Vitamin D and diabetes: Practical prescribing

In recent years, there has been a considerable resurgence in interest in vitamin D, with increasing recognition of its role as an important sterol hormone that appears to exert pleiotropic effects beyond bone and calcium homeostasis. Adequate sun exposure is the major source of vitamin D production, providing approximately 80–90% of the body’s requirement. The remainder is obtained from dietary sources, especially fish, cod liver oil and egg yolks.

Vitamin D biosynthesis
During exposure to sunlight, ultraviolet B radiation converts cutaneous 7-dehydrocholesterol into previtamin D3 and subsequently into vitamin D3 (colecalciferol). Vitamin D2 (ergocalciferol) is obtained from dietary sources and both vitamin D2 and D3 follow a common synthesis pathway in the body. Vitamin D is biologically inactive and undergoes a two-stage activation process via 25-hydroxylation in the liver to 25-hydroxycolecalciferol (25(OH)D) and 1-alpha-hydroxylation in the kidney to yield 1,25-dihydroxycolecalciferol D [1,25(OH)₂D] sometimes known as 1,25 dihydroxyvitamin D: the biologically active form of vitamin D. 1,25(OH)₂D binds with vitamin D receptors (VDRs), which are present on almost all cells of the body, and regulates expression of a gamut of genes that control cell growth, differentiation, immune function, muscle function and calcium homeostasis.

Vitamin D deficiency
As vitamin D is involved in cell growth and homeostasis, it is believed that vitamin D deficiency may negatively impact on skeletal, as well as extra-skeletal, health (Holick and Chen, 2008).

A 25(OH)D concentration below 30 nmol/L  is defined as vitamin D deficiency. A concentration above 50 nmol/L is regarded by some as an adequate concentration, whilst others recommend that 25(OH)D concentration should be above 75 nmol/L (Holick et al, 2011; National Osteoporosis Society, 2013 [see Table 1]). With respect to replacement of vitamin D from dietary sources, the literature suggests that for every 1000 International Units (IU)* of vitamin D ingested, there is a 25 nmol/L increment in 25(OH)D (Holick and Chen, 2008).

*400 IU vitamin D=10 μg vitamin D

Measuring vitamin D deficiency
Vitamin D status is assessed through the measurement of its metabolite, 25(OH)D concentrations in plasma. This molecule has a greater half-life (21–30 days) versus the active hormone 1,25 dihydroxyvitamin D, which has a relatively short half-life of 4–15 hours and is present in much lower pmol/L concentrations (National Osteoporosis Society, 2013). Similarly, 1,25 dihydroxyvitamin D concentrations may be normal despite vitamin D deficiency (National Osteoporosis Society, 2013).

There has also been debate regarding the appropriate assay to use for the measurement of 25(OH)D. Generally, tandem mass spectrometry is more sensitive than immunoassays, with assay costs at £5 per test in the NHS (Nottinghamshire Area Prescribing Committee, 2013). However, guidance suggests that measuring 25(OH)D is not necessary in all patients as, in certain people, vitamin D deficiency is highly predictable, obviating the need to measure 25(OH)D concentrations (National Osteoporosis Society, 2013).

Association between vitamin D deficiency and diabetes
Both vitamin D deficiency and diabetes are major global public health concerns (Holick and Chen, 2008; Lam and LeRoith, 2012). In recent decades, there has been an exponential increase in research on vitamin D with some of these data implicating vitamin D in glucose metabolism (Holick and Chen, 2008). Epidemiological data have implicated lower 25(OH)D concentrations as a risk factor for the development of type 1 diabetes (Zipitis and Akobeng, 2008; Pittas and Dawson-Hughes, 2010), type 2 diabetes (Knekt et al, 2008) and gestational diabetes (Alzaim and Wood, 2013).

Vitamin D deficiency and type 2 diabetes
A meta-analysis of observational studies revealed that a 25(OH)D concentration >25 ng/mL (>62.5 nmol/L) lowered the risk of developing type 2 diabetes by 43% when compared to a 25(OH)D concentration <14 ng/mL (<35 nmol/L; Mitri et al, 2011). Moreover, vitamin D supplementation of >500 IU daily in people with hypovitaminosis D attenuated the risk of developing type 2 diabetes by 13% (Mitri et al, 2011).

Confirmation of these encouraging findings was supported by another meta-analysis, which demonstrated a significant inverse correlation between 25(OH)D concentrations and the risk of type 2 diabetes; with every 4 ng/mL (10 nmol/L) increment in 25(OH)D concentration, the attendant risk of type 2 diabetes reduced by 4% (Song et al, 2013).

Vitamin D deficiency and type 1 diabetes
Some studies have noted that lower levels of vitamin D are associated with the development of type 1 diabetes (Bener et al, 2009; Borkar et al, 2010; Gorham et al, 2012), but this has not been universally accepted (Bierschenk et al, 2009).

There is a limited amount of data available linking vitamin D deficiency with type 1 diabetes; however, there has been some research taking into account the seasonality at the time of a person’s birth and diabetes diagnosis. An increased risk of developing type 1 diabetes was found if offspring were born in spring (Vaiserman et al, 2007), although type 1 diabetes has been found to be more commonly diagnosed during the winter months (Fishbein et al, 1982). Mohr et al (2008) found that there was an association between vitamin D deficiency and type 1 diabetes with increasing latitude – there was a higher incidence of type 1 diabetes at higher latitudes.

Genetic link
Interestingly, polymorphisms of VDR genes and genes regulating vitamin D metabolism have been associated with the development of diabetes (Takiishi et al, 2010). Specifically, there is accumulating evidence that points to an association between polymorphisms of VDR genes and risk for type 2 diabetes. In particular, FokI polymorphism of the VDR gene has been linked with increased risk of developing type 2 diabetes, especially in individuals of Asian ethnicity (Li et al, 2013). Also, allelic variations of three specific genes that are involved in vitamin D metabolism – DHCR7/NADSYN1 rs12785878 and CYP27B1 rs4646536 – have been associated with the development of islet autoimmunity – a harbinger of type 1 diabetes (Frederiksen et al, 2013).

Potential mechanisms associating vitamin D and diabetes
Type 1 diabetes
Type 1 diabetes is an autoimmune disease in which there is destruction of pancreatic beta-cells by auto-reactive immune cells. The presence of VDRs and the 1-alpha-hydroxylase enzyme on various immune cells, as well as beta-cells, implicates a possible immunomodulatory role of vitamin D against the development of type 1 diabetes (Prietl et al, 2013). Moreover, vitamin D exerts direct effects on pancreatic beta-cell functioning and survival, and improves insulin secretion. (Riachy et al, 2006; Mitri et al, 2013)

Type 2 diabetes
In type 2 diabetes, vitamin D deficiency is associated with impaired glucose-mediated insulin secretion (Takiishi et al, 2010; Wolden-Kirk et al, 2011). It has also been suggested that calcium influx associated with insulin release may be impaired in vitamin D deficiency (Bourlon et al, 1999) and that vitamin D may moderate the pro-inflammatory environment associated with type 2 diabetes. Vitamin D and its active metabolite 1,25[OH]₂D have been shown to reduce the expression of pro-inflammatory cytokines such as interleukin (IL)-1, IL-6 and TNF-alpha (Cohen-Lahav et al, 2007; Giulietti et al, 2007).

Although there is evidence that is generally supportive of the observational data relating vitamin D deficiency with the risk of type 2 diabetes, there have been no conclusive data to demonstrate a role for vitamin D in the prevention of type 2 diabetes.

Table 2 summarises some of the postulated antidiabetic effects of vitamin D in type 1 and type 2 diabetes. Despite these observations, studies have failed to show an association between vitamin D deficiency and beta-cell autoimmunity (Reinert-Hartwall et al, 2014).

Efficacy of vitamin D treatment in diabetes
The incidence of vitamin D deficiency is high in the British population (Pearce and Cheetham, 2010), yet studies suggest there are even higher rates of vitamin D deficiency in people with type 1 diabetes (Janner et al, 2010; Daga et al, 2012) and type 2 diabetes (Saedisomeolia et al, 2013). Recommended daily allowances for vitamin D to improve bone health and well-being, and avoid deficiency are illustrated in Table 3. Bearing this in mind, plus the putative benefits of vitamin D in the pathological processes involved in diabetes, it would seem reasonable that increasing concentrations of 25(OH)D in individuals with hypovitaminosis D might prove effective in the prevention and treatment of type 1 and type 2 diabetes.

However, there are conflicting data relating to the prevention and treatment of both type 1 and type 2 diabetes with vitamin D. Although studies have revealed possible preventative effects and improved glycaemic control in type 1 diabetes (Aljabri et al, 2010), others have failed to show any positive effect. Similarly, some have reported improved glycaemic control and reduced risk of progressing to type 2 diabetes (Talaei et al, 2013; Dutta et al, 2014) whilst others have reported no change in HOMA-IR or HbA_1c in a deficient population with doses of approximately 1000–2000 IU daily (Heshmat et al, 2012; Breslavsky et al, 2013; Ryu et al, 2014; Strobel et al, 2014). A recent meta-analysis revealed there was no overall benefit for multiple outcomes, including diabetes, following vitamin D therapy (Theodoratou et al, 2014).

Limitations of vitamin D studies
It is difficult to draw firm conclusions regarding specific outcomes as caution needs to be exercised in the interpretation of such meta-analytic data. Studies relating to vitamin D therapy are particularly heterogeneous regarding drug therapy used (ergocalciferol or colecalciferol), doses used, participant selection and the criteria defining vitamin D deficiency. Studies that are included in meta-analyses that have used ineffectually low doses or people with potentially normal vitamin D concentration create bias towards the null effect. Thus, we await the results of well-structured randomised studies to determine the extra-skeletal effects of vitamin D supplementation particularly among people with diabetes (Manson et al, 2012; Menon et al, 2013).

Musculo-skeletal benefits of vitamin D treatment
Whilst there are no clear indications for the use of vitamin D in the prevention and treatment of diabetes, there may be a number of reasons why vitamin D therapy should be considered in people with diabetes. Vitamin D deficiency is associated with a plethora of symptoms ranging from muscle weakness, arthralgia and depression to skeletal deformity as a results of rickets, and it is clear that symptoms attributable to vitamin D deficiency resolve with appropriate vitamin D supplementation.

As vitamin D deficiency is more common among people with diabetes, if symptoms are potentially attributable to vitamin D deficiency, then this should be investigated and treated according to appropriate guidance (National Osteoporosis Society, 2013).

Bone quality and fracture risk
Poorer bone quality with increased fracture risk is recognised in people with diabetes (Saito et al, 2014) despite lower FRAX scores (Carnevale et al, 2014) and apparently adequate bone mineralisation (Hamilton et al, 2013). While increased fracture risk is likely to be multifactorial, vitamin D deficiency may contribute particularly among older people (Lechleitner et al, 2013).

Evidence points to vitamin D therapy in doses at, or above, 800 IU daily for individuals aged 65 and over being beneficial for musculo-skeletal health with regard to the prevention of falls (Bischoff-Ferrari et al, 2009; Panel on Prevention of Falls in Older People et al, 2011) and fractures (Bischoff-Ferrari et al, 2012). Hip fracture in particular is associated with high morbidity and mortality (Kanis et al, 2003) and prevention may partly contribute to reduced mortality seen with vitamin D supplementation in recent meta-analyses (Zheng et al, 2013; Chowdhury et al, 2014). Thus, consideration should be given to identifying and treating those people with diabetes who have vitamin D deficiency.

Myopathy
Myopathy is a well-recognised complication associated with type 1 diabetes and type 2 diabetes, is often multifactorial and may also contribute to the increased risk of falls and fractures in people with diabetes. Although proximal myopathy is associated with vitamin D deficiency and the development of osteomalacia, vitamin D deficiency may be overlooked as a cause of myopathy in people with diabetes. Yet, following treatment, improvements can be expected (Kocián, 1992; Prabhala et al, 2000; Sitta et al, 2009).

Improved muscle performance as a result of vitamin D treatment may be a consequence of changes in muscle fibre type and function (Sanders et al, 2014), as well as improved muscle energy utilisation (Rana et al, 2014). Vitamin D deficiency has also been associated with statin-induced myalgia (a type of myopathy).

As the majority of people with type 2 diabetes receive statins, studies suggest that vitamin D deficiency may contribute to further myopathy, and vitamin D replacement has been reported to reduce symptoms (Ahmed et al, 2009; Glueck et al, 2011; Palamaner Subash Shantha et al, 2014).

Vitamin D prescribing: As simple as it seems?
A number of guidelines have been developed to guide investigation and treatment of vitamin D deficiency (Holick et al, 2011). In the UK, a popular guideline is that of the National Osteoporosis Society (2013), which offers a pragmatic approach to the investigation and treatment of the condition: Vitamin D and Bone Health: A Practical Clinical Guideline for Patient Management.

There has been much debate regarding treatment with either the plant-derived vitamin D2 (ergocalciferol) or the animal-derived vitamin D3 (colecalciferol). However, colecalciferol is the agent of choice, as it is the molecule synthesised in humans, appears to achieve better and longer-lasting 25(OH)D concentrations than ergocalciferol and has more published data relating to its effects (National Osteoporosis Society, 2013).

Oral therapy is the mainstay of treatment, being cheaper and more convenient than parenteral therapy (the latter being reserved for those with malabsorption). Similarly, treatment of vitamin D deficiency does not usually require the co-prescription of calcium as the majority of people with diabetes, including older people, have adequate dietary calcium intake (Francis, 2008), unless patients are felt to have inadequate dietary calcium intake. It is important to note, however, that 1-alpha colecalciferol should not be considered to treat vitamin D deficiency as it is associated with a high incidence of hypercalcaemia and is reserved for people with renal failure incapable of 1-alpha-hydroxylation.

What dosage to prescribe?
High intermittent doses, such as a cumulative dose of vitamin D 300000 IU over a 5- to 10-week period, are advocated for “rapid” correction of deficiency, but there is rarely a requirement for rapid correction. Hence, daily doses of 3000–4000 IU over a 10- to 12-week period are an acceptable alternative. This approach may be preferable, particularly in older people, where high intermittent doses have been paradoxically associated with increased risk of falls and fractures (Sanders et al, 2010; Turner et al, 2013).

It is also important to appreciate that following treatment of vitamin D deficiency, maintenance treatment is required to prevent relapse of deficiency with daily doses of 800–2000 IU or weekly 20000 IU doses being appropriate. Such approaches will achieve and maintain a target 25(OH)D concentration above 30 µg/L in the majority of people. Higher doses of colecalciferol may be required for obese people as vitamin D is stored in fat tissue, hence there is a requirement for larger replacement doses with increased fat mass. Similarly, higher doses of colecalciferol  may be required when co-prescribed with certain drugs such as anticonvulsants, owing to increased metabolism.

What vitamin D product should be prescribed?
Once a decision to treat with vitamin D is made, it is important to give consideration to the choice of vitamin D product. Recent UK data reveal that many prescriptions for vitamin D are fulfilled by unlicensed vitamin D products (Davies and Poole, 2014). Superficially this may seem unimportant, yet there are quality issues relating to unlicensed vitamin D products that may expose the patient to risk. Studies reveal large errors in stated versus actual dose with significant under- and over-estimation (Garg et al, 2013; LeBlanc et al, 2013). The former may be associated with lack of resolution of symptoms, and the latter associated with potential toxicity. With regard to the latter, cases of vitamin D toxicity requiring hospitalisation with potentially life-threatening hypercalcaemia are reported in adults (Koutkia et al, 2001; Araki et al, 2011; Benemei et al, 2013) and children (Kara et al, 2014) due to errant manufacture of unlicensed vitamin D preparations.

Thus, with NHS Prescription Cost Analysis Data from the UK (Davies and Poole, 2014) revealing that licensed vitamin D products are cheaper than the comparable unlicensed alternatives, it is important to ensure the quality of the product for both economic and safety reasons. Indeed, the Medicines and Healthcare products Regulatory Agency (2009) stipulates that unlicensed products should not be used where licensed products exist.

Conclusion
Many pathophysiological studies implicate vitamin D deficiency in the development and progression of type 1 and type 2 diabetes; however, the evidence in support of a therapeutic role for vitamin D in improving glycaemic control or preventing diabetes is confused. However, in a group at high risk of falls and fractures, such as older people with diabetes, vitamin D should be considered as a generally safe and effective therapy in doses at, or in some cases above, 800 IU daily. Indeed, the beneficial effects of vitamin D therapy appear most marked in the older person, offering potentially reduced mortality.

Large well-designed randomised controlled trials in people with vitamin D deficiency may well answer remaining questions on its association with diabetes. Nonetheless, with a higher prevalence of vitamin D deficiency amongst the diabetes population, evidence does support the use of vitamin D therapy in the treatment of the symptomatic individual.