This paper will discuss the interaction between the host, the agent, and the environment as depicted in Figure 1. The environmental factor being considered is UVB radiation induced vitamin D or supplemental vitamin D. Five autoimmune diseases will be discussed. Multiple sclerosis, type 1 diabetes, rheumatoid arthritis, autoimmune disease of the thyroid, and inflammatory bowel disease.

In the last twenty years, the importance of vitamin D in the role of a hormone has been shown to influence numerous other diseases including cancer by increasing apoptosis in cancer cells and protecting DNA in normal cells [16]. Its effect on the immune system and infections is only beginning to be understood, and much higher doses of vitamin D may be needed to be effective in combating viruses, bacteria, and fungi [17]. Vitamin D is now recognized to be crucial in beta-defensin production in Crohn's disease [18].

Multiple environmental factors may play a role in multiple sclerosis. An example is the Epstein-Barr virus infections [22]. The risk of developing multiple sclerosis following infectious mono is increased significantly for more than 30 years following infection [23]. Even after 10 years following infection, the risk has been defined as at least four-fold [24]. Sunlight and vitamin D may be protective, and MS demonstrates vitamin D sensitivity [25].

Evidence shows that MS correlates positively with higher latitude, with latitudes >37.5 degrees from the equator having significantly higher rates of MS [26–29]. The time of the first exacerbation after disease onset shows seasonal variation with 76% of exacerbations occurring in winter [30]. However, in another study, the likelihood and intensity of MS disease activity correlated positively with spring and summer (March to August) along with increased temperature and UVR [31]. Relapses of MS have a biphasic pattern with peaks in early spring when vitamin D levels are low and late fall when levels are declining [32]. MS also correlates positively with season of birth, with a significantly higher incidence of MS in those born in May, corresponding to low vitamin D levels in the winter months prior to giving birth [33]. As well, MS correlates inversely with altitude, with higher elevation (>2000 Meters) receiving more intense solar radiation having lower rates of MS [34]. Adiposity has been associated with lower vitamin D levels [35, 36], and a higher body mass index (BMI) has been associated with higher incidence of MS in adolescent women but not in adult women [37]. Multiple sclerosis correlates positively with skin color where sun avoidance is more likely (fair skin phenotype) [38]. In Norway, the intake of vitamin D in costal communities is estimated to be three times the average intake of those living inland and this is inversely associated with MS [39].

In the past rubella infection has been implicated as an inducer of type I diabetes. Several studies now indicate infection with enteroviruses seem to be linked to the induction of islet cell destruction and development of autoantibodies [45, 46]. Stress in the mother during pregnancy has been associated with elevated islet autoantibodies in cord blood [47].

The following is a summary of the evidence that type 1 diabetes is a vitamin D-sensitive disease. Type I diabetes varies by latitude with higher incidence in latitudes further from the equator (both north or south of the equator) except where local weather may influence this [48]. A yearly cyclical pattern (consistent over 20 years) of type 1 diabetes in Newfoundland reveals a peak incidence in winter [49]. Greater exposure to erythemal UVB radiation is negatively correlated with the incidence of type 1 diabetes [50]. The birth months of March to June correlate with an increased incidence of diabetes in Britain with prenatal exposure to low vitamin D levels during the winter months [51]. The average yearly temperature also correlates inversely with the incidence of type 1 diabetes [52]. Higher BMI as well as lower plasma levels of 25(OH)D correlate directly with development of type 1 diabetes in young adults [53, 54]. Maternal vitamin D levels have been shown to correlate inversely to the presence of islet autoantibodies in the offspring [55]. Supplementation of vitamin D in early childhood appears to reduce the chances of developing type 1 diabetes [56].

The association of infections and RA is still being debated in the literature. There is some evidence that RA-like diseases may result from the immunologic interaction between the host and bacterial peptidoglycans. Several agents relative to RA have been entertained such as heat shock proteins, bacterial IgG FC-binding proteins and rheumatoid factors [64]. Retroviruses and enteropathogenic bacteria continue to be intensively discussed candidates [65]. IgM antibodies to parvovirus B19 indicating recent infection correlate with juvenile idiopathic arthritis [66]. Certainly smoking has been cited as being a major risk factor both in RA and SLE as well as a risk factor for RF-positive and anti-citrulline antibody titers [67]. The risk only diminishes slowly after several years of cessation of smoking.

Environmental factors are numerous and include iodine excess and deficiency, selenium deficiency, oral contraceptive use, parity, low birth weight, seasonal variation, allergy, smoking, radiation, viral and bacterial infections, and so forth [78]. Overall, data suggests that cumulative cigarette consumption significantly increases the risk of developing AITD [80]. Hashimoto's thyroiditis is associated with vitamin D insufficiency with AITD patients having significantly lower levels than controls [81, 82]. In AITD 72% of patients had 25(OH)D levels less than 25nmol/L compared to 30.6% in healthy individuals. In patients with Hashimoto's thyroiditis, 79% had levels less than 25nmol/L compared to 52% in normal controls.

An environmental trigger shown to increase the risk of development of CD is smoking. Many other triggers have been proposed such as infectious gastroenteritis, oral contraceptives, invasive E. coli, and antibiotics [86]. Vitamin D insufficiency is associated with inflammatory bowel disease [87, 88]. In another study 25(OH)D levels were lower in those with severe disease activity and less sun exposure [89].

A study in France has shown that there is an increasing incidence of UC with northern latitude [90]. Smaller studies did not show any seasonal pattern for IBD, but a larger study showed an increase incidence in those born in the first half of the year [91] while others show an increased incidence of exacerbations in summer. There is no information on altitude or temperature. Obesity is more common in Crohn's disease at the time of diagnosis [92]. Vitamin D deficiency is very prevalent in IBD [93] and the effects of low-dose supplementation are poor [94]. A clinical trial using 1200IU of vitamin D₃ for 12 months resulted in a reduced risk of relapse from 29% to 13%, although this did not quite reach statistical significance [95]. Other trials are currently under way.

Vitamin D has been shown to improve outcomes and prevent some autoimmune diseases if taken early in life as discussed previously in the paper.

Treatment of ITP usually includes medications such as corticosteroids, splenectomy danazol, and various immune suppressant therapies [97]. The use of vitamin D₃ in ITP has not been described in the literature.

This example describes a patient who had refractory ITP who has been treated in the past with a splenectomy, danazol, and prednisone rescue during intercurrent illness. A review of the history of this case revealed a 48-year-old female who was found to have a very low platelet count in 1998, which remained persistent over time. After consultation with a hematologist, the diagnosis of ITP was made. Her platelets continued to drop so she had a splenectomy, which improved her platelet count, but it never achieved normal. Despite the use of danazol, her platelet count never normalized. She had frequent episodes of low platelets as low as 8×10⁹ with intercurrent illnesses such as colds or flus. In 2006 she was found to have an inadequate level of 25(OH)D of 65nmol/L. Her platelet count at that time was 8×10⁹ after a viral illness. She was treated successfully with a tapering dose of Prednisone. She was started on vitamin D₃ 2000IU daily after this episode and during the next two years while she was on this dose she did not have any flus or colds and her platelet count never fell below 44×10⁹and was usually from 70 to 80×10⁹. This was quite out of the normal for her since she had at least one low episode a year. After being on this dose for two years, a neighbor in her building where she resided suggested that she was going to become toxic on this dose and she stopped her vitamin D. About three months later, she again had an upper respiratory infection (URI) and her platelets dropped only to 50×10⁹ and she was started on prednisone with recovery of her platelets to 140×10⁹. She was seen after this, and it was recommended that she resume her vitamin D₃ at a higher dose of 4000IU daily. She was feeling quite well, and her platelets remained above 70 and continued to rise so she discontinued her danazol. Again, she had no further episodes of flu or colds for the next two years and her platelets did not drop below 70. Her vitamin D level on this dose was 88nmol/L after 4 months. She phoned one day that she was sick with a URI, and it was suggested that she take 10,000IU of vitamin D for a 3 days and have her platelets checked. She had her platelets checked after being on this dose for two days, and they were in the normal range much higher than they had been for years. The platelet count was 248×10⁹. Her vitamin D level was 99nmol/L at this time. She continues on the 4000IU vitamin D, and she continues off the danazol and remains well. Her latest platelet count was 318×10⁹.

This case presents a significant response to vitamin D₃ in ITP. As well, it demonstrates a recurrent failure with lack of vitamin D and restoration of a normal platelet count on a higher dosing of vitamin D₃, which did not result in toxicity. The vitamin D₃ rescue with 10,000IU of vitamin D₃ appeared to result in a similar response as that of prednisone used in past treatments over the years. ITP has been shown to have spontaneous remission in some people; however, it is uncommon in older patients. Is it possible that restoration of vitamin D levels results in some of these cases of spontaneous resolution? At this time this is not known. Restoring vitamin D to a level that is safe appears to be sensible supportive therapy.

The link between various autoimmune diseases and vitamin D in various ecological, population, and case control studies is summarized in Table 4. Interventional studies when available have been cited throughout this paper. Although not all linkages are positive in each disease, the ones that have been studied the most, such as MS and type 1 diabetes, are quite striking. Studies done in type 1 diabetes were done in infants, and supplementation during the prenatal and first year of life may result in a significant reduction of morbidity later in life. Vitamin D may very well be a significant factor in preventing the loss of tolerance to self and resultant autoimmune disease.