A placebo-controlled trial of gabapentin for painful HIV-associated sensory neuropathies

Background: Painful HIV-associated sensory neuropathies (HIV-SN) are a common complication of HIV infection. The pathogenesis is unknown and the treatment very limited. Gabapentin (GBP) is effective in painful diabetic neuropathy and postherpetic neuralgia and its effectiveness on painful HIV-SN has been reported anecdotally. Design: Multicenter, prospective, randomised, double-blind, placebo-controlled study. Methods: Patients were followed for a 1-week screening, a 4-week double-blind and a 2-week open treatment phase. GBP was initiated at 400 mg/d, titrated over 2 weeks to 1200 mg/d, and then either maintained at this level or - if not beneficial - titrated to 2400 mg/d. After 4 weeks the medication was unblinded and the patient had the choice to begin, to maintain or to increase GBP to 3600 mg/d. The primary outcome measure was an improvement in median pain on the Visual Analogue Scale (VAS) from the screening week compared to the 4^th treatment week. A secondary efficacy measure was the median sleep score (VAS). Results: 15 patients received GBP and 11 placebo. In each group one patient dropped out during the double-blind phase. Median pain (GBP 5.1; placebo 4.7) and sleep score (GBP 4.5; placebo 5.6) did not differ between both groups at baseline. In the GBP-group there was a significant decrease of the pain to 2.85 (-44.1 %) as well as of the sleep VAS to 2.3 (-48.9 %). No significant decrease in the pain (median VAS = 3.3, -29.8 %) as well as in the sleep score (median VAS = 4.95, -11.6%) was observed in the placebo-group. GBP was generally well tolerated. The most frequent side effect was somnolence reported in 80 % of GBP-treated patients. Conclusions: GBP was more effective than placebo in reducing pain and sleep interference in patients with HIV-SN.