Systematic Review of Data from Randomized Controlled Trials on the Efficacy, Safety and Tolerability of Drugs Used to Treat Painful Diabetic Neuropathy
Authors: John Edelsberg, Claudia Lord and Gerry Oster
Abstract: Numerous drugs are used in clinical practice to treat patients with painful diabetic peripheral neuropathy (DPN). Their comparative efficacy, safety, and tolerability have not been well-documented. Objective. To evaluate systematically data from reports of placebo-controlled randomized trials (RCTs) on the efficacy, safety, and tolerability of drugs used to treat painful DPN. Methods. Using data from all published English-language reports of placebo-controlled RCTs of drugs for painful DPN, we compiled information on: (1) placebo-corrected % reduction in pain intensity from randomization to end of active treatment; (2) relative risk (RR) of withdrawal due to lack of efficacy; (3) RR of adverse events (AEs); and (4) RR of withdrawal due to AEs. When estimates were available from more than one trial for a given agent, we pooled data using meta-analysis. Significance of treatment effects versus placebo was based on Z scores. Results. We identified 31 RCTs evaluating 14 different drugs in patients with painful DPN. Most trials were small (<200 patients). All agents were found to significantly reduce pain intensity (range: 8.0% [oxcarbazepine] to 26.4% [CR oxycodone]). RR of withdrawal due to lack of efficacy was significantly lower only for CR oxycodone and tramadol. Analysis of AEs was hampered by spotty and inconsistent reporting and wide variation in sample sizes. Conclusion. While all drugs that we examined were efficacious in painful DPN, available trial data provide only limited guidance regarding optimal therapy, due to small sample sizes, incomplete reporting of AEs, and no published reports of placebo-controlled RCTs of tricyclic antidepressants.