The use of highly active antiretroviral therapy (ART) has led to a significant reduction in HIV-related morbidity and mortality. However, both HIV infection and ART itself have been associated with the development of numerous acute and long-term toxicities [1–3]. The mechanisms underlying these toxicities are still not fully understood [4,5].
Some studies claim that ART is the main factor responsible for bone deterioration. tenofovir disoproxil fumarate (TDF), an acyclic nucleotide analog of adenosine monophosphate, is a core component of many ART regimens. Although TDF is highly effective against HIV, it has also been associated with a significant decrease in bone mineral density (BMD) with antiretroviral initiation , and also with fracture and osteomalacia . A large observational study based on the Veterans Cohort showed that TDF treatment significantly increased risk of osteoporotic fracture .
Although TDF-treated patients commonly show lower BMD than those not treated with TDF . However, even though there is an increased fractured risk, not all patients exposed to TDF will experience fracture. Hence, it is important to be able to identify patients with greater fracture risk. Bone health, as measured by BMD, has been recommended as a measurement for guiding clinical decisions. BMD measured by dual-energy X-ray absorptiometry (DXA) is the gold standard for diagnosing osteoporosis because of their reproducibility, large normative data, noninvasive nature, little time requirement for procedure, and minimal radiation exposure. However, it only provides information about the quantity of bone tissue, and thereby is only partly predictive of fracture risk . On the other hand, bone mechanical strength (or bone fragility) integrates various features in addition to BMD, such as bone microarchitecture and bone tissue quality.
The methods used to clinically evaluate bone have been enhanced by incorporating new techniques into already available diagnostic tools, improving our ability to identify patients with decreased bone strength. For instance, the trabecular bone score (TBS) is a noninvasive method based on DXA images of the lumbar spine. This method, which is an independent predictor of fracture risk, provides skeletal information about bone microarchitecture that is not fully captured by standard BMD measurements . Compared with DXA, TBS is capable of more accurately identifying patients with fractures .
Bone mechanical properties at the tissue level can be directly measured in patients using a recently developed technique called reference point indentation (RPI) (Fig. 1). RPI assesses the resistance of bone to the penetration of a microscopic probe, and hence, bone tissue mechanical performance. In previous validation studies, RPI has been shown to be superior to DXA in identifying patients with fractures [13,14]. We previously used bone RPI to show that HIV-positive patients have lower bone strength than HIV-negative controls, independent of BMD . Moreover, BMSi in those studies was poorly correlated with BMD .
As the combined use of these three techniques offers a more comprehensive estimate of bone strength, the principal goal of the current study was to analyse the changes induced by TDF during the first year of treatment in the components of bone strength such as bone density, trabecular microarchitecture, and tissue-level bone quality in a cohort of ART-naive HIV patients initiating TDF-containing ART. As secondary goals, we also sought to determine the factors associated with bone material strength (BMS) at baseline and at follow-up,
Patients and methods
The study population consisted of consecutively-recruited ART-naive HIV patients who were about to start HIV treatment. We considered as ineligible those patients who had previously been received treatments that could potentially affect the bone, such as systemic glucocorticoids or antiosteoporotic medications. We also excluded patients who had previously been diagnosed with chronic kidney disease, chronic endocrine conditions, malabsorption syndrome, advanced liver disease, neoplasia, and bone diseases. This study was approved by the local Clinical Research Ethics Committee, and all participants gave written informed consent (Ethics Committee number 2013/5250/I).
All patients received elvitegravir/cobicistat daily as the main treatment with TDF/emtricitabine as the backbone, and were followed up during 48 weeks.
Patients had scheduled visits before treatment (i.e. baseline, visit 0), and at 24 (visit 1), and 48 weeks (visit 2). At the first visit, the patient's clinical history was recorded, and a general physical examination was performed. Lateral spinal X-rays were taken and assessed by two independent observers to detect any vertebral fractures, defined as deformities of grade I or above (a loss of >20% of vertebral height). Wherever a documented history or radiological report was available, we also recorded any peripheral fractures. At each visit, lab testing, BMD assessment, TBS, and bone microindentation were performed. All patients were recommended to have an enforced calcium diet and vitamin D supplementation if necessary.
Bone mineral density
Bone mineral density was measured at the lumbar spine and hip using DXA (Hologic QDR 4500 SR; Hologic, Inc, Bedford, Massachusetts, USA). Osteopenia was defined as a T-score between −1 and −2.5 SD, and osteoporosis as a T-score −2.5 or less, considering the lowest of lumbar spine, femoral neck, and total hip. Low BMD was defined as osteopenia or osteoporosis.
Trabecular bone score
Trabecular bone score is an analytical tool that captures information on trabecular microarchitecture by taking novel gray-level texture measurements of DXA images of the lumbar spine. TBS is related to other 3D bone characteristics such as trabecular number, trabecular separation, and connectivity density [11,17]. A TBS score above 1.35 is considered normal; a value between 1.35 and 1.20 denotes partially degraded bone microarchitecture; and a value below 1.20 denotes strongly degraded bone, or weak, fracture-prone microarchitecture.
We evaluated TBS using iNsight v2.1 (Med.Imaps, Merignac, France) at the same regions where lumbar spine BMD was measured.
Bone microindentation measurements
Bone microindentation was measured using an Osteoprobe instrument (Active Life Scientific, Santa Barbara, California, USA) according to a protocol described previously . Bone microindentation yields a dimensionless quantifiable parameter called BMS index (BMSi), which is positively correlated with bone tissue quality.
To minimize interobserver variation, all measurements for this study were taken by the same investigator (R.G.F.). As previously described, the microindentation procedure is minimally invasive, well tolerated, painless, and takes less than 5 min. The software provides immediate results. Contraindications for this technique included local skin infection, significant local oedema, and/or thick subcutaneous adipose tissue at the site of indentation.
We used chemiluminescent immunoassays (CLIA) to determine various laboratory parameters specific to bone (based on fasting blood samples). Each immunoassay had an inter-assay coefficient of variation, iCV, of 10%. Specifically, we measured levels of intact parathyroid hormone (iPTH; Siemens, Munich, Germany), bone alkaline phosphatase (Roche Diagnostics, Indianapolis, Indiana, USA), amino propeptide of type I collagen (PINP, Roche Diagnostics), collagen type I C-telopeptide (CTX, Roche Diagnostics), serum 25-hydroxyvitamin D (Roche Diagnostics).
We calculated sample size as previously described [18,19], with alpha and beta risks of 0.05 and 0.2, respectively. In a two-sided t-test, we required at least 35 participants to detect a statistically significant difference of at least 5% in BMSi between paired measurements. We assumed the SD to be 6, and anticipated a dropout rate of 10%.
Statistical significance was determined using the Mann–Whitney U-test for continuous variables and the chi-square test for categorical variables. We analyzed paired observations using the Wilcoxon rank t-test. We used the Spearman correlation coefficient to test for correlation between the BMSi, TBS, and BMD with laboratory values obtained after treatment and their respective baseline values.
After checking for normal distribution using the Shapiro–Wilk W-test, we used Student's paired t-test to evaluate significant changes in BMSi between visits. We also assessed changes in BMD, TBS, and BMSi over time (i.e. over the three visits) using repeated measures analysis of variance (ANOVA).
BMSi values were plotted against femoral neck BMD, age, P1NP, and CTX, and analyzed by linear regression.
We fit a multivariable linear regression analysis to determine the independent effect of BMD and other factors such as bone turnover markers and baseline CD4+ T cells on changes in BMSi (48-w to baseline). The terms age, sex, BMI were included in the model to ensure that the effect on change in BMSi was not confounded with these variables.
We performed all analyses using Stata/IC 14 (College Station, Texas, USA).
Our study included 40 HIV patients, 33 men (82.5%), and seven women (17.5%). The baseline characteristics of the patients are shown in Table 1. The median age of the patients at baseline was 38 years [interquartile range (IQR) 31.7–44.2], and the median time as HIV diagnosis was 1 year (IQR: 0–4). Mean BMI was 23.6 (SD 2.6) kg/m2. Before initiation of treatment, 18 patients (45%) already showed low BMD as measured by DXA, of whom 2 (5%) presented osteoporosis. We observed no significant differences in the main variables studied between patients with low BMD and patients with normal BMD.
Changes in bone turnover, density, microarchitecture, and material properties with antiretroviral therapy initiation
Markers of bone turnover (resorption and formation) significantly increased during the first year of treatment (see Table 2): amino propeptide of type 1 collagen increased from 50.6 (19.8) to 76.6 (25.8) ng/ml [relative difference (diff.) 51.38%, P < 0.001]; bone alkaline phosphatase from 16.7 (14.2) to 19.9 (7.98) μg/ml (diff. 19.16%, P = 0.087); and type I collagen C-telopeptide from 0.309 (0.114) to 0.436 (0.137) ng/ml (diff. 41.1%, P < 0.001) .
Table 2 also shows changes in surrogate markers of bone strength (BMD, TBS, and BMSi). As expected, we observed a significant decrease in spine BMD from 0.984 (0.11) to 0.958 (0.11) g/cm2 (−3.25%, P < 0.001) and femoral neck BMD from 0.837 (0.12) to 0.805 (0.11) g/cm2 (−3.82%, P = 0.016) between baseline and 48 weeks of treatment. We found no significant differences in total hip BMD at any time during patient follow-up [0.954 (0.11) versus 0.943 (0.12); −1.15%, P = 0.547].
Bone microarchitecture score at the spine, as measured by TBS, also significantly decreased from 1.357 (0.09) to 1.322 (0.09; −2.5%, P = 0.011) between baseline to 48 weeks of treatment.
In contrast to the BMD and TBS results, BMSi values were significantly higher than at baseline [89.04 (4.2) versus 86.07 (6.1); 3.49%, P < 0.001] after 48 weeks of TDF-based ART treatment, indicating improved bone material properties (Fig. 1)
Bone material strength index bivariate correlation
In terms of bone tissue quality, measured by microindentation, our bivariate model showed that BMSi at baseline was correlated with age (r = −0.28, P = 0.07), CD4+ nadir cell count (r = 0.365, P = 0.02), TBS at baseline (r = 0.372, P = 0.03), and hip BMD (r = 0.388, P = 0.014), but not with vitamin D levels, longer time of HIV infection. Figure 2 shows the correlation between BMSi and CD4+ cell count and hip BMD.
No correlation was found between change in any of the three markers of bone health (BMD, TBS, BMSi) and changes in bone turnover markers.
Multivariable linear regression analysis
In a multivariate linear regression analysis, lower TBS (ß-coefficient, −20.6; P = 0.01) and BMSi (ß-coefficient, −0.59; P < 0.001) at baseline were associated with greater BMSi increases at 48-w from baseline, after adjusting for age, sex, and BMI. No statistical significant influence of CD4+ nadir cell count, baseline GFR, baseline vitamin D levels was found.
Gain in bone tissue quality
Table 3 shows differences in the groups that experience positive change of BMSi after 48-weeks of TDF (increase BMSi) and the group that presents a negative change (decline BMSi). There were no differences in age, sex, nadir CD4+ cell count between groups. But BMSi at baseline and TBS at baseline were significantly higher in those who experienced declined BMSi.
During this study, we did not encounter any complications related to the techniques used.
To our knowledge, this is the first comprehensive study to assess different levels of bone health in naive HIV patients, who start ART. We not only assessed BMD, but also examined quantity, structure, and tissue quality. TDF-based treatment regimens are known to induce a greater loss in BMD than other types of treatment . Here, we show that not all bone parameters influencing fracture risk are affected by TDF-based ART in the same way. Although BMD and trabecular microarchitecture decrease in a comparable manner, bone tissue quality, as measured by microindentation, is maintained or even improves after 48 weeks of TDF-based ART.
Our data support the finding that BMD declines during the first year of TDF-based HIV treatment (average 3% during the first 48 weeks), as observed in other studies. For instance, a randomized clinical trial showed a 2–6% decline in BMD within 48 weeks after ART initiation . The use of TDF, and to a lesser extent, protease inhibitors, is associated with accelerated decline of BMD in ART-naive patients . BMD shows a significantly greater decline in patients treated with TDF than those who undergo other ART regimens. McComsey et al. demonstrated that, after 96 weeks of treatment, patients taking ABC–lamivudine (3TC) showed a 1.3 and 3.3% decrease in spine and hip BMD, respectively. In contrast, those receiving TDF–emtricitabine (FTC) treatment showed a decrease of 2.6 and 4.0%, respectively.
Two prospective studies of patients beginning their first ART regimen showed that TDF-based regimens led to a significantly greater decrease in both spine and hip BMD than abacavir (ABC)-containing regimens [6,23]. Moreover, decreased BMD has been observed both in HIV-positive and HIV-negative individuals undergoing TDF treatment. In HIV-negative individuals, TDF-based preexposure prophylaxis (PrEP) was associated with a statistically significant decrease in BMD by week 24, albeit with more stable values thereafter . The mechanisms underlying the effects of TDF on BMD remain unclear, although chronic abnormal phosphaturia at least partly explains the progressive decline in bone density decline that is associated with TDF therapy .
The incidence of fracture among HIV-infected individuals is only partly correlated with reduced BMD, age, and sex, suggesting that other factors are important, such as altered bone microstructure and/or bone material. In the EuroSIDA study, which investigated the incidence of fracture and associated risk factors, a multivariable analysis showed that fracture was associated with older age, lower BMI, Caucasian ethnicity, and intravenous drug use as the mode of transmission. Furthermore, TDF exposure was found to be associated with fracture incidence whenever assessed as a dichotomous variable (ever/never use of TDF). However, after multivariable adjustment, cumulative TDF exposure was not associated with increased risk of fractures , raising the possibility of individual susceptibility to TDF-induced fractures.
Classically, bone deterioration has been measured in terms of osteopenia and osteoporosis, as defined by BMD, but other techniques such as TBS can measure other aspects of bone quality. We found that TBS, a measure of bone microarchitecture, also significantly decreased after starting ART. Our data suggest that TDF affects not only the quantity of bone but also its organization. A significant decrease in BMD in turn leads to loss of trabeculae. Compared with BMD, TBS is more closely correlated with fracture. Gazzola et al. found that individuals with and without fracture had a mean TBS of 1.226 and 1.338, respectively, but reported no significant difference in BMD. Similarly, Ciullini et al. found that fracture was more closely correlated to TBS than to BMD in patients with HIV.
Another key component of bone strength is bone tissue quality. We measured this parameter in vivo using microindentation, and found, surprisingly, that BMSi significantly increased during the first year of TDF-based treatment. Unlike the observed decline in BMD and TBS, this indicates an improvement in bone tissue quality. In a previous study, we found that HIV patients had a poorer bone tissue quality than HIV-negative individuals, despite having similar BMD . We also found that patients undergoing long-term TDF treatment showed worse material properties than those receiving long-term ABC treatment . Nonetheless, these differences were quite low and only appeared after adjusting for other covariates. This suggests that TDF has a lower impact on bone tissue quality than expected according to changes in BMD.
As microindentation is more sensitive than densitometry for detecting early bone changes [27,28], and can explain bone fractures not attributable to loss of BMD , we think our results suggest that not all patients exposed to TDF experience bone deterioration. With BMD by DXA, we are measuring the mineral component of bone. We are quantifying the amount of mineral content. It is well known that TDF through tubular damage, phosphate loss, and the counteregulatory mechanisms could induce mineral bone loss detected by DXA . However, microindentation provides new information on BMSi that was previously unavailable. Fractures in both trabecular bone and cortical bone begin with the separation of mineralized collagen fibrils and the initiation of cracks. Moreover, this technique creates cracks that are very similar to those observed in experimental bone fractures . Precisely it could be hypothesized that once starting TDF, the tubular damage and the counterregulatory mechanism induced lead to an increased bone loss. However, at the same time, the control of HIV infection, along with the reduction of the inflammatory state and OPG/RANK/RANKL system changes could lead to microenvironmental changes at bone , recovering some of its physiological functions (i.e. bone turnover producing normal bone), independently of bone turnover .
We hypothesize that some patients are more prone to bone impairment than others. Ability to identify patients at higher risk of deteriorating bone tissue quality and fracture is an important milestone. Sosa et al. found that combining microindentation and BMD measurements increases the ability to discriminate, which patients are more prone to fracture. As we are measuring the different components of bone strength with these techniques, a combination of all of them would likely improve bone resistance to fracture in ART-treated patients. Probably baseline measurement with them should be considered, and further larger studies should determine how and how often the follow-up assessments should be done.
In the interpretation of the opposite results between these techniques, we should consider two possibilities: first, the TDF-fracture association is actually a spurious finding because of biases inherent to the retrospective analyses concluding that there is such a risk. In that case, BMSi would be an interesting marker to monitor the bone toxicity; second that TDF exposure indeed increases fracture risk. In this case, either the BMSi is not an optimal predictor of fracture risk, or it is being measured too early. There might be a temporary increase (maybe owing to control of viremia) followed by a long-term decline. In any case, longer and larger studies are needed to answer these questions.
Our study has some limitations, including the fact that it is a single-center study with a limited sample size. Also, microindentation is a new technique, and although we have extensive experience in using it, this longitudinal study should be replicated by other centers who are also using this technique. Also, not only larger sample sizes but also studies with longer follow-up are needed to elucidate the long-term trajectories of TBS and BMSi. Another limitation of our study is that it may be subject to selection bias: the patients in our study were mainly healthy, and did not have any nutritional deficiencies, or other ‘classical[Combining Diaeresis] risk factors for fracture and low BMD. Nonetheless, our findings can be fully explained by HIV, and by the patient–ART interaction. Another key limitation is the fact that all patients were receiving elvitegravir/cobicistat/TDF/emtricitabine, and therefore, our finding may not be generalizable to other regimens.
In conclusion, we present a longitudinal study in which ART-naive HIV patients are assessed prospectively, and bone tissue quality is assessed directly. We find that even while BMD and TBS decrease after 48 weeks of TDF treatment, bone tissue quality improved in most patients. As the different components of bone strength behave differently, an overarching approach to assess bone toxicity in TDF-treated patients is needed.
Funding: This work was supported by a European Regional Development Fund (ERDF) award and by the Instituto de Salud Carlos III (ISCIII) under grant agreement PI16/01860 from the Spanish Ministry of Health. T.T.B. is supported in part by NIH (NIAID) K24 AI120834.
Conflicts of interest
There are no conflicts of interest.
1. Casado JL, Santiuste C, Vazquez M, Bañón S, Rosillo M, Gomez A, et al. Bone mineral density decline according to renal tubular dysfunction and phosphaturia in tenofovir-exposed HIV-infected patients
2. Güerri-Fernandez R, Vestergaard P, Carbonell C, Knobel H, Avilés FF, Castro AS, et al. HIV infection is strongly associated with hip fracture risk, independently of age, gender, and comorbidities: a population-based cohort study
. J Bone Miner Res
3. Prieto-Alhambra D, Güerri-Fernández R, De Vries F, Lalmohamed A, Bazelier M, Starup-Linde J, et al. ‘HIV infection and its association with an excess risk of clinical fractures and: a nation-wide case-control study
. J Acquir Immune Defic Syndr
4. Lagathu C, Cossarizza A, Béréziat V, Nasi M, Capeau J, Pinti M. Basic science and pathogenesis of ageing with HIV: potential mechanisms and biomarkers
2017; 31 (suppl 2):S105–S119.
5. Hunt PW, Lee SA, Siedner MJ. Immunologic biomarkers, morbidity, and mortality in treated hiv infection
. J Infect Dis
2016; 214 (suppl 2):S44–S50.
6. Stellbrink H-J, Orkin C, Arribas JR, Compston J, Gerstoft J, Van Wijngaerden E, et al. Comparison of changes in bone density and turnover with abacavir-lamivudine versus tenofovir-emtricitabine in HIV-infected adults: 48-week results from the ASSERT study
. Clin Infect Dis
7. Woodward C, Hall A, Williams I, Madge S, Copas A, Nair D, et al. Tenofovir-associated renal and bone toxicity
. HIV Med
8. Bedimo R, Maalouf NM, Zhang S, Drechsler H, Tebas P. Osteoporotic fracture risk associated with cumulative exposure to tenofovir and other antiretroviral agents
9. Moyle GJ, Stellbrink H-J, Compston J, Orkin C, Arribas JR, Domingo P, et al. ASSERT Team. 96-week results of abacavir/lamivudine versus tenofovir/emtricitabine, plus efavirenz, in antiretroviral-naive, HIV-1-infected adults: ASSERT study
. Antivir Ther
10. Hans D, Durosier C, Kanis JA, Johansson H, Schott-Pethelaz A-M, Krieg M-A. Assessment of the 10-year probability of osteoporotic hip fracture combining clinical risk factors and heel bone ultrasound: the EPISEM Prospective Cohort of 12,958 elderly women
. J Bone Miner Res
11. Hans D, Goertzen AL, Krieg M-A, Leslie WD. Bone microarchitecture assessed by TBS predicts osteoporotic fractures independent of bone density: the Manitoba study
. J Bone Miner Res
12. Ciullini L, Pennica A, Argento G, Novarini D, Teti E, Pugliese G, et al. Trabecular bone score (TBS) is associated with sub-clinical vertebral fractures in HIV-infected patients
. J Bone Miner Metab
13. Diez-Perez A, Güerri R, Nogues X, Cáceres E, Peña MJ, Mellibovsky L, et al. Microindentation for in vivo measurement of bone tissue mechanical properties in humans
. J Bone Miner Res
14. Malgo F, Hamdy NAT, Papapoulos SE, Appelman-Dijkstra NM. Bone material strength as measured by microindentation in vivo is decreased in patients with fragility fractures independently of bone mineral density
. J Clin Endocrinol Metab
15. Güerri-Fernández R, Molina D, Villar-García J, Prieto-Alhambra D, Mellibovsky L, Nogués X, et al. HIV infection is associated with worse bone material properties, independently of bone mineral density
. J Acquir Immune Defic Syndr
16. Duarte Sosa D, Fink Eriksen E. Women with previous stress fractures show reduced bone material strength Microindentation measurements in a retrospective case-control study of 60 sub- jects
. Acta Orthop
17. Silva BC, Leslie WD, Resch H, Lamy O, Lesnyak O, Binkley N, et al. Trabecular bone score: a noninvasive analytical method based upon the DXA image
. J Bone Miner Res
18. Güerri-Fernández R, Molina-Morant D, Villar-García J, Herrera S, González-Mena A, Guelar A, et al. Bone density, microarchitecture and tissue quality following long-term treatment with tenofovir/emtricitabine or abacavir/lamivudine
. J Acquir Immune Defic Syndr
19. Güerri-Fernández R, Molina D, Villar-García J, Prieto-Alhambra D, Mellibovsky L, Nogués X, et al. Brief report: HIV infection is associated with worse bone material properties, independently of bone mineral density
. J Acquir Immune Defic Syndr
20. Moran CA, Neale Weitzmann M, Ofotokun I. Bone loss in HIV infection
. Curr Treat Options Infect Dis
21. Brown TT, McComsey GA, King MS, Qaqish RB, Bernstein BM, da Silva BA. Loss of bone mineral density after antiretroviral therapy initiation, independent of antiretroviral regimen
. J Acquir Immune Defic Syndr
22. Brown TT, Qaqish RB. Antiretroviral therapy and the prevalence of osteopenia and osteoporosis: a meta-analytic review
23. McComsey GA, Kitch D, Daar ES, Tierney C, Jahed NC, Tebas P, et al. Bone mineral density and fractures in antiretroviral-naive persons randomized to receive abacavir-lamivudine or tenofovir disoproxil fumarate-emtricitabine along with efavirenz or atazanavir-ritonavir: AIDS Clinical Trials Group A5224 s, a substudy of ACTG A5202
. J Infect Dis
24. Mulligan K, Glidden DV, Anderson PL, Liu A, McMahan V, Gonzales P, et al. Preexposure Prophylaxis Initiative Study Team. Effects of emtricitabine/tenofovir on bone mineral density in HIV-negative persons in a randomized, double-blind, placebo-controlled trial
. Clin Infect Dis
25. Borges AH, Hoy J FE. Antiretrovirals, fractures, and osteonecrosis in a large European HIV cohort [abstract 46]
. Conference on Retroviruses and Opportunistic Infections (CROI)
, 22–25 February 2016; Boston, USA.
26. Gazzola L, Savoldi A, Bai F, Magenta A, Dziubak M, Pietrogrande L, et al. Assessment of radiological vertebral fractures in HIV-infected patients: clinical implications and predictive factors
. HIV Med
27. Duarte Sosa D, Vilaplana L, Güerri R, Nogués X, Wang-Fagerland M, Diez-Perez A, et al. Are the high hip fracture rates among norwegian women explained by impaired bone material properties?
. J Bone Miner Res
28. Mellibovsky L, Prieto-Alhambra D, Mellibovsky F, Güerri-Fernández R, Nogués X, Randall C, et al. Bone tissue properties measurement by reference point indentation in glucocorticoid-induced osteoporosis
. J Bone Miner Res
29. Fantner GE, Hassenkam T, Kindt JH, Weaver JC, Birkedal H, Pechenik L, et al. Sacrificial bonds and hidden length dissipate energy as mineralized fibrils separate during bone fracture
. Nat Mater
30. Brown TT, Ross AC, Storer N, Labbato D, McComsey GA, Guaraldi G. Bone turnover, OPG/RANKL, and inflammation with antiretroviral initiation: comparison of tenofovir-vs. non tenofovir regimens
. Antivir Ther
31. Sosa DD, Eriksen EF. Reduced bone material strength is associated with increased risk and severity of osteoporotic fractures. An impact microindentation study
. Calcif Tissue Int