This report was to evaluate the efficacy of lipoic acid, prostaglandin El and methylcobalamin (L + P + M) for the treatment of diabetic peripheral neuropathy (DPN) in comparison with that of prostaglandin El plus methylcobalamin (P + M), in order to provide the basis and reference for clinical rational drug use. Randomized controlled trials (RCTs) of L + P + M for DPN published up to 3rd August, 2014 were searched. A random or fixed effect model was used to analyze outcomes which were expressed as risk ratios (RRs) or mean difference (MD) with a 95% confidence interval (CI). Eighteen RCTs with 1410 participants were included. Clinical efficacy of L + P +M therapy was significantly better than P + M therapy (fifteen trials; RR 1.32, 95% CI 1.24-1.41, P<0.00001, I-2=32%). As compared with P + M therapy, the pooled effects of L + P + M therapy on nerve conduction velocities (NCVs) were (fifteen trials; MD 4.70, 95% CI 3.77-5.63, P < 0.00001, I-2 = 79%) for median MNCV, (thirteen trials; MD 4.73, 95% CI 3.69-5.77, P < 0.00001, I-2 = 85%) for median SNCV, (sixteen trials; MD 4.22, 95% CI 3.32-5.12, P < 0.00001, I-2 = 83%) for peroneal MNCV, (fourteen trials; MD 3.09,95% CI 2.04-4.14, P < 0.00001,12 = 82%) for peroneal SNCV. There was no serious adverse events associated with drugs intervention. L + P +M therapy was superior to P + M therapy for improvement of clinical efficacy and NCVs in DPN patients. These findings should be further verified by high-quality RCTs. (C) 2015 Elsevier Ireland Ltd. All rights reserved.