Local insulin improves wound healing: a Systematic Review and Bayesian Network Meta-analysis

Background: Wounds are a significant health issue that requires reliable and safe strategies to promote repair. Clinical trials have demonstrated that local insulin promotes healing in acute and chronic wounds (i.e., reductions of 7-40% vs. placebo). However, the trials' sample sizes have prevented drawing solid conclusions. Furthermore, no analysis has focused on safety concerns, (i.e. hypoglycaemia). Under the hypothesis that local insulin promotes healing through pro-angiogenic effects and cellular recruitment, the aim of this systematic review and network meta-analysis (NMA) was to assess its safety and relative effectiveness using a Bayesian approach. Methods: Medline, CENTRAL, EMBASE, Scopus, LILACS, and grey literature sources were searched for human studies assessing the local use of insulin vs. any comparator since inception to October 2020. Data on glucose changes and adverse events, wound and treatment characteristics, and healing outcomes was extracted, and an NMA conducted. Results: A total of 949 reports were found, of which 23 (n = 1,240 patients) were included in the NMA. The studies evaluated six different therapies, and most comparisons were against placebo. NMA showed -1.8 mg/dL blood glucose change with insulin and a lack of reported adverse events. Statistically significant clinical outcomes identified include reduction in wound size (-27%), increased healing rate (23 mm/day), lower PUSH scores (-2.7), -10 days to attain complete closure, and an OR = 20 for complete wound closure with insulin use. Likewise, significantly increased neo-angiogenesis (+30 vessels/mm 2) and granulation tissue (+25%) were also found. Conclusion: Local insulin promotes wound healing without significant adverse events.