The current life expectancy of patients living with HIV (PLHIV) is increasing and approaching the life expectancy of the general population.1 Thus, current challenges for PLHIV are disorders associated with older age such as osteoporosis. It is estimated that osteoporotic fractures account for 0·5 million euros excess cost in PLHIV in Denmark where there is 6200 PLHIV,2 which makes it more costly than acute myocardial infarction and stroke in PLHIV.3 Thus, in larger populations, the expense would increase proportionally. The aim of this systematic review is to investigate the fracture risk and longitudinal changes in bone mineral density (BMD) in PLHIV and to compare BMD in PLHIV with non-HIV infected controls in randomized controlled trials and observational studies. Based on this evidence, we aim to calculate the expected fracture rates and compare with the observed to determine whether BMD explains the fracture risk. Furthermore, we aim to investigate the effects of antiosteoporotic treatment on bone health in PLHIV in randomized controlled trials and observational studies and collate current guidelines on osteoporosis in PLHIV in an attempt to determine an optimal guideline for the management of osteoporosis in PLHIV.
In this systematic review and meta-analysis, we adhered to the PRISMA guidelines4 and are registered at PROSPERO (CRD42018103760).
Data Sources, Searches, and Eligibility Criteria
A systematic literature search was conducted on July 5, 2017, and updated on June 22, 2018. The databases Medline at PubMed and EMBASE were explored using the search terms in free-text: “HIV” and “fracture,” “HIV” and “bone turnover,” and “HIV” and “bone mineral density.” Free-text was applied to collect all records including most recent that are not indexed in the thesaurus. The updated search from EMBASE is presented in Appendix 1, Supplemental Digital Content, http://links.lww.com/QAI/B391. The systematic review aimed to investigate the fracture risk, BMD, and longitudinal changes of BMD in PLHIV compared with controls, as well as the effects of antiosteoporotic treatment in PLHIV. Furthermore, we sought to identify current guidelines for the management of osteoporosis in PLHIV. Studies using the same population and solely reporting on bone turnover markers were excluded. Studies with PLHIV aged 18–99 years were included. No restrictions were made to geographic area, language of the publication, sex, body mass index (BMI), use of pharmaceuticals, or potential comorbidities. Only studies reporting BMD results of dual-energy x-ray absorptiometry (DXA) of the hip and/or lumbar spine (LS) were included. When evaluating the prevalence of vertebral fractures (VFs), only studies with tests of diagnostic accuracy were included in the pooled analysis (eg, x-ray or computed tomography of the spine). No restriction was made on publication date or language.
Data Extraction and Quality Assessment
The papers were screened and selected independently by 2 authors (J.S.-L. and S.B.R.). If disagreement occurred, this was solved by discussion and adjudicated by a third author if necessary.
Each full text paper was assessed to determine its eligibility. Data on design of the study, follow-up period, BMD, fracture rates, and number, age, BMI, gender, and CD4 cell count on the included subjects were extracted and tabulated independently by one of the 2 authors.
Each study was graded for the level of evidence for prognostic studies or levels of evidence for therapeutic studies.5
The fracture rates [odds ratio, relative risk (RR), and hazard ratio], prevalence of VF and the mean values of BMD and 95% confidence intervals (CIs), SDs, or SEs of BMD were collected. SD was used for the analyses. From the data, SD was collected or SD was calculated from SE or CI. If SD, SE, or CI were not given in the data, medians and interquartile range were presented; SD was calculated from interquartile range if the data were assumed to be normally distributed. If estimates were not given in text but presented in figures, they were extracted from the figures. Studies not presenting the estimates described above were not included in the meta-analysis. A common weighted estimate was calculated using random-effects model. A pooled analysis was only performed if 3 or more studies, or study subgroups, were available. Heterogeneity among studies was calculated by I2 analysis. The possibility of a publication bias was evaluated visually by funnel plot. Comparison analyses of specific treatments were performed [eg, antiretroviral treatment (ART) and antiosteoporotic treatment]. The RevMan 5.3 software program was applied to make these calculations and graphics. To investigate BMD differences between PLHIV and controls and calculate expected fracture rates, estimates of SD for BMD were obtained from the BMD reference database in American men and women.6 To calculate expected fracture risks, the BMD was converted applying the estimates of Marshall et al7 The expected fracture risk was calculated as cz, where c is a constant and z is the observed z-score difference between patients with HIV and controls. Metaregression analysis was performed by STATA 8 to investigate whether BMI differences explained differences in BMD.
Figure 1 displays selection of included papers. In the systematic review, 142 papers were included, and of these, 87 papers provided data that were eligible for meta-analysis. The studies were heterogeneous in the characteristics of the study populations, although studies mainly reported on young- or middle-aged PLHIV (younger than age 60). Tables 1–5, Supplemental Digital Content, http://links.lww.com/QAI/B391, provide characteristics of the included studies.
Fracture Risk in PLHIV
Forty-two papers were included with details on fracture risk in PLHIV. Studies assessing prevalence of VF in PLHIV are presented in Table 1, Supplemental Digital Content, http://links.lww.com/QAI/B391, and studies assessing the risk of any fracture in PLHIV are presented in Table 2, Supplemental Digital Content, http://links.lww.com/QAI/B391. The number of PLHIV in the studies ranged from 92 to 96,235 and the mean age in the PLHIV ranged from 36 to 57 years, and most studies reported on a majority of men. The evidence level for reporting of the prevalence of VF was III and for the risk of fracture between III-I dependent on the specific study. Twenty-one papers were included in meta-analysis. Twelve studies reported that the prevalence of VF ranged from 4.1% to 47%; however, 10 of these studies reported prevalence between 12% and 27%. In pooled analysis, the prevalence of VF in PLHIV was 22% based on 438 fractures. Supplementary Figure 1 presents the prevalence of VF by the mean age of PLHIV in the studies. Because of lack of data, we did not analyze the risk of incident VF in PLHIV. In the pooled analysis of 9 studies, reporting on 64,633 PLHIV and 4,223,060 controls, the risk of any fracture was 1.53 (95% CI: 1.46 to 1.61) fold increased (Fig. 2). Similarly, in PLHIV, the risk of a fragility fracture was 1.51 (95% CI: 1.41 to 1.63) compared with controls (Fig. 2), and omitting the 2 case-control studies8,9 from this analysis did not change the risk (RR = 1.50, 95% CI: 1.39 to 1.61). The risk of a hip fracture was assessed in 9610 PLHIV, and in the pooled analysis, the risk was 4.09 (95% CI: 3.03 to 5.52) compared with controls (Fig. 2). The risk of hip fracture was also increased when omitting the case-control study8 from this analysis (RR = 4.05, 95% CI: 2.99 to 5.49). The heterogeneity of the studies was high (77%) for the analyses of any fracture and fragility fracture, whereas it was low (0%) for the analysis of hip fracture.
Bone Mineral Density in PLHIV Compared With Controls and Expected Relative Fracture Risks
Twenty-eight papers were included with BMD at the hip and LS measured by DXA in PLHIV compared with controls. Studies assessing BMD in PLHIV and controls are presented in Table 3, Supplemental Digital Content, http://links.lww.com/QAI/B391. The number of PLHIV in the studies ranged from 16 to 328, and the number of controls ranged from 14 to 402. The mean age ranged from 24 to 61 years and 25 to 62 years for PLHIV and controls, respectively. Most studies reported predominantly on men and BMI were in general between 22 and 28 kg/m2 for both PLHIV and controls. The evidence level for reporting on the difference in BMD between PLHIV and controls was III, due to the cross-sectional nature of the studies. Twenty-four papers were included in the meta-analysis. In a pooled analysis of 3628 PLHIV and 4108 controls, the LS BMD was significantly lower in PLHIV −0.04 (95% CI: −0.05 to −0.03) g/cm2 (see Figure 2, Supplemental Digital Content, http://links.lww.com/QAI/B391). Similarly, BMD was lower at the total hip (TH) in PLHIV compared with controls −0.04 (95% CI: −0.05 to −0.02) g/cm2 (see Figure 3, Supplemental Digital Content, http://links.lww.com/QAI/B391). In a metaregression analysis, differences in BMI between PLHIV and controls did not influence observed differences in BMD at the LS (β = −14.3, 95% CI: −56.2 to 27.6) or TH (β = −8.41, 95% CI: −120 to 103). The metaregression was based on 22 and 16 studies for LS and TH, respectively.
The differences in BMD were converted into Z-scores, and the expected RRs of fracture were calculated. Table 1 displays the expected and observed fracture risks in PLHIV.
Longitudinal Changes in Bone Mineral Density in PLHIV
Fifty-five papers were included with DXA-measured BMD at the hip and LS at baseline and longitudinal follow-up in PLHIV. Studies assessing longitudinal changes of BMD in PLHIV are presented in Table 4, Supplemental Digital Content, http://links.lww.com/QAI/B391. Thirty-six papers were included in the meta-analysis. Within 1 year, the average changes in LS and TH in PLHIV compared with baseline were −1.47% (95% CI: −2.10 to −0.85) and −1.53% (95% CI: −2.05 to −1.02), respectively. Studies assessing BMD in the period 1–2 years after baseline report changes of −1.45% (95% CI: −1.56 to −1.35) and −1.85% (95% CI: −2.57 to −1.13) for LS and TH, respectively. Studies evaluating changes after 2 years or more compared with baseline report changes in BMD of −1.17% (95% CI: −2.00 to −0.34) and −2.34% (95% CI: −3.21 to −1.46) for LS and TH, respectively. No publication bias was observed for any of the analyses. For the following sections investigating the effect of ART on longitudinal changes in BMD, the analyses do not take duration of ART or efficacy of ART (cell count and viral load into account).
Longitudinal Changes in Bone Mineral Density in ART-Naive PLHIV Initiating ART Treatment
Figure 4, Supplemental Digital Content, http://links.lww.com/QAI/B391 displays the longitudinal changes of BMD within 1 year in ART-naive PLHIV initiating treatment. In treatment-naive subjects, BMD changed by −2.68% (95% CI: −3.11 to −2.25) and −2.66% (95% CI: −3.18 to −2.13) at the LS and TH within 1 year compared with baseline, respectively, and BMD changed by −1.77% (95% CI: −1.89 to −1.66) and −2.51% (95% CI: −3.28 to −1.73) at the LS and TH within 1–2 years compared with baseline, respectively, and by −2.03% (95% CI: −2.85 to −1.21) and −2.79% (95% CI: −3.82 to −1.77) after 2 years or more compared with baseline for the LS and TH, respectively. When further restricting to treatment-naive initiating tenofovir disoproxil fumarate (TDF) treatment, the BMD changed by −3.08% (95% CI: −3.44 to −2.72) and −3.14% (95% CI: −4.02 to −2.26) at the LS and TH within 1 year compared with baseline, respectively, and BMD changed by −2.72% (95% CI: −2.90 to −2.54) and −3.45% (95% CI: −3.75 to −3.15) at the LS and TH within 1 to 2 years compared with baseline, respectively, and by −3.25% (95% CI: −4.45 to −2.05) and −3.33% (95% CI: −3.65 to −3.02) after 2 years or more compared with baseline at the LS and TH, respectively. No publication bias was observed for any of the analyses.
Longitudinal Changes in Bone Mineral Density in PLHIV Continuing ART
Figure 5, Supplemental Digital Content, http://links.lww.com/QAI/B391 displays the longitudinal changes of BMD within 1 year in PLHIV continuing ART. In PLHIV continuing ART, BMD did not change significantly at either LS; 0.21% (95% CI: −0.39 to 0.81) or TH; −0.09% (95% CI: −0.41 to 0.59) within 1 year compared with baseline. However, after 1–2 years, BMD changed by −0.27% (95% CI: −0.49 to −0.05) and −0.56% (95% CI: −1.73 to −0.61) at the LS and TH compared with baseline, respectively, and by 1.93% (95% CI: −1.52 to 5.39) and −0.73% (95% CI: −1.43 to −0.02) within 1–2 years compared with baseline for the LS and TH, respectively. When further restricting PLHIV receiving TDF treatment, the BMD changed by −0.67% (95% CI: −1.13 to −0.20) and −0.35% (95% CI: −0.63 to −0.07) within 1 year compared with baseline for the LS and TH, respectively. It was not possible to conduct analyses of longer time spans due to too few studies. No publication bias was observed for any of the analyses.
Abacavir vs. TDF
A sensitivity analysis of randomized controlled trials (RCTs) comparing the effect of abacavir and TDF on BMD was performed, and the studies included displayed evidence grade I. The studies are displayed in Table 4, Supplemental Digital Content, http://links.lww.com/QAI/B391. The BMD was reduced significantly less with abacavir treatment compared with TDF after 48 and 96 weeks at the LS by 1.18% (95% CI: 0.80 to 1.57) and 1.37% (95% CI: 0.58 to 2.15), respectively, and at the TH by 1.75% (95% CI: 1.44 to 2.06) and 1.40% (95% CI: 0.75 to 2.05), respectively. Results after 48 weeks are displayed in Figure 6, Supplemental Digital Content, http://links.lww.com/QAI/B391. No publication bias was observed for any of the analyses.
Tenofovir Alafenamide vs. TDF
A sensitivity analysis of RCTs comparing the effect of tenofovir alafenamide (TAF) and TDF on BMD was performed, and the studies included displayed evidence grade I–II. The studies are displayed in Table 4, Supplemental Digital Content, http://links.lww.com/QAI/B391. The BMD was reduced significantly less with TAF treatment compared with TDF after 48 weeks and 96 weeks at the LS by 1.57% (95% CI: 1.45 to 1.69) and 1.90% (95% CI: 1.65 to 2.15), respectively, and at the TH by 2.00% (95% CI: 1.88 to 2.11) and 2.66% (95% CI: 2.52 to 2.79), respectively. Results after 48 weeks are displayed in Figure 7, Supplemental Digital Content, http://links.lww.com/QAI/B391. No publication bias was observed for any of the analyses.
Antiosteoporotic Therapies in PLHIV
A pooled analysis was only possible for changes in LS BMD by alendronate treatment in studies investigating antiosteoporosis treatment in PLHIV. The characteristics of these studies are displayed in Table 5, Supplemental Digital Content, http://links.lww.com/QAI/B391. In general, the evidence level was II–I; however, relatively small study populations were used (ranging from 21 to 87 subjects). Figure 8, Supplemental Digital Content, http://links.lww.com/QAI/B391 displays a pooled analysis of 3 RCTS for changes in LS with alendronate treatment where a 3.54% (95% CI: 1.83 to 5.26) increase in BMD was observed compared with placebo or no study drug. No publication bias was observed for the analysis. A study reported significantly improvement in BMD at the LS and TH after 2 doses of zoledronate (8.9% and 3.8% increase, respectively),10 and the effect of zoledronate persisted for an additional 5 years.11 Also, at the initiation of TDF-containing ART in treatment-naive PLHIV, coadministration of zoledronate improved BMD with 11% compared with placebo at the LS after 48 weeks.12 A single dose of zolendronate was as effective as 2 doses with a 2-year follow-up.13 A recent study demonstrated that adding zoledronate to ongoing treatment with TDF was superior at improving BMD at both LS and TH compared with switching from TDF to another ART (4.4% and 1.9% difference, respectively).14
Guidelines for the Management of Osteoporosis in PLHIV
Table 6, Supplemental Digital Content, http://links.lww.com/QAI/B391, presents 7 identified guidelines for the management of osteoporosis in PLHIV.
In general, the guidelines recommend using DXA15–19 or the Fracture Risk Assessment Tool (FRAX)20 for screening for osteoporosis in men above the age of 50 years and in postmenopausal women, whereas one suggests screening at older age.21 Two guidelines suggest DXA for osteoporosis in PLHIV aged 40–50 years with FRAX major osteoporosis fracture risk >10%.15,17 Four guidelines provides recommendation of treatment, and in general, the suggestions follow osteoporosis guidelines with treatment at T-score ≤−2.5, fragility fractures, and, for some regions, FRAX score ≥20% for major osteoporosis fractures or ≥3 for hip fractures. The European Aids Clinical Society (EACS) guideline recommends TAF as first line instead of TDF in case of severe osteopenia.18 Two guidelines suggest treatment with alendronate or zoledronate.15,17
The evidence reported in this systematic review and meta-analysis is summarized in Table 2. In PLHIV, the risks of any fracture and a fragility fracture are increased 1.5-fold, whereas the risk of a hip fracture is increased 4-fold compared with uninfected controls. The prevalence of VF is 22% and is similar to the reported prevalence of VF in uninfected women.22 The increased fracture risk in PLHIV is not caused by reductions in BMD as this only explains a 15% increase in fragility fractures. These findings underline a bone deficiency in PLHIV besides what is reflected in BMD. Also, FRAX that applies common fracture risk factors underestimates fracture risk in PLHIV with23,24 and without25 BMD evaluation. In the present meta-analysis, we observed a rapid decline in BMD at the LS and TH after initiation of ART. This BMD decrease was not more pronounced in TDF users. The decline seemed to stabilize after 1 year, but in PLHIV continuing TDF, the annual decrease after the first year was 0.67% and 0.35% at the LS and TH, respectively. However, we observed differences between treatments as abacavir and TAF are superior to TDF in terms of preserving BMD. Treatment for 48 weeks with abacavir or TAF resulted in a clinically relevant reduction in bone loss compared with TDF. The differences in BMD at the spine and hip between patients treated with TAF and TDF continued to increase for up to 96 weeks and stayed stable for up to 144 weeks. For abacavir, the differences seen after 48 weeks compared with TDF remained for up to 96 weeks.26 Other factors than HIV may explain the increased fracture risk in PLHIV. Infections with hepatitis B and C are common in PLHIV,27,28 and hepatitis B and C infections have independently of HIV been associated with a significantly increased fracture risk.29,30 In PLHIV, hepatitis B coinfection31 and hepatitis C coinfection32 were associated with an increased fracture risk. Awareness of osteoporosis in hepatitis B or C coinfected PLHIV is therefore warranted. PLHIV are at an increased risk of falls that may impact the fracture risk.33 This falls risk may be due to a reduced muscle mass in PLHIV on ART.34,35 Besides strategies for antiosteoporotic treatment in PLHIV, there is a need for fall prevention.
Antiosteoporosis treatment showed beneficial effects on BMD in PLHIV. Alendronate increased LS BMD by 3.5%, whereas the effect on hip BMD is uncertain. Zoledronate is less widely investigated in PLHIV but shows impressive increases in BMD in patients continuing ART10 with effects remaining after 7-year follow-up. Furthermore, one dose of zoledronate abolished the decrease in BMD in treatment-naive PLHIV initiating ART.12 Importantly, adding zoledronate in patients continuing TDF was superior to switch of treatment from TDF to another ART.14 Bone anabolic therapies have not been investigated in PLHIV. Therefore, antiosteoporosis treatment is a cornerstone in treatment of osteoporosis in PLHIV because the effects of alendronate on LS are of a larger magnitude than ART switch, and zoledronate is superior to ART switch.
Current guidelines for the management of osteoporosis in PLHIV follow general osteoporosis guidelines with the exception that the EACS guideline recommends TAF as first line instead of TDF in case of severe osteopenia or osteoporosis.18 We recommend increased focus on a timely diagnosis and treatment of osteoporosis in PLHIV because the reported increased fracture rates are seen already at a relatively young age and expected to increase when PLHIV reach older age (>60).36–38 Also, in PLHIV younger than age 40 years, the BMD decrease was comparable with older PLHIV. As BMD or FRAX do not reflect, the increased fracture risk in PLHIV and the observed fracture risk in PLHIV correspond to a 1 SD decrease in BMD corresponding to a 1 unit decrease in T-score7; initiation of osteoporosis prevention at higher T-scores than −2.5 should be considered. This is similar to the situation in patients treated with glucocorticoids. Glucocorticoids affect fracture risk more than what can be explained by reductions in BMD, and therefore, most guidelines recommend antiosteoporosis treatment at a T-score of −1 or −1.5.39,40
Based on the current evidence on fracture rates and DXA measurements of the LS or total hip, we recommend acknowledging HIV and ART as establish risk factors for osteoporosis and therefore including DXA and information about bone-healthy lifestyle in the management of PLHIV. In PLHIVs aged 40 years or more with T-score at the spine or hip below −1.5, we suggest switching from TDF-containing ART to other regimens applicable to the patients. In addition, if other risk factors for fracture are present, specific osteoporosis prophylaxis should be initiated with zoledronate or alendronate. Zoledronate is superior to switching ART in protecting against bone loss and is not subject to compliance issues that may be an issue with alendronate. In some cases, bone anabolic treatment should be considered, but this should follow national guidelines. Besides antiosteoporotic treatment, efforts should aim at environmental factors such as smoking cessation, decreased alcohol intake, optimal nutritional intake including calcium and vitamin D, increased physical activity, and screening for risk factors for falls such as orthostatic hypotension. Figure 3 displays our suggestion of a guideline to the screening and management of osteoporosis in PLHIV.
The strengths of this meta-analysis are the number of studies and thus the number of PLHIV evaluated. Overall, the studies were in agreement with the overall result and neither selection bias nor publication bias seem present. However, in studies assessing VF and BMD in PLHIV, a selection bias may be present, and the most severely affected patients or the most mildly affected patients may not have been included, which could influence the results. However, we expect this would result in a further decline in BMD and detection of more VFs. The prevalence of VF was equally high in the populations irrespective of mean ages (age 43–57 years). Assessment of VF may be influenced by the different techniques used to evaluate VF; however, most studies used lateral x-ray of the spine (8 of 11). Furthermore, the assessment of VF is dependent on the observer, which may influence results. Furthermore, the studies reporting prevalence on VF are from different geographic areas and have different ratios of men to women in their populations. Thus, the results may be influenced by the heterogeneity of the studies. Nevertheless, the results suggest an age-independent relatively high prevalence of VF in PLHIV. The studies assessing fracture risk, BMD compared with controls, and longitudinal changes in BMD are limited by the observational design; however, randomized trials on these effects are not possible. The estimated fracture risk is based on populations reporting BMD, whereas studies reporting fracture risk did not report BMD results. This may limit our results; however, as the analysis is based on LS BMD from 3628 PLHIV, the studies in a general showing a slightly lower BMD in PLHIV and fracture data from 64,633 PLHIV, we find it very likely that our findings are representative of the total PLHIV population. The present analyses were not able to adjust for confounding by factors as comorbidities, falls, pharmaceutical use, lean mass, and lifestyle that may influence results. However, the literature available reports a consistently increased fracture risk in PLHIV.
In conclusion, fracture risk is increased in PLHIV and is not sufficiently explained by BMD reduction. New fracture predictors are needed in PLHIV, and BMD and bone microarchitecture should be investigated further. As the fracture risk is increased, we recommend optimization of current guidelines with earlier initiation of osteoporosis prophylaxis and treatment.
1. Antiretroviral Therapy Cohort Collaboration. Survival of HIV-positive patients starting antiretroviral therapy between 1996 and 2013: a collaborative analysis of cohort studies. Lancet HIV. 2017;4:e349–e356.
2. Statens Serum Institute. EPI-NYT: Overvågning Og Forebyggelse Af Smitsomme Sygdomme. HIV 2016. 2017. Available at: https://mailchi.mp/ssi/epi-nyt-uge-36-2017-emne-hiv-2016?e=127eacf50f
. Accessed July 31, 2018.
3. Hjalte F, Calara PS, Blaxhult A, et al. Excess costs of non-infectious comorbidities among people living with HIV-estimates from Denmark and Sweden. AIDS Care. 2018;30:1–9.
4. Moher D, Altman DG, Liberati A, et al. PRISMA statement. Epidemiology. 2011;22:128; author reply 128.
5. Burns PB, Rohrich RJ, Chung KC. The levels of evidence and their role in evidence-based medicine. Plast Reconstr Surg. 2011;128:305–310.
6. Kelly TJ. Bone mineral density reference databases for American men and women. J.Bone Miner.Metab. 1990;5(suppl1):249.
7. Marshall D, Johnell O, Wedel H. Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures. BMJ. 1996;312:1254–1259.
8. Prieto-Alhambra D, Guerri-Fernandez R, De Vries F, et al. HIV infection and its association with an excess risk of clinical fractures: a nationwide case-control study. J Acquir Immune Defic Syndr. 2014;66:90–95.
9. Prior J, Burdge D, Maan E, et al. Fragility fractures and bone mineral density in HIV positive women: a case-control population-based study. Osteoporos Int. 2007;18:1345–1353.
10. Bolland MJ, Grey AB, Horne AM, et al. Annual zoledronate increases bone density in highly active antiretroviral therapy-treated human immunodeficiency virus-infected men: a randomized controlled trial. J Clin Endocrinol Metab. 2007;92:1283–1288.
11. Bolland MJ, Grey A, Horne AM, et al. Stable bone mineral density over 6 years in HIV-infected men treated with highly active antiretroviral therapy (HAART). Clin Endocrinol (Oxf). 2012;76:643–648.
12. Ofotokun I, Titanji K, Lahiri CD, et al. A single-dose zoledronic acid infusion prevents antiretroviral therapy-induced bone loss in treatment-naive HIV-infected patients: a phase IIb trial. Clin Infect Dis. 2016;63:663–671.
13. Negredo E, Bonjoch A, PÃ©rez-Ãlvarez N, et al. A randomized open label study for comparing two doses of zoledronic acid in HIV infected patients. Top Antivir Med 2014;22:402.
14. Hoy JF, Richardson R, Ebeling PR, et al. Zoledronic acid is superior to tenofovir DF-switching for low bone mineral density in adults with HIV: a randomised trial. AIDS. 2018;32:1967–1975.
15. Brown TT, Hoy J, Borderi M, et al. Recommendations for evaluation and management of bone disease in HIV. Clin Infect Dis. 2015;60:1242–1251.
16. McComsey GA, Tebas P, Shane E, et al. Bone disease in HIV infection: a practical review and recommendations for HIV care providers. Clin Infect Dis. 2010;51:937–946.
17. Negredo E, Bonjoch A, Clotet B. Management of bone mineral density in HIV-infected patients. Expert Opin Pharmacother. 2016;17:845–852.
18. EACS. GUIDELINES Version 9.0. 2017. Available at: http://www.eacsociety.org/files/guidelines_9.0-english.pdf
. Accessed June 26, 2018.
19. Aberg JA, Gallant JE, Ghanem KG, et al. Primary care guidelines for the management of persons infected with HIV: 2013 update by the HIV medicine association of the Infectious Diseases Society of America. Clin Infect Dis. 2014;58:e1–34.
20. BHIVA. British HIV Association Guidelines for the Routine Investigation and Monitoring of Adult HIV-1-Positive Individuals 2016. 2016. Available at: http://www.bhiva.org/documents/Guidelines/Monitoring/2016-BHIVA-Monitoring-Guidelines.pdf
. Accessed June 26, 2018.
21. Bolland MJ, Grey A, Reid IR. Skeletal health in adults with HIV infection. Lancet Diabetes Endocrinol. 2015;3:63–74.
22. Ballane G, Cauley JA, Luckey MM, et al. Worldwide prevalence and incidence of osteoporotic vertebral fractures. Osteoporos Int. 2017;28:1531–1542.
23. Peters BS, Perry M, Wierzbicki AS, et al. A cross-sectional randomised study of fracture risk in people with HIV infection in the probono 1 study. PLoS One. 2013;8:e78048.
24. Yang J, Sharma A, Shi Q, et al. Improved fracture prediction using different FRAX adjustments in HIV-infected women. AIDS. 2018;32:1699–1706.
25. Yin MT, Shiau S, Rimland D, et al. Fracture prediction with modified-FRAX in older HIV-infected and uninfected men. J Acquir Immune Defic Syndr. 2016;72:513–520.
26. Arribas JR, Thompson M, Sax PE, et al. Brief report: randomized, double-blind comparison of tenofovir alafenamide (TAF) vs tenofovir disoproxil fumarate (TDF), each coformulated with elvitegravir, cobicistat, and emtricitabine (E/C/F) for initial HIV-1 treatment: week 144 results. J Acquired Immune Defic Syndr. 2017;75:211–218.
27. Freitas SZ, Teles SA, Lorenzo PC, et al. HIV and HCV coinfection: prevalence, associated factors and genotype characterization in the Midwest region of Brazil. Rev Inst Med Trop Sao Paulo. 2014;56:517–524.
28. Singh KP, Crane M, Audsley J, et al. HIV-hepatitis B virus coinfection: epidemiology, pathogenesis, and treatment. AIDS. 2017;31:2035–2052.
29. Byrne DD, Newcomb CW, Carbonari DM, et al. Risk of hip fracture associated with untreated and treated chronic hepatitis B virus infection. J Hepatol. 2014;61:210–218.
30. Hansen AB, Omland LH, Krarup H, et al. Fracture risk in hepatitis C virus infected persons: results from the DANVIR cohort study. J Hepatol. 2014;61:15–21.
31. Byrne DD, Newcomb CW, Carbonari DM, et al. Increased risk of hip fracture associated with dually treated HIV/hepatitis B virus coinfection. J Viral Hepat. 2015;22:936–947.
32. Bedimo R, Maalouf NM, Lo Re V III. Hepatitis C virus coinfection as a risk factor for osteoporosis and fracture. Curr Opin HIV AIDS. 2016;11:285–293.
33. Erlandson KM, Allshouse AA, Jankowski CM, et al. Risk factors for falls in HIV-infected persons. J Acquir Immune Defic Syndr. 2012;61:484–489.
34. Debroy P, Lake JE, Sim M, et al. Lean mass declines consistently over 10 years in people living with HIV on antiretroviral therapy, with patterns differing by sex. Antivir Ther. 2019.
35. Erlandson KM, Zhang L, Ng DK, et al. Risk factors for falls, falls with injury, and falls with fracture among older men with or at risk of HIV infection. J Acquir Immune Defic Syndr. 2019;81:e117–e126.
36. Triant VA, Brown TT, Lee H, et al. Fracture prevalence among human immunodeficiency virus (HIV)-infected versus non-HIV-infected patients in a large U.S. healthcare system. J Clin Endocrinol Metab. 2008;93:3499–3504.
37. Gonciulea A, Wang R, Althoff KN, et al. An increased rate of fracture occurs a decade earlier in HIV+ compared with HIV-men. AIDS. 2017;31:1435–1443.
38. Guerri-Fernandez R, Molina-Morant D, Villar-Garcia J, et al. Bone density, microarchitecture, and tissue quality after long-term treatment with tenofovir/emtricitabine or abacavir/lamivudine. J Acquir Immune Defic Syndr. 2017;75:322–327.
39. Canalis E, Mazziotti G, Giustina A, et al. Glucocorticoid-induced osteoporosis: pathophysiology and therapy. Osteoporos Int. 2007;18:1319–1328.
40. Van Staa TP, Laan RF, Barton IP, et al. Bone density threshold and other predictors of vertebral fracture in patients receiving oral glucocorticoid therapy. Arthritis Rheum. 2003;48:3224–3229.