The review addressed a clear question with well-defined inclusion criteria. Only two databases were searched, so relevant trials may have been missed. The search was restricted to published articles, so publication bias could not be ruled out. As there were no language restrictions, the risk of language bias was small. Appropriate steps were taken during the study selection and quality assessment stages to minimise the risk of reviewer error and bias. It was unclear whether similar steps were taken during the data extraction process.
The quality of included trials was assessed using an appropriate tool and the results showed good quality. There were no conflicts of interest in the current review, but all included trials received funding from a manufacturer of alpha lipoic acid. It was unclear whether suitable steps were taken in the meta-analysis to enable multiple intervention groups with the same comparator group to be entered into the meta-analysis. There was evidence of significant statistical heterogeneity for two of the analyses, so it may not have been appropriate to combine the trials in a meta-analysis. Only a small number of trials with relatively few patients were available for meta-analysis, which potentially undermined the strength of the results.
Given the possibility of publication bias, the small number of available trials and the possibility of weaknesses in the statistical analysis, the reliability of the authors' conclusions is unclear.