Selenium supplementation to improve bone health in postmenopausal women: the SeMS three-arm RCT

Notes

Permissions

Chapter 1 Introduction

Parts of this report are based on Walsh et al. 1 © 2021 The Authors. Published by Elsevier Ltd. This is an Open Access article under the CC BY 4.0 license.

One in two women and one in five men aged > 50 years will have a fragility fracture. Fractures lead to pain, disability, loss of independence and increased mortality. This is a huge health burden for affected individuals, the NHS and social care, and it is increasing as the population ages.

About 30% of women aged > 65 years are osteopenic (i.e. they have below-average bone density) and are at risk of developing osteoporosis and fractures. More than 50% of all fractures in postmenopausal women occur in those with osteopenia. These women are generally not given osteoporosis treatment at present because of the individual risk–benefit ratio. Previously, these women could have been offered hormone replacement therapy (HRT) for bone protection, but adverse effects of HRT have limited its use. Bisphosphonates are the mainstay of osteoporosis treatment for women at higher risk, but use of these medications has declined because of physician and patient wariness of adverse events. 2 More costly treatments, such as teriparatide and denosumab, are restricted to patients with more severe osteoporosis who do not respond to bisphosphonate treatment. Calcium and vitamin D supplements are generally recommended for osteopenic adults, but effects on bone density are small and may not outweigh the risk of adverse effects. 3

Therefore, there is a need for an effective, safe, well-tolerated, inexpensive and widely applicable preventative option for osteopenic women.

Selenium is a chemical element present in several human proteins. Twenty-five human selenoproteins have been identified. 4 Known functions of selenoproteins include thyroid hormone synthesis (iodothyronine deiodinases) and anti-oxidants (thioredoxin reductases and glutathione peroxidases). 5,6 Selenoproteins are anti-inflammatory and anti-oxidant; they reduce levels of interleukin 6 (IL-6) and reactive oxygen species, both of which are potent stimuli for bone resorption. 7–10

Selenium is obtained from diet, particularly seafood, meat and cereals. The main determinant of food selenium content is soil selenium content. The recommended adequate intake for adults aged > 50 years in the UK is 75 µg/day for men and 60 µg/day for women,11 but in the UK the mean intake is only 40 µg/day. 12 Selenium intakes have been declining in the UK in the past few decades and are generally low in Europe compared with the USA. The main reason for the decreasing intake in the UK is a change in the source of flour for bread-making from North America (which contains higher selenium) to Europe. More recently, the levels of selenium in UK soils have declined because of changes in fertiliser practice (e.g. replacing single superphosphate with triple superphosphate) and reduced industrial emissions. 13

Studies of all-cause mortality suggest that the optimum range of serum selenium for human health is between about 120 and 150 µg/l. Most adults in the UK have serum selenium between 80 and 100 µg/l. 12

We previously reported that, in 1144 older women from the UK, France and Germany, higher serum selenium or selenoprotein P (SePP) was associated with higher bone mineral density of the lumbar spine and total hip, and lower biochemical markers of bone turnover. 14 High bone turnover is the principal mechanism of osteoporotic bone loss. We also noted associations of selenium levels with balance and grip strength. Selenium status was inversely related to thyroid hormone status (selenium is required for thyroid hormone synthesis), but the associations of selenium with bone measures were independent of thyroid hormones.

It is plausible that selenium could affect bone metabolism. Selenoproteins are expressed in osteoblasts and osteoclasts, and are found in the bone microenvironment. 7,15 Selenoproteins are anti-inflammatory and reduce IL-6, a potent stimulus for bone resorption. 8 Selenoproteins are anti-oxidant and reduce reactive oxygen species; these also stimulate bone resorption via increased RANK-L signalling. 15 Oxidative stress markers are associated with high bone resorption markers and lower bone mineral density (BMD). 16,17 An increase in reactive oxygen species has been proposed as a key mechanism by which sex hormone deficiency causes age-related bone loss through the same RANK pathway. 18 Therefore, it is possible that selenium could directly antagonise the cellular mechanism of postmenopausal osteoporosis.

There is experimental animal evidence to support the hypothesis that selenium has a role in bone biology and reduces bone turnover. Selenium-deficient mice have poorer bone microarchitecture, higher bone resorption markers and higher inflammatory markers than selenium supplemented mice. 19 Selenium-deficient rats have poor bone microarchitecture and abnormal skeletal growth. 20,21

Selenium status has also been associated with BMD in men in the Netherlands,22 and higher selenium intake was associated with lower hip fracture risk in older adults in the USA,23 but there was no association with BMD in postmenopausal Turkish women. 24 In a US study of hip fracture risk in women aged ≥ 65 years, the counties that had the highest rates were those situated in a belt across the south of the USA, and the lowest rates were in the north. 25 By contrast, a current map of soil selenium content in the USA shows that the highest level of selenium content is in the north of the USA and the lowest level is in a belt across the south of the USA (http://mrdata.usgs.gov/geochem/doc/averages/se/usa.html; accessed 5 February 2021).

Endemic selenium deficiency in humans has been associated with the osteoarthropathy Kashin–Beck disease. 26

Several other age-related disorders are linked to inadequate selenium status, including poor cognitive function and reduced muscle strength. 27 Selenium may be an independent predictor of mortality among older community-dwelling adults. 28 Selenium supplementation with coenzyme Q10 reduced cardiovascular mortality and markers of inflammation, increased IGF-1 (insulin-like growth factor 1) and altered microRNA expression in older Swedish adults. 29–31 In the Nutritional Prevention of Cancer trial, selenium supplementation reduced all-cancer risk in people with lower baseline serum selenium,32 and meta-analyses generally find a beneficial effect of selenium on cancer risk. 12

The possible adverse effects of selenium supplementation are thyroid dysfunction (because some selenoproteins are involved in thyroid hormone synthesis) and increased risk of type 2 diabetes mellitus. 33

We hypothesised that, in a relatively selenium-deficient population such as the UK, selenium supplementation would decrease bone turnover by reducing the action of reactive oxygen species on osteoclast activity and may improve muscle function. In the longer term, both of these actions could have benefits with regard to reducing fracture risk.

The aim of the study was to determine if selenium supplementation is beneficial for bone health and muscle function in postmenopausal women.

The objectives of the study were to determine if selenium supplementation in postmenopausal women with osteopenia:

• decreases bone turnover

• improves physical function score and grip strength

• is safe (particularly for thyroid function and diabetes)

• decreases markers of oxidative stress and inflammation.

Chapter 2 Methods

Chapter 3 Results

We recruited 120 women between January 2017 and April 2018. One hundred and fifteen women completed follow-up and were included in the ITT analysis (Figure 1).

The participants’ baseline characteristics are shown in Tables 1–3. The groups were generally well balanced, and the mean baseline serum selenium was 79.4 µg/l. All participants had baseline serum selenium < 120 µg/l and so were included in the primary analysis.

FIGURE 1.

The CONSORT (Consolidated Standards of Reporting Trials) flow diagram. PP, per protocol.

Nine participants had missing data for baseline or week 26 NTX/Cr, so 106 were included in the primary end-point analysis. Baseline characteristics for participants with and participants without complete primary end-point data were similar.

The sample size calculation assumed a correlation between baseline and week 26 NTX/Cr measurement of 0.7. In the ITT population, the Pearson correlation between baseline and week 26 NTX/Cr was 0.62 [95% confidence interval (CI) 0.49 to 0.73]. The residuals from the model were not normally distributed, so NTX/Cr was log-transformed and the treatment group differences were back-transformed so that they could be presented as a ratio.

The primary end point (urine NTX/Cr) did not differ between treatment groups after 26 weeks or 13 weeks (Figures 2 and 3 and Table 4). Eighty-six participants were included in the per-protocol analysis, and NTX/Cr did not differ between treatment groups after 26 weeks or 13 weeks (Table 5).

FIGURE 2.

NTX/Cr 26 weeks: ITT analysis.

FIGURE 3.

NTX/Cr 13 weeks: ITT analysis.

Mean serum selenium increased from baseline to 26 weeks in the treatment groups; 78.8 µg/l (95% CI 73.5 to 84.2 µg/l) to 105.7 µg/l (95%CI 99.5 to 111.9 µg/l) in the 200 µg group, and 79.3 µg/l (95% CI 74.2 to 84.4 µg/l) to 96.2 µg/l (95% CI 90.7 to 101.6 µg/l) in the 50 µg group. There was no change in the placebo group (Table 6 and Figure 4). Mean serum SePP increased from baseline to 26 weeks in the treatment groups; 5.15 mg/l (95% CI 4.71 to 5.60 mg/l) to 6.03 mg/l (95% CI 5.54 to 6.51 mg/l) in the 200 µg group and 5.21 mg/l (95% CI 4.73 to 5.70 mg/l) to 6.25 mg/l (95% CI 5.79 to 6.70 mg/l) in the 50 µg group. There was no change in the placebo group (see Table 6 and Figure 5).

FIGURE 4.

Serum selenium by treatment group at 13 weeks and 26 weeks.

FIGURE 5.

Serum SePP by treatment group at 13 weeks and 26 weeks.

A linear regression model was fitted with log(week 26 NTX/Cr) as the dependent variable and log (baseline NTX/Cr), baseline selenium and treatment group as the independent variables. The interaction between treatment group and baseline selenium was not statistically significant (p = 0.465), suggesting that treatment group did not modify the relationship between baseline selenium and week 26 NTX/Cr.

There were no differences between treatment groups in any of the other biochemical markers of bone turnover (PINP, CTX or OC) at 26 weeks or 13 weeks (Table 7).

There was a small statistically significant but not clinically relevant difference in lumbar spine BMD T-score at 26 weeks in the 50 µg group (T-score difference 0.2) compared with the placebo group and the 200 µg group (T-score difference –0.1). Total hip BMD did not differ between treatment groups at 26 weeks (Table 8).

There was a statistically significant but small (0.5/12) unfavourable difference in the SPPB score in the 50 µg group compared with the placebo group at 26 weeks, but there was no difference between the 200 µg group and the placebo group; overall, there was no significant treatment effect (p = 0.08). Grip strength did not differ between treatment groups (Table 9).

Measurements of anti-oxidant activity and inflammation did not differ between treatment groups (Table 10). The majority of IL-6 measurements were below the limit of detection of 1.6 ng/l (74/110 at baseline, 71/110 at week 13 and 74/108 at week 26), so no further analysis was conducted on the IL-6 measurements.

The safety assessments for diabetes and thyroid function did not differ between treatment groups. There was a small difference in glycated haemoglobin (HbA_1c) between 200 µg and placebo groups at 26 weeks (–1.0 mmol/mol), but this is not clinically significant (Table 11). Seven participants were withdrawn from treatment at week 13 because of abnormal TSH (200 µg, n = 1; 50 µg, n = 3; placebo, n = 3), and one was withdrawn because of abnormal blood glucose (in the 50 µg group).

Analyses were repeated in the per-protocol group for all efficacy and safety end points, and the results did not differ from those for the ITT population.

The study group were generally vitamin D replete. There was a small difference in 25OHD between 200 µg and placebo groups at 26 weeks, but this is not clinically significant (Table 12).

The dietary intake of vitamin D, calcium and selenium assessed by a 7-day diet diary were similar in all three treatment groups (Table 13). The dietary selenium intake decreased between baseline and 26 weeks in all three groups.

The number and severity of adverse events and the systems affected by adverse events were similar across treatment groups (Tables 14 and 15).

There were three serious adverse events: a non-ST elevation myocardial infarction at week 18 (in the 50 µg/day group); a diagnosis of bowel cancer after routine screening at week 2 (in the placebo group); and a pulmonary embolus due to metastatic bowel cancer at week 4 (in the 200 µg/day group). All SAEs were judged by the principal investigator as unrelated to trial medication.

Chapter 4 Discussion

We have conducted a well-powered, randomised, double-blind, placebo-controlled study of the effects of selenium supplementation on musculoskeletal health in postmenopausal women. We found no effect on biochemical markers of bone turnover, BMD or physical function with Selenase 200 µg or 50 µg daily. None of the end-point results differed between the ITT analysis and the per-protocol analysis. This is an important result because it is from the first randomised controlled trial of selenium supplementation for musculoskeletal health, to our knowledge.

Serum selenium in the 200 µg treatment group increased from 80 µg/l to 105 µg/l. Mortality data suggest that the optimum range for serum selenium is 120–150 µg/l. Although serum selenium did not reach this range, based on the correlation of serum selenium and NTX/Cr in our previous study, an increase of 30% should be enough to demonstrate some change in bone markers if there was any effect. Biochemical markers of bone turnover are dynamic and respond to bone active agents within a few weeks. For example, bone markers decrease by about 20% within 2 weeks of starting calcium supplements. 45 Selenium at 200 µg/day has been shown to be effective in Graves’ eye disease and cancer prevention studies, so there is good evidence that this dose is high enough to be biologically active in humans. It is possible that higher dose supplements would have an effect on bone, but there was no dose–response effect across the two doses we studied. In addition, higher doses may increase the risk of adverse effects. 33

There was a small increase in lumbar spine BMD in the 50 µg group, but, in the absence of any effect on bone turnover or any BMD effect in the 200 µg group, this is likely to be a spurious result.

There were enough promising epidemiological, observational and pre-clinical data to suggest that selenium might have beneficial effects on musculoskeletal health.

Higher selenium status is associated with BMD in men in the Netherlands. 22 Higher dietary selenium intake is associated with lower hip fracture risk in older adults in the USA23 and higher BMD in middle-aged and older adults in China46 and Europe. 14,22 However, there was no association with BMD in postmenopausal Turkish women. 24

Lower serum selenium and dietary selenium are associated with lower muscle mass and poorer muscle function in older adults. 27,47–49

The proposed mechanism of action of selenium to reduce reactive oxygen species, and therefore reduce the pro-resorptive drive to osteoclasts, was plausible. However, we saw no effect at all on markers of bone resorption. It may be that selenium status is a marker for other factors acting on bone health or that a single factor approach is ineffective and selenium is part of a more complex system that is not yet fully characterised.

The population in this study was generally representative of postmenopausal women in the UK, in terms of BMI, BMD, vitamin D status and calcium intake. Their dietary selenium intake at baseline was higher than expected for the UK, but at 26 weeks it was more typical. 50 We do not know if this is a true change in dietary behaviour over the course of the study; we might speculate that participants reduced their dietary selenium intake because they were receiving a supplement. However, selenium is a ubiquitous nutrient, and it would be difficult to reduce it in isolation.

We studied women only, and it is possible that the effects of selenium on bone would be different in men. However, postmenopausal women have higher bone resorption than men, and we hypothesised that selenium could act particularly through one of the resorption pathways activated by oestrogen deficiency. Therefore, in the absence of any effect in women, we do not think that an effect in men is likely.

We conclude that selenium supplementation at these doses is not beneficial for musculoskeletal health in postmenopausal women. Other trials have demonstrated benefit in cancer prevention, so selenium may have benefits for human health. However, it is not likely to be effective for treatment of osteoporosis and reduction in fracture risk.

Chapter 5 Patient and public involvement

The Sheffield Lay Advisory Panel for Bone Research has contributed to this study. The panel was established in 2009 and has made valuable contributions to these aspects of research in the Academic Unit of Bone Metabolism since then. It has received training in research methods, research governance and grant application processes. The panel was consulted about trial design and contributed to the grant application, protocol and recruitment strategy.

The panel received updates on the study’s progress at its monthly meetings; during these meetings, panel members had the opportunity to discuss the study with investigators and other members of the study team.

Now that the study is complete, the Sheffield Lay Advisory Panel for Bone Research will write a lay summary of the results, which will be sent to study participants and publicised through the University of Sheffield, Sheffield Teaching Hospitals and the Royal Osteoporosis Society.

Acknowledgements

We thank Fatma Gossiel for managing the biochemical measurements, Margaret Paggiosi for the bone densitometry, Julie Walker, Angela Green and Jill Thompson for nursing and clerical support, Marian Schini for sub-investigator support, Kim Ryalls for pharmacy management, and Aimee Card for governance support. We would also like to thank the chairpersons and members of the TSC and DMC. The study was conducted in and with the support of the Sheffield NIHR Clinical Research Facility. The study was also supported by the Sheffield School of Health and Related Research, with advice on study implementation from Professor Cindy Cooper.

References

1. Walsh JS, Jacques RM, Schomburg L, Hill TR, Mathers JC, Williams GR, et al. Effect of selenium supplementation on musculoskeletal health in older women: a randomised, double-blind, placebo-controlled trial [published online ahead of print March 23 2021]. Lancet Healthy Longevity 2021. https://doi.org/10.1016/S2666-7568(21)00051-9.
2. Jha S, Wang Z, Laucis N, Bhattacharyya T. Trends in media reports, oral bisphosphonate prescriptions, and hip fractures 1996-2012: an ecological analysis. J Bone Miner Res 2015;30:2179-87. https://doi.org/10.1002/jbmr.2565.
3. Reid IR, Bolland MJ. Calcium and/or vitamin D supplementation for the prevention of fragility fractures: who needs it?. Nutrients 2020;12. https://doi.org/10.3390/nu12041011.
4. Moghadaszadeh B, Beggs AH. Selenoproteins and their impact on human health through diverse physiological pathways. Physiology 2006;21:307-15. https://doi.org/10.1152/physiol.00021.2006.
5. Köhrle J, Jakob F, Contempré B, Dumont JE. Selenium, the thyroid, and the endocrine system. Endocr Rev 2005;26:944-84. https://doi.org/10.1210/er.2001-0034.
6. Hatfield DL, Tsuji PA, Carlson BA, Gladyshev VN. Selenium and selenocysteine: roles in cancer, health, and development. Trends Biochem Sci 2014;39:112-20. https://doi.org/10.1016/j.tibs.2013.12.007.
7. Jakob F, Becker K, Paar E, Ebert-Duemig R, Schütze N. Expression and regulation of thioredoxin reductases and other selenoproteins in bone. Methods Enzymol 2002;347:168-79. https://doi.org/10.1016/s0076-6879(02)47015-2.
8. Duntas LH. Selenium and inflammation: underlying anti-inflammatory mechanisms. Horm Metab Res 2009;41:443-7. https://doi.org/10.1055/s-0029-1220724.
9. Vaananen HK, Zhao H, Mulari M, Halleen JM. The cell biology of osteoclast function. J Cell Sci 2000;113:377-81.
10. Battin EE, Brumaghim JL. Antioxidant activity of sulfur and selenium: a review of reactive oxygen species scavenging, glutathione peroxidase, and metal-binding antioxidant mechanisms. Cell Biochem Biophys 2009;55:1-23. https://doi.org/10.1007/s12013-009-9054-7.
11. Scientific Advisory Committee on Nutrition . SACN Position Statement on Selenium and Health 2013. www.gov.uk/government/uploads/system/uploads/attachment_data/file/339431/SACN_Selenium_and_Health_2013.pdf (accessed 17 September 2020).
12. Rayman MP. Selenium and human health. Lancet 2012;379:1256-68. https://doi.org/10.1016/S0140-6736(11)61452-9.
13. Rayman MP. Food-chain selenium and human health: emphasis on intake. Br J Nutr 2008;100:254-68. https://doi.org/10.1017/S0007114508939830.
14. Hoeg A, Gogakos A, Murphy E, Mueller S, Köhrle J, Reid DM, et al. Bone turnover and bone mineral density are independently related to selenium status in healthy euthyroid postmenopausal women. J Clin Endocrinol Metab 2012;97:4061-70. https://doi.org/10.1210/jc.2012-2121.
15. Zhang Z, Zhang J, Xiao J. Selenoproteins and selenium status in bone physiology and pathology. Biochim Biophys Acta 2014;1840:3246-56. https://doi.org/10.1016/j.bbagen.2014.08.001.
16. Cervellati C, Bonaccorsi G, Cremonini E, Romani A, Fila E, Castaldini MC, et al. Oxidative stress and bone resorption interplay as a possible trigger for postmenopausal osteoporosis. Biomed Res Int 2014;2014. https://doi.org/10.1155/2014/569563.
17. Hahn M, Conterato GM, Frizzo CP, Augusti PR, da Silva JC, Unfer TC, et al. Effects of bone disease and calcium supplementation on antioxidant enzymes in postmenopausal women. Clin Biochem 2008;41:69-74. https://doi.org/10.1016/j.clinbiochem.2007.10.010.
18. Manolagas SC. From estrogen-centric to aging and oxidative stress: a revised perspective of the pathogenesis of osteoporosis. Endocr Rev 2010;31:266-300. https://doi.org/10.1210/er.2009-0024.
19. Cao JJ, Gregoire BR, Zeng H. Selenium deficiency decreases antioxidative capacity and is detrimental to bone microarchitecture in mice. J Nutr 2012;142:1526-31. https://doi.org/10.3945/jn.111.157040.
20. Min Z, Zhao W, Zhong N, Guo Y, Sun M, Wang Q, et al. Abnormality of epiphyseal plate induced by selenium deficiency diet in two generation DA rats. APMIS 2015;123:697-705. https://doi.org/10.1111/apm.12404.
21. Moreno-Reyes R, Egrise D, Neve J, Pasteels JL, Schoutens A. Selenium deficiency-induced growth retardation is associated with an impaired bone metabolism and osteopenia. J Bone Miner Res 2001;16:1556-63. https://doi.org/10.1359/jbmr.2001.16.8.1556.
22. Beukhof CM, Medici M, van den Beld AW, Hollenbach B, Hoeg A, Visser WE, et al. Selenium status is positively associated with bone mineral density in healthy aging European men. PLOS ONE 2016;11. https://doi.org/10.1371/journal.pone.0152748.
23. Zhang J, Munger RG, West NA, Cutler DR, Wengreen HJ, Corcoran CD. Antioxidant intake and risk of osteoporotic hip fracture in Utah: an effect modified by smoking status. Am J Epidemiol 2006;163:9-17. https://doi.org/10.1093/aje/kwj005.
24. Arikan DC, Coskun A, Ozer A, Kilinc M, Atalay F, Arikan T. Plasma selenium, zinc, copper and lipid levels in postmenopausal Turkish women and their relation with osteoporosis. Biol Trace Elem Res 2011;144:407-17. https://doi.org/10.1007/s12011-011-9109-7.
25. Jacobsen SJ, Goldberg J, Miles TP, Brody JA, Stiers W, Rimm AA. Regional variation in the incidence of hip fracture. US white women aged 65 years and older. JAMA 1990;264:500-2. https://doi.org/10.1001/jama.1990.03450040096038.
26. Moreno-Reyes R, Suetens C, Mathieu F, Begaux F, Zhu D, Rivera MT, et al. Kashin-Beck osteoarthropathy in rural Tibet in relation to selenium and iodine status. N Engl J Med 1998;339:1112-20. https://doi.org/10.1056/NEJM199810153391604.
27. Lauretani F, Semba RD, Bandinelli S, Ray AL, Guralnik JM, Ferrucci L. Association of low plasma selenium concentrations with poor muscle strength in older community-dwelling adults: the InCHIANTI Study. Am J Clin Nutr 2007;86:347-52. https://doi.org/10.1093/ajcn/86.2.347.
28. Lauretani F, Semba RD, Bandinelli S, Ray AL, Ruggiero C, Cherubini A, et al. Low plasma selenium concentrations and mortality among older community-dwelling adults: the InCHIANTI Study. Aging Clin Exp Res 2008;20:153-8. https://doi.org/10.1007/BF03324762.
29. Alehagen U, Alexander J, Aaseth J. Supplementation with selenium and coenzyme Q10 reduces cardiovascular mortality in elderly with low selenium status. A secondary analysis of a randomised clinical trial. PLOS ONE 2016;11. https://doi.org/10.1371/journal.pone.0157541.
30. Alehagen U, Johansson P, Aaseth J, Alexander J, Brismar K. Increase in insulin-like growth factor 1 (IGF-1) and insulin-like growth factor binding protein 1 after supplementation with selenium and coenzyme Q10. A prospective randomized double-blind placebo-controlled trial among elderly Swedish citizens. PLOS ONE 2017;12. https://doi.org/10.1371/journal.pone.0178614.
31. Alehagen U, Johansson P, Aaseth J, Alexander J, Wågsäter D. Significant changes in circulating microRNA by dietary supplementation of selenium and coenzyme Q10 in healthy elderly males. A subgroup analysis of a prospective randomized double-blind placebo-controlled trial among elderly Swedish citizens. PLOS ONE 2017;12. https://doi.org/10.1371/journal.pone.0174880.
32. Duffield-Lillico AJ, Reid ME, Turnbull BW, Combs GF, Slate EH, Fischbach LA, et al. Baseline characteristics and the effect of selenium supplementation on cancer incidence in a randomized clinical trial: a summary report of the Nutritional Prevention of Cancer Trial. Cancer Epidemiol Biomarkers Prev 2002;11:630-9.
33. Rayman MP, Blundell-Pound G, Pastor-Barriuso R, Guallar E, Steinbrenner H, Stranges S. A randomized trial of selenium supplementation and risk of type-2 diabetes, as assessed by plasma adiponectin. PLOS ONE 2012;7. https://doi.org/10.1371/journal.pone.0045269.
34. Eastell R, Hannon RA, Garnero P, Campbell MJ, Delmas PD. Relationship of early changes in bone resorption to the reduction in fracture risk with risedronate: review of statistical analysis. J Bone Miner Res 2007;22:1656-60. https://doi.org/10.1359/jbmr.07090b.
35. Hurst R, Collings R, Harvey LJ, King M, Hooper L, Bouwman J, et al. EURRECA-Estimating selenium requirements for deriving dietary reference values. Crit Rev Food Sci Nutr 2013;53:1077-96. https://doi.org/10.1080/10408398.2012.742861.
36. Guralnik JM, Simonsick EM, Ferrucci L, Glynn RJ, Berkman LF, Blazer DG, et al. A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission. J Gerontol 1994;49:M85-94. https://doi.org/10.1093/geronj/49.2.m85.
37. Minneci C, Mello AM, Mossello E, Baldasseroni S, Macchi L, Cipolletti S, et al. Comparative study of four physical performance measures as predictors of death, incident disability, and falls in unselected older persons: the insufficienza Cardiaca negli Anziani Residenti a Dicomano Study. J Am Geriatr Soc 2015;63:136-41. https://doi.org/10.1111/jgs.13195.
38. Veronese N, Bolzetta F, Toffanello ED, Zambon S, De Rui M, Perissinotto E, et al. Association between Short Physical Performance Battery and falls in older people: the Progetto Veneto Anziani Study. Rejuvenation Res 2014;17:276-84. https://doi.org/10.1089/rej.2013.1491.
39. Rogers A, Glover SJ, Eastell R. A randomised, double-blinded, placebo-controlled, trial to determine the individual response in bone turnover markers to lasofoxifene therapy. Bone 2009;45:1044-52. https://doi.org/10.1016/j.bone.2009.07.089.
40. Eastell R, Barton I, Hannon RA, Chines A, Garnero P, Delmas PD. Relationship of early changes in bone resorption to the reduction in fracture risk with risedronate. J Bone Miner Res 2003;18:1051-6. https://doi.org/10.1359/jbmr.2003.18.6.1051.
41. Champely S. Pwr: Basic Functions for Power Analysis. R Package Version 1.2-2 n.d. https://CRAN.R-project.org/package=pwr (accessed 17 September 2020).
42. Frison L, Pocock SJ. Repeated measures in clinical trials: analysis using mean summary statistics and its implications for design. Stat Med 1992;11:1685-704. https://doi.org/10.1002/sim.4780111304.
43. Hoeflich J, Hollenbach B, Behrends T, Hoeg A, Stosnach H, Schomburg L. The choice of biomarkers determines the selenium status in young German vegans and vegetarians. Br J Nutr 2010;104:1601-4. https://doi.org/10.1017/S0007114510002618.
44. Hollenbach B, Morgenthaler NG, Struck J, Alonso C, Bergmann A, Köhrle J, et al. New assay for the measurement of selenoprotein P as a sepsis biomarker from serum. J Trace Elem Med Biol 2008;22:24-32. https://doi.org/10.1016/j.jtemb.2007.11.003.
45. Naylor KE, Jacques RM, Paggiosi M, Gossiel F, Peel NF, McCloskey EV, et al. Response of bone turnover markers to three oral bisphosphonate therapies in postmenopausal osteoporosis: the TRIO study. Osteoporos Int 2016;27:21-3. https://doi.org/10.1007/s00198-015-3145-7.
46. Wang Y, Xie D, Li J, Long H, Wu J, Wu Z, et al. Association between dietary selenium intake and the prevalence of osteoporosis: a cross-sectional study. BMC Musculoskelet Disord 2019;20. https://doi.org/10.1186/s12891-019-2958-5.
47. Perri G, Mendonça N, Jagger C, Walsh J, Eastell R, Mathers JC, et al. Dietary selenium intakes and musculoskeletal function in very old adults: analysis of the Newcastle 85+ Study. Nutrients 2020;12. https://doi.org/10.3390/nu12072068.
48. Chen YL, Yang KC, Chang HH, Lee LT, Lu CW, Huang KC. Low serum selenium level is associated with low muscle mass in the community-dwelling elderly. J Am Med Dir Assoc 2014;15:807-11. https://doi.org/10.1016/j.jamda.2014.06.014.
49. Beck J, Ferrucci L, Sun K, Walston J, Fried LP, Varadhan R, et al. Low serum selenium concentrations are associated with poor grip strength among older women living in the community. Biofactors 2007;29:37-44. https://doi.org/10.1002/biof.5520290104.
50. Public Health England . National Diet and Nutrition Survey. Results from Years 7 and 8 (combined) of the Rolling Programme (2014 2015 to 2015 2016) 2018.

Appendix 1 Trial flow chart

Appendix 2 Schedule of procedures

List of abbreviations

25OHD

25-hydroxyvitamin D

BCE

bone collagen equivalents

BMD

bone mineral density

BMI

body mass index

CI

confidence interval

CTX

C–terminal cross-linking telopeptide of type I collagen

CV

coefficient of variation

DMC

Data Monitoring Committee

DXA

dual-energy X-ray absorptiometry

HbA1c

glycated haemoglobin

HRT

hormone replacement therapy

hsCRP

highly sensitive C-reactive protein

IL-6

interleukin 6

IMP

Investigational Medical Product

ITT

intention to treat

NTX

N–terminal cross-linking telopeptide of type I collagen

OC

osteocalcin

PINP

procollagen type I N propeptide

REC

Research Ethics Committee

SAE

serious adverse event

SAP

statistical analysis plan

SD

standard deviation

SePP

selenoprotein P

SERM

selective oestrogen receptor modulator

SPPB

short physical performance battery

TSC

Trial Steering Committee

TSH

thyroid-stimulating hormone