Association between diabetes and thiamine status - A systematic review and meta-analysis

Abstract

BackgroundThiamine (vitamin B1) is an essential cofactor in glucose metabolism, but it remains unclear whether thiamine status is lower in individuals with diabetes compared to individuals with normal glucose metabolism. AimsWe conducted a systematic review and meta-analysis to study whether the circulating concentrations of various thiamine analytes differ between people with and those without diabetes. MethodsPubMed and the Cochrane Central Register of Controlled Trials were searched according to the study protocol. The standardized mean difference (SMD) and 95 % confidence intervals (CI) of thiamine markers between individuals with and without diabetes were used as effect size (random effects model). Subgroup analysis considered albuminuria as an additional variable. ResultsOut of the 459 articles identified, 24 full-texts were eligible for the study, 20 of which qualified for the data analysis and four were evaluated for coherence. Compared to controls, individuals with diabetes showed lower concentrations of thiamine (pooled estimate SMD [95 % CI]: −0.97 [−1.89, −0.06]), thiamine monophosphate (−1.16 [−1.82, −0.50]), and total thiamine compounds (−1.01 [−1.48, −0.54]). Thiamine diphosphate (−0.72 [−1.54, 0.11] and erythrocyte transketolase activity (−0.42 [−0.90, 0.05]) tended to be lower in persons with diabetes than in controls without reaching statistical significance. Subgroup analysis showed that individuals with diabetes and albuminuria had lower thiamine levels than the controls (−2.68 [−5.34, −0.02]). ConclusionsDiabetes is associated with lower levels of various thiamine markers, suggesting that individuals with diabetes may have higher thiamine requirements than those without diabetes, but well-designed studies are required to confirm these findings.