Objective: 25-hydroxyvitamin D (25[OH]D) has received attention being involved in exerting effects in diabetes-associated mild cognitive impairment (MCI). We aimed to examine the relationship between 25(OH)D and the risk of incident MCI, and the exact mechanism to regulate MCI in diabetes.

Methods: This cross-sectional study was conducted from 103 patients with type 2 diabetes (aged 40-75years), focusing on cognitive function. After adjustment for age, gender and education level, the partial correlation analysis were used to estimate the relationship of 25(OH)D level with multiple neuropsychological tests and metabolism profile, including glucose and lipid metabolism, blood pressure, coagulation and thyroid function. The mediating effects of metabolism profile on the relationship 25(OH)D with cognitive function were were performed by the simple mediating effect model of Model 4.

Results: These participants were finally divided into two groups basing on MoCA: MCI group (n=31) and healthy group (n=72). MCI group showed a lower 25(OH)D level and unfavorable metabolic profiles characterized by higher levels of HbA1c and LPa, but lower levels of INR and FT3 (all p<0.05). The partial correlation analysis found there were positive correlation between 25(OH)D and MoCA, VFT, CDT, LMT, AVLT-IR and AVLT-DR (respectively, r=0.023, p=0.0; 02; r=0.272, p=0.006; r=0.280, p=0.005; r=0.241, p=0.016; r=0.319, p=0.001; r=0.242 p=0.015), primarily reflecting memory, execution and attention. Decreased 25(OH)D level was also correlated with lower FT3 and higher HbAc1 and LPa (respectively, r=0.361, p<0.001; r=-0.223, p=0.026; r=-0.260, p=0.009). In addition, the association between 25(OH)D and MoCA was found to be mediated by FT3 and LPa, with the mediation effect calculated to be 25.57% and 16.69%.

Conclusion: Decreased 25(OH)D increased the risk of incident MCI, especially memory, execution and attention dysfunction in diabetes, and their relationship were may mediated by FT3 and LPa.


W.Zhu: None. S.Wang: None. T.Niu: None. K.Liu: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.