Diabetes (Type II) reduced by single injection of 300,000 IU of vitamin D3 – RCT March 2014

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

Sigrid Jehlea, Alessia Lardi, Barbara Felixa, Henry N. Hulterb, Christoph Stettlerc, Reto Krapfa
a Department of Medicine, Kantonsspital Bruderholz, University of Basel, Switzerland
b Department of Medicine, University of California, San Francisco, USA
c Division of Endocrinology and Diabetology, Insel University Hospital, Berne, Switzerland

 Download the PDF from VitaminDWiki

OBJECTIVE: Vitamin D (D3) status is reported to correlate negatively with insulin production and insulin sensitivity in patients with type 2 diabetes mellitus (T2DM). However, few placebo-controlled intervention data are available. We aimed to assess the effect of large doses of parenteral D3 on glycosylated haemoglobin (HbA1c) and estimates of insulin action (homeostasis model assessment insulin resistance: HOMA-IR) in patients with stable T2DM.

MATERIALS AND METHODS: We performed a prospective, randomised, double-blind, placebo-controlled pilot study at a single university care setting in Switzerland. Fifty-five patients of both genders with T2DM of more than 10 years were enrolled and randomised to either 300,000 IU D3 or placebo, intramuscularly. The primary endpoint was the intergroup difference in HbA1c levels. Secondary endpoints were: changes in insulin sensitivity, albuminuria, calcium/phosphate metabolism, activity of the renin-aldosterone axis and changes in 24-hour ambulatory blood pressure values.

RESULTS: After 6 months of D3 supply, there was a significant intergroup difference in the change in
HbA1c levels (relative change [mean ± standard deviation]

• +2.9% ± 1.5% in the D3 group vs
• +6.9% ± 2.1% the in placebo group, p = 0.041)

as HOMA-IR

• decreased by 12.8% ± 5.6% in the D3 group and
• increased by 10% ± 5.4% in the placebo group (intergroup difference, p = 0.032).

Twenty-four-hour urinary albumin excretion decreased in the D3 group from 200 ± 41 to 126 ± 39, p = 0.021). There was no significant intergroup difference for the other secondary endpoints.

CONCLUSIONS: D3 improved insulin sensitivity (based on HOMA-IR) and affected the course of HbA1c positively compared with placebo in patients with T2DM.

Clinical trial registration number at ClinicalTrials.gov: NCT01585051

Discussion from the PDF

This study examined a population of slightly D3 deficient (defined as <50 nmol/l [24]), metabolically stable, long-standing (>10 years) T2DM patients with adequate baseline blood pressure and acceptable glycaemic control (HbA1c 7.1% ± 1.0%, table 1). The main findings were: first, HbA1c showed a differential course during treatment with D3, with a significantly smaller increase in the treatment group compared with placebo. Second, markers of insulin resistance were significantly reduced in individuals treated with D3 compared with placebo.
HOMA-IR has been shown to correlate closely with analysis of insulin sensitivity by the euglycaemic insulin clamp method [25]. Based on this calculation, D3 administration ameliorated insulin resistance and significantly limited the rise in HbA1c as compared to placebo during this 6-month intervention trial. The amelioration of insulin resistance could theoretically be indirect via the reported anti-inflammatory effects of D3 [26], but this thesis was not supported by changes in hsCRP levels. However, both groups exhibited normal baseline hsCRP values, suggesting that systemic inflammatory activity was low and rendering demonstration of a putative inhibitory effect more difficult. Other studies examining the effects of D3 in patients at risk for diabetes or normal subjects have failed to demonstrate a significant effect of the intervention on insulin sensitivity [9–11]. Thus, the effect of D3 may be limited to establish T2DM and may depend on the degree of insulin resistance.
It had been planned to enrol all patients between October and November to limit the contribution of skin synthesis of D3 (the recruited subjects all live ~47o N latitude). However, as a result of patient factors (holidays, professional engagements, etc.) enrolment could be completed only at the end of December. The last subjects completed the protocol in July 2010, thereby natural sun exposure increased D3 in both groups. The effect of D3 administration may have been mitigated by the fact that the placebo group exhibited a “spontaneous” increase in 25(OH)-D3 levels most probably owing to increased sun exposure in the spring and early summer. A small but significant reduction of C-peptide levels was noted in the D3 arm relative to placebo. This finding is consistent with the observed tendency to a relative reduction of fasting glucose levels in the D3 arm and may thus reflect a secondary consequence of improved insulin sensitivity.
Previously, 1,25(OH)2-D3 has been shown to inhibit renin gene transcription and vitamin D receptor knockout mice demonstrate hypertension [27, 28]. However, this study in T2DM with well-controlled blood pressure did not show evidence for a detectable inhibitory effect of D3 on the activity of the renin/aldosterone system on the basis of the analysis of plasma renin, plasma aldosterone and 24-hour urinary excretion rates of tetrahydro-aldosterone. Also, there was no intergroup treatment difference in the 24-hour ambulatory systolic and diastolic blood pressure measurements.
The observation of a significant decrease in urinary albumin excretion in the D3 group is of interest in view of the association of low D3 status with albuminuria [29] and is confirmatory evidence for the possible retarding effect of D3 agonists on progression of glomerular injury [30].
Our study cannot conclusively answer the question as to whether the observed effects of D3 administration are due to changes in 25(OH)-D3 or 1,25(OH)2-D3, although the increase in 1,25(OH)2-D3 was limited to the intervention group. The increase in 1,25(OH)2-D3 and the decrease in intact PTH are responsible – at least in part – for the significant increase in FGF-23. However, the role of higher circulating levels of 25(OH)-D3 also requires consideration as osteoblasts exposed to 25(OH)-D3 have been shown to produce 1,25(OH)2-D3 locally in a paracrine/autocrine fashion and, thereby, to increase the synthesis of FGF-23 [31].
The D3-induced rise in FGF-23 in this study might be viewed adversely since injection of pharmacological amounts of murine FGF-23 into myocardium induced left ventricular hypertrophy in mice [32], and elevated FGF-23 levels have been reported to be independently associated with total mortality in a prospective patient cohort with incident end-stage renal disease [33, 34]. The role of FGF-23 in the incidence of coronary heart disease (CHD) in the general population is unclear. However, there is substantial reason to consider that incident CHD is not dependent on FGF-23 levels in the general population. In a prospective, nested, case-control cohort study, from the 51,529-subject Health Professionals Follow-up Study, within the subset with no history of CHD (mean serum creatinine 1.0 mg/dl) no association was found between baseline FGF-23 levels and subsequent nonfatal myocardial infarction and fatal CHD events [35]. Nevertheless, other epidemiological data have shown that FGF-23 concentration is a risk factor for increased all-cause and cardiovascular mortality in Swedish community dwelling adults [36].
Interestingly, it has been demonstrated that insulin-resistant T2DM patients exhibit an impaired FGF-23 and PTH response to an acute phosphate load, sufficient to result in a supernormal hyperphosphataemic response [37]. Higher postprandial serum phosphate in T2DM might account, at least in part, for the systemic vascular calcification observed in this disorder and its status as a cardiovascular risk factor, as a function of duration of diabetes [38], and thus it is possible that higher FGF-23 levels might mitigate diabetic vascular disease, at least in subjects without chronic kidney disease. Since FGF-23 is produced in osteoblasts and osteocytes, and osteocyte density is reduced in experimental diabetes [39], it has been suggested that T2DM may be a state of relative FGF-23 hyporesponsiveness [37]. In fact, osteoblast-specific deletion of the insulin receptor in mice results in a phenotype of systemic insulin resistance and obesity that is mediated in part by osteoblastic endocrine dysfunction characterised by diminished secretion of under-carboxylated osteocalcin [40]. The findings that D3 induced increases in FGF-23 levels in T2DM, as previously reported in non-diabetics [41], and that D3 can improve the course of HBA1c and insulin sensitivity of peripheral tissues (HOMA-IR), raises the possibility that D3 therapy in T2DM, in addition to improving insulin sensitivity for glucose homeostasis in muscle cells and hepatocytes, might also result in increased FGF-23 levels via a similar insulin-sensitising action in osteocytes/osteoblasts.
The strengths of the present study are the placebo-controlled, prospective study design and the fact that no potentially confounding concomitant medication changes were made during the observation period. The chief limitation of this study is the relatively small number of participants and single centre location. In addition, the “spontaneous” rise in 25(OH)-D3 levels in the control population may have narrowed the differences and we cannot assume that the effects are dependent on different D3 doses. Also, we cannot exclude an additional effect of clandestine use of D3 in the placebo group as a result of the subjects’ interest in the study hypothesis.
Our results encourage the design and conduct of studies that further explore the roles of D3 and D3 analogues on glycaemic control in T2DM patients. Future studies should – among many other points – establish the dose-response characteristics, examine the best analogue of D3 with regard to the benefit/harm ratio and evaluate the effects in larger study populations and over longer time periods. In view of our study results, the effects of D3 on beta-cell function and insulin secretion merit special attention. In addition, the relevance of increased FGF-23 on cardiovascular morbidity should be evaluated in diabetes.
In summary and conclusion, D3 improved insulin sensitivity (based on HOMA-IR) and affected the course of HbA1c positively compared to placebo in patients with T2DM, but did not weaken, and was well tolerated.