Objectives This retrospective cross-sectional study aimed to analyze the effects of minerals and vitamins on insulin resistance in a multivariate manner. Methods The medical records of 301 patients who presented to our clinic between January 2013 and 2023 were retrospectively reviewed. Patients were divided into two groups based on their HOMA-IR levels: non-insulin resistant (n = 201) and insulin resistant (n = 100). Age, gender, Body Mass Index (BMI), fasting blood sugar, insulin, triglyceride, ALT, AST, uric acid and ferritin, HbA1c, homocysteine, urine iodine, free T3, T4, total cholesterol, magnesium, vitamin D, zinc, B12, folate, and TSH levels were recorded. Results BMI, fasting blood sugar, insulin, triglyceride, aspartate transaminase (AST), alanine transaminase (ALT) uric acid and ferritin levels were significantly higher in the insulin resistance group (p < 0.05). High-density lipoprotein (HDL) level was significantly lower in the insulin resistance group (p < 0.05). Correlations between insulin resistance and BMI (r = 0.412; p < 0.01), fasting blood sugar (r = 0.374; p < 0.01), insulin (r = 0.799; p < 0.01), HbA1c (r = 0.276; p < 0.01), HDL (r = − 0.247; p < 0.01), triglyceride (r = 0.236; p < 0.01), ALT (r = 0.289; p < 0.01), AST (r = 0.269; p < 0.01), uric acid (r = 0.316; p < 0.01) and ferritin (r = 0.186; p < 0.01) were significant. Ferritin had a significant effect on insulin resistance at univariate (B = 0.005; p < 0.01) and multivariate (B = 0.005; p < 0.01) levels. Effects of Mg, Vitamin D, zinc, B12, folate and TSH were statistically insignificant (p > 0.05). ROC analysis results showed that the area under the curve (AUC) for ferritin was 0.614, indicating modest discriminatory ability. For ferritin 55.75 cutoff value, sensitivity was 61.0% and specificity was 61.7%. Conclusions Although significant associations between vitamins and minerals and IR were reported in univariate analysis, we found a significant association only between ferritin and IR in multivariate analysis. Therefore, it is beneficial to evaluate studies evaluating IR with vitamins and minerals by including all cofounders.

Multivariate analysis of the effects of minerals and trace elements on insulin resistance

期刊论文
出版模式：

开放获取
发表日期：

2026年4月29日

查看全文

Arzu Güntürk,
Müzeyyen Eryılmaz,
Kadir Yılmaz

摘要

Objectives

This retrospective cross-sectional study aimed to analyze the effects of minerals and vitamins on insulin resistance in a multivariate manner.

Methods

The medical records of 301 patients who presented to our clinic between January 2013 and 2023 were retrospectively reviewed. Patients were divided into two groups based on their HOMA-IR levels: non-insulin resistant (n = 201) and insulin resistant (n = 100). Age, gender, Body Mass Index (BMI), fasting blood sugar, insulin, triglyceride, ALT, AST, uric acid and ferritin, HbA1c, homocysteine, urine iodine, free T3, T4, total cholesterol, magnesium, vitamin D, zinc, B12, folate, and TSH levels were recorded.

Results

BMI, fasting blood sugar, insulin, triglyceride, aspartate transaminase (AST), alanine transaminase (ALT) uric acid and ferritin levels were significantly higher in the insulin resistance group (p < 0.05). High-density lipoprotein (HDL) level was significantly lower in the insulin resistance group (p < 0.05). Correlations between insulin resistance and BMI (r = 0.412; p < 0.01), fasting blood sugar (r = 0.374; p < 0.01), insulin (r = 0.799; p < 0.01), HbA1c (r = 0.276; p < 0.01), HDL (r = − 0.247; p < 0.01), triglyceride (r = 0.236; p < 0.01), ALT (r = 0.289; p < 0.01), AST (r = 0.269; p < 0.01), uric acid (r = 0.316; p < 0.01) and ferritin (r = 0.186; p < 0.01) were significant. Ferritin had a significant effect on insulin resistance at univariate (B = 0.005; p < 0.01) and multivariate (B = 0.005; p < 0.01) levels. Effects of Mg, Vitamin D, zinc, B12, folate and TSH were statistically insignificant (p > 0.05). ROC analysis results showed that the area under the curve (AUC) for ferritin was 0.614, indicating modest discriminatory ability. For ferritin 55.75 cutoff value, sensitivity was 61.0% and specificity was 61.7%.

Conclusions

Although significant associations between vitamins and minerals and IR were reported in univariate analysis, we found a significant association only between ferritin and IR in multivariate analysis. Therefore, it is beneficial to evaluate studies evaluating IR with vitamins and minerals by including all cofounders.