Commentary by Michael R. McClung, MD, FACP
The average serum level of 25-(OH)D in healthy US adults who do not take vitamin D supplements or experience extensive sun exposure is at or lower than 20 ng/ml (50 nmol/L), a value below which impaired skeletal and muscle function has been demonstrated. Thus, many if not most healthy adults may benefit from vitamin D supplementation.
Controversy, or at least uncertainty, exists about the appropriate target for serum 25-(OH)D levels. The 2010 Institute of Medicine (IOM) report suggested that a target of 20 ng/ml was appropriate for healthy adults. This recommendation was based on the limited evidence available documenting fracture risk reduction in randomized, prospective trials. It is also consistent with the even more limited evidence for reduction in fall frequency or cancer incidence where the serum levels achieved in the treated groups were about 20 ng/ml. Cross-sectional studies with physiologic endpoints suggest that a serum level of 30 ng/ml may be a reasonable target.
Regardless of the target chosen, we have very little evidence about the dose of vitamin D supplement required to have most individuals achieve that level. The study by Gallagher et al provides important information to address this question. In a carefully designed dose-ranging study in healthy postmenopausal women, baseline values ranged between 8 ng/ml and 20 ng/ml. Daily supplements of 800 IU vitamin D (and extrapolated results with 600 IU/d) resulted in a serum level of 25-(OH)D of 20 ng/ml in more than 95% of subjects, consistent with the IOM recommendations. A serum level of 30 ng/ml was achieved by almost all subjects with a daily dose of 1,600 IU. Little additional increase in 25-(OH)D occurred with intakes greater than 2,400 IU daily. Substantial variation in serum levels achieved by individuals taking the same dose was evident; 25-(OH)D values in subjects taking 800 IU daily ranged between 20 ng/ml and 50 ng/ml. As expected, 25-(OH)D levels while receiving supplements were modestly lower in obese subjects, but all of the obese individuals reached a serum level of 20 ng/ml on 1,600 IU daily.
The Gallagher et al data do not resolve the uncertainty of the optimal target for serum 25-(OH)D or optimal dose of supplement. We may learn that increasing serum levels higher than 30 ng/ml is clinically beneficial. Many observational studies demonstrate a correlation between decreased serum 25-(OH)D levels and many medical problems. However, recent meta-analyses suggest that the relationship between serum 25-(OH)D and mortality is U-shaped, with a progressive increase in the risk of death observed with values greater than 50 ng/ml. The Gallagher et al data support the recommendation, made in the 2010 NAMS Position Statement, to provide supplements of 800 IU to 1,000 IU of vitamin D daily. This is a reasonable position until we have solid evidence that achieving higher serum levels results in clinical benefit. Even with the modest doses evaluated by Gallagher et al, some patients experienced hypercalcemia and hypercalciuria, adding an important element of caution to the use of higher doses of vitamin D. In my opinion, the Gallagher et al data also add support to the sentiment that routine testing of serum 25-OH D is not necessary or justified in most healthy adults. The pretest probability is high that low values will be found in healthy adults and that a daily intake of 800 IU to 1,000 IU daily will result in serum levels within the current target range of 20 ng/ml to 50 ng/ml in virtually all adults. Testing is appropriate in patients with confounding problems such as gastric surgery or small bowel abnormalities, drugs that increase hepatic clearance of vitamin D (older anti-seizure medications, cimetidine), and patients with disordered calcium metabolism like hyperparathyroidism.
Enthusiasm among many practitioners for use of and the potential benefits of high doses of vitamin D supplements has outstripped the available evidence supporting that usefulness. Our recommendations to our patients must be based on solid evidence, not on hopes or hypotheses. We need more studies such as this one that address the practicalities of these issues, and then we all need to pay attention to those results.
From the NAMS First to Know e-newsletter released May 24, 2012
For more, please visit http://www.menopause.org/news.aspx