The Effect of Neurontin on Pain Management in the Acutely Burned Patient

The study was conducted in a 16-bed American Burn Association certified burn unit. Patients age >18 years old, with at least a 5% burn injury and an expected length of stay (LOS) of 48 hours, were approached for enrollment in this prospective, placebo controlled randomized study. Patients who were pregnant, lactating, prisoners or who had renal insufficiency were excluded. After consent, patients were assigned to either placebo or Gp by random numbers generated in Microsoft Excel (2010). Both the drug and the placebo were over-encapsulated to appear identical. The placebo pills contained starch. The research clinical pharmacist was the only unblinded staff member and did not participate in clinical care of the patients.

Following randomization, patients received a loading dose of study drug on day one and began three times a day (TID) dosing per the dose escalation schedule the following day. At discharge, patients were given a three day taper per the dose de-escalation schedule Patients were assessed for completion of psychosocial adjustment (Brief Symptom Inventory, BSI, and Sickness Inventory Profile, SIP) at their first clinic visit.

Agents used for pain control included: acetaminophen, non-steroidal anti-inflammatories, morphine instantaneous release and morphine extended release. In the case of allergies or ineffectiveness, other agents were occasionally used. Short acting morphine was ordered every two hours prn and hydromorphone was ordered every four hours as needed. All were converted to morphine equivalents.

The study was powered to detect a 22% difference in opioid consumption between the two groups based on the work by Cuignet et al. It was estimated that a total of 50 patients were needed to achieve an alpha of 0.05 and a beta of 0.80 to detect the difference in the primary endpoint.

For statistical purposes, conversion tables were used to convert all opioid medications into morphine equivalents with 1 morphine equivalents (ME)=30mg oral morphine. The primary outcome variable (morphine consumption) were adjusted for days past injury. The BSI and SIP scales were scored according to study directions.

Both an intention to treat and actual treatment analysis were performed using Stata 11.2 for Windows (Stata Corp. College Station, Texas, U.S.A.). Continuous variables between groups were analyzed with the students T test. Categorical variables were analyzed with the Chi Square test or Fischer Exact Test where appropriate. A random effects model adjusting for confounders was used to assess the effect of treatment on the outcome measures. The study was approved by the University's Institutional Review Board and was registered with the clinical trials association (200909736).

Inclusion criteria

• All admitted patients with LOS expected to be > 48 hours (usually burn injury > 5%)
• > 18 years of age
• Thermal injury to skin

Exclusion criteria

• Prisoners
• Pregnant or nursing women
• Children <18 years of age
• Frostbite or non thermal injury to skin
• Renal insuffiency (creatinine clearance < 60mL/min) or failure (on renal replacement)
• Expected length of stay < 48 hours (this usually includes burn <5%