Background
Significant pain from HIV-associated sensory neuropathy (HIV-SN) affects ∼40% of HIV infected individuals treated with antiretroviral therapy (ART). The prevalence of HIV-SN has increased despite the more widespread use of ART. With the global HIV prevalence estimated at 33 million, and with infected individuals gaining increased access to ART, painful HIV-SN represents a large and expanding world health problem. There is an urgent need to develop effective pain management strategies for this condition.

Method and Findings
Objective: To evaluate the clinical effectiveness of analgesics in treating painful HIV-SN. Design: Systematic review and meta-analysis. Data sources: Medline, Cochrane central register of controlled trials, www.clinicaltrials.gov, www.controlled-trials.com and the reference lists of retrieved articles. Selection criteria: Prospective, double-blinded, randomised controlled trials (RCTs) investigating the pharmacological treatment of painful HIV-SN with sufficient quality assessed using a modified Jadad scoring method. Review methods: Four authors assessed the eligibility of articles for inclusion. Agreement of inclusion was reached by consensus and arbitration. Two authors conducted data extraction and analysis. Dichotomous outcome measures (≥30% and ≥50% pain reduction) were sought from RCTs reporting interventions with statistically significant efficacies greater than placebo. These data were used to calculate RR and NNT values.

Results
Of 44 studies identified, 19 were RCTs. Of these, 14 fulfilled the inclusion criteria. Interventions demonstrating greater efficacy than placebo were smoked cannabis NNT 3.38 95%CI(1.38 to 4.10), topical capsaicin 8%, and recombinant human nerve growth factor (rhNGF). No superiority over placebo was reported in RCTs that examined amitriptyline (100mg/day), gabapentin (2.4g/day), pregabalin (1200mg/day), prosaptide (16mg/day), peptide-T (6mg/day), acetyl-L-carnitine (1g/day), mexilitine (600mg/day), lamotrigine (600mg/day) and topical capsaicin (0.075% q.d.s.).

Conclusions
Evidence of efficacy exists only for capsaicin 8%, smoked cannabis and rhNGF. However,rhNGF is clinically unavailable and smoked cannabis cannot be recommended as routine therapy. Evaluation of novel management strategies for painful HIV-SN is urgently needed.