Abstract
Background:
Epidemiological studies suggest that vitamin D status may have an influence on lipid profile.
Patients and methods:
We studied consecutive type 2 diabetic patients with vitamin D deficiency (serum 25-hydroxyvitamin D (25(OH)D) lower than 20 ng/ml). They were treated with 16,000 IU of calcifediol orally once a week for a minimum of 8 weeks.
Results:
A total of 28 patients were treated for a mean time of 84.1 days (range 56-120 days). All patients achieved serum levels of 25(OH)D higher than 20 ng/ml. There was significant reduction in total cholesterol (172.1 ± 32.4 versus 164.4 ± 27.3 mg/dl, p = 0.04). There were nonsignificant reductions in low-density lipoprotein (LDL) cholesterol, non-high-density lipoprotein (non-HDL) cholesterol and triglycerides. There was no change in HDL cholesterol.
Conclusions:
Correction of vitamin D deficiency in type 2 diabetic patients decreases total cholesterol. Our results do not rule out reductions in LDL cholesterol, non-HDL cholesterol and triglycerides.
Introduction
There is great interest in extra-skeletal effects of vitamin D [Hossein-nezhad and Holick, 2013; Wolden-Kirk et al. 2012]. Epidemiological studies have shown a link between type 2 diabetes mellitus and vitamin D deficiency [Mattila et al. 2007]. Other studies have also shown that vitamin D deficiency is associated with increased risk of cardiovascular disease [Wang et al. 2008]. Levels of serum cholesterol are a strong predictor of cardiovascular risk. Observational studies have demonstrated that high levels of vitamin D are associated with a favorable lipid profile, whereas low levels of vitamin D are associated with an atherogenic lipid profile [Wang et al. 2012]. However, intervention studies show contradictory results [Jorde and Grimnes, 2011]. Therefore, further studies are needed to confirm or refute these possible effects. In the present study, we investigated the effect on lipid profile of vitamin D supplementation in type 2 diabetic patients with vitamin D deficiency.
Patients and methods
We consecutively studied all outpatients with type 2 diabetes mellitus with drug treatment attending two internal medicine offices of a first-level hospital in Cáceres (Extremadura, Southern Spain). We excluded patients treated with insulin and/or vitamin D. Serum levels of 25-hydroxyvitamin D (25(OH)D) were determined and patients with a level lower than 20 ng/ml were included in the study. Patients were treated with 16,000 IU of calcifediol (25-hydroxycholecalciferol, Hidroferol® 266 µg, Faes Farma) orally once a week for a minimum of 8 weeks. No changes in lipid-lowering or antidiabetic treatment were allowed during the study period. After the treatment period with calcifediol, we determined the same parameters using the same analytical methods.
The Charlson index was used to assess comorbidity [Charlson et al. 1987]. Glomerular filtration rate was calculated using the MDRD-4 IDMS formula. Serum levels of 25(OH)D and fasting insulin were determined by chemiluminescence. High-sensitivity C-reactive protein (hs CRP) was determined by turbidimetric immunoassay. Total cholesterol, high-density lipoprotein (HDL) cholesterol and triglycerides were determined by automated conventional enzymatic methods. Low-density lipoprotein (LDL) cholesterol was calculated using the Friedewald’s formula [Friedewald et al. 1972]. Non-HDL cholesterol was calculated by the subtraction: total cholesterol – HDL cholesterol. Statistical analysis was performed using Student’s t test for paired samples. Statistical significance was defined as a p value lower than 0.05.
Results
A total of 41 type 2 diabetic patients without insulin therapy were screened between January and June 2013: 28 patients had a serum level of 25(OH)D lower than 20 ng/ml and were included in the study; 9 (32.1%) had a serum level of 25(OH)D lower than 10 ng/ml. Characteristics of the patients in the series are described in Table 1. Mean treatment time with calcifediol was 84.1 ± 27.1 days (range 56–120 days). After treatment, all patients achieved a serum level of 25(OH)D higher than 20 ng/ml and 25 patients (89.3%) higher than 30 ng/ml. No patient developed hypercalcemia (serum calcium higher than 10.2 mg/dl) after treatment. Table 2 shows the parameters studied before and after treatment.
Characteristics of patients.
BMI, body mass index; GF, glomerular filtrate (MDRD-4 IDMS); DPP, dipeptidyl peptidase; GLP, glucagon-like peptide. Values in brackets are percentages.
Changes of pre-treatment and post-treatment parameters.
CI, confidence interval; LDL, low-density lipoprotein; HDL, high-density lipoprotein; hs CRP, high-sensitivity C-reactive protein.
Discussion
The serum level of 25(OH)D is the best method to determine vitamin D status [Hossein-nezhad and Holick, 2013; Wolden-Kirk et al. 2012]. A consensus has defined vitamin D deficiency as a serum level of 25(OH)D lower than 20 ng/ml [Holick et al. 2011], although there is controversy. We choose this level of 25(OH)D to select our patients. Vitamin D can be administered using different schemes to achieve an adequate serum level of 25(OH)D [Hossein-nezhad and Holick, 2013]. In our study the scheme was 16,000 IU calcifediol orally once a week and we achieved a significant elevation of serum levels of 25(OH)D, higher than 20 ng/ml in all cases, in spite of about a third of our patients had a severe deficiency (serum level of 25(OH)D lower than 10 ng/ml). It is remarkable that no case of hypercalcemia was documented after treatment. Vitamin D intoxication can cause hypercalcemia, but it is rare and a serum level of 25(OH)D up to 100 ng/ml is considered reasonably safe [Hossein-nezhad and Holick, 2013].
The results of the available studies about the effects of the administration of vitamin D on lipid profile are not definitive. Drawing conclusions is complex due to the variety of designs, if only patients with vitamin D deficiency or insufficiency were included, the serum level of 25(OH)D to define deficiency or insufficiency, schemes of vitamin D administration and achieved increases in serum levels of 25(OH)D. In a study of 36 type 2 diabetic patients with normal serum levels of 25(OH)D, administration for 6 months of 40,000 U weekly of cholecalciferol brought about no modification of the studied lipid parameters (total cholesterol, LDL cholesterol, HDL cholesterol and triglycerides) [Jorde and Figenschau, 2009]. In another study of 92 type 2 diabetic patients with a mean serum level of 25(OH)D of 13 ng/ml, administration of 2,000 IU daily of vitamin D3 for 18 months was associated with a significant decrease in levels of total cholesterol and LDL cholesterol, without changes in levels of triglycerides and HDL cholesterol and only a mean increase of 8 ng/ml in serum level of 25(OH)D was achieved [Al-Daghri et al. 2012]. One study of 151 adults with vitamin D deficiency and high cardiovascular risk showed that 50,000 U weekly of cholecalciferol for 8 weeks entailed no effect in lipid profile (total cholesterol, LDL cholesterol, HDL cholesterol and triglycerides) [Ponda et al. 2012]. In a group of 24 type 2 diabetic patients with a mean serum level of 25(OH)D of 11.8 ng/ml, administration of 1,000 U daily of cholecalciferol for 12 months had no effect on levels of total cholesterol, LDL cholesterol, HDL cholesterol and triglycerides, but the mean serum level of 25(OH)D after treatment was only 17.6 ng/ml [Breslavsky et al. 2013]. Another study of 100 type 2 diabetic patients (only 24% of the participants were vitamin D deficient) demonstrated that administration of 50,000 U weekly of cholecalciferol for 8 weeks did no change serum levels of total cholesterol, LDL cholesterol, HDL cholesterol or triglycerides [Talaei et al. 2013]. In 87 obese type 2 diabetic patients with vitamin D deficiency, administration of 6,000 U daily of cholecalciferol for 3 months and 3,000 U daily for other 3 months showed no effect on lipid parameters [Sadiya et al. 2015].
The only meta-analysis conducted to elucidate the effects of vitamin D supplementation on lipid profile, with 12 intervention studies and 1,346 patients, showed a statistically significant increase in the levels of LDL cholesterol, a trend toward increasing levels of total cholesterol and a nonsignificant reduction in levels of HDL cholesterol and triglycerides [Wang et al. 2012]. However, this meta-analysis includes heterogeneous studies regarding basal and post-treatment serum levels of 25(OH)D, vitamin D dose, treatment duration and characteristics of patients [Wang et al. 2012].
Our study found a statistically significant reduction in levels of total cholesterol and a statistically nonsignificant trend to decrease levels of LDL cholesterol, non-HDL cholesterol and triglycerides. Levels of non-HDL cholesterol represent an estimate of the total number of atherogenic particles in plasma, correlate well with apolipoprotein B levels and can provide a better estimate of cardiovascular risk in diabetic patients [Robinson et al. 2009]. We found no change in HDL cholesterol levels. Some authors suggest that possible extra-skeletal benefits of vitamin D would require levels of 25(OH)D higher than 28-32 ng/ml [Hossein-nezhad and Holick, 2013]. Almost 90% of our patients reached these levels.
It has been suggested that vitamin D could have an anti-inflammatory effect [Guillot et al. 2010]. We have found no effect on hs CRP, a marker of subclinical systemic inflammation. Similarly, other studies have shown no effect on hs CRP with vitamin D treatment [Ponda et al. 2012; Breslavsky et al. 2013; Sadiya et al. 2015].
We are not able to rule out that with a larger sample or with a longer treatment we would have obtained other statistically significant results. Our study is small and the findings cannot be considered definitive. The design of our study did not include a control group: we prefer a design of paired samples due to the investigated subject. Finally, it is noteworthy that there is a selection bias in our study because patients attending in internal medicine typically have higher age and comorbidity than type 2 diabetic patients from the general population. Additional studies with larger samples are needed to confirm or refute our results.
Footnotes
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.