Abstract
Objective
Epidemiologic studies suggest that vitamin D status plays a role in glycaemic control in patients with type 2 diabetes. However, intervention studies yielded inconsistent results. The aim of this study is to systematically review the effect of vitamin D supplementation on glycaemic control in patients with type 2 diabetes.
Methods
Systematic review and meta-analysis. We searched Medline, Embase and the Cochrane Library for RCTs examining the effect of vitamin D supplementation on glycaemic control in patients with type 2 diabetes. A random-effects model meta-analysis was performed to obtain a summarized outcome of vitamin D supplementation on HbA1c, fasting glucose and homeostasis model assessment – insulin resistance (HOMA-IR).
Results
Twenty-three RCTs were included in this systematic review representing a total of 1797 patients with type 2 diabetes. Mean (± s.d.) change in serum 25-hydroxyvitamin D varied from 1.8 ± 10.2 nmol/L to 80.1 ± 54.0 nmol/L. Nineteen studies included HbA1c as outcome variable. Combining these studies no significant effect in change of HbA1c was seen after vitamin D intervention compared with placebo. A significant effect of vitamin D supplementation was seen on fasting glucose in a subgroup of studies (n = 4) with a mean baseline HbA1c ≥ 8% (64 mmol/mol) (standardized difference in means: 0.36; 95% CI: 0.12–0.61, P = 0.003).
Conclusions
Current evidence of RCTs does not support short-term vitamin D supplementation in a heterogeneous population with type 2 diabetes. However, in patients with poorly controlled diabetes, a favourable effect of vitamin D is seen on fasting glucose.
Introduction
Vitamin D is a key factor for the maintenance of calcium and bone homeostasis. Over the past decade, vitamin D has attracted substantial interest due to its extra-skeletal roles in various disease conditions, including diabetes mellitus (1). This interest has arisen due to the identification that most cells, including the pancreatic beta-cells, contain the vitamin D receptor (VDR). Most of these cells also have the capability to produce the biologically active form of vitamin D: 1,25-dihydroxyvitamin D for paracrine functions (1, 2, 3). Furthermore, vitamin D is known to have immuno-modulatory and anti-inflammatory effects, which could improve peripheral insulin resistance by altering low-grade chronic inflammation that has been implicated in insulin resistance in type 2 diabetes mellitus (3, 4, 5).
Observational studies have demonstrated a link between vitamin D deficiency and the onset of and progression of type 2 diabetes (6, 7, 8, 9). Furthermore, low vitamin D status is associated with future macrovascular events in patients with type 2 diabetes mellitus (10). This association may be the result of the link between vitamin D status and renin-angiotensin system (11), endothelial function (12), blood pressure (13) or chronic inflammation (4).
A recent meta-analysis performed in 2012 by George et al. (14) has demonstrated a weak positive effect of vitamin D supplementation on fasting glucose and insulin resistance in patients with type 2 diabetes mellitus. However, overall, the authors concluded that there was insufficient evidence of a beneficial effect to recommend vitamin D supplementation as a means of improving glycaemic control in patients with type 2 diabetes, impaired fasting glucose or normal glucose tolerance. Inconsistency in these results may be due to the different study populations (normal glucose tolerance, impaired glucose tolerance and type 2 diabetes), small sample sizes and different dosage regimes of vitamin D supplementation. Additionally, a meta-analysis published by Seida et al. in 2014, which included RCTs among adults with normal glucose tolerance, prediabetes and/or type 2 diabetes, demonstrated no effect of vitamin D supplementation on improving glucose homeostasis and preventing diabetes including only RCTs. Definitive conclusion could not be drawn in the context of heterogeneity, short-term follow-up duration and variable risk of bias (15). Due to the ongoing increased interest in the effect of vitamin D on glycaemic control in type 2 diabetes, many more studies have been published since these meta-analyses were performed.
Taken together, it is still unclear whether vitamin D supplementation has a beneficial effect on glycaemic control in patients with type 2 diabetes mellitus. We present an up to date analysis of the effect of vitamin D supplementation on glycaemic indices (HbA1c, insulin resistance and fasting glucose) in patients with type 2 diabetes mellitus.
Methods
Search strategy and selection criteria
A systematic literature search (MEDLINE, Embase and The Cochrane Library) was performed to identify articles from January 1976 to 15 October 2015 that assessed the effect of vitamin D supplementation on glycaemic indices in patients with type 2 diabetes. The search terms included type 2 diabetes mellitus AND (vitamin D OR vitamin D deficiency OR vitamin D2 OR vitamin D3 OR cholecalciferol OR ergocalciferol). References of the retrieved articles were scanned for additional studies. The objective was to systematically review the evidence that vitamin D could improve glycaemic indices (HbA1c, insulin resistance and fasting glucose) in patients with type 2 diabetes. One author (YK-P) performed an initial screening of titles and abstracts. Full-text articles of the selected titles were screened using the inclusion criteria described below. If there was a doubt to whether a particular article should be included, the author discussed the article with the last author (SS) until consensus was reached.
We included randomized controlled trials (RCT) in the following groups: vitamin D supplementation versus placebo, vitamin D supplementation and calcium supplementation versus calcium alone and/or placebo. Additional inclusion criteria were the following: (i) the study population consisted of patients with type 2 diabetes; (ii) supplementation of vitamin D2 (ergocalciferol) or vitamin D3 (cholecalciferol) for intervention; (iii) HbA1c or parameters of glycaemic control (fasting glucose, fasting insulin or homeostatic model assessment – insulin resistance (HOMA-IR)) had to be a primary or secondary outcome; (iv) the authors report data of an original clinical study (i.e. no review, commentary, case reports, or editorial); (v) study performed in adults ≥18 years; (vi) published in English. We excluded studies carried out using 1,25-dihydroxyvitamin D and studies performed in patients other than type 2 diabetes mellitus, or patients on dialysis.
Quality assessment and data extraction
The quality of selected articles was assessed by two reviewers using a checklist from the Dutch Cochrane Collaboration (Fig. 1) (16). The checklist consists of 11 criteria, with each criterion having three answer options: yes (adequate information/approach); no (no adequate information/approach); or little information. Each criteria answered with yes scored one point, we considered a total score ≥9 points as a good quality study.
Quality checklists randomized controlled trials (RCTs).
Citation: European Journal of Endocrinology 176, 1; 10.1530/EJE-16-0391
Data were extracted by one author (YK-P) and controlled by the last author (SS) using a self-composed form including the following items of studies included: country, design, publication year, participants, therapy duration, type and dose of vitamin D supplementation, primary outcome, baseline and change in serum 25-hydroxy vitamin D (25(OH)D) and parameters of glycaemic control (HbA1c, fasting glucose, fasting insulin and homeostasis model of assessment – insulin resistance (HOMA-IR)). For studies lacking a reported standard deviation of change in outcome between baseline and follow-up, we derived standard deviation of change as the mean of the baseline and follow-up standard deviations for each treatment group. This method was used successfully in the meta-analysis from George et al. performed on this subject (14).
Statistical analysis
To obtain a summarized outcome of the effect of vitamin D supplementation on glycaemic control, we compared the mean change between baseline and follow-up of each variable of the intervention and control group. Studies in which the mean change and/or standard deviation was not reported, or could not be derived, were excluded from the meta-analysis. If a study included more than two groups, we used the data of the group in which the highest dose of vitamin D supplementation was given for the meta-analysis compared with placebo. If studies compared both vitamin D and/or calcium supplementation versus placebo, the data of the group with solely vitamin D supplementation were used for the meta-analysis.
The results of the included studies were pooled and meta-analyses were carried out using random-effects models, as some heterogeneity of outcome was expected. To compare the intervention and placebo groups, the results are presented as between-group standardized mean differences with 95% CI. Subgroup analyses were performed for studies with a baseline mean serum 25(OH)D <50 nmol/L and <30 nmol/L, and for studies having a mean baseline HbA1c ≥8% (64 mmol/L) in the intervention group. We assessed statistical heterogeneity between studies with I 2 statistic (with 95% CIs). The I 2 is the proportion of total variation contributed by between-study variation. In general, I 2 values greater than 60–70% indicate the presence of substantial heterogeneity (17). In the presence of heterogeneity between studies, we assessed potential publication bias using formal tests: the funnel plot and Egger test (18). Meta-analyses were performed using comprehensive meta-analysis version 3.0 (http://www.meta-analysis.com). A P value <0.05 was considered to be statistically significant.
Results
Selected articles
The initial systematic search yielded 1489 articles. Of those, 328 were duplicates and 1074 articles were excluded based on abstract and title. The most common reasons for exclusion of these articles were no inclusion of patients with type 2 diabetes or no intervention with vitamin D. Eighty-seven articles were selected for full-text review as shown in Fig. 2. Finally, 23 trials were selected for quality assessment and included in this systematic review.
Flow chart of literature search.
Citation: European Journal of Endocrinology 176, 1; 10.1530/EJE-16-0391
Description of the studies
Twenty-three RCTs representing a total of 1797 patients with type 2 diabetes were included in this systematic review. The quality assessment of the studies resulted in 14 out of 23 studies having a good quality (Appendix 1, see section on Appendix given at the end of this article) (12, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31). The main characteristics and main outcomes of the included studies are given in Table 1. All studies had a randomized controlled trial design, of which 18 studies used a placebo for control (12, 19, 20, 22, 23, 24, 25, 27, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38), three studies compared vitamin D fortified yoghurt versus plain yoghurt (21, 28, 39), one study used oral calcium supplementation for control (26), and one study used vitamin C supplementation for control (40). Apart from two studies, which solely included post-menopausal women (21, 33), all studies included both men and women. Mean age varied from 44 to 67 years (24, 25). Mean HbA1c varied from 6.2% (44 mmol/mol) (20) to 8.7% (71 mmol/mol) (28) in the intervention group. Six studies had a mean baseline HbA1c ≥8% (64 mmol/mol) in the intervention group (27, 28, 32, 33, 36, 40). Different assays were used for measurement of serum 25(OH)D with most studies using an enzyme-immunoassay (12, 20, 21, 23, 24, 26, 27, 29, 30, 31, 32, 33, 34, 35, 36, 37), three studies measure serum 25(OH)D using high-performance liquid chromatography (22, 28, 39), two studies used a radio-immunoassay method (25, 38), one study used a competitive protein-binding assay (19), and one study did not report the method of measurement (40).
Summary of the intervention studies included in this systematic review.
| Reference | Location | Cohort T2DM (n) | Intervention | Control | Duration | Primary outcome | 25(OH)D nmol/L before and after treatment* | Baseline Hba1c (%)* | Main results |
|---|---|---|---|---|---|---|---|---|---|
| (32) | Saudi Arabia | 183, 25(OH)D <75 nmol/L | Vitamin D3 45 000 IU/week | Placebo | 3 months | Metabolic parameters | 25.3 ± 15.8 to 82.8 ± 31.7 | 8.5 ± 1.6 | ↓ diastolic blood pressure= HbA1c, fasting glucose, lipid profile |
| (19) | Israël | 47 | Vitamin D3 1000 IU/day | Placebo | 12 months | Metabolic parameters | 29.5 ± 27.2 to 43.9 ± 28.7 | 7.3 ± 1.1 | = HbA1c, fasting glucose, insulin, HOMA-IR, lipid profile |
| (33) | Brazil | 38 post-menopausal women, 25(OH)D <75 nmol/L | Vitamin D3 6600 IU/week | Placebo | 3 months | Metabolic parameters and muscle strength | 55.5 ± 9.9 to 57.4 ± 10.5 | 8.2 ± 2.1 | ↑ handgrip strength= HbA1c, fasting glucose, insulin, lipid profile |
| (20) | Australia | 50 T2DM duration <1 year | Vitamin D3 10 000 IU/day for 2 weeks followed by 6000 IU/day for 6 months | Placebo | 6 months | Change in C-peptide | 59 (42–75) to 128 (111–146) | 6.2 (6.0–6.6) | = C-peptide = HbA1c, fasting glucose, insulin, HOMA-IR |
| (34) | Iran | 51, non- insulin | Vitamin D3 400 IU/day | Placebo | 14 weeks | HbA1c, TNF-α, leptin | 21.5 ± 23.7 to 46.4 ± 35.1 | 6.8 ± 0.4 | = HbA1c ↑ serum leptin ↓ TNF-α |
| (35) | Iran | 42, non-insulin | Vitamin D3 300 000 IU single dose | Placebo | 3 months | Glycaemic parameters | 117.3 ± 86.7 to 173.2 ± nr | 6.5 ± 0.9 | = HbA1c ↑ HOMA-IR, fasting glucose |
| (21) | Iran | 59 post-menopausal women, non-insulin | Vitamin D3 fortified yoghurt (2000 IU/day) | Plain yoghurt | 12 weeks | Metabolic parameters | 62.2 ± 24.6 to 86.8 ± 26.7 | 7.2 ± 1.3 | = HbA1c ↓ fasting glucose, insulin, HOMA-IR, lipid profile |
| (22) | Switzerland | 55 T2DM duration >10 years | Vitamin D3 300 000 IU single dose i.m. | Placebo | 6 months | Change in HbA1c | 36.0 ± 18.1 to 84.9 ± 16.0 | 7.0 ± 1.1 | ↓ HOMA-IR = fasting insulin and glucose Significantly less increase in HbA1c in the intervention group |
| (36) | Norwegian | 36, insulin treatment | Vitamin D3 40 000 IU/week | Placebo | 6 months | Glycaemic parameters | 60.0 ± 14.0 to 118.3 ± nr | 8.0 ± 1.3 | = HbA1c, HOMA-IR, lipid levels |
| (23) | Denmark | 16, 25(OH)D <50 nmol/L | Vitamin D3 11 200 IU/day for 2 weeks followed by 5600 IU/day for 10 weeks | Placebo | 12 weeks | Glycaemic parameters† | 31.0 ± 13.6 to 104.9 ± 53.7 | nr | = insulin sensitivity, HbA1c, lipid profile, 24 h blood pressure |
| (24) | Netherlands | 261, non-insulin | Vitamin D3 50 000 IU/month | Placebo | 6 months | HbA1c | 60.6 ± 23.3 to 101.4 ± 27.6 | 6.8 ± 0.5 | = HbA1c, HOMA-IR, lipid levels ↓ HbA1c (subgroup: 25(OH)D ≤30 nmol/L) |
| (37) | Iran | 60 | Vitamin D3 50 000 IU/week | Placebo | 12 weeks | Glycaemic parameters | 83.9 ± 52.0 to 164.0 ± 57.0 | 7.7 ± 0.4 | ↓ HbA1c in male subjects |
| (39) | Iran | 90 | 1. Vitamin D3 fortified yoghurt (1000 IU/day) 2. Vitamin D3 + Ca fortified yoghurt (1000 IU/500 mg/day) | Plain yoghurt | 12 weeks | Metabolic parameters | 44.4 ± 28.7 to 77.7 ± 28.6 | 7.4 ± 1.8 | ↓ HbA1c, HOMA-IR, fasting glucose and insulin, BMI = lipid levels |
| (25) | India | 28, non-insulin | Vitamin D3 300 000 IU single dose i.m. | Placebo | 4 weeks | Glycaemic parameters; OGTT | 37.2 ± 16.9 to 103.8 ± 30.5 | 7.6 ± 0.6 | = HbA1c, HOMA-IR, fasting glucose, insulin |
| (26) | Korea | 158, non-insulin, 5(OH)D <50 nmol/L | Vitamin D3 1000 IU/day + Ca 100 mg bid | Ca 100 mg bid | 24 weeks | Glycaemic parameters | 27.0 ± 12.7 to 75.4 ± 27.0 | 7.3 ± 0.6 | = HbA1c, HOMA-IR |
| (27) | United Arab Emirates | 8,7 25(OH)D <50 nmol/L, BMI > 30 | Vitamin D3 6000 IU/day for 3 months followed by 3000 IU/day for 3 months | Placebo | 6 months | Metabolic parameters | 28.5 ± 9.5 to 62.3 ± 20.8 | 8.3 ± 1.3 | = HbA1c, fasting glucose, lipid levelsSubgroup 25(OH)D <30 nmol/L: no difference |
| (28) | Iran | 100, non-insulin | Vitamin D3 fortified doogh (1000 IU/day + 340 mg Ca/day) | Plain doogh (340 mg Ca) | 12 weeks | Metabolic parameters, endothelial biomarkers | 38.5 ± 20.2 to 72.0 ± 23.5 | 8.7 ± 1.8 | ↓ fasting glucose, insulin, lipid profile, endothelial biomarkers= HbA1c |
| (40) | US | 37 | Vitamin D3 1200 IU/day | Vitamin C 500 mg/day | 12 weeks | HbA1c | Nr | 8.6 ± 1.2 | = HbA1c (total group) ↓ HbA1c (subgroup: HbA1c ≥9.0%) |
| (38) | Germany | 86, non-insulin | Vigantol oil (vitamin D3 1904 IU/day) | Placebo | 6 months/12 months | Glycaemic parameters | 30.2 ± nr to 87.4 ± nr | nr | = HbA1c, HOMA-IR, fasting insulin and glucose |
| (12) | UK | 34, 25(OH)D <50 nmol/L | Vitamin D3 100 000 IU single dose | Placebo | 8 weeks | Endothelial function | 40.2 ± 10.3 to 63.1 ± nr | 7.5 ± 1.6 | ↑ FMD brachial artery = HbA1c |
| (29) | Iran | 118, 25(OH)D <75 nmol/L | 1. Vitamin D3 50 000 IU/week 2. Ca 1000 mg/day 3. Vitamin D3 50 000 IU/week + Ca 1000 mg/day | Placebo | 8 weeks | Metabolic parameters | 28.0 ± 13.9 to nr | 6.6 ± 0.8 | ↓ HbA1c, HOMA-IR, fasting glucose and insulin, LDL-cholesterol in Calcium + Vitamin D group. No change in the vitamin D group |
| (30) | UK | 61 | Vitamin D3 single dose: 1. 100 000 IU 2. 200 000 IU | Placebo | 16 weeks | Metabolic parameters | 48.0 ± 21.0 to 76.0 ± 30.0 | 6.9 ± 0.8 | = HbA1c, HOMA-IR, lipid levels |
| (31) | China | 100, 25(OH)D <75 nmol/L | Vitamin D3 5000 IU/day | Placebo | 12 weeks | Endothelial function | 52.7 ± 11.0 to 146.3 ± nr | 7.4 (6.8–8.5) | = FMD, HbA1c, lipid levels |
25(OH)D, 25-hydroxy vitamin D; BMI, body mass index; Ca, calcium; FMD, flow-mediated dilatation; HOMA-IR, homeostatic model assessment – insulin resistance; nr, not reported; OGTT, oral glucose tolerance test; T2DM, type 2 diabetes mellitus.
Baseline values of the intervention groups; ↑ increasement; ↓ decreasement; †hyperinsulinaemic – euglycaemic clamp method.
A wide variety was seen in mean baseline serum 25(OH)D in the intervention group, with the lowest value of 21.5 ± 23.7 nmol/L (34) and a highest value of 117.3 ± 86.7 nmol/L (35). Four studies included only vitamin D-deficient (serum 25(OH)D <50 nmol/L) patients (12, 23, 26, 27). Many different intervention regimes were used. The mean change in serum 25(OH)D between the intervention and control groups is summarized for each study in Fig. 3. Except for two studies performed by Breslavsky (19) and Cavalcante (33), all studies observed a significant increase in serum 25(OH)D in the intervention group compared with the placebo group, with an overall mean difference: 30.2 nmol/L; 95% CI: 23.1–37.3, P < 0.01). Five studies could not be included in this analysis due to missing data (20, 29, 31, 38, 40).
Mean change from baseline in serum 25(OH)D (nmol/L) between intervention and control.
Citation: European Journal of Endocrinology 176, 1; 10.1530/EJE-16-0391
The effect on HbA1c
Nineteen studies reported sufficient data for inclusion in the meta-analysis to measure the overall effect of vitamin D on HbA1c (12, 19, 20, 21, 22, 24, 25, 26, 27, 28, 29, 30, 32, 34, 35, 36, 37, 39, 40). Four studies were excluded from this analysis by the following reasons: (i) no post-intervention HbA1c of the control group (33); (ii) glycaemic control was measured by hyperinsulinaemic–euglycaemic clamp method (23); (iii) no baseline HbA1c was available in the intervention group (38), and (iv) no standard deviations were reported (31). The total number of included patients was 1475, of whom 755 were included in the treatment group and 720 in the placebo group. One out of these 19 studies reported a significant reduction in HbA1c after vitamin D intervention compared with placebo (39). In a study among 118 patients who were randomized to either vitamin D with or without calcium, or placebo, a significant decrease in HbA1c was seen in the vitamin D plus calcium group versus placebo. However, this study failed to reach a significant reduction in HbA1c in the group treated solely with vitamin D supplementation (29). A pilot RCT performed by Soric et al. (40) showed a trend towards a greater reduction in the mean change of HbA1c in the vitamin D group compared with the control group; however, this difference was not statistically significant. In a subgroup analysis among patients with an HbA1c > 9.0%, a significantly greater reduction in HbA1c was observed in the intervention group (mean change: −1.4%; 95% CI: −2.4 to −0.4, P = 0.01) compared with placebo (40). In our own study population, a significant effect of vitamin D supplementation on HbA1c was seen in patients with a serum 25(OH)D level ≤30 nmol/L (n = 19, mean change: −0.34%; 95% CI: −0.65 to −0.04, P = 0.02) (24). Furthermore, Nasri et al. (37) reported a significant difference in HbA1c between the intervention and control groups only in male patients.
Based on a random-effect meta-analysis, comparing the mean change in HbA1c from baseline between the intervention and placebo groups, no overall effect was seen on HbA1c after vitamin D intervention (standardized difference in means: 0.12; 95% CI: −0.03 to 0.26, P = 0.11) (Fig. 4A). Heterogeneity was present (I 2 = 42%, P = 0.03); however, there was no evidence for publication bias (Egger’s test: P = 0.38).
Meta-analysis of effects on HbA1c in all studies (A) and in studies with a baseline mean serum 25(OH)D <50 nmol/L (B) or <30 nmol/L (C), and mean baseline HbA1c ≥8% (D).
Citation: European Journal of Endocrinology 176, 1; 10.1530/EJE-16-0391
Including only the studies with a mean baseline 25(OH)D <50 nmol/L did not change the effect of vitamin D intervention on HbA1c (standardized difference in means: 0.14; 95% CI: −0.07 to 0.35, P = 0.20) (Fig. 4B) (12, 19, 22, 23, 25, 27, 28, 29, 30, 32, 34, 38, 39). In addition, no difference was seen, including only the studies with a mean baseline serum 25(OH)D <30 nmol/L (standardized difference in means: 0.02; 95% CI: −0.18 to 0.23, P = 0.82) (Fig. 4C). Including the studies with a baseline mean HbA1c ≥8% (64 mmol/mol) a trend towards a positive effect of vitamin D supplementation was seen, but this was not significant (standardized difference in means: 0.14; 95% CI: −0.05 to 0.33, P = 0.14) (Fig. 4D).Furthermore, inclusion of the studies which were labelled as good quality did not alter the results (standardized difference in means: 0.01; 95% CI: −0.12 to 0.14, P = 0.90) (Fig. 5). Heterogeneity was not present (I 2 = 1%).
Meta-analysis of effects on HbA1c in studies labelled as good quality.
Citation: European Journal of Endocrinology 176, 1; 10.1530/EJE-16-0391
The effect on fasting glucose
Of the 23 studies that were included in the systematic review, 13 reported fasting glucose as primary or secondary outcome measure (19, 20, 21, 22, 23, 24, 26, 27, 28, 32, 35, 36, 39). Three studies reported a significant reduction in fasting glucose level after vitamin D supplementation (21, 28, 39).
A pooled meta-analysis including 1180 patients (vitamin D: n = 608; controls: n = 572) comparing the mean change in fasting glucose between baseline and follow-up for both groups did not reveal an overall effect of vitamin D supplementation on fasting glucose (between-group standardized mean difference: 0.09; 95% CI: −0.11 to 0.28, P = 0.39, I 2 = 60%) (Fig. 6A). No evidence for publication bias was found using a funnel plot and Egger’s test (P = 0.97). Including only the good quality studies did not alter the effect on fasting glucose. A pooled meta-analysis with the inclusion of the studies with a mean baseline HbA1c ≥8% (64 mmol/mol) shows a significant effect of vitamin D on fasting glucose (standardized difference in means: 0.36; 95% CI: 0.12–0.61, P = 0.003, I 2 = 0%) (Fig. 6B).
Meta-analysis of effects on fasting glucose in all studies (A) and in studies with a baseline mean HbA1c ≥8% (B).
Citation: European Journal of Endocrinology 176, 1; 10.1530/EJE-16-0391
The effect on insulin resistance
Thirteen studies reported data on insulin resistance, of which twelve studies used the HOMA-IR to quantify insulin resistance (19, 20, 21, 22, 24, 25, 26, 29, 30, 35, 36, 39), and one study measured insulin resistance through hyperinsulinaemic–euglycaemic clamp method (23). Two studies observed a significant reduction in insulin resistance after vitamin D supplementation (21, 39), and one study found a negative effect of vitamin D supplementation on insulin resistance compared with placebo (35).
Twelve studies were compared in a random-effects meta-analysis model, demonstrating no significant effect of vitamin D supplementation on insulin resistance compared with controls (between-group standardized difference in means: 0.23; 95% CI: −0.06 to 0.53, P = 0.12; I 2 = 77%, P = 0.04) (Fig. 7). No evidence for publication bias was found using a funnel plot and Egger’s test (P = 0.26). Inclusion of the studies which were qualified as good did not alter the results. Only one study reported data of HOMA-IR with a baseline HbA1c ≥8%. The study performed by Kampmann et al. (23), which measured insulin resistance by using the hyperinsulinaemic–euglycaemic clamp method – the golden standard – did not find a positive effect of vitamin D on glycaemic control in 16 patients with type 2 diabetes.
Meta-analysis of effects on HOMA-IR.
Citation: European Journal of Endocrinology 176, 1; 10.1530/EJE-16-0391
Discussion
Our systematic review and meta-analysis examined the effect of vitamin D supplementation on glycaemic indices in patients with type 2 diabetes mellitus. Combining all studies, no effect was seen of vitamin D supplementation on the parameters of glycaemic control (i.e. HbA1c, fasting glucose and HOMA-IR) in patients with type 2 diabetes. Including only studies with a mean baseline serum 25(OH)D <50 nmol/L or <30 nmol/L did not change these results. Including only the studies with a mean baseline HbA1c ≥8% (64 mmol/mol) revealed a significant effect of vitamin D supplementation on fasting glucose.
The main challenge of this systematic review was the heterogeneity between the studies. To level for this challenge, we included only RCTs. However, still heterogeneity was present with a wide variety of intervention schemes and follow-up duration used in the studies included, which resulted in a varying increase in serum 25(OH)D as was shown in Fig. 2. To resolve the problem of heterogeneity, we made a quality assessment of all included studies. Including only good quality studies did not alter the effect of vitamin D supplementation on glycaemic indices.
Still no consensus has been reached in the optimal value of serum 25(OH)D and the best supplementation regime. Nowadays, vitamin D deficiency is commonly defined by a serum 25(OH)D <30 nmol/L. This threshold level has been confirmed by the Institute of Medicine at the end of 2010 and the Endocrine Society Guideline (41, 42). Optimal serum 25(OH)D is defined as a level above 50 nmol/L according to the Institute of Medicine and above 75 nmol/L according to the Endocrine Society.
A possible explanation for the lack of effect found in most studies could be an underrepresentation of vitamin D-deficient patients. It is possible that vitamin D could only be effective in vitamin D-deficient patients, and especially in those with poor glycaemic control (43, 44). This hypothesis was confirmed in the study performed by Soric et al. (40), who showed a 1.4% decrease in HbA1c in patients with a baseline HbA1c level ≥9.0% after 12 weeks with a daily consumption of 2.000 IU vitamin D in contrast to patients with a HbA1c <9.0% where no effect on glycaemic control was seen after vitamin D treatment. Additionally, in our previous RCT among 275 patients with type 2 diabetes, in 19 patients with a serum 25(OH)D below 30 nmol/L, a significant decrease in HbA1c level was seen after 6 months of vitamin D supplementation compared with placebo (24). Another important note is the wide range in follow-up duration between the studies. As HbA1c is representing the glycosylated haemoglobin which has a lifetime around 100 days, a follow-up duration of more than three months is favourable.
Of interest is the possibility that vitamin D could only be beneficial in patients with normal glucose tolerance or impaired glucose tolerance. The pathogenesis of type 2 diabetes consists of a progressive insulin resistance, which is initially compensated by enhanced insulin secretion by the pancreatic beta-cells. At the time of onset of type 2 diabetes, the beta-cell mass is reduced by 25–50% (45). The direct effect of vitamin D on the pancreatic β-cell might be negligible at this time. In this line, our systematic review including only studies examining patients with type 2 diabetes is a limitation of this study.
Individual variability explained by vitamin D receptor polymorphisms may also play a role in the study results. Earlier research demonstrated an association between vitamin D receptor polymorphisms and the risk for type 2 diabetes, suggesting that timing of vitamin D supplementation is critical (46, 47). In addition, a study by Wang et al. demonstrated that the vitamin D-binding protein polymorphism, and thus vitamin D bioavailability, was moderately associated with increased susceptibility to type 2 diabetes in Asians, but not in Caucasians, suggesting that ethnicity might be a potential factor associated with heterogeneity (48).
Another relevant note is the different vitamin D assays that were used in the included studies. Much discussion is going on about the comparability and accuracy of the different assays, which raises concerns (49). Most of the studies included in this review used an enzyme-immunoassay method for measurement of serum 25(OH)D, where the liquid chromatography--mass spectrometry (LC–MS) method is the golden standard.
The strength and limitations of our study need to be mentioned. First, as our initial search was performed by only one author, all eligible studies may not have been included. However, our negative findings suggest that unpublished studies (which also tend to be negative) would be very unlikely to alter our conclusions. We found no evidence for publication bias from the funnel plots. For the meta-analysis, we performed a quality assessment according to the checklist of the Dutch Cochrane Collaboration, which has some limitations, especially when trying to decide on the relative importance of the different criteria (16). Another note is that we did not have access to all original data, which is the best method to perform a meta-analysis. A strength of our study is that we included only RCTs to assess the strength of evidence and limit the role of bias.
In conclusion, current evidence of RCTs does not support short-term vitamin D supplementation in a heterogeneous population with type 2 diabetes. However, in patients with poorly controlled diabetes a favourable effect of vitamin D is seen on fasting glucose. Future research in this subgroup is highly of interest.
Appendix
This is linked to the online version of the paper at http://dx.doi.org/10.1530/EJE-16-0391.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this review.
Funding
No external funding was sought for preparation of this manuscript.
Author contribution statement
Each author contributed in a substantial way to the work described in the manuscript and its preparation. S Simsek initiated the systematic review. Y Krul-Poel performed the systematic search and data extraction, S Simsek checked the search. Y Krul-Poel and M ter Wee performed the statistical analyses. Y Krul-Poel wrote the first version of the manuscript and all authors revised the paper critically. Y Krul-Poel was responsible for the figures. All the authors were involved in the approval of the final version to be published.
References
- 1↑
Rosen CJ, Adams JS, Bikle DD, Black DM, Demay MB, Manson JE, Murad MH & Kovacs CS. The nonskeletal effects of vitamin D: an Endocrine Society scientific statement. Endocrine Reviews 2012 33 456–492. (doi:10.1210/er.2012-1000)
- 2↑
Bland R, Markovic D, Hills CE, Hughes SV, Chan SL, Squires PE & Hewison M. Expression of 25-hydroxyvitamin D3-1alpha-hydroxylase in pancreatic islets. Journal of Steroid Biochemistry and Molecular Biology 2004 89–90 121–125. (doi:10.1016/j.jsbmb.2004.03.115)
- 3↑
Zehnder D, Bland R, Williams MC, McNinch RW, Howie AJ, Stewart PM & Hewison M. Extrarenal expression of 25-hydroxyvitamin d(3)-1 alpha-hydroxylase. Journal of Clinical Endocrinology and Metabolism 2001 86 888–894. (doi:10.1210/jc.86.2.888)
- 4↑
Chagas CE, Borges MC, Martini LA & Rogero MM. Focus on vitamin D, inflammation and type 2 diabetes. Nutrients 2012 4 52–67. (doi:10.3390/nu4010052)
- 5↑
Hewison M. An update on vitamin D and human immunity. Clinical Endocrinology 2012 76 315–325. (doi:10.1111/j.1365-2265.2011.04261.x)
- 6↑
Grimnes G, Emaus N, Joakimsen RM, Figenschau Y, Jenssen T, Njolstad I, Schirmer H & Jorde R. Baseline serum 25-hydroxyvitamin D concentrations in the Tromso Study 1994–95 and risk of developing type 2 diabetes mellitus during 11 years of follow-up. Diabetic Medicine 2010 27 1107–1115. (doi:10.1111/j.1464-5491.2010.03092.x)
- 7↑
Mitri J, Muraru MD & Pittas AG. Vitamin D and type 2 diabetes: a systematic review. European Journal of Clinical Nutrition 2011 65 1005–1015. (doi:10.1038/ejcn.2011.118)
- 8↑
Pittas AG, Lau J, Hu FB & Dawson-Hughes B. The role of vitamin D and calcium in type 2 diabetes. A systematic review and meta-analysis. Journal of Clinical Endocrinology and Metabolism 2007 92 2017–2029. (doi:10.1210/jc.2007-0298)
- 9↑
Song Y, Wang L, Pittas AG, Del Gobbo LC, Zhang C, Manson JE & Hu FB. Blood 25-hydroxy vitamin D levels and incident type 2 diabetes: a meta-analysis of prospective studies. Diabetes Care 2013 36 1422–1428. (doi:10.2337/dc12-0962)
- 10↑
Joergensen C, Gall MA, Schmedes A, Tarnow L, Parving HH & Rossing P. Vitamin D levels and mortality in type 2 diabetes. Diabetes Care 2010 33 2238–2243. (doi:10.2337/dc10-0582)
- 11↑
Ferder M, Inserra F, Manucha W & Ferder L. The world pandemic of vitamin D deficiency could possibly be explained by cellular inflammatory response activity induced by the renin-angiotensin system. American Journal of Physiology: Cell Physiology 2013 304 C1027–C1039. (doi:10.1152/ajpcell.00403.2011)
- 12↑
Sugden JA, Davies JI, Witham MD, Morris AD & Struthers AD. Vitamin D improves endothelial function in patients with type 2 diabetes mellitus and low vitamin D levels. Diabetic Medicine 2008 25 320–325. (doi:10.1111/j.1464-5491.2007.02360.x)
- 13↑
Vimaleswaran KS, Cavadino A, Berry DJ, Jorde R, Dieffenbach AK, Lu C, Alves AC, Heerspink HJ, Tikkanen E & Eriksson J. Association of vitamin D status with arterial blood pressure and hypertension risk: a mendelian randomisation study. Lancet Diabetes and Endocrinology 2014 2 719–729. (doi:10.1016/s2213-8587(14)70113-5)
- 14↑
George PS, Pearson ER & Witham MD. Effect of vitamin D supplementation on glycaemic control and insulin resistance: a systematic review and meta-analysis. Diabetic Medicine 2012 29 e142–e150. (doi:10.1111/j.1464-5491.2012.03672.x)
- 15↑
Seida JC, Mitri J, Colmers IN, Majumdar SR, Davidson MB, Edwards AL, Hanley DA, Pittas AG, Tjosvold L & Johnson JA. Clinical review: effect of vitamin D3 supplementation on improving glucose homeostasis and preventing diabetes: a systematic review and meta-analysis. Journal of Clinical Endocrinology and Metabolism 2014 99 3551–3560. (doi:10.1210/jc.2014-2136)
- 16↑
Dutch Cochrane Centre. Checklist Quality Assessment RCT, Cohort and Case-Control studies. (available at: http://netherlands.cochrane.org/beoordelingsformulieren-en-andere-downloads 2016). 1-1-0016.
- 17↑
Higgins J, Thompson S, Deeks J & Altman D. Statistical heterogeneity in systematic reviews of clinical trials: a critical appraisal of guidelines and practice. Journal of Health Services Research & Policy 2002 7 51–61. (doi:10.1258/1355819021927674)
- 18↑
Egger M, Davey SG, Schneider M & Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997 315 629–634. (doi:10.1136/bmj.315.7109.629)
- 19↑
Breslavsky A, Frand J, Matas Z, Boaz M, Barnea Z & Shargorodsky M. Effect of high doses of vitamin D on arterial properties, adiponectin, leptin and glucose homeostasis in type 2 diabetic patients. Clinical Nutrition 2013 32 970–975. (doi:10.1016/j.clnu.2013.01.020)
- 20↑
Elkassaby S, Harrison LC, Mazzitelli N, Wentworth JM, Colman PG, Spelman T & Fourlanos S. A randomised controlled trial of high dose vitamin D in recent-onset type 2 diabetes. Diabetes Research and Clinical Practice 2014 106 576–582. (doi:10.1016/j.diabres.2014.08.030)
- 21↑
Jafari T, Faghihimani E, Feizi A, Iraj B, Javanmard SH, Esmaillzadeh A, Fallah AA & Askari G. Effects of vitamin D-fortified low fat yogurt on glycemic status, anthropometric indexes, inflammation, and bone turnover in diabetic postmenopausal women: a randomised controlled clinical trial. Clinical Nutrition 2016 35 67–76. (doi:10.1016/j.clnu.2015.02.014)
- 22↑
Jehle S, Lardi A, Felix B, Hulter HN, Stettler C & Krapf R. Effect of large doses of parenteral vitamin D on glycaemic control and calcium/phosphate metabolism in patients with stable type 2 diabetes mellitus: a randomised, placebo-controlled, prospective pilot study. Swiss Medical Weekly 2014 144 w13942. (doi:10.4414/smw.2014.13942)
- 23↑
Kampmann U, Mosekilde L, Juhl C, Moller N, Christensen B, Rejnmark L, Wamberg L & Orskov L. Effects of 12 weeks high dose vitamin D3 treatment on insulin sensitivity, beta cell function, and metabolic markers in patients with type 2 diabetes and vitamin D insufficiency – a double-blind, randomized, placebo-controlled trial. Metabolism 2014 63 1115–1124. (doi:10.1016/j.metabol.2014.06.008)
- 24↑
Krul-Poel YH, Westra S, ten Boekel E, ter Wee MM, van Schoor NM, van Wijland H, Stam F, Lips PT, Simsek S.Effect of vitamin D supplementation on glycemic control in patients with type 2 diabetes (SUNNY trial): a randomized placebo-controlled trial. Diabetes Care 2015 38 1420–1426. (doi:10.2337/dc15-0323)
- 25↑
Parekh D, Sarathi V, Shivane VK, Bandgar TR, Menon PS & Shah NS. Pilot study to evaluate the effect of short-term improvement in vitamin D status on glucose tolerance in patients with type 2 diabetes mellitus. Endocrine Practice 2010 16 600–608. (doi:10.4158/EP09300.OR)
- 26↑
Ryu OH, Lee S, Yu J, Choi MG, Yoo HJ & Mantero F. A prospective randomized controlled trial of the effects of vitamin D supplementation on long-term glycemic control in type 2 diabetes mellitus of Korea. Endocrine Journal 2014 61 167–176. (doi:10.1507/endocrj.EJ13-0356)
- 27↑
Sadiya A, Ahmed SM, Carlsson M, Tesfa Y, George M, Ali SH, Siddieg HH & Abusnana S. Vitamin D supplementation in obese type 2 diabetes subjects in Ajman, UAE: a randomized controlled double-blinded clinical trial. European Journal of Clinical Nutrition 2014 69 707–711. (doi:10.1038/ejcn.2014.251)
- 28↑
Shab-Bidar S, Neyestani TR, Djazayery A, Eshraghian MR, Houshiarrad A, Gharavi A, Kalayi A, Shariatzadeh N, Zahedirad M & Khalaji N et al. Regular consumption of vitamin D-fortified yogurt drink (Doogh) improved endothelial biomarkers in subjects with type 2 diabetes: a randomized double-blind clinical trial. BMC Medicine 2011 9 125. (doi:10.1186/1741-7015-9-125)
- 29↑
Tabesh M, Azadbakht L, Faghihimani E, Tabesh M & Esmaillzadeh A. Effects of calcium-vitamin D co-supplementation on metabolic profiles in vitamin D insufficient people with type 2 diabetes: a randomised controlled clinical trial. Diabetologia 2014 57 2038–2047. (doi:10.1007/s00125-014-3313-x)
- 30↑
Witham MD, Dove FJ, Dryburgh M, Sugden JA, Morris AD & Struthers AD. The effect of different doses of vitamin D(3) on markers of vascular health in patients with type 2 diabetes: a randomised controlled trial. Diabetologia 2010 53 2112–2119. (doi:10.1007/s00125-010-1838-1)
- 31↑
Yiu YF, Yiu KH, Siu CW, Chan YH, Li SW, Wong LY, Lee SW, Tam S, Wong EW & Lau CP et al. Randomized controlled trial of vitamin D supplement on endothelial function in patients with type 2 diabetes. Atherosclerosis 2013 227 140–146. (doi:10.1016/j.atherosclerosis.2012.12.013)
- 32↑
Al-Zahrani MK, Elnasieh AM, Alenezi FM, Almoushawah AA, Almansour M, Alshahrani F, Rahman SU & Al-Zahrani A. A 3-month oral vitamin D supplementation marginally improves diastolic blood pressure in Saudi patients with type 2 diabetes mellitus. International Journal of Clinical and Experimental Medicine 2014 7 5421–5428.
- 33↑
Cavalcante R, Maia J, Mesquita P, Henrique R, Griz L, Bandeira MP & Bandeira F. The effects of intermittent vitamin D3 supplementation on muscle strength and metabolic parameters in postmenopausal women with type 2 diabetes: a randomized controlled study. Therapeutic Advances in Endocrinology and Metabolism 2015 6 149–154. (doi:10.1177/2042018815578998)
- 34↑
Ghavamzadeh S, Mobasseri M & Mahdavi R. The effect of vitamin D supplementation on adiposity, blood glycated hemoglobin, serum leptin and tumor necrosis factor-alpha in type 2 diabetic patients. International Journal of Preventive Medicine 2014 5 1091–1098.
- 35↑
Heshmat R, Tabatabaei-Malazy O, Abbaszadeh-Ahranjani S, Shahbazi S, Khooshehchin G, Bandarian F & Larijani B. Effect of vitamin D on insulin resistance and anthropometric parameters in type 2 diabetes; a randomized double-blind clinical trial. Daru 2012 20 10. (doi:10.1186/2008-2231-20-10)
- 36↑
Jorde R & Figenschau Y. Supplementation with cholecalciferol does not improve glycaemic control in diabetic subjects with normal serum 25-hydroxyvitamin D levels. European Journal of Nutrition 2009 48 349–354. (doi:10.1007/s00394-009-0020-3)
- 37↑
Nasri H, Behradmanesh S, Maghsoudi AR, Ahmadi A, Nasri P & Rafieian-Kopaei M. Efficacy of supplementary vitamin D on improvement of glycemic parameters in patients with type 2 diabetes mellitus; a randomized double blind clinical trial. Journal of Renal Injury Prevention 2014 3 31–34. (doi:10.12861/jrip.2014.10)
- 38↑
Strobel F, Reusch J, Penna-Martinez M, Ramos-Lopez E, Klahold E, Klepzig C, Wehrle J, Kahles H & Badenhoop K. Effect of a randomised controlled vitamin D trial on insulin resistance and glucose metabolism in patients with type 2 diabetes mellitus. Hormone and Metabolic Research 2014 46 54–58. (doi:10.1055/s-0033-1358453)
- 39↑
Nikooyeh B, Neyestani TR, Tayebinejad N, Alavi-Majd H, Shariatzadeh N, Kalayi A, Zahedirad M, Heravifard S & Salekzamani S. Daily intake of vitamin D- or calcium-vitamin D-fortified Persian yogurt drink (doogh) attenuates diabetes-induced oxidative stress: evidence for antioxidative properties of vitamin D. Journal of Human Nutrition and Dietetics 2013 27 (Supplement 2) 276–283. (doi:10.1111/jhn.12142)
- 40↑
Soric MM, Renner ET & Smith SR. Effect of daily vitamin D supplementation on HbA1c in patients with uncontrolled type 2 diabetes mellitus: a pilot study. Journal of Diabetes 2012 4 104–105. (doi:10.1111/j.1753-0407.2011.00164.x)
- 41↑
Holick MF, Binkley NC, Bischoff-Ferrari HA, Gordon CM, Hanley DA, Heaney RP, Murad MH & Weaver CM. Evaluation, treatment, and prevention of vitamin D deficiency: an Endocrine Society clinical practice guideline. Journal of Clinical Endocrinology and Metabolism 2011 96 1911–1930. (doi:10.1210/jc.2011-0385)
- 42↑
Ross AC, Manson JE, Abrams SA, Aloia JF, Brannon PM, Clinton SK, Durazo-Arvizu RA, Gallagher JC, Gallo RL & Jones G et al. The 2011 report on dietary reference intakes for calcium and vitamin D from the Institute of Medicine: what clinicians need to know. Journal of Clinical Endocrinology and Metabolism 2011 96 53–58. (doi:10.1210/jc.2010-2704)
- 43↑
Krul-Poel YH, Westra S & Simsek S. Response to comment on Krul-Poel et al. Effect of vitamin D supplementation on glycemic control in patients with type 2 diabetes (SUNNY Trial): a randomized placebo-controlled trial. Diabetes Care 2015 38 1420–1426. Diabetes Care 2015 38 e169. (doi:10.2337/dci15-0011)
- 44↑
Krul-Poel YH, Westra S & Simsek S.Response to comment on Krul-Poel et al. Effect of vitamin D supplementation on glycemic control in patients with type 2 diabetes (SUNNY trial): a randomized placebo-controlled trial. Diabetes Care 2015 38 1420–1426. Diabetes Care 2015 38 e191–e192. (doi:10.2337/dci15-0005)
- 45↑
Mathieu C. Vitamin D and diabetes: where do we stand? Diabetes Research and Clinical Practice 2015 108 201–209. (doi:10.1016/j.diabres.2015.01.036)
- 46↑
Jia J, Ding H, Yang K, Mao L, Zhao H, Zhan Y & Shen C. Vitamin D receptor genetic polymorphism is significantly associated with risk of type 2 diabetes mellitus in Chinese Han population. Archives of Medical Research 2015 46 572–579. (doi:10.1016/j.arcmed.2015.09.006)
- 47↑
Oh JY & Barrett-Connor E. Association between vitamin D receptor polymorphism and type 2 diabetes or metabolic syndrome in community-dwelling older adults: the Rancho Bernardo Study. Metabolism 2002 51 356–359. (doi:10.1053/meta.2002.29969)
- 48↑
Wang G, Li Y, Li L, Yu F, Cui L, Ba Y, Li W & Wang C. Association of the vitamin D binding protein polymorphisms with the risk of type 2 diabetes mellitus: a meta-analysis. BMJ Open 2014 4 e005617. (doi:10.1136/bmjopen-2014-005617)
- 49↑
Carter GD. Accuracy of 25-hydroxyvitamin D assays: confronting the issues. Current Drug Targets 2011 12 19–28. (doi:10.2174/138945011793591608)
