In the present study, we investigated the association of dietary Mg intake with HGI in a large population-based health and nutrition survey using the NHANES data. We found a negative correlation between dietary Mg intake and HGI. After adjusting for various potential confounding factors, including age, sex, race, education level, family income, BMI, smoking, alcohol drinking, Hb, CR, hypertension, hypercholesterolemia, diabetes, cardiovascular disease, and daily energy intake, the correlation was still significant. We found a linear dose-response relationship when dietary Mg intake exceeded 412 mg/d, which implied that there is a threshold effect on the inverse association of dietary Mg intake with HGI. According to the 2020–2025 American dietary guidelines17, the adequate intakes of Mg were 360 mg/d (aged 18 year), 310 mg/d (aged 19–30 years) and 320 mg/d (aged 31 + years) for females; 410 mg/d (aged 18 year), 400 mg/d (aged 19–30 years) and 420 mg/d (aged 31 + years) for males. Of the 4249 participants, only 869 (20.45%) met the criteria. In stratified analysis by diabetes, we found that dietary Mg intake was negatively linear correlated with HGI in both diabetic and non-diabetic individuals. However, for non-diabetic individuals, we found an L-shaped dose-response relationship, which means that when the plateau is reached, even if the dietary intake of Mg is increased, the protective effects on HGI will not be increased.
The prevalence of hypomagnesemia ranges between 14 and 48% in diabetic patients, and these patients with hypomagnesemia show a more rapid disease progression and have an increased risk for diabetes complications 18,19. Epidemiological studies have shown that dietary Mg intake is inversely associated with the risk of diabetes in a dose-response manner 20. Moreover, clinical studies have demonstrated that oral Mg supplementation reduces HbA1c among subjects with diabetes and diabetic foot ulcers 21,22. Serum Mg levels are tightly controlled between 0.7 and 1.05 mmol/L in healthy humans. Mg deficiency in diabetic patients is mainly caused by a low intake and impaired renal function. In our study, to exclude the potential effect of renal impairment on Mg wasting, serum CR is adjusted and a significant negative correlation still exists.
Although current strategies in the management of glycemia mainly rely on HbA1c. However, a discrepancy between HbA1c and other assessments of glycemia is well reported 10,23−25. Thus, Hempe et al. proposed the HGI as the difference between the measured HbA1c and the predicted HbA1c derived from blood glucose estimations 26. Trial conducted by AleCardio has shown a positive relationship between HGI and mortality after additional adjustment for HbA1c 27. Similarly, after adjusting for observed HbA1c, our study also showed a significant relationship between dietary Mg intake and HGI, which indicated that HGI is independent of HbA1c in these scenarios. On one hand, the mechanisms of Mg on glucose metabolism could explain the negative association between dietary Mg intake and HGI. For example, Mg-ATP complex is key regulators of glucokinase, PI3K/Akt kinase, and ATP-sensitive K+ channels; thus, Mg deficiency results in decreased Mg-ATP complex level, which induces hyperinsulinemia and decreased insulin sensitivity 28. Mg may also affect glucose metabolism by the modulation of inflammatory responses, such as interleukin-6, interleukin-10, and tumor necrosis factor-alpha in diabetic, prediabetic, and non-diabetic subjects 29–31. On the other hand, factors that impact red blood cell lifespan and the glucose gradient across the red cell membrane are known to affect HbA1c independently of glycemia. An animal study showed that new erythrocytes rapidly develop biochemical and morphologic abnormalities with aging in a magnesium-deficient plasma environment 31. Paolisso et a., find that the impairment of insulin-induced erythrocyte magnesium accumulation is correlated to impaired insulin-mediated glucose disposal in type 2 diabetic patients 32.
Our study has several strengths. Firstly, this study included nationally representative individuals, allowing us to generalize our findings to a broader population and avoiding selection bias. Secondly, the dose-response relationship between dietary Mg intake and HGI was explored, which allows nutritionists to provide dietary recommendations. Thirdly, we adjusted many confounding factors and made a sensitivity analysis. Nevertheless, several limitations should be addressed. Firstly, this was a cross-sectional observational study, thus the directional causality cannot be ascertained. Secondly, the potential effects from unmeasured or residual confounding cannot be ruled out. Thirdly, the data obtained through 24-h dietary recall might have recall bias.
In conclusion, in the representative sample of the US adult population, we found that a higher intake of dietary Mg was associated with a decreased HGI, independent of traditional risk factors, even HbA1c. These findings required confirmation in prospective studies before dietary recommendation.