To ascertain whether vitamin D supplementation or vitamin D deficiency in infancy could affect the development of type I diabetes, a birth-cohort study was done in Oulu and Lapland, Finland.  All infants born in 1996 were studied (n = 12,055).  Data were collected on vitamin D supplementation and on the presence of suspected rickets during the first year of life.  The primary outcome measured was the diagnosis of type I diabetes by the end of 1997 (30 year follow-up).  Of the 10,366 children included in the analysis, 81 were diagnosed with type I diabetes.  Vitamin D supplementation was associated with a decreased frequency of this disease.  Children who took the recommended 2000 IU of vitamin D on a daily basis had a rate ratio of 0.22 of developing the disease, as compared with those who received no vitamin D.  The rate ratio in those who received a lesser amount of vitamin D supplementation was 0.12. Children suspected of having rickets during the first year of life had a rate ratio of 3.0 as compared with those without such diagnosis.  The authors concluded that vitamin D supplementation was associated with a reduced risk of type I diabetes. 

First Editor’s Comments: This is a very provocative study implicating the deficiency of one hormone (vitamin D) on the development of another hormone deficiency (insulin).  The mechanisms of such association were thought to be related to the triggering of an immune process resulting from the lack of vitamin D.  This is consistent with data from animal studies, and with the observation that cod liver oil supplementation during pregnancy is associated with a reduced rate of type I diabetes in the offspring.¹ The Eurodiab study also showed that vitamin D supplementation in early childhood may prevent this disease.²  However, only 0.3% of infants in the Eurodiab study were not given vitamin D during the first year of life, thus the comparative population was rather small. The increased prevalence of this disease (3x) among children in this Finnish study, who were suspected of having rickets, is impressive.  However the data are not very compelling since there was no radiologic or biochemical evidence of rickets presented.

The infants who took 2000 IU of vitamin D as a daily supplement had a 78% lower risk of developing diabetes.  This dose of vitamin D, however, is high and not recommended by most authorities. (The Committee of Nutrition of the American Academy of Pediatrics, among others, state that an adequate intake of this vitamin is 200 IU per day.)  Others have recommended dosages ranging from 400 to 1000u per day,³ where there may be lack of sunlight exposure, particularly during the long winter months in the northern hemisphere.  Although there is no single recommendation for the amount of vitamin D supplemented, exposure to the sun usually will satisfy the requirements to prevent rickets and vitamin D deficiency.  As little as 1 minimal erythemal dose (MED) of sunlight is equivalent to ingesting about 10,000 IU of vitamin D.  Simple exposure of hands and face two or three times per week provides a third to a half of the MED (about 5 minutes for fair-skinned people) is more than adequate.  Moreover, sunlight is without risk of hypervitaminosis D which may occur when large amounts of vitamin D supplements are ingested.  Thus, caution should be exercised to the possible temptation of increasing vitamin D supplementation in an attempt to prevent type I diabetes.  Further studies are needed and other studies should be undertaken to ascertain why there is a high prevalence of type I diabetes among other populations who also are exposed to insufficient sunlight such as those in Finland. 

Fima Lifshitz, MD

References

	Stene L, et al. Diabetologia 2000;43:1093-1098.
	The EURODIAB Substudy 2 Study Group Diabetologia 1999;42:51-54.
	Canadian Pediatric Society CMAJ 1988;138:229-230.

Second Editor’s Comment: In the early 19^th Century, cod liver oil was given to prevent rickets.  The classical role of vitamin D in the prevention of rickets is to assist absorption of calcium and phosphate. Vitamin D also appears to play a role in preventing some cancers and autoimmune diseases.  Ideally, in a study such as the one reported here, evaluation would include plasma 25(OH) D or 1,25(OH) 2D₃ concentrations. When sun exposure is limited, as in northern Finland, supplementation or dietary intake is an important source of vitamin D. Breast milk does not contain enough vitamin D to cover an infant’s needs.  The role of vitamin D in the pathogenesis of type 1 diabetes certainly deserves follow-up.  If vitamin D does impair the immune system functioning in infancy, there may be other long-term effects.  Interesting as well, Finland has the highest incidence of type 1 diabetes in the world.

Judith Hall, MD