First Report®: 31st Annual Meeting of the American Society for Bone and Mineral Research Denver, CO; September 11-15, 2009

Decreased Serum Vitamin B12 Concentration is Associated With Greater Fall Risk Among Older Disabled Women: The Women’s Health and Aging Study

Denver, CO—Robert McLean, DSc, Hebrew SeniorLife Institute for Aging Research and Harvard Medical School, Boston, MA, and colleagues examined the associations of serum concentrations of folate, vitamin B12, methylmalonic acid (MMA) and homocysteine with the risk of falling among older disabled women participating in the Women’s Health and Aging Study. Results presented at the ASBMR annual meeting suggest that low vitamin B12 status may be associated with a greater risk of falls among older disabled women. Authors concluded that further studies are needed to explore this relation among a general population of older men and women.

While studies suggest that low B vitamin status and elevated homocysteine concentration are associated with increased fracture risk among older persons, the mechanism is unclear as low bone mass does not fully account for the observed relations. Folate and vitamin B12 deficiencies can lead to lower-extremity sensory and proprioceptive impairment, both of which increase the propensity for falling. An association of these nutrients with fall risk could support a mechanism whereby B vitamins and homocysteine increase fracture risk.

In 1992, non-fasting blood samples were collected from 657 women at baseline (mean age at baseline, 77 yr; range 65-100 yr). Radiodilution assay measured folate (nmol/L) and vitamin B12 (pmol/L), and MMA (nmol/L) and homocysteine (mmol/L) were assessed via mass spectrometry. Quartiles of serum measures were created (Q1 = low). Self-report ascertained the number of incident falls after baseline at six follow-up examinations occurring generally at 6-month intervals. Negative binomial regression calculated the relative rate (RR) of falls, and 95% confidence intervals (CI), over follow-up for quartiles of each serum measure, adjusting for age at baseline (yr) and the number of falls reported in the year prior to baseline (0, 1, ≥ 2). Median follow-up time was 3.1 years (range 0.5-5.6). Falling at least twice over follow-up was reported in 41% of women. Those in vitamin B12 Q1 and Q2 had a 40-57% increased rate of falls as compared to Q4 (referent). Participants in homocysteine Q4 tended to have a 25% increased rate of falls as compared to Q1 (referent), although the CI included the null value. According to the authors, neither folate nor MMA were associated with the rate of falls.
___________________________________________________________________________________________________________________________________

Low Vitamin D Predicts Bone Loss in Breast Cancer Patients Taking Aromatase Inhibitors

Denver, CO—Baseline 25-hydroxyvitamin D (25OHD) level is a predictor of aromatase inhibitor (AI)–associated bone loss, and women with low 25OHD are at risk for bone loss at 6 months of AI therapy, according to results presented at the ASBMR annual meeting. Researchers concluded that future studies are needed to determine whether aggressive vitamin D replacement will be able to prevent bone loss in women taking AIs for breast cancer. According to researchers, suboptimal vitamin D levels had been reported in as much as 88% of women with breast cancer who were on AI therapy. However, there are no studies investigating whether suboptimal 25OHD levels are an important determinant of bone loss in patients on AI.

Bone mineral density (BMD) was measured by dual energy x-ray absorptiometry (DXA) and 25OHD was measured by radioimmunoassay. Participants were advised to take 1200 mg of calcium and 800 IU of vitamin D daily. 161 postmenopausal women with breast cancer participated; 110 came for their 6-month follow-up, and 71 came for their 1-year follow-up. At 6 months, bone loss was 1.3±3.2% on the spine, 1.7±3.9% on the femoral neck, 0.76±2.3 % on the total hip, 1.2±3.4% on the trochanter, and 0.92±3.2% on the intertrochanter. At the end of 1 year, bone loss was 1.7±3.6% on the spine, 2.5%±4.7% on the femoral neck, 0.87±3.4% on the total hip, 1.5±5.0% on the trochanter, and 0.7±4.0% on the intertrochanter. The mean baseline 25OHD level was 22.4±0.84 ng/ml; 77.6% of participants had 25OHD levels that were less than the level considered adequate (ie, < 30 ng/ml), while 22.4% had levels of ≥ 30 ng/ml. Mean 25OHD value was 31.3±1.3 ng/ml at 6 months and 34±1.8 ng/ml at 1 year. These values represent a significant improvement from the baseline value. At 6 months, 52% of women had 25OHD levels of ≥ 30 and only 48% had levels < 30, while 62% of women had 25OHD level of ≥ 30 and only 38% had levels < 30 at 1 year. Analysis of BMD changes according to 25OHD levels showed that women with levels < 30 at baseline had significant bone loss in the femoral neck and trochanter as compared to those with levels ≥ 30. Simple correlation analysis also showed a positive correlation between changes in BMD at the femoral neck at 6 months with the baseline vitamin D (r = 0.29; P = 0.01), but not with the 6-month vitamin D level. According to researchers, analysis of BMD changes in the limited number of patients who came for 1-year follow-up showed no significant correlation between vitamin D levels at any time point and BMD changes.
_____________________________________________________________________________________________________________

Relationship Between Change in Bone Mineral Density and Nonvertebral Fracture Risk Reduction In Osteoporosis Trials: Results From an Updated Meta-Analysis

Denver, CO—A previous meta-analysis published in the Journal of Clinical Endocrinology & Metabolism (October 2002) by Hochberg et al reported that increases in hip and spine bone mineral density (BMD) were associated with reduced risk of nonvertebral fracture (NVF) in women with postmenopausal osteoporosis (PMO). Marc Hochberg, MD, MPH, University of Maryland School of Medicine, Baltimore, and colleagues extended this analysis to include recently published data in order to enhance the understanding of the relationship between BMD change and NVF risk and to estimate the reduction in NVF risk with clinically relevant ibandronate dosage.

Researchers concluded that treatment-related increases in spine or hip BMD were associated with reduced risk of NVF in women with PMO. High-dose ibandronate treatment has been shown to produce increases in BMD of a magnitude that were estimated by these models to be associated with significant reductions in NVF risk. These reductions in NVF risk were consistent with those demonstrated in two previous meta-analyses of ibandronate using individual patient data (Harris et al, 2008 and Cranney et al, 2009).

This meta-analysis included 23 randomized, placebo-controlled clinical trials of antiresorptive treatments (raloxifene, alendronate, risedronate, calcitonin, etidronate, estrogen, zoledronic acid, and ibandronate) in women with PMO. Poisson regression models were developed to examine the relationship between mean percent change in BMD from baseline to 1 year and treatment (active vs placebo) with risk of NVF. Separate models were constructed for spine and hip BMD. Relative risk (RR) of NVF for 1% change in BMD and associated 95% confidence interval (CI) were calculated as measures of strength of association and precision.

Results showed that when change in hip or spine BMD was added to a base model with treatment, the effect of treatment was no longer statistically significant. This suggests that the effect of treatment is mediated by BMD change. In the final models (BMD change only), increased hip or spine BMD was associated with reduced NVF risk (P < 0.05). For a 1% increase in spine BMD, the RR for NVF was 0.935 (95% CI, 0.913, 0.957), and for hip BMD was 0.906 (95% CI, 0.876, 0.936). In a prior pooled analysis, mean 1-year change in BMD with ibandronate regimens providing annual cumulative exposure ≥ 10.8 mg was 4.7% at the spine and 2.5% at the hip. Using these BMD data and the final models, the estimated RR of NVF for ibandronate was 0.731 (95% CI, 0.646, 0.816) for spine BMD and 0.782 (95% CI, 0.711, 0.852) for hip BMD.
______________________________________________________________________________________________________________________________________________

Teriparatide Effects on Bone Markers and Bone Microarchitecture in Osteoporotic Women: 18-Month Evaluation by High-Resolution pQCT

Denver, CO—Teriparatide [recombinant human PTH, rhPTH(1-34)] is the first bone anabolic agent with proven antifracture efficacy. It is unknown whether its effect on bone structure, especially on cortical compartments, is related to weight-bearing environment of bone sites. The aim of this study was to provide additional clues in this topic as researchers assessed teriparatide effects on 3D bone structural parameters in vivo by high-resolution peripheral quantitative computed tomography (HR-pQCT) at both distal radius and tibia.

Ten postmenopausal women (74.3±8.8 yr) with established osteoporosis and at least two vertebral fragility fractures, regardless of any previous osteoporosis treatment, were measured by HR-pQCT at baseline and at 1, 3, 6, 12, and 18 months of teriparatide treatment (20 microg/d). Serum bone alkaline phosphatase (bALP) and C terminal collagen type I telopeptides (CTx) as well as lumbar and femoral bone mineral densities assessed by DXA, were obtained at the same time points. The 18-month results showed a 5% increase in lumbar spine BMD (P = 0.06). The CTx levels were significantly increased (P = 0.01) as early as the 3rd month of treatment, and the bALP levels were not increased enough to reach statistical significance. HR-pQCT analysis revealed no significant alteration of bone microarchitecture at the radius level, suggesting that this site was preserved during the treatment. At the tibia site, cortical mineral density decreased by 5.8% (P < 0.001) and cortical thickness and area decreased even more importantly (11.7% and 11.5%; P < 0.001, respectively). The periosteal diameter slightly but significantly increased (0.4%; P = 0.02). The tibia trabecular area was enlarged (0.6%; P = 0.01), whereas trabecular density and structural parameters were preserved. Finite element analysis demonstrated no clear improvement in strength for all patients. Thus, it remains unclear whether the modest increase in periosteal diameter might compensate for the radial thinning of the cortical envelope.

Researchers concluded that, overall, these results demonstrate for the first time interactions between teriparatide effects and bone strain in osteoporotic patients. The unexpected effects at the tibia must be confirmed in a larger study, as they may have implication in selecting patients for this treatment.