Madam

The new Institute of Medicine (IOM) guidelines⁽¹⁾ for vitamin D are a step in the right direction to indicate a greater amount of vitamin D is needed than previously thought; however, there are a number of shortcomings and unanswered questions.

First, the minimum daily requirement has tripled from 5 to 15 μg/d for bone health. This information is welcome. This would bring most people in the general population to a 25-hydroxyvitamin D (25(OH)D; the metabolite measured for status) level >50 nmol/l, according to this report. However, this is not an adequate cut-off since maximum absorption of Ca improves up to about 80 nmol/l⁽Reference Heaney, Dowell and Hale²⁾, which would in turn improve bone health. Parathyroid hormone (PTH) levels increase rapidly with 25(OH)D levels <50 nmol/l, but there are clinical studies that show a gradual rise in PTH with levels of 25(OH)D <78 nmol/l⁽Reference Chapuy, Preziosi and Maamer³⁾. Thus, the cut-off should be 80 nmol/l, not 50 nmol/l, and many researchers across the world would agree with this. The Canadian Osteoporosis Society recommends achieving >75 nmol/l with 20 μg of vitamin D daily but acknowledges that intakes up to 50 μg/d are required⁽Reference Hanley, Cranney and Jones⁴⁾. Dental health would be improved in all people with levels above 20 μg/d, as 10 or 15 μg/d did not show any benefit⁽Reference Schwalfenberg⁵⁾. This has been known since the 1930s and 1940s but has not been addressed.

Second, to say that most people have adequate levels from diet, even for bone health, using the conservative cut-off of 50 nmol/l is certainly not true. This is especially so in Canada where the latitude and long winters contribute to the low vitamin D levels. Two studies show that many population groups in Canada have very low levels of vitamin D and about 18% of Canadians have levels below 40 nmol/l⁽Reference Genuis, Schwalfenberg and Hiltz^6, Reference Schwalfenberg, Genuis and Hiltz⁷⁾. In the Canadian Health Measures Survey, respondents who were not white had 25(OH)D levels 20 nmol/l lower than those of white European origin⁽Reference Hanley, Cranney and Jones⁴⁾. Supplementation with vitamin D at 50 μg/d in a nursing home setting, where levels average about 35–40 nmol/l because of little or no sun exposure, did not achieve levels over 80 nmol/l in 6% of the population studied and did not result in any toxic levels or elevation of Ca⁽Reference Schwalfenberg and Genuis⁸⁾.

Third, the IOM did not address the needs in pregnancy. The Canadian Pediatric Society recommends all pregnant women take 50 μg/d, which is only reasonable since this group has very low vitamin D levels and consequences are grave if vitamin D levels are not adequate⁽⁹⁾. Low vitamin D levels have been associated with pre-eclampsia⁽Reference Bodnar, Catov and Simhan¹⁰⁾ and bacterial vaginosis in pregnancy⁽Reference Bodnar, Krohn and Simhan¹¹⁾. The use of 50 μg/d in the first year of life has been shown to reduce the development of type 1 diabetes by 80% over the next 30 years⁽Reference Hypponen, Laara and Reunanen¹²⁾. The dose recommended for pregnant women in the IOM report is only 15 μg/d, which would be inadequate.

Finally, the increase of the upper tolerable level of vitamin D from 50 to 100 μg/d is very welcome and will result in the ability to perform studies that use appropriate doses of vitamin D for the bone and for some, but not all, non-bone effects. However much of the rhetoric surrounding the release of the report concerned risk of taking vitamin D supplements, a risk that is not based on good evidence. The IOM committee had some concerns about the U-shaped curve in a number of studies where levels above 100–125 nmol/l showed a possible increase in cardiovascular deaths⁽Reference Melamed, Michos and Post^13, Reference Michaelsson, Baron and Snellman¹⁴⁾. Many researchers question this effect, believing that high levels of Ca as well as high levels of vitamin D get us into trouble. With adequate levels of vitamin D, the need for more than 800 mg Ca daily for bone health is really questionable⁽Reference Steingrimsdottir, Gunnarsson and Indridason¹⁵⁾.

What should physicians do? The Canadian Cancer Society has already suggested the use of 25 μg vitamin D daily for all Canadian adults for the prevention of cancer⁽¹⁶⁾. This should be expanded to include all Canadian children over the age of 1 year. Those 12 months and younger should take only the 10 μg/d dose recommended. Will vitamin D at this level have an impact on cancer prevention? The effect of vitamin D on reducing risk of bowel cancer is well established in the literature and it is surprising that this was totally ignored by the IOM. Even the WHO's ultra conservative body, the International Agency for Research on Cancer, has recently accepted that insufficiency of vitamin D increases colon cancer⁽¹⁷⁾.

But physicians should not fall victim to ‘more is better’ or use inappropriately high doses of vitamin D. The use of 12 500 μg yearly for osteoporosis resulted in an increase in falls/fractures⁽Reference Sanders, Stuart and Williamson¹⁸⁾. This dose makes no physiological sense; such a high level of vitamin D would result in some displacement from the vitamin D-binding receptor of the active hormone, causing a rise in the free active hormone to a degree that the body would actively degrade the hormone, resulting in a transient lowering of 1,25-dihydroxyvitamin D₃ and thus give the opposite result. This kind of dosing should never be recommended again. However, the use of 250 μg/d, which is physiological, has been shown to improve outcomes when used as an adjunct for tuberculosis therapy⁽Reference Schwalfenberg⁵⁾. One should remember that sanatoriums were considered part of therapy for tuberculosis only 50 years ago.

The recommendations that seem to make the most sense and which come from both ends of the life cycle are between 20 and 50 μg of vitamin D daily. It seems Canadians need to lead the way. How much longer do we need to see needless morbidity and mortality? In looking at the North American continent as a whole the Canadian average blood levels may provide a better model for northern US states than US averages, which must be weighted by the large southern population. The savings in health-care dollars each year in Canada⁽Reference Grant, Schwalfenberg and Genuis¹⁹⁾ and the USA⁽Reference Grant, Garland and Holick²⁰⁾ have been estimated in the billions and would save many lives.