European Journal of Cardiovascular Medicine

Diabetes mellitus is one of the most common non communicable disease in modern world. It is estimated to have 422 million people suffering from diabetes, majority from low and middle income countries.¹ According to National Family Health Survey (NFHS) – 5, 15.6 % of men and 13.5 % of women are living with high blood sugar level (>140 mg/dl) among adults age 15 year and above in India.²

Vitamin D is a fat soluble vitamin playing vital role in calcium homeostasis. In last few decades, its role has been studied and described in conditions other than calcium homeostasis. Sometimes, vitamin D is stated as a hormone due to its widespread physiological role. Role of vitamin D is described in diabetes mellitus in multiple studies owing to its facilitation/inhibition of transcription factor and immune-modulator property. Abnormal lipid profile parameters are well described in patients of diabetes mellitus.

Aim: To evaluate relation of vitamin D level with glycemic control and lipid profile in patients of type 2 diabetes mellitus.

This was an analytical observational case study conducted for 18 months on patients attending OPD and IPD of Department of Medicine at Government institute of medical sciences, Gautam Buddha nagar (UP). All patients of age 30-70 year of age presenting to OPD and IPD of Department of Medicine with type 2 diabetes mellitus formed the study group. Patients with established chronic kidney disease, chronic liver disease, glucocorticoid therapy, anti-epileptic drugs, taking vitamin D, on diuretics, not willing, Patients with acute complications like diabetic ketoacidosis or hyperglycaemic hyperosmolar state, with severe anaemia, requiring repeated blood transfusions, diagnosed with carcinoma. All routine lab parameters were done for patients included in the study. That included haemoglobin level, fasting and post-prandial blood glucose, HbA1c level, kidney function test (serum urea, serum creatinine, serum uric acid), urine albumin to creatinine ratio (UACR), fasting lipid profile, fasting insulin and 25-hydroxy vitamin D level. Patients were assessed for diabetic neuropathy by clinical examination. Patients were assessed for diabetic retinopathy by fundus examination. BMI was calculated using weight and height. Statistical analysis Continuous variables were measured by using mean and standard deviation. For categorical variables, frequency was measured. Continuous variables were compared by Pearson correlation test with measurement of correlation coefficient. For comparison between subgroups, Fisher’s exact test was used. For comparison of mean values among subgroups, ANOVA test was used. Regression analysis was done for comparison of vitamin D with selected variables. Analysis was done by STATA version 12 software. Ethical consideration This study was initiated after grant of approval from GIMS scientific research committee (GSRC) and GIMS institutional ethics committee (IEC).

119 patients of type 2 diabetes mellitus were included and evaluated with socio-economic data, biochemical profile and clinical examination. Mean age of patients was 49.5 ± 11 year. Out of these, 67 patients were male and 52 patients are female. Mean duration of diabetes was 6.5 ± 5.18 year.

In anthropometric data, patients had mean weight of 65.5 ± 10.2 kilogram, mean height of 155.7 ± 8.67 centimetres and mean BMI of 27.1 ± 4.21 kg/m².

Table 1: Distribution of patients as per BMI

In biochemical profile, patients had mean HbA1c of 9 ± 2.3 %. Mean fasting and post-prandial blood glucose were 183 ± 39.17 mg/dl and 247.6 ± 67.66 mg/dl, respectively. Mean serum 25(OH) D level was 20.1 ± 16.22 ng/ml.

Table 2: Biochemical profile of patients

Correlation of serum 25(OH) D level was done with selected variables. A significant positive correlation was seen between vitamin D level and age (r = 0.1824, p = 0.0471). HbA1c and fasting plasma glucose were found to be negatively correlated with vitamin D level (r = -0.0994, p = 0.2824) and (r = -0.1401, p = 0.1285), respectively, however, not significant. There was negative but non-significant relation found between vitamin D level and total cholesterol (r = -0.0134, p = 0.8987), serum triglyceride level (r = -0.0310, p = 0.7690) and serum LDL (r = -0.0149, p = 0.8873). A positive correlation was present between vitamin D level and serum HDL (r = 0.0296, p = 0.7782).

Table 3: Correlation of vitamin D level with different parameters, r (Pearson correlation coefficient) and p-value are mentioned.

For comparison, patients were divided into different subgroups on the basis of different variables. Patients with serum 25(0H) D <=20 ng/ml were considered as vitamin D deficient while serum 25 (OH) D > 20 ng/ml were considered as vitamin D sufficient.

Figure 1 shows in Pdf.

Figure 1: Scatter diagram that illustrates negative correlation of vitamin D level with HbA1c i.e. poor glycaemic control is noted in vitamin D deficient individuals.

Figure 2 shows in Pdf.

 Figure 2: illustrates distribution of patients as per vitamin D level. 77 patients (64.71 %) were vitamin D deficient while 42 patients (35.29 %) were vitamin D sufficient.

Three groups were formed on the basis of HbA1c. Patients with HbA1c <= 7% were included in well control group. HbA1c of 7-8 % formed poor control group. HbA1c > 8 % were included in uncontrolled group.

Figure 3 shows in Pdf.

Figure 3: Categories of glycaemic control as per HbA1c values. 26 patients (21.85 %) formed well controlled group. 18 patients (15.13 %) formed poor control group. 75 patients (63.03 %) formed uncontrolled group.

Fisher’s exact test was used for comparison between vitamin D deficient and vitamin D sufficient groups for glycaemic control and other variables. Out of 77 patients of vitamin D deficiency, 54 patients (70.12 %)  had uncontrolled glycaemic status, while 21 patients (50 %) out of 42 patients with vitamin D sufficiency had uncontrolled glycaemic status indicating that vitamin D deficiency has significant association with poor glycaemic control. (Fisher’s exact = 0.024)

Table 4: illustrates association between vitamin D and glycaemic control. Vitamin D deficient group had significant association with poor glycaemic control. (Fisher’s exact = 0.024)

For ANOVA test, patients were categorised in three categories based on serum 25(0H) D values- vitamin D deficient (<=10 ng/ml), vitamin D insufficient (11-20 ng/ml) and vitamin D sufficient (>20 ng/ml). Mean of selected variables were compared among these three groups by ANOVA test.

Mean HbA1C in vitamin D deficient, insufficient and sufficient groups were 9.8 ± 2.6%, 8.6 ± 1.6% and 8.7 ± 2.4% respectively. Vitamin D deficiency was found to be associated with poor glycaemic control, but association was near significant. (P-value = 0.0535) Mean serum triglyceride level in vitamin D deficient, insufficient and sufficient groups were 206 ± 147.3 mg/dl, 152.7 ± 64 mg/dl and 147.8 ± 72.6 mg/dl respectively. Vitamin D deficiency was found to be associated with elevated serum triglyceride level, but association was near significant.  (P-value = 0.0510)

Table 5: illustrates comparison of mean values of different variables among three subgroups based on vitamin D level by ANOVA test. Significant association was found between vitamin D deficiency and young age. (P-value = 0.0100)   Near significant association was found between vitamin D deficiency and poor glycaemic control and hypertriglyceridemia. (P-value = 0.0535 and 0.0510, respectively).

The present study aims at analysis of association of vitamin D level with glycaemic control and lipid profile in type 2 diabetes mellitus. Discussion about the study is described in following headings –

1. Vitamin D and glycaemic control
2. Vitamin D and lipid profile

Vitamin D and glycaemic control

In present study, vitamin D showed negative correlation with HbA1c and significant association was noted between vitamin D deficiency and poor glycaemic control. Vitamin D and HbA1c also showed negative linear regression on regression analysis.

Li Z et al³ in 2021 performed a study in which they used haemoglobin glycation variation index for glycaemic control. Vitamin D deficiency was found to be significantly associated with poor glycaemic control, same as in present study.

Vitamin D and Lipid profile

In present study, vitamin D showed negative correlation with total cholesterol, serum triglyceride and serum LDL levels. It also showed positive correlation with serum HDL level. Near significant association was present between vitamin D deficiency and increased serum triglyceride level.

Jafari T et al⁴ in 2016 conducted a meta analysis of randomized controlled trials enrolling 17 studies. They concluded that vitamin D supplementation had significant lowering effect on serum total cholesterol and serum LDL level but negligible effect on serum triglyceride and serum HDL level.

Shenoy V et al⁵ in 2014 performed a cross sectional study on evaluation of vitamin D level and lipid profile in euglycemic individuals. They concluded that a significant inverse correlation was present between vitamin D and HbA1c. A significant inverse correlation was found between vitamin D and serum total cholesterol and between vitamin D and total cholesterol/HDL ratio. A significant direct correlation was present between vitamin D and serum HDL.

• Vitamin D deficiency in type 2 diabetes mellitus is associated with poor glycaemic control; however causality could not be established based on present study.

• Vitamin D deficiency may be related to dyslipidaemia in type 2 diabetes mellitus. Vitamin D deficiency is associated with hypertriglyceridemia.

1. World health organization. Diabetes-fact sheet [Internet]. 2021. Available from: https://www.who.int/news-room/fact-sheets/detail/diabetes.
2. International institute for population sciences (IIPS). National family health survey (NFHS) - 5 [Internet]. 2021. Available from: https://main.mohfw.gov.in/basicpage-14
3. Li Z, Wang F, Jia Y, Guo F, Chen S. The relationship between hemoglobin glycation variation index and Vitamin D in Type 2 diabetes mellitus. Diabetes, Metabolic Syndrome and Obesity: Targets and Therapy. 2021;14:1937.
4. Jafari T, Fallah AA, Barani A. Effects of vitamin D on serum lipid profile in patients with type 2 diabetes: a meta-analysis of randomized controlled trials. Clinical nutrition. 2016 Dec 1;35(6):1259-68.
5. Shenoy V, Datta P, Prabhu K, Singh K. Association between vitamin D, fasting blood glucose, HbA1c and fasting lipid profile in euglycemic individuals. J Res Diabetes. 2014 May 14;2014:a1-8.