Vitamin D Supplements: Is Bone Loss by pQCT Really Negative?

To the Editors: Recently Calgary researchers re-analyzed results by sex(1) of a 3-year trial in a mixed group of healthy adults aged 55 to 70 years. The study examined the effect of vitamin D supplementation on volumetric bone mineral density (vBMD) and other indicators by HR-pQCT.(2) Daily supplementation was 400 IU, 4000 IU, and 10,000 IU, but there was no placebo group. The latter two doses are the upper limits set by the Institute of Medicine (IOM) and the Endocrine Society, respectively. Baseline calcifediol averaged ~32 ng/mL and over 90% of the subjects had calcifediol >20 ng/mL; ie, the subjects were not abnormal. There were no significant decreases of vBMD in males at the three doses. Total vBMD in females decreased at the radius site by −1.8%, −3.8%, and −5.4% for the three doses, respectively, with corresponding tibia decreases about one-half those at the radius; ie, −1.0%, −2.0%, and −3.2%. Despite these decrements for the separate female group there was no significant decrease for failure load at both sites (calculated from finite element analysis of pQCT) nor for areal BMD (aBMD) of the proximal femur by dual-energy X-ray absorptiometry (DXA). The researchers cautioned that vitamin D supplements at the higher levels could be dangerous for use to prevent bone loss in the postmenopausal decades. That caution was not supported by the results. Moreover these higher doses are almost never used by clinicians for prevention of osteoporosis but they have been used in treating acute respiratory distress.(3) Currently larger supplements of vitamin D are also being debated for prevention of coronavirus disease 2019 (COVID-19) infection in high-risk groups and for its treatment in hospitalized patients.(4, 5) There are several reasons that the results do not warrant that warning. First, there was no control group of unsupplemented women. Moreover longitudinal results (5 years) from an earlier study by the group(6) showed that over 3 years, total vBMD of the radius in untreated women declined by 2.9%, while failure load decreased 0.9%. A study from Denmark over 3 years using the same HR-pQCT showed losses in unsupplemented females of 2.3% for radius vBMD and 2.7% for failure load.(7) Total radius vBMD loss in both studies was not significantly below that of women receiving 4000 IU suggesting that: (i) there was no significant skeletal effect of 4000 IU, other than stabilization of failure load; and (ii) the 400 IU dose to women with adequate calcifediol slightly reduced the average vBMD loss of ~2.6% in unsupplemented females to 1.8%. The stability of both femur aBMD and failure load associated with D supplementation indicates that both 400 IU and 4000 IU were no worse than and perhaps superior to a lack of supplementation. Second, it is unclear what caused the anomalous vBMD loss at 10,000 IU to be double the loss of unsupplemented women despite the stability of both femoral aBMD and calculated fracture load. Femur aBMD is not suspect because bone measurements (DXA and/or pQCT at the femur, spine, radius, and tibia) are stabile or show small increases with vitamin D supplementation but no significant loss.(8) Something possibly improved failure load, a well-validated index of strength that is not itself in question.(9) Studies show strength correlates better with bone mass of the total section than with derivative indices (size, porosity, and vBMD) of subareas (r ~0.95 versus 0.6 to 0.8) but it also depends on bone size. Failure load could be partially maintained in postmenopausal women, despite the usual bone loss through endosteal resorption, given augmentation of their usual (~6%) periosteal apposition.(10) If we assume the radius in those receiving 10,000 IU showed an annual 1% increase of its ~240 mm2 area (2.4 mm2) that would amount to 1-mm along the radius margin. Theoretically, that area increase, which decreases vBMD by 1% without change of mass, could stabilize or increase fracture load. It is unclear if that 1-mm change could be reliably measured with a 61-mm voxel if only because of partial volume effects. Unfortunately the failure to report bone mass and size has become a common "error of omission" in pQCT analyses that focus on vBMD and a myriad of derivative indices. Thus the vBMD "loss" of 5.8% with 10,000 IU, which amounted to only 1% annually above the 2.6% in unsupplemented women, most likely reflected in part stimulated periosteal apposition. Vitamin D supplementation could even be a positive for strength rather than of no import if an increase of bone size (and possibly cortical area) did occur. Studies using pQCT should address the question by providing both mass and areas for the total, cortical, and trabecular zones. Clinicians considering use of vitamin D supplementation at the upper limits need not be concerned because the Calgary study showed no decrease of bone strength and aBMD over 3 years despite questionable decreases of vBMD. Much greater concerns are the known risks of (i) hypercalciuria associated with calcium intake; and (ii) an increased rate of falls in the most elderly, particularly with bolus dosing. The author states that he has no conflicts of interest.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.