Bone fractures are a major public health problem, especially for older adults, and vitamin D has long been seen as key to supporting bone health. But data on the effects of vitamin D supplementation on fractures has not been consistent, leading to questions about who needs tests for 25-hydroxyvitamin D, the proper serum levels, and differing recommendations.
In 2021, the United States Preventive Services Task Force found insufficient evidence to recommend screening for vitamin D deficiency for asymptomatic, community-dwelling adults who are not pregnant. More recently, a report from investigators in the large Vitamin D and Omega-3 Trial (VITAL) concluded that vitamin D supplementation performed no better than placebo at lowering fracture risk in a very large, randomized controlled trial (RCT) in generally healthy midlife and older adults (N Engl J Med 2022; doi: 10.1056/NEJMoa2202106).
“For the general population, widespread hydroxyvitamin D testing could be curtailed,” said the study’s first author, Meryl LeBoff, MD, chief of the Calcium and Bone Section in the Endocrine Division at Brigham and Women’s Hospital and professor of medicine at Harvard Medical School. “In the largest RCT of supplemental vitamin D versus placebo on incident fractures and bone health outcomes with 5.3 years of follow up, supplemental vitamin D did not reduce fracture risk in generally healthy midlife and older adults enrolled nationwide.”
Endocrinology and laboratory medicine experts do not yet agree on how the study results should affect hydroxyvitamin D testing. Meanwhile, laboratorians’ desire to provide accurate results should prompt detailed attention to both the study and differences between testing methods and platforms, given a continuing lack of standardization of hydroxyvitamin D tests and methods, experts said.
THE LARGEST VITAMIN D STUDY
The recent paper detailed an ancillary analysis in the VITAL study to determine whether 2000 IU/day of vitamin D would result in a lower risk of fractures than placebo among generally healthy men 50 years and older and women 55 and older in the United States. Notably, recruitment did not enroll participants on the basis of vitamin D deficiency, low bone mass, or osteoporosis.
Participants were diverse. They came from all 50 states, including almost equal numbers of men and women. And 20% were Black, among whom vitamin D deficiency is more common. The participants completed initial questionnaires about demographic characteristics, medical history, and medication and supplement use, plus annual questionnaires about adherence to the trial regimen, use of nontrial supplements, medications, major illnesses, and development of osteoporosis or its risk factors, physical activity, and falls.
Almost 17,000 participants gave baseline blood samples and about 6,000 gave follow-up samples. Quest Diagnostics measured 25-hydroxyvitamin D via liquid chromatography-tandem mass spectrometry assays calibrated to Centers for Disease Control and Prevention (CDC) standards.
Fractures occurred in 5.9% of patients who took vitamin D supplements, versus 6.0% of those who got placebos. Researchers found no difference in treatment effect according to age, sex, race or ethnic group, body mass index, or serum 25-hydroxyvitamin D levels. Stratification of patients by baseline 25-hydroxyvitamin D levels yielded no significant differences in fracture incidence according to various levels. Post hoc analysis found insufficient power to evaluate both effects of vitamin D supplements on fracture outcomes in the small number of participants with baseline 25-hydroxyvitamin levels less than 12 ng/mL and in participants on osteoporosis drugs or who had a history of fragility fractures at study entry.
Although the authors note the large, diverse study population and 90% vitamin D adherence as strengths, they also point to limitations. These include evaluation of only one vitamin D dose in a trial that was not designed to test effects of vitamin D supplementation in vitamin D-deficient patients, plus the fact that only 2.4% of participants had severe deficiency, with 25-hydroxyvitamin D levels less than 12 ng/mL.
CHANGES AHEAD?
Because of these limitations, LeBoff noted that the results apply only to midlife and older healthy adults who were not preselected for osteoporosis, severe vitamin D deficiency, and gastric and intestinal disorders that can interfere with the normal uptake of vitamin D.
Joely Straseski, PhD, DABCC, FADLM, predicted that these limitations mean the paper will have little effect on the volume of vitamin D testing in labs for now. She is a professor of pathology at the University of Utah School of Medicine and medical director of endocrinology at ARUP Laboratories in Salt Lake City.
A common misconception is that clinical decisions about testing and supplements should be made uniformly across patient populations, said Anastassios G. Pittas, MD, MS, cochair of the Endocrine Society’s vitamin D guidelines writing committee and chief of endocrinology at Tufts Medical Center in Boston. Decisions about vitamin D testing and supplementation should depend on the clinical reasons for them—such as risk for fractures or diabetes—and individual patients' risk for low vitamin D status, he added.
Forthcoming, updated vitamin D guidelines from the Endocrine Society will include population and condition-specific recommendations, Pittas noted. He was the primary investigator of the Vitamin D and Type 2 Diabetes Study (D2d), which showed that supplementation had significant benefit in diabetic patients with blood 25-hydroxyvitamin D levels less than 12 ng/mL (N Engl J Med 2019; doi: 10.1056/NEJMoa1900906). This finding “highlights the difference between supplementing people with low 25-hydroxyvitamin D levels versus further supplementing those whose levels are thought to be adequate,” Pittas said.
The new guideline authors will review the most recent VITAL paper and others before issuing updated vitamin D recommendations in the spring of 2024, said Marie Demay, MD, chair of the writing committee and professor of medicine at Harvard Medical School in Boston. Current guidelines issued in 2011 noted insufficient evidence to recommend screening individuals who are not at risk for deficiency or prescribing vitamin D for a cardiovascular protection benefit (J Clin Endocrinol Metab 2011; doi: 10.1210/jc.2011-0385). The updated Endocrine Society guidelines likely will include population and condition-specific recommendations, Pittas added.
The 2022 American Diabetes Association (ADA) guidelines note that D2d also showed no significant benefit of vitamin D versus placebo on the progression to type 2 diabetes in individuals at high risk. But post hoc analysis and meta-analysis suggest a potential benefit in specific populations. ADA calls for more research on patient characteristics and clinical indicators for vitamin D (Diabetes Care 2022; doi: 10.2337/dc22-S001).
Both Straseski and LeBoff agree that some risk factors warrant vitamin D testing. These include osteoporosis, taking osteoporosis medications, and gastrointestinal issues that drive malabsorption issues. Straseski noted that clinicians also may consider consider testing long-term hospital inpatients, nursing home and long-term care patients, and incarcerated or institutionalized people, all of whom do not get much sunlight.
A CALL FOR LESS TESTING
An editorial that accompanied the paper said that combined with other VITAL findings, the latest study delivers “a decisive verdict” against widespread 25-hydroxyvitamin D testing in the general population and treatment to a targeted serum level (N Engl J Med 2022; doi: 10.1056/NEJMe2205993). As a whole, the editorial said, VITAL findings have called into question vitamin D supplementation’s effects on preventing cancer or cardiovascular disease, improving cognitive function, reducing atrial fibrillation, changing body composition, reducing migraine frequency, improving stroke outcomes, decreasing age-related macular degeneration, or reducing knee pain.
The editorial also called for reconsidering vitamin D insufficiency and deficiency definitions. The current paper’s findings show there is “is no justification for measuring 25-hydroxyvitamin D in the general population or treating to a target serum level,” said coauthor Clifford J. Rosen, MD, in an interview. Rosen, who is director of clinical and translational research at the Maine Medical Center Research Institute in Scarborough, Maine, called for reevaluation of what constitutes an optimum level of vitamin D in serum, adding that no data from an RCT shows that a minimum level of 30 ng/mL prevents chronic disease.
Guidelines differ on 25-hydroxyvitamin D levels. Institutes of Medicine recommendations issued in 2010 estimated that a level of 20 ng/mL or higher was adequate for good bone health and consider a level below 20 as a deficiency. In 2011, the Endocrine Society urged a higher minimum blood level of vitamin D at 30 ng/mL.
The VITAL study does not provide sufficient evidence for changes to cutoffs or supplementation, Demay said. The parent study was neither designed nor powered to answer any of the ancillary study questions. The study population was not known to be at risk for cardiovascular disease or fractures and had already achieved a blood 25-hydroxyvitamin D level of about 31 ng/mL, so the study “cannot answer the question of whether broad testing and supplementation is unnecessary,” she added. “Furthermore, it is not clear whether screening and testing in routine clinical practice had contributed to the VITAL study population having achieved that level to begin with.”
The question of appropriate cutoffs is important and requires an answer from more definitive studies, Straseski noted. She is also concerned about harmonization and standardization of vitamin D tests. While tests that rely on mass spectrometry now compare well, immunoassays are not standardized across vendor platforms. CDC’s Vitamin D Standardization Certification Program can help vendors and laboratories improve analytical accuracy and reliability of vitamin D tests by looking at their analytical biases and precision, and grant certification to those that meet related criteria, Straseski said.
“Labs should understand the effects of the lack of standardization and the differences between platforms,” Straseski advised. “We still need to provide good, accurate results for those patients at high risk of fracture and vitamin D deficiency.”
Deborah Levenson is a freelance writer in College Park, Maryland. +Email: [email protected]