Vitamin D for painful diabetic neuropathy: a systematic review and meta-analysis of randomised controlled trials

BackgroundDiabetes mellitus contributes increasingly to the Global Burden of Disease [GBD]; particularly in high and increasingly in low-and middle-income countries. Strategies to prevent and mitigate the impact are a public health priority. Painful diabetic neuropathy (PDN) is a syndrome of sensory disorders caused by both type 1 and type 2 diabetes mellitus. Available treatments include antidepressant medications and strong analgesics. These are often only partly effective and associated with significant side effect profiles. There is a need for effective treatments with low toxicity. Vitamin D has been proposed as potential therapeutic and biologically plausible agent. Non-randomised studies demonstrate benefit, but are subject to biases. There is a need for robust evidence derived from randomised data to inform patient care in this debilitating complication of diabetes. Review aimTo synthesise randomised controlled trials (RCTs) of Vitamin D supplementation and its effects on painful diabetic neuropathy. Review methodsA range of databases [Medline, EMBASE, Web of Science, Cochrane Library, CINHAL, EBSCO and Google Scholar] were searched from inception to March 2025, with backwards and forward citation searches to identify eligible studies. RCTs comparing Vitamin D with placebo in patients with diabetes [type 1 or 2] and PDN were sought. The primary outcome was pain as measured using a validated pain measure or measure of PDN. A fixed effects meta-analysis of continuous pain data was conducted, with standardisation between studies to calculate a standardised mean difference [SMD] between Vitamin D and placebo. Small study and publication bias was tested using an Egger plot, and study quality was assessed using the Cochrane Risk of Bias [RoB] tool. ResultsFour eligible studies were identified and three of these studies [comprising 260 participants] provided meta-analysable data. There was a statistically significant short-term benefit for vitamin D (pooled standardised mean difference =-0.70; 95%CI-0.95 to-0.45). There was moderate between study heterogeneity, and there was an intermediate level of heterogeneity (I = 54.9%). No studies reported medium-or long-term outcomes. The quality of studies was variable (either low or moderate risk of bias), with poor concealment of allocation as the most important design limitation. Two of the studies had been prospectively registered, making it difficult to check for bias in one study due to potential selective reporting of outcomes. Despite conducting an Egger Funnel Plot, it was not possible to exclude the influence of small study and selective publication bias. DiscussionBased on a meta-analysis of non-registered small size studies there was evidence of short-term reduction in pain symptoms in painful diabetic neuropathy. This benefit needs to be confirmed in fully powered RCTs with a longer duration of follow up. Vitamin D remains a viable low-cost treatment option for PDN, but more research is needed before this can confidently be recommended for routine patient care.