ORIGINAL ARTICLE Normative Data for Impact Microindentation for Australian Men: Cross-Sectional Data From the Geelong Osteoporosis Study Pamela Rufus-Membere,1 Kara L Holloway-Kew,1 Mark A Kotowicz,1,2,3 Adolfo Diez-Perez,4 and Julie A Pasco1,2,3 1 School of Medicine, Deakin University, Geelong, Victoria, Australia 2 Department of Medicine-Western Health, Melbourne Medical School, The University of Melbourne, Melbourne, Victoria, Australia 3 Barwon Health, Geelong, Victoria, Australia 4 Department of Internal Medicine, Hospital del Mar-IMIM, Autonomous University of Barcelona and CIBERFES, Instituto Carlos III, Madrid, Spain ABSTRACT Impact microindentation (IMI) is a novel technique for assessing the bone material strength index (BMSi) in vivo. However, no studies have presented normative data for BMSi. The aim of this study was to develop such normative data using a population-based sample of men, randomly selected from electoral rolls for the Barwon Statistical Division in southeastern Australia to participate in the Gee- long Osteoporosis Study. BMSi was measured on the tibial plateau using an OsteoProbe in 405 men (ages 33 to 96 years) during the period 2016 to 2019. Associations between BMSi, age, and anthropometry were examined using linear regression models. BMSi values ranged from 49.0 to 100.5. BMSi was negatively correlated with age (r = −0.152, p = 0.002), weight (r = −0.103, p = 0.039), and BMI (r = −0.187, p < 0.001), and positively correlated with height (r = +0.107, p = 0.032). Mean  SD BMSi was 82.6  7.0 for the whole group, and ranged from 85.6  6.0 for ages 30 to 39 years to 79.8  6.6 for ages 80+ years. This study provides normative data that can be used to calculate T- and Z-scores for BMSi. These data will be useful for identifying men with low BMSi. Further research is warranted to derive optimal cut points for BMSi that discriminate fracture risk. © 2020 The Authors. JBMR Plus published by Wiley Periodicals LLC. on behalf of American Society for Bone and Mineral Research. KEY WORDS: BONE MATERIAL STRENGTH INDEX; IMPACT MICROINDENTATION; NORMATIVE DATA; OSTEOPOROSIS; POPULATION-BASED Introduction It is important to note that the application of inappropriate refer- ence data for clinical assessment of BMSi may bias estimates of frac- ture risk.(12) To our knowledge, no studies to date have reported I mpact microindentation (IMI) uses a novel handheld device, the OsteoProbe, to assess cortical bone in vivo using a minimally invasive method.(1) Other devices capable of measuring tissue normative data for BMSi. The aim of this study was to develop age-related reference ranges for BMSi using an Australian, material properties are mainly intended for application in surgical population-based sample of men. settings.(2) The methodology and unique clinical significance of IMI have been previously reported.(3–5) During IMI, the more suscep- Participants and Methods tible the bone to fracture, the further the test probe will indent the bone. The technology quantifies bone microindentation distance in Study participants relation to a reference material and expresses the ratio as a bone material strength index (BMSi). Hence, lower BMSi values are likely Participants for this study were men from the Geelong Osteopo- to be associated with an increased risk of fracture. Previous studies rosis Study (GOS), a population-based cohort study situated in a have evaluated BMSi in relation to fragility fractures,(6–8) chronic kid- geographically well-defined region in southeastern Australia, ney disease,(9) and type 2 diabetes.(10,11) The clinical nature of such known as the Barwon Statistical Division.(13) The male arm of samples and, sometimes, opportunistic samples used as controls, the GOS commenced in 2001 with recruitment of 1540 men have failed to inform the BMSi values in the underlying populations. 20 to 92 years old and follow-up assessments every few years. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. Received in original form August 7, 2019; revised form May 29, 2020; accepted June 4, 2020. Accepted manuscript online June 15, 2020. Address correspondence to: Pamela Rufus-Membere, PhD, Epi-Centre for Healthy Ageing, IMPACT Institute, School of Medicine, Deakin University (Barwon Health), PO BOX 281, Geelong Victoria 3220, Australia. Email: pamela.r@deakin.edu.au JBMR® Plus (WOA), Vol. 4, No. 9, September 2020, e10384. DOI: 10.1002/jbm4.10384 © 2020 The Authors. JBMR Plus published by Wiley Periodicals LLC. on behalf of American Society for Bone and Mineral Research. 1 of 6 n At least 99% of men recruited for the GOS were white. The inclu- measures. Precision was calculated as the mean (expressed as %) sion criterion was a listing on the electoral rolls for the region; the of SD/mean for two sets of indentations for 10 participants. only exclusion criterion was inability to provide informed con- We have previously reported the opinions of the men in the sent to participate in the study. Electoral roll registration is com- GOS regarding their experience with the OsteoProbe, and pulsory in Australia; thus, it is a comprehensive sampling frame showed that participants tolerated the procedure well, demon- for adults aged over 18 years. A detailed description of the GOS strating the high feasibility of performing IMI measures.(15) study design, recruitment, and attrition have been published elsewhere.(13,14) Of the 1540 men recruited at baseline, 424 had Statistical analysis died before the current follow-up, 25 had left the region, 8 were BMSi values were normally distributed. Means and SDs were calcu- unable to provide informed consent, and 188 were unable to be lated for each age group: 30 to 39, 40 to 49, 50 to 59, 60to 69, 70 to contacted. Of the remaining 895, 233 declined participation cit- 79, and 80+ years. Relationships between BMSi values and age, ing personal reasons (n = 112), time constraints (n = 42), old weight, height, and BMI were described using Pearson’s correlation. age (n = 48), or illness (n = 31). Further, at the time of writing, Multivariable linear regression models were developed to deter- 32 men had not provided clinical measures and 63 men were mine the regression equation and prediction intervals for describ- yet to schedule a follow-up visit. Thus, data for this cross- ing the relationship between BMSi and covariates, and for sectional analysis were generated for 567 men (ages 33 to identifying the best predictors for BMSi. The residuals for the regres- 96 years) assessed in the current follow-up phase, 2016 to sion models were visualized for normality. Statistical analyses were 2019. The study was approved by the Human Research Ethics performed using Minitab V.18 (State College, PA, USA) http://www. Committee at Barwon Health (Geelong, Victoria, Australia). All minitab.com/en-us/products/minitab/. participants provided written informed consent. Results Measures Height and weight were measured to the nearest 0.1 cm and Participant characteristics 0.1 kg, respectively, and BMI (kg/m2) calculated. Areal BMD Of 567 potential participants in this current follow-up, 405 under- (g/cm2) was measured at the femoral neck using DXA (Lunar; went IMI testing. Reasons for nonmeasurement in 162 men were Prodigy, Madison, WI, USA). A participant’s prior fracture was needle phobia (n = 22), existing skin infections (n = 45), excessive defined as any low-trauma fracture equivalent to a fall from a soft tissues around the midtibial region (n = 84), discomfort standing height or less, excluding fractures of the toe, skull, fin- (pressure, not pain) after the first indentation (n = 6), inability ger, and face, occurring during adulthood (age ≥20 years). Frac- to provide informed consent (n = 2), and three participants did tures were radiologically verified where possible. not provide any reasons for declining. Compared with partici- IMI was measured using the OsteoProbe RUO (Active Life Tech- pants, nonparticipants were older (mean  SD, 70.3  15.9 ver- nologies, Santa Barbara, CA, USA). The indentation site on the ante- sus 64.2  11.9 years, p < 0.001) and had greater mean BMI rior surface of the midtibia was determined by measuring the (30.2  5.4 versus 27.0  3.2 kg/m2, p < 0.001). midpoint from the medial border of the tibial plateau to the distal Participant characteristics are given in Table 1. BMSi values edge of the medial malleolus.(4) Following disinfection of the area ranged from 49.0 to 100.5. Mean BMSi in the presence or and administration of local anesthetic, the OsteoProbe was inserted absence of prior fracture was (mean  SD): 80.0  5.5 versus through the skin and periosteum until reaching the surface of the 82.6  5.9 (p = 0.023). No associations were detected between bone at the anterior face of the midtibia. At least 11 indentations BMSi and BMD (r = 0.000, p = 0.986). were performed for each participant, of which the first measure- ment was systematically disregarded to ensure sufficient penetra- Associations between BMSi and age tion of the probe through the periosteum, followed by 10 valid test indentations. Two trained operators conducted the IMI mea- Mean and SD BMSi values for the whole group and according to surements following internationally recognized recommendations age-decades are presented in Table 2. The highest mean BMSi for using the OsteoProbe(4); however, the majority (92.7%) were was observed for the youngest age group, 30 to 39 years completed by a single operator. The mean (SD) BMSi of partici- (85.6  6.0), and the lowest for the oldest age group, 80 years pants measured by each of the operators was 83.0  7.0 and and older (79.8  6.6). 82.7  6.9. This was not significantly different (p = 0.801). The coef- There was a weak negative relationship between BMSi and age ficient of variation (CV) for microindentation was 2% for repeated across the age range examined (r = −0.152, p = 0.002) and the slope Table 1. Characteristics of All Participants and by Age Groups (n = 405) Age group (years) n Weight (kg)a Height (cm)a BMI (kg/m2)a Femoral neck BMD (g/cm2)a Prior fractures (n%)b 30–39 26 80.4  8.7 178.6  5.5 25.2  2.7 1.031  0.148 1 (0.2) 40–49 56 83.3  14.2 178.3  8.0 26.2  3.7 1.035  0.136 4 (1.0) 50–59 87 83.0  11.2 175.3  6.1 26.9  3.2 1.011  0.138 10 (2.5) 60–69 112 81.8  11.2 173.4  6.1 27.2  3.3 0.963  0.139 11 (2.7) 70–79 81 82.1  10.3 173.2  5.7 27.3  2.9 0.944  0.124 12 (3.0) 80+ 43 77.0  9.7 170.2  7.8 26.6  3.0 0.897  0.115 5 (1.2) All 405 81.7  11.3 174.4  6.9 26.8  3.2 0.971  0.132 43 (10.6) a Data are shown as mean  SD. b Fractures were 3 clinical vertebra, 3 hip, 2 foot, 4 elbow, 5 ankle, 5 humerus, 8 tibia, and 13 rib. n 2 of 6 RUFUS-MEMBERE ET AL. JBMR Plus (WOA) of the graph indicates that for each decade increase in age, there Discussion was a mean 0.8 decrease in BMSi units (Fig. 1A, Table 3). This study presents normative BMSi data for a population-based sample of men, randomly selected from a geographically well- Associations between BMSi and anthropometry defined region in southeastern Australia. BMSi values ranged BMSi was positively correlated with height (r = +0.107, p = 0.032) from 49.0 to 105.0. and negatively correlated with weight (r = −0.103, p = 0.039) and In previous clinical studies in patients with specific diseases, BMI (r = −0.187, p < 0.001) (Fig. 1B–D). In multivariable models, BMSi ranges of 60.0 to 87.0 and 70.0 to 89.0 have been reported the best predictors of BMSi were age and BMI, which explained for women,(8,9,16–18) and for samples comprised of men and 3.7% of the variance (Table 3). BMD did not contribute to the women,(6,19–22) respectively. For control samples in clinical stud- models. ies, BMSi ranges of 75.0 to 90.0 and 75.0 to 93.0 have been reported for samples involving women(8,9,16–18,23) and men and women combined,(6,19–22) respectively. In population-based studies in women,(24–27) BMSi values ran- Table 2. Bone Material Strength Index (BMSi) by Age Group and ged from 52.9 to 93.3. However, these studies spanned a rela- for All Participantsa tively narrower age range (75 to 80 years), and there is a dearth Age group (years) n BMSi of population-based BMSi data for men. Moreover, in our study, 30–39 26 85.6  6.0 we did not exclude participants on the basis of disease or medi- 40–49 56 82.0  7.6 cation use. The importance of using a large, randomly selected 50–59 87 83.8  6.6 sample has been previously been reported for developing nor- 60–69 112 82.9  5.7 mative data for the broader population.(12) 70–79 81 82.0  8.3 We report a small, age-related decline in BMSi for each age- 80+ 43 79.8  6.6 decade increase: BMSi declined by approximately 0.8 units. Such All 405 82.6  7.0 a decline seems reasonable, given that the main determinants of bone strength (bone mass and bone structure) also deteriorate a Data are shown as mean  SD. with increasing age.(28,29) The differences in bone material Fig. 1. (A–D) Associations between bone material strength index and age, weight, height, and BMI for men. Regression lines and 95% prediction intervals are shown. JBMR® Plus NORMATIVE DATA FOR IMPACT MICROINDENTATION 3 of 6 n Table 3. Linear Regression Model for Bone Material Strength Index With Age, Weight, Height, and BMI β coefficient SE p Value R2 Model 1 Age (years) −0.08 0.02 0.002 1.3% Constant 87.33 1.56 <0.001 Model 2 Age (years) −0.07 0.02 0.008 3.7% BMI −0.37 0.11 0.001 Constant 96.52 3.05 <0.001 Model 3 Age (years) −0.06 0.03 0.031 3.6% Weight (kg) −0.12 0.03 0.001 Height (cm) 0.16 0.06 0.007 Constant 67.20 10.10 <0.001 properties observed with increasing age in this study might be length,(37) as well as bone structure, particularly in long bones.(38) explained by the age-related accumulation of microcracks and Thus, further studies are warranted to investigate the relation- a deterioration in bone microarchitecture.(30,31) Consistent with ship between BMSi and different aspects of anthropometry. our data, Malgo and colleagues(21) reported an inverse associa- The major strength of this study is that data were obtained from tion between age and BMSi for 90 patients (male and female) a randomly selected sample from the population of southeastern with low bone mass (r = −0.539, p < 0.001). In contrast, Duarte Australia; thus, it will be relevant for the underlying population. Sosa and colleagues(23) and Popp and colleagues(17) reported We have also provided mean and SD values for each decade, which no association between age and BMSi in their case–control stud- can be used to calculate T- and Z-scores for BMSi. Furthermore, we ies of postmenopausal women investigating fractures and explored the associations between BMSi, age, and anthropometry bisphosphonate treatment. Moreover, we have previously in the largest sample, and widest age range of men to date. Not- reported no correlation between age and BMSi in a subset of this withstanding, loss to follow-up and exclusions for BMSi testing lim- cohort; this is likely to be a type 2 error as the number of partic- ited our ability to explore associations across the full range of BMI, ipants was smaller than in this analysis.(15) The relatively smaller as excessive accumulation of soft tissue at the measurement site number of men in the 40 to 49 years category may also explain was the most common reason for nonparticipation. why the same mean BMSi values were observed in men aged In the only published study evaluating geographical variation in 40 to 49 years and those aged 70 to 79 years. BMSi, significant differences in BMSi were observed between coun- Thus, the lack of consistency in the literature may be partly tries, with BMSi higher in healthy Spanish women than in healthy explained by the number of participants, the range of ages, Norwegian women.(23) This indicates that the observations from and sampling frames of the participants studied. this study may not be generalizable to other populations, as there BMSi was negatively correlated with weight and BMI, and pos- may be differences in BMSi values between geographical areas. itively correlated with height. Sundh and colleagues(25) also Comparable data will be needed for other geographical regions. reported a negative correlation (r = −0.17, p = 0.01) between Moreover, although the assessment of only male subjects may limit BMSi and BMI in a population-based study of 202 women the transferability of these findings, prior studies(6,7,20,21,39) suggest between 75 and 80 years of age; Rudang and colleagues(24) that sex is not associated with varying BMSi values. reported a weak negative correlation (r = −0.14, p = 0.04) To the best of our knowledge, this study provides the first nor- between BMSi and weight in a sample of 211 women between mative data for BMSi in men; these data will be useful for identi- 75 and 80 years of age, and no association with height. Popp fying BMSi deficits in men. Further research is warranted to and colleagues(17) observed no association between BMSi and derive optimal cut points for BMSi that discriminate fracture risk BMI (r = −0.08, p = 0.31) in a sample of 153 postmenopausal in this population. women. To the best of our knowledge, no studies have explored the relationship between BMSi, weight, and BMI in men only. Reports on the effects of high BMI and adiposity on fracture Disclosures risk are equivocal. Some studies suggest that a high BMI may be associated with increased fracture risk.(32–34) After BMD PGR is supported by Deakin University Postgraduate Industry adjustment, obesity appears to contribute to increased fracture Research Scholarship. KLH-K is supported by an Alfred Deakin Post- risk,(35) suggesting that other mechanisms such as poor bone doctoral Research Fellowship. AD-P owns shares of Active Life Scien- quality or increased falls risk may be responsible for the deleteri- tific, Inc., the manufacturer of the RPI device. MAK and JAP are ous effects of excessive adipose tissue.(25) Moreover, a site- recipients of grants from the NHMRC; KLH-K, MAK, and JAP are recip- dependent association between obesity and fracture has been ients of a grant from Amgen-GSK OA-ANZBMS and Amgen Australia. reported in women, with higher fractures occurring at sites with more cortical bone, including the ankle and upper arm.(25,33,36) Conversely, Premaor and colleagues(34) have reported a Acknowledgments decreased fracture risk in most sites in men with obesity. Few studies have investigated the relationship between BMSi and The Geelong Osteoporosis Study was supported by grants from height. At this stage, we have no clear explanation about the cor- the National Health and Medical Research Council (NHMRC; pro- relation between BMSi and height based on our results. jects 299831, 628582), Amgen-GSK OA-ANZBMS, and Amgen However, various mechanisms through which height may Australia, but they played no role in the collection or interpreta- influence bone properties include hip axis and moment arm tion of data. n 4 of 6 RUFUS-MEMBERE ET AL. JBMR Plus (WOA) Authors’ roles: PGR performed the indentation measurements 16. Farr JN, Drake MT, Amin S, Melton LJ, McCready LK, Khosla S. In vivo in the presence of another trained observer (KLH-K) and drafted assessment of bone quality in postmenopausal women with type 2 diabetes. J Bone Miner Res. 2014;29(4):787–95. the manuscript. KLH-K assisted with taking measurements and administered questionnaires. AD-P assisted with training to use 17. Popp KL, Caksa S, Martinez-Betancourt A, Yuan A, Tsai J, Yu EW, et al. Cortical bone material strength index and bone microarchitecture in the OsteoProbe device and provided advice on measurement postmenopausal women with atypical femoral fractures. J Bone technique. MAK and JAP conceived and designed the study. Miner Res. 2019;34(1):75–82. JAP secured ethics approval. All authors interpreted the data, 18. Nogués X, Prieto-Alhambra D, Güerri-Fernández R, et al. Fracture dur- and guided and reviewed the manuscript. All authors read and ing oral bisphosphonate therapy is associated with deteriorated approved the final manuscript. bone material strength index. Bone. 2017;103:64–9. 19. Malgo F, Hamdy NAT, Papapoulos SE, Appelman-Dijkstra NM. Impact microindentation: consistency of serial measurements and alter- Peer Review ations in patients with Paget’s disease of the tibia. J Bone Miner Res. 2017;32(12):2375–80. The peer review history for this article is available at https:// 20. Aasarød KM, Mosti MP, Stunes AK, et al. Impaired skeletal health in patients with chronic atrophic gastritis. Scand J Gastroenterol. publons.com/publon/10.1002/jbm4.10384. 2016;51(7):774–81. 21. Malgo F, Hamdy NAT, Rabelink TJ, et al. Bone material strength index as measured by impact microindentation is altered in patients with References acromegaly. Eur J Endocrinol. 2017;176(3):339–47. 22. Güerri-Fernández R, Molina D, Villar-García J, et al. Brief report: HIV infec- 1. Randall C, Bridges D, Guerri R, et al. Applications of a new handheld tion is associated with worse bone material properties, independently of reference point indentation instrument measuring bone material bone mineral density. J Acquir Immune Defic Syndr. 2016;72(3):314–8. strength. J Med Device. 2013;7(4):410051–6. 23. Duarte Sosa D, Vilaplana L, Güerri R, et al. Are the high hip fracture 2. Yuehuei H, Robert A. Mechanical testing of bone and the bone- rates among Norwegian women explained by impaired bone mate- implant interface. mechanical testing of bone and the bone-implant rial properties? J Bone Miner Res. 2015;30(10):1784–9. interface. 1st ed. Boca Raton, FL: CRS Press; 1999. 24. Rudäng R, Zoulakis M, Sundh D, et al. Bone material strength is asso- 3. Herrera S, Diez-Perez A. Clinical experience with microindentation ciated with areal BMD but not with prevalent fractures in older in vivo in humans. Bone. 2017;95:175–82. women. Osteoporos Int. 2016;27(4):1585–92. 4. Diez-Perez A, Bouxsein ML, Eriksen EF, et al. Technical note: recom- 25. Sundh D, Rudang R, Zoulakis M, Nilsson AG, Darelid A, Lorentzon M. A mendations for a standard procedure to assess cortical bone at the high amount of local adipose tissue is associated with high cortical tissue-level in vivo using impact microindentation. Bone Reports. porosity and low bone material strength in older women. J Bone 2016;5:181–5. Miner Res. 2016;31(4):749–57. 5. Allen MR, McNerny EMB, Organ JM, Wallace JM. True gold or pyrite: a 26. Nilsson AG, Sundh D, Johansson L, et al. Type 2 diabetes mellitus is review of reference point indentation for assessing bone mechanical associated with better bone microarchitecture but lower bone mate- properties in vivo. J Bone Miner Res. 2015;30(9):1539–50. rial strength and poorer physical function in elderly women: a 6. Malgo F, Hamdy NAT, Papapoulos SE, Appelman-Dijkstra NM. Bone population-based study. J Bone Miner Res. 2017;35(5):1062–71. material strength as measured by microindentation in vivo is decreased in patients with fragility fractures independently of bone 27. Johansson L, Sundh D, Zoulakis M, et al. The prevalence of vertebral mineral density. J Clin Endocrinol Metab. 2015;100(5):2039–45. fractures is associated with reduced hip bone density and inferior peripheral appendicular volumetric bone density and structure in 7. Malgo F. Bone material strength index as measured by impact micro- older women. J Bone Miner Res. 2018;33(2):250–60. indentation is low in patients with fractures irrespective of fracture site. Osteoporos Int. 2017:28(8):2433–7. 28. Tommasini SM, Nasser P, Jepsen KJ. Sexual dimorphism affects tibia size and shape but not tissue-level mechanical properties. Bone. 8. Sosa DD, Eriksen EF. Reduced bone material strength is associated 2007;40(2):498–505. with increased risk and severity of osteoporotic fractures. An impact microindentation study. Calcif Tissue Int. 2017;101(1):34–42. 29. Nyman JS, Roy A, Tyler JH, Acuna RL, Gayle HJ, Wang X. Age-related factors affecting the postyield energy dissipation of human cortical 9. Furst JR, Bandeira LC, Fan WW, et al. Advanced glycation endpro- bone. J Orthop Res. 2007;25(5):646–55. ducts and bone material strength in type 2 diabetes. J Clin Endocrinol Metab. 2016;101(6):2502–10. 30. Burr DB, Forwood MR, Fyhrie DP, Martin RB, Schaffler MB, Turner CH. Bone microdamage and skeletal fragility in osteoporotic and stress 10. Pérez-Sáez MJ, Herrera S, Prieto-Alhambra D, et al. Bone density, fractures. J Bone Miner Res. 1997;12(1):6–15. microarchitecture, and material strength in chronic kidney disease patients at the time of kidney transplantation. Osteoporos Int. 31. Zioupos P. Accumulation of in-vivo fatigue microdamage and its rela- 2017;28(9):2723–7. tion to biomechanical properties in ageing human cortical bone. J Microsc. 2001;201(2):270–8. 11. Holloway-Kew KL, Rufus-Membere P, Anderson KB, et al. Bone mate- rial strength index is associated with prior fracture in men with and 32. Gonnelli S, Caffarelli C, Nuti R. Obesity and fracture risk. Clin Cases without moderate chronic kidney disease. Bone. 2020;133:115241. Miner Bone Metab. 2014;11(1):9–14. 12. Henry MJ, Pasco JA, Seeman E, Nicholson GC, Sanders KM, 33. Johansson H, Kanis JA, Odén A, et al. A meta-analysis of the associa- Kotowicz MA. Assessment of fracture risk: value of random tion of fracture risk and body mass index in women. J Bone Miner population-based samples—the Geelong osteoporosis study. J Clin Res. 2014;29(1):223–33. Densitom. 2001;4(4):283–9. 34. Premaor MO, Compston JE, Fina Avilés F, et al. The association 13. Pasco JA, Nicholson GC, Kotowicz MA. Cohort profile: Geelong Oste- between fracture site and obesity in men: a population-based cohort oporosis Study. Int J Epidemiol. 2012;27:1585–92. study. J Bone Miner Res. 2013;28(8):1771–7. 14. Markanday S, Brennan SL, Gould H, Pasco JA. Sex-differences in rea- 35. Nielson CM, Marshall LM, Adams AL, et al. BMI and fracture risk in sons for non-participation at recruitment: Geelong Osteoporosis older men: The osteoporotic fractures in men study (MrOS). J Bone Study. BMC Res Notes. 2013;6:104. Miner Res. 2011;26(3):496–502. 15. Rufus-Membere P, Holloway-Kew KL, Diez-Perez A, Kotowicz MA, 36. Compston JE, Flahive J, Hosmer DW, et al. Relationship of weight, Pasco JA. Feasibility and tolerability of bone impact microindentation height, and body mass index with fracture risk at different sites in testing: a cross-sectional, population-based study in Australia. BMJ postmenopausal women: the global longitudinal study of osteoporo- Open. 2018;8:e023959. sis in women (GLOW). J Bone Miner Res. 2014;29(2):487–93. JBMR® Plus NORMATIVE DATA FOR IMPACT MICROINDENTATION 5 of 6 n 37. Faulkner KG, Cummings SR, Black D, Palermo L, Glüer C-C, for by cortical porosity of relatively thinner cortices. J Bone Miner Genant HK. Simple measurement of femoral geometry predicts Res. 2013;28(9):2017–26. hip fracture: the study of osteoporotic fractures. J Bone Miner 39. Guerri-Fernandez R, Molina-Morant D, Villar-Garcia J, Herrera S, Res. 1993;8(10):1211–7. Gonzalez-Mena A, Guelar A, et al. Bone density, microarchitecture, and 38. Bjørnerem Å, Bui QM, Ghasem-Zadeh A, Hopper JL, Zebaze R, tissue quality after long-term treatment with tenofovir/emtricitabine Seeman E. Fracture risk and height: an association partly accounted or abacavir/Iamivdine. J Acquir Immune Defic Syndr. 2017;75:322–7. n 6 of 6 RUFUS-MEMBERE ET AL. JBMR Plus (WOA)