Although emerging clinical evidence supports that magnesium deficiency is a risk factor for the development of type 2 diabetes, there are sparse studies concerning the dynamic change of serum magnesium with the risk of diabetes and its early stages. In this nested case-control study, we performed a 75-g oral glucose tolerance test or a standardized steamed bread meal test in 178 subjects with incident glucose metabolism impairment (33 with type 2 diabetes and 145 with prediabetes) and 178 matched controls at baseline and at 3-year follow-up and determined the associations between baseline serum magnesium levels as well as changes in serum magnesium levels at follow-up and odds of prediabetes and diabetes. After adjusting for potential confounders, the odds ratios of risk for prediabetes and type 2 diabetes in the highest quartile of serum magnesium levels were 0.22 (95 % confidence intervals [CI] 0.10-0.49; p for trend <0.001) and 0.02 (95 % CI 0.00-0.29; p for trend = 0.009), respectively, as compared with the lowest quartile. In addition, a significant decline in the serum magnesium level was detected in type 2 diabetes cases (p = 0.015) at 3 years as compared with at baseline. These results suggest that a low magnesium level is an independent risk factor for prediabetes and type 2 diabetes, and that the reduction of serum magnesium is associated with type 2 diabetes in a southern Chinese population.
Keywords: Diabetes; Dynamic change; Magnesium; Prediabetes.