Liquid chromatography-mass spectrometry reveals lower vitamin D dose needed for bone health

ORLANDO, Fla. — The recommended daily allowance for vitamin D intake decreases by 50% when serum 25-hydroxyvitamin D is measured using a newer immunoassay, according to study findings presented here.

“It has been suggested by the Institute of Medicine, the World Health Organization and the European Food Safety Authority that a serum 25-(OH)D level of 20 ng/L is what is needed to protect bone health,” Lynette M. Smith, PhD, a biostatistician with the division of endocrinology at Creighton University School of Medicine in Omaha, Nebraska, said during a presentation. “Previously, we reported in a dose-response study that 800 IU per day would meet the requirement for 97.5% of the population ... but this result was based on the Diasorin immunoassay. Recently, we have reanalyzed the screening and 12-month data using liquid chromatography mass spectrometry (LC-MS/MS), which has become the gold standard for 25-(OH)D measurements.”

In a randomized, double blind, placebo-controlled study, Smith, J. Christopher Gallagher, MD, professor and director of the Bone Metabolism Unit at Creighton University School of Medicine, and colleagues analyzed data from 163 postmenopausal, white women aged 57 to 85 years with a 25-(OH)-D level 20 ng/mL or less (mean age, 67 years; mean BMI, 30.2 kg/m²). Researchers randomly assigned women to one of seven different vitamin D dose groups: 400, 800, 1,600, 2,400, 3,200, 4,000 or 4,800 IU daily, or to receive a placebo for 1 year. Calcium intake was assessed via 7-day food diaries and increased to 1,200 mg daily if necessary. Researchers measured the change in serum 25-(OH)D across doses over the course of the study using both the Diasorin immunoassay and LC-MS/MS, which measures multiple vitamin D metabolites after chromatography separation.

J. Christopher Gallagher

Researchers then looked at the agreement between the two techniques, Smith said, using scatterplots as well as Bland-Altman difference plots, and calculated limits of agreements for the two methods. Researchers then re-estimated the dose-response curve based on the LC-MS/MS data using a mixed effects model and formulated 95% prediction limits for the 12-month value for 25-(OH)D based on 1,000 bootstrap samples to estimate the recommended daily allowance (RDA). The primary outcome was to estimate the RDA for vitamin D intake based on exceeding a serum 25-(OH)D value of 20 ng/mL as measured by LC-MS/MS after 1 year of treatment.

Researchers found that measurements using both techniques were correlated (r = .93); however, compared with LC-MS/MS measurements, the Diasorin immunoassay underestimated serum 25-(OH)D concentration on average by 3.7 ng/mL below levels of 30 ng/mL, Smith said. While both assays showed a curvilinear increase in serum 25-(OH)D with dose that best fit a quadratic model, the quadratic curve was higher using LC-MS/MS. In addition, bootstrap prediction intervals showed a vitamin D dose of 400 IU daily would exceed the serum 25-(OH)D level of 20 ng/mL; a vitamin D dose of 800 IU would exceed a 25-(OH)D level of 30 ng/mL.

“When the samples were remeasured by LC-MS/MS, the 25-(OH)D values read 4 ng/mL higher, and this shifted the dose response curve higher,” Gallagher told Endocrine Today. “As a result, we found that 400 IU daily vitamin D would meet the RDA, based on a serum 25-(OH)D level of 20 ng/mL.”

“In estimating the RDA for vitamin D intake, we need to consider the method of measuring serum 25-(OH)D in this calculation,” Smith said while presenting the findings. “Based on the Diasorin assay system, the RDA for bone health was 800 IU per day, whereas with LC-MS we found 400 IU per day would be adequate, or potentially 1,600 IU per day if looking at a cutoff of 30 ng/mL. This has important ramifications for public health recommendations.” – by Regina Schaffer

Reference:

Gallagher JC, et al. OR07-4. Presented at: The Endocrine Society Annual Meeting; April 1-4, 2017; Orlando, Fla.

Disclosures: Gallagher and Smith report no relevant financial disclosures.