter. In the post-operative period, nursing staff assessed patient pain using the VAS at 4-hour intervals from 06:00 until 22:00 for the duration of hospitalization for a minimum of 4 daily scores recorded per patient. After the third post-operative day, but before discharge, the parent or guardian of each subject was asked to complete an IRB-approved survey to measure parent demographics and parental satisfaction with the patient's hospitalization and pain control.

Each patient received standardized medications according to our multimodal pain protocol. Following hospital admission, patients in both groups received one 12.5 mg/kg dose of intravenous (IV) acetaminophen. Patients in the experimental group received one 15 mg/kg dose of liquid gabapentin while the control group received a placebo, formulated to match the volume, color, taste, and smell of the experimental medication. The gabapentin or placebo was prepared by the OHSU research pharmacists so that providers and investigators remained blinded to treatment assignment. Several intraoperative anesthetic medications were given to subjects in both groups including: IV ketamine at 5mcg/kg/min for 120 minutes and IV Ketorolac 0.5mg/kg up to 15mg. Intra-operative IV propofol and IV hydromorphone were titrated to desired effect.In the post-operative period, gabapentin was administered to the experimental group at 10mg/kg PO q8h, beginning at Phase II and continued through postoperative day four. The control group received equivalent volume doses of placebo at the same intervals. Post-operative medication was administered according to the following protocol for both groups: ketorolac continued at 0.5mg/kg up to 15mg IV q6h for 12 total doses. Once ketorolac doses were complete, the patient may have received Ibuprofen 10mg/kg up to 600mg PO as needed. Hydromorphone was given through patient-controlled analgesia (PCA) at a basal dose of 0.002mg/kg/hr for 24 hours and demand dose of 0.004mg/kg with an 8-minute lockout. Once basal PCA was discontinued, administration of oxycodone 0.1-0.2mg/kg PO up to 15mg PO q4h as needed supplemented the PCA demand dose. If the patient tolerated PO oxycodone without emesis, the PCA hydromorphone was discontinued after 24 hours, but a rescue dose of hydromorphone 0.002mg IV q4 was available if needed. Other as needed medications included diazepam 0.15mg/kg up to 5mg PO q6h for muscle spasms, ondansetron 0.1mg/kg up to 4mg IV q12h for nausea, and IV acetaminophen 12.5mg/kg up to 1000mg q6h. Acetaminophen 12.5mg/kg up to 650mg PO q6h hours was administered after the IV was removed. All patients received one Senokot-S tablet and Miralax 0.8 g/kg up to 17g daily for bowel regimen.

For the entire hospitalization, nursing staff monitored vital signs and assessed sedation using the standardized Pasero Opioid-induced Sedation Scale (POSS) protocols at 4-hour intervals.16 Any POSS score of 3 or greater resulted in more frequent monitoring of respiratory status and sedation level, decreased opioid dosing, and administration of naloxone as needed. All patients were routinely monitored for known adverse gabapentin drug reactions including: peripheral edema, nausea/emesis, viral disease, ataxia, dizziness, nystagmus, somnolence, hostile behavior, fatigue and fever, Stevens-Johnson syndrome, drug hypersensitivity reactions, drug induced coma/seizure, and suicidal thoughts. Any perceived adverse reaction would have resulted in the gabapentin or placebo being stopped at the clinicians' discretion

Inclusion Criteria:

• Patients of age 10-19 with an American Society of Anesthesiologists patient classification of I to III undergoing surgery to correct idiopathic or neurogenic scoliosis.

Exclusion Criteria:

• Patients who require a surgical approach or technique differing from posterior spinal fusion and/or have allergies to any of the standardized or experimental study medications: acetaminophen, gabapentin, hydromorphone, ketorolac or oxycodone.

Adverse event data was collected over the time period that patients spent in the hospital during their procedure (operative day through discharge day). Average hospital stay was 4.5 days with a standard deviation of 0.7 days for the gabapentin arm and 0.9 days for the placebo arm.