Vitamin D Consensus Remains Elusive Despite Recommendations

Nancy A. Melville

October 22, 2012

MINNEAPOLIS, Minnesota — The report by the Institute of Medicine on the dietary intake of vitamin D was designed to help clear up confusion, but it hasn't.

The report, which was discussed here at the American Society for Bone and Mineral Research (ASBMR) 2012 Annual Meeting, states that supplementation levels of 600 to 800 IU/day will meet the needs of 97% of the population. However, there is little consensus on this; other guidelines recommend up to 1500 IU/day.

As clinicians continue to grapple with conflicting guidelines, bone expert Neil Binkley, MD, associate director of the Institute on Aging and codirector of the Osteoporosis Clinical Center and Research Program at the University of Wisconsin, Madison, offered a few unofficial guidelines of his own to in keep in mind when advising patients.

He urged clinicians to remember that the IOM report represents a "big picture" perspective, which might have little to do with the patient sitting across from you.

"The IOM report even stated that a margin of safety for public recommendations is prudent. I think most clinicians would agree with that." The recommendations, therefore, should not be regarded as being etched in stone, he asserted.

"Diagnostic end points are simply lines in the sand, and soft diagnostic cut-points are necessary to practice medicine in all fields," he explained.

Interpreting 25(OH)D Levels

Consider, for instance, the fine line between osteoporosis and osteopenia, Dr. Binkley said.

"A 65-year-old woman with a T-score of –2.4 has osteopenia, but her 10-year fracture probability is 13%. The same woman with a T-score of –2.6 has crossed the line in the sand to osteoporosis, but her fracture risk over 10 years is 14%."

"Similarly, our patients with 25-hydroxyvitamin D (25[OH]D) levels of 29 ng/mL are not different from those with 31 ng/mL."

Although serum 25(OH)D levels are currently recognized as the gold standard for defining vitamin D levels, the physiologic effect of different levels on different people is not well understood.

"We can titrate thyroid-stimulating hormone levels to optimal levels of thyroid replacement, but this is not the case with 25(OH)D. This is simply a blood level," Dr. Binkley emphasized.

Assay Variability

In addition, he noted, assays should be approached with a healthy dose of skepticism.

For one thing, assays have substantial variability. "The tests are administered by a person with an instrument; neither is perfect, and there will be much variability," he said.

He described a study in which he and his colleagues sent serum samples to 8 laboratories that used various 25(OH)D assay methods (Clin Chim Acta. 2010;411:1976-1982). They found significant interlaboratory variability.

The Vitamin D External Quality Assessment Scheme (DEQAS) is currently working to standardize lab assessments and reduce that variability. This is difficult because the matrix effect introduces many confounders to throw off assay readings, Dr. Binkley explained. Among the confounders are 24,25(OH)D (present in 10% of the 25[OH]D serum levels), 25(OH)D3-sulfate, and the metabolites of both ergocalciferol and cholecalciferol.

Clinical Application

Although standardization efforts should help address many of the issues related to the assessment of 25(OH)D, Dr. Binkley likened the current situation to the very early stage of lipid analysis. "I would suggest we are, today, in 25(OH)D measurement where we were 50 or 60 years ago with lipid measurement," he said.

"We need to link outcomes to blood levels that are achieved and we need to understand what analytes to measure. Until we do that, meta-analyzing data isn't going to answer the question of how much is enough for your patient or mine," he noted.

There is little disagreement about the risks posed by levels of vitamin D that are too low (e.g., rickets, osteomalacia, and fractures) or too high (e.g., hypercalcemia, hypercalciuria, and fractures).

"Too little is bad and too much is bad, so it's important to let moderation and clinical judgment be your guide," he said.

It is also important to factor in the known variability and potential confounders when assessing the vitamin D status of a patient. "If you get a measurement for a patient that is, for instance, 20 ng/mL, recognize that the real value is likely somewhere between 10 and 35 ng/mL," Dr. Binkley said.

"For many patients, 1000 to 2000 IU of vitamin D daily is required to maintain a 25(OH)D level at 30 ng/mL or above," Dr. Binkley said. "In my opinion, vitamin D inadequacy is common, but I think fixing this is cheap and virtually side-effect free."

Bone specialist Ian Reid, MD, professor of medicine and endocrinology at Auckland Medical School in New Zealand, warned against pushing levels too high. He cited a study in which women who received a single annual dose of 500,000 IU of cholecalciferol were at a significantly increased risk for fracture (JAMA. 2010;303:1815-1822). The significance of this study is that if you take a group of people who are at about 20 ng/mL and you put them up to 40 ng/mL, you increase fractures and falls by about 20%," Dr. Reid said.

Dr. Binkley countered that "there is a difference between the administration of 500,000 IU/year of vitamin D and daily administration." "I don't think we understand the mechanisms by which a blast of vitamin D leads to an increased risk of falls and fractures immediately following that," he said.

Nevertheless, Dr. Reid said he supports a more conservative approach. "I think the evidence is that we should be happy if people are around 15 to 20 ng/mL," he said. "Those who are lower should be supplemented with 400 to 800 IU/day. This approach is safe and does not make assumptions about the benefit of higher levels of vitamin D that have not yet been demonstrated in clinical trials."

Dr. Binkley and Dr. Reid have disclosed no relevant financial relationships.

American Society for Bone and Mineral Research (ASBMR) 2012 Annual Meeting. Presented October 14, 2012.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....