Osteoporosis is common in HIV-infected individuals with prevalence estimates approximately 3-fold higher than non-HIV-infected individuals.1 The etiology of this increased risk is likely multifactorial. Traditional osteoporosis risk factors such as hypogonadism, low body weight, smoking, and alcohol use are more prevalent among HIV-infected individuals2-5 and likely play a causative role.
Antiretroviral therapy (ART) and chronic infection with HIV have also been implicated in the pathogenesis of reduced bone mineral density (BMD) in HIV-infected patients. HIV viral proteins, including Vpr and gp120, and inflammatory cytokines, such as tumor necrosis factor-α (TNF-α) and interleukin 6 (IL-6), both stimulate osteoclast activity.6-8 In experimental models evaluating the effect of nucleoside/tide reverse transcriptase inhibitors (NRTI) on bone metabolism, zidovudine (ZDV) has been shown to induce osteoclastogenesis9,10 [and tenofovir (TDF) has been shown to impair bone mineralization].11 In vitro models have shown variable effects of individual protease inhibitors (PIs) on osteoclast and osteoblast function.8,12,13
To best understand the relative contributions of HIV infection and its treatment to decreased BMD observed in HIV-infected patients, longitudinal studies are required. Randomized prospective studies of HIV-infected subjects initiating ART may be particularly informative in determining whether ART is associated with reductions in BMD and whether some regimens have a greater impact than others. In Gilead 903, in which antiretroviral-naive HIV-infected persons were randomized to either TDF or lamivudine (3TC) compared with stavudine (d4T)/3TC along with efavirenz (EFV), BMD decreased overall by approximately 2% over the 144-week study interval, with more pronounced decreases in the TDF/3TC arm.14 Two recent randomized trials, comparing antiretroviral initiation with a PI-containing regimen and a non-PI-containing regimen, showed larger decreases in BMD in those randomized to PIs.15,16 In contrast, in a substudy of AIDS Clinical Trials Group 384 in which antiretroviral-naive persons were randomized to receive nelfinavir (NFV), EFV, or both, in combination with ZDV/3TC or didanosine/d4T, there was an average reduction in BMD of 2.25% over 96 weeks, without a difference between those receiving NFV or EFV.17 Changes in BMD in this study were most pronounced in those with the lowest CD4+ T-cell counts, suggesting an antiretroviral or immunologic effect, regardless of the type of therapy.
Study 613 is a 96-week prospective study in antiretroviral-naive subjects who were randomized to initiate either a PI lopinavir/ritonavir (LPV/r)-based regimen or a nonnucleoside reverse transcriptase inhibitor (NNRTI) EFV-based regimen in combination with ZDV/3TC. Those randomized to LPV/r and who met protocol-specified criteria were later simplified to LPV/r monotherapy. The goals of this analysis were (1) to evaluate changes in BMD prospectively in antiretroviral-naive persons initiating a PI versus an NNRTI-based regimen; (2) to determine whether removal of the ZDV/3TC backbone impacts BMD; and (3) to identify factors associated with BMD loss over 96 weeks, including systemic inflammation.
A total of 155 antiretroviral-naive HIV-infected subjects were recruited between January 2004 and July 2004 from 33 investigational sites in 5 countries (United States, Canada, United Kingdom, France, and Spain). Eligibility criteria included plasma HIV-1 RNA ≥1000 copies per milliliter, absence of documented resistance to study drugs, and no evidence of acute illness. Subjects who were treated for an AIDS-defining opportunistic infection within 30 days of initiating study drug or those who had a significant medical history including oncologic, endocrinologic, metabolic, or hepatic disease that would have, in the opinion of the investigator, adversely affected his/her participation in this study were excluded. The study was approved by the institutional review board or ethics committee at each center, and all enrolled subjects provided written informed consent. Virologic and body composition outcomes of this trial have been previously reported.18-20 The study was registered at http://clinicaltrials.gov, number NCT00075231.
Study Design and Intervention
Eligible subjects were randomized 1:2 to receive EFV 600 mg once a day or LPV/r 400/100 mg (soft gelatin capsules) twice a day. All subjects received coformulated ZDV/3TC (300/150 mg) twice a day. All drugs were administered open label. Between weeks 24 and 48, subjects in the LPV/r group who demonstrated 3 consecutive plasma HIV-1 RNA values <50 copies per milliliter discontinued ZDV/3TC. The primary outcome measure of the study was HIV-1 RNA of <50 copies per milliliter at week 96.
Total body BMD was assessed by dual-energy x-ray absorptiometry (DXA) at baseline, 24, 48, 72, and 96 weeks at each of the clinical sites. A single DXA scanner at each site was calibrated with a standardized phantom, and all scans were read centrally at Tufts University (Boston, MA). Two-hour 75-g oral glucose tolerance tests were performed at baseline, week 24, and week 96. Soluble tumor necrosis factor-α receptors (sTNFRs I and II) were measured using a commercially available enzyme-linked immunosorbent assay kit (R&D Systems, Minneapolis, MN) and analyzed at a central laboratory (ICON Central Laboratories, Farmingdale, NY) from serum samples collected at baseline, 24, 48, and 96 weeks. For both assays, the sensitivity of the assay is less than 1 pg/mL and the coefficients of variation are 2.6%-8.8%. The reference ranges for sTNFR I and sTNFR II are 749-1966 and 1003-3170 pg/mL, respectively. These markers had been measured because of their association with lipoatrophy,21 a secondary end point of the original trial.18
Subjects who remained on study through 96 weeks with DXA scans available were included in the analysis. Sensitivity analyses included all subjects with at least 1 post-baseline DXA scan. Associations between baseline total BMD and other baseline factors were assessed by linear regression (for quantitative baseline factors) or 1-way analysis of variance (for categorical baseline factors). Changes from baseline to each study visit in total BMD and sTNFRs I and II were assessed by 1-way analysis of variance with treatment group as the factor. A mixed-effects regression model was also used to estimate the rate of change in BMD. Akaike information criterion was used to select the covariance structure for the mixed-effects model. We assessed the relationship between sTNFRs I and II and subsequent changes in total BMD in 2 ways. First, we used linear regression to determine whether baseline concentrations of sTNFR I or sTNFR II were associated with the 0- to 96-week change in total BMD. Second, because sTNFR I or sTNFR II decreased quickly with ART initiation, we used linear regression to determine whether week 24 sTNFR I or sTNFR II were associated with the 24- to 96-week changes in total BMD. Last, associations with a 5% decrease in total BMD through 96 weeks were assessed by logistic regression, using a stepwise selection procedure with a P value of 0.05 to enter and remain in the model. A 5% loss in total BMD over 96 weeks was considered potentially clinically significant because it is the approximate total BMD bone loss over 2 years during the menopause transition in the general population.22 Factors tested included demographics, weight, standard chemistry lab parameters (eg, alkaline phosphatase), HIV-1 RNA, CD4+ T-cell count, smoking/alcohol history (user/ex-user vs nonuser and light drinker/ex-drinker (<2 drinks/d) or nondrinker vs moderate or heavy drinker/ex-drinker (≥2 drinks/d), respectively), body composition variables (eg, trunk fat and limb fat), homeostasis model assessment of insulin resistance,23 sTNFRs I and II, and glucose and insulin area under the curve (using the trapezoidal rule) during a 2-hour oral glucose tolerance test.
Of the 104 subjects randomized to LPV/r + ZDV/3TC, 92 achieved 3 consecutive HIV-1 RNA levels <50 copies per milliliter, then simplified to LPV/r monotherapy, and were followed for a median of 68 weeks on monotherapy. The majority of subjects (55%) simplified to LPV/r monotherapy at week 24, and 89% had simplified by week 32. Seventy-four LPV/r-treated and 32 EFV-treated subjects had DXA scans available through 96 weeks.
Demographics and baseline characteristics were comparable between treatment groups, with the exception of lower mean age in the EFV group (Table 1). Of the 106 subjects included in the analysis, there were 23 women (16 LPV/r, 7 EFV), of whom 9 were older than 40 years at entry and 4 were older than 50 years. Mean total BMD at baseline for all 106 subjects was 1.17 ± 0.09 g/cm2 (LPV/r) and 1.19 ± 0.12 g/cm2 (EFV) [P = not significant (NS)].
Associations With Baseline BMD
In a multivariable analysis, lower baseline total BMD was statistically significantly associated with lower weight, non-black race, and lower baseline HIV-1 RNA levels (Table 2). Although higher baseline HIV-1 RNA level was associated with black race, the positive association between baseline HIV-1 RNA and baseline BMD remained statistically significant after adjustment for race. Significant univariable relationships between sTNFRs I and II and total BMD were no longer present after including HIV-1 RNA in the multivariable model.
Changes From Baseline in BMD
After 24 weeks (ie, before discontinuation of ZDV/3TC in the LPV/r group), total BMD decreased by 0.7% [95% confidence interval (CI): −1.2% to −0.2%] in the LPV/r group and 0.6% (95% CI: −1.3% to 0.1%) in the EFV group from baseline (P = 0.79 for the difference between groups). After 96 weeks, similar mean decreases from baseline in BMD were also observed by both treatment groups: −2.5% (95% CI: −3.4% to −1.4%) in the LPV/r group and −2.3% (95% CI: −3.8% to −0.8%) in the EFV group (P = 0.86 for the difference between groups). Within treatment group, changes from baseline to 96 weeks in BMD were statistically significant (P < 0.01 for each group). Mean decreases to each visit and the fitted model from a mixed-effects regression analysis are shown in Figure 1. In a sensitivity analysis that included all subjects with any post-baseline BMD data, similar results were observed, with a significant time trend and no significant treatment group effect (data not shown). No alteration in the rate of BMD change was observed when subjects in the LPV/r group simplified to LPV/r monotherapy (Fig. 1). Changes in BMD over 96 weeks were similar in men (−2.4%) and women (−2.2%) (P = 0.85).
Changes From Baseline in Inflammatory Markers
At 24 weeks, sTNFR I and sTNFR II decreased significantly in both the LPV/r and EFV groups, without significant between-group differences (Fig. 2). At 24 weeks, 1% (0% LPV/r, 3% EFV; P = NS) were above the reference range for sTNFR I and 10% (10% LPV/r, 10% EFV; P = NS) were above the reference range for sTNFR II compared with 6% and 60%, respectively, at baseline.
At 96 weeks, the initial decreases observed in sTNFR I had waned, whereas for sTNFR II, the initial decreases were maintained over time (Fig. 2), with an average decrease below baseline sTNFR II concentrations of 1225 pg/mL (95% CI: −1502 to −948 pg/mL) in the LPV/r group and 1105 pg/mL (95% CI: −1526 to −683 pg/mL) in the EFV group (P = 0.64 for the difference between groups).
The Relationship Between Inflammatory Markers and Subsequent Change in BMD
Baseline sTNFR I was not associated with 0- to 96-week changes in total BMD (P = 0.31), whereas baseline sTNFR II was marginally associated with subsequent changes in total BMD (r = −0.18, P = 0.06). However, after adjustment for baseline CD4+ T-cell count, this marginally significant relationship was no longer present (P = 0.69). Similarly, although week 24 sTNFR I concentrations were not correlated with 24- to 96-week total BMD change, higher week 24 sTNFR II was associated with larger decreases in 24- to 96-week total BMD (r = −0.22, P = 0.028). However, after adjustment for week 24 CD4+ T-cell count, the relationship between sTNFR II concentration and subsequent BMD change was no longer statistically significant (P = 0.47). Changes from baseline to 24 weeks in CD4+ T-cell count, sTNFR I, and sTNFR II were not associated with subsequent changes in total BMD (24-96 weeks) (data not shown).
Baseline Factors Associated With >5% Decrease in BMD Over 96 Weeks
Sixteen subjects in the 2 arms (15%) demonstrated a >5% decrease in total BMD over 96 weeks. Subjects with lower baseline CD4+ T-cell count, with higher baseline fasting glucose and of non-black race, demonstrated a higher risk for >5% decrease in BMD (Table 2). Changes in BMD were not correlated to baseline BMD or other parameters of glucose metabolism.
In this randomized controlled trial, we found that ART-naive HIV-infected subjects beginning ART had significant reductions in total BMD over 96 weeks, irrespective of whether they received ZDV/3TC/EFV or an LPV/r simplification strategy. Because DXA evaluations were performed at 24-week intervals, we were also able to observe that the removal of ZDV/3TC had no appreciable impact on the bone loss trajectory in those randomized to the simplification strategy. Taken together, these findings suggest that significant reductions in total BMD are observed with ART initiation, independent of the ART regimen or strategy.
Although reduced BMD is common in HIV-infected patients,1 the contributing factors are not clear. In contrast to longitudinal studies of treated HIV-infected patients, which have generally showed stability of BMD over time,24-26 studies investigating BMD changes with ART initiation have uniformly demonstrated significant reductions over the intervals studied.14-17 Our finding of a 2.5% loss of total BMD over 96 weeks is similar in magnitude to the previous studies. In a large randomized trial of ART-naive HIV-infected patients initiating d4T/3TC/EFV or TDF/3TC/EFV, Gallant et al14 reported a 1.0%-2.2% BMD loss at the spine and a 2.4%-2.8% BMD loss at the hip at 144 weeks, with most of the bone loss occurring in the first 48 weeks. More recently, Tebas et al17 reported a 2.5% loss in total bone mineral content over 96 weeks in patients initiating NFV, EFV, or both combined along with ZDV/3TC or d4T/3TC. In 2 studies, which compared PI-containing and NNRTI-containing regimens, Duvivier et al16 reported an average 4.1% bone loss at the spine and a 2.8% decrease in the hip over 48 weeks in the combined arms, whereas Bonnet et al15 reported ∼0.8% decrease in lumbar spine BMD over 9 months. It should be noted that some of the differences in the point estimates may be due to the site examined for BMD change (eg, spine vs hip, vs total BMD) and other study characteristics, including study duration, population, and intervention.
The mechanisms underlying the BMD loss with ART initiation are not clear. Members of the PI class have been implicated in the pathogenesis of bone loss,12,27 and some cross-sectional studies suggest a higher prevalence of osteopenia and osteoporosis in HIV-infected patients receiving PIs.1,28 However, longitudinal studies of treatment-experienced HIV-infected patients treated with various PIs24-26 have shown stability of BMD over time.
In studies of HIV-infected persons initiating ART, results have been mixed regarding the differential impact of PIs on BMD. Duvivier et al16 found that lumbar spine BMD decreased by 4.9% over 48 weeks in patients initiating a regimen containing either LPV/r or indinavir/ritonavir (n = 55) compared with a 1.5% loss in those receiving a PI-sparing NNRTI-based regimen (EFV or nevirapine) (n = 15). Interestingly, the BMD loss was similar at the hip between the groups (2.8% vs 2.7%, respectively), with a trend toward a greatest loss in those receiving a combination of NRTIs and PIs (P = 0.1 vs NRTI/NNRTI). Similarly, Bonnet et al15 recently reported a larger lumbar spine BMD decline in subjects initiating PIs after 9 months (either fosamprenavir/r or atazanavir/r) compared with those initiating an NNRTI-containing regimen. There were no differences in the changes in total BMD between the 2 arms. Using total bone mineral content as an outcome measure, Tebas et al17 found a nonsignificant trend toward increased bone loss in NFV-treated patients compared with those treated with EFV (P = 0.08).
In the present study, we found that total body BMD declined similarly over the first 24 weeks in those randomized to ZDV/3TC/LPV/r vs ZDV/3TC/EFV, and after 96 weeks, BMD had declined similarly in those who de-intensified to LPV/r monotherapy compared with those who remained on EFV-based therapy, arguing against a specific effect of LPV/r. These data stand in contrast to a recent, small, nonrandomized, prospective study of treatment-naive HIV-infected patients, ART initiation with LPV/r was independently associated with BMD change at the lumbar spine after 1 year of treatment.29 In our study, BMD was assessed using total body DXA, and it is possible that the change in BMD may have been different between the arms if site-specific DXA had been used because some,15,16 but not all,14 previous studies have shown the greatest decline with ART initiation at the lumbar spine.
One novel aspect of our study design is the ability to determine whether the removal of ZDV/3TC from the regimen had any appreciable effect on the trajectory of the change in BMD. In vitro studies have shown that ZDV exposure increases osteoclast activity and therefore may be associated with increased bone resorption.9 Consistent with this observation, in a randomized switch study, subjects who continued receiving d4T or ZDV therapy had a significant decrement in total body BMD over the 104-week observation period, whereas BMD in those who switched to abacavir remained stable.30 However, in our study, after ART simplification to LPV/r monotherapy at 24 weeks, we observed no evidence of an alteration in the trajectory of BMD changes in the LPV/r group, suggesting that ZDV/3TC does not have an additive effect on total BMD loss. However, a randomized trial comparing simplification to LPV/r monotherapy with continuation of ZDV/3TC/LPV/r would be required to definitively assess the effect of ZDV/3TC discontinuation.
Our finding of total body BMD loss, irrespective of the regimen used, suggests that changes in immunologic or virologic factors with ART initiation may play a role. Chronic HIV infection is associated with systemic inflammation, which may have adverse effects on BMD. Cytokines such as TNF-α and IL-6 are potent stimulators of osteoclast activity,6 and serum concentrations have been associated with BMD loss in elderly subjects in the general population.31 Among HIV-infected patients, TNF-α activity has been associated with markers of bone resorption.32
A unique feature of our study was that in addition to BMD, soluble receptors of TNF-α were measured at baseline and at 24-week intervals thereafter, to understand whether changes in BMD were associated with systemic inflammation. Both sTNFR I and sTNFR II rapidly decreased with ART initiation by 24 weeks in both arms, but only sTNFR II remained suppressed at 96 weeks. Interestingly, although we found that baseline and 24-week sTNFR II concentrations were associated with subsequent changes in BMD, after adjustment for concomitant CD4+ T-cell count, the associations were no longer observed. In other populations, the effect of TNF-α on bone loss is in part mediated by RANKL (receptor activator of nuclear factor κB ligand)/RANK signaling, which can be blocked by osteoprotegerin.33 We did not measure concentrations of osteoprotegerin/RANKL or other cytokines, such as IL-6, which have been associated with reduced BMD.31 In addition, systemic markers of TNF-α activity were measured in our study. It is possible that local effects of TNF-α may be operative, which are not measurable in the systemic circulation.
Regardless of the relative contributions of antiretrovirals or change in inflammatory or virologic factors, the 2.5% loss in total BMD over 96 weeks may be clinically significant. In longitudinal studies in healthy young adults, BMD decreases yearly from 0% to 1%, depending on the site studied with more profound changes in the total hip compared with the hip or total body. For example, the 2-year change in BMD in men between 20 and 49 years of age in a large longitudinal study was −0.8% in the hip, −0.3% in the lumbar spine, and +0.4% for the total body,34 with similar findings in premenopausal women.22 During the early menopausal transition, the loss of BMD accelerates with annual rates of bone loss between 1.2% and 1.6%.22 The magnitude of total body bone loss that we observed with ART initiation, therefore, is larger than what would be expected by aging alone, even in older adults,35 and approaches annual losses seen in late menopause.22 Further studies are needed to determine if this degree of bone loss is associated with an increased risk of fracture and the extent to which these decreases stabilize over time.
We identified several factors that may be useful in identifying patients who are at increased risk of having a potentially clinically significant decrease in BMD with ART initiation. As in the general population,36 white and Hispanic patients were found to be at an increased risk of having a >5% decline in BMD over the study interval. In addition, those patients with the lowest CD4+ T-cell count were at increased risk of bone loss, consistent with cross-sectional studies showing a higher prevalence of osteopenia and osteoporosis in patients with the lowest nadir CD4+ T-cell counts.37,38 Although subjects with low body weight had the lowest BMD at baseline, low body weight was not associated with an increased risk of bone loss over the study interval, as might have been predicted.
We also found that higher fasting glucose at baseline was independently associated with total BMD loss over the study interval. Although some studies have shown associations between abnormal glucose metabolism and low bone density in both the general population and HIV-infected patients,39,40 these findings have not been consistent across studies and the biologic plausibility has not been clearly established. In addition to fasting glucose values, we also measured fasting insulin and glucose and insulin 2 hours after a 75-g oral glucose challenge. Because similar findings were not found using other measures of glucose metabolism, it is likely that this represents a chance finding.
Our analysis had additional limitations that should be highlighted. Because this study was not designed to evaluate BMD changes prospectively in subjects treated with ART, we did not perform DXA scans at the spine and hip. Future studies investigating changes in BMD with ART initiation should include these site-specific evaluations. In addition, potentially clinically useful T scores and Z scores can be generated with these studies. Second, although the duration of the study interval was typical of ART-naive treatment studies, this duration of therapy provides only an initial evaluation of BMD changes. Some studies have suggested that BMD may decrease and plateau after initiation of ART,14 our study was not able to assess long-term effects of ART initiation on BMD. Next, there were other important factors that were not assessed in the study, such as hypovitaminosis D that are common in the HIV populations41 and may contribute to BMD changes. Future studies should examine the relationship between vitamin D and BMD loss, as this is a potentially modifiable risk factor. Finally, similar to many other ART-naive trials, the proportion of women was low. Although the BMD trajectory was similar between men and women using the limited data available, changes in BMD with ART initiation need to be addressed in women because the risk of osteoporotic fracture is higher in women compared with men in both HIV-infected and HIV-uninfected populations.42
In summary, similar decreases in total BMD over 96 weeks occurred in ART-naive subjects receiving either EFV-based regimen or LPV/r-based regimen (∼2.4%), which were not altered by simplification to LPV/r monotherapy and were not independently related to systemic measures of TNF-α activity. These data suggest that the loss of BMD with ART initiation occurs independently of the ART regimen used. Non-black subjects and those with lower nadir CD4+ T-cell count may be at an increased risk of more pronounced BMD loss. Further investigation is required to determine whether these changes in BMD are associated with an increased risk of fracture.
The authors would like to thank the patients who agreed to participate in this clinical trial and the investigators for their management of the study protocol and patient care. We would like to thank the site's study coordinators for facilitating the study procedures and data collection. The authors wish to acknowledge the following Abbott Study Team members for the operational management of the study: Kevin Niemi, Karen Wikstrom, Karmin Robinson, Lisa McCartney, Mary Woulfe, Marion Dehaan, Danielle Grant, Florence McMillan, Debbie Tokimoto, and Jaime Baldner. The authors also acknowledge Hamani Henderson of Abbott Laboratories for assistance with the development of this article. We would also like to thank all the Abbott and Covance employees who helped monitor the sites. Investigators: Jose R. Arribas: Hospital La Paz; Nicholaos C. Bellos: Southwest Infectious Disease Associates; D. William Cameron: University of Ottawa at The Ottawa Hospital; Rafael Campo: University of Miami School of Medicine; Paul J. Cimoch: Orange County Center for Special Immunology; Bonaventura Clotet: Fundacio IrsiCaixa, Hospital Germans Trias I Pujol; Calvin Cohen: Community Research Initiative of New England; Laurent Cotte: Hôpital Hôtel-Dieu; Pere Domingo: Hospital de la Santa Creu i Sant Pau; Fernando Dronda: Hospital Ramón y Cajal; Philippa Easterbrook: The Caldecot Centre Kings College Hospital; Charles Farthing: AIDS Healthcare Foundation Research Center; Joseph C. Gathe Jr: Therapeutic Concepts; Norbert Gilmore: McGill University Health Center; Charles B. Hicks: Duke University Medical Center; Margaret Johnson: Royal Free Hospital; Veronique Joly: Hôpital Bichat; Harold P. Katner: Mercer University School of Medicine; Harold Kessler: Rush University Medical Center, Chicago; Clifford Leen: Western General Hospital; George F. McKinley: St. Luke's-Roosevelt Hospital; Robert Myers: Body Positive, Inc; David M. Parenti: The George Washington University Medical Center; Gerald Pierone Jr: Treasure Coast Infectious Disease Consultants; Frederico Pulido: Hospital 12 de Octubre; Anita Rachlis: Sunnybrook Health Sciences Centre; Robert R. Redfield: University of Maryland School of Medicine, Institute of Human Virology; Rafael Rubio: Hospital 12 de Octubre; Patricia D. Salvato: Diversified Medical Practice; Richard Styker: Abbott (formerly Tower ID Medical Associates); Sharon Walmsley: University of Toronto; Ed Wilkins: North Manchester General Hospital; Peter Wolfe: Lightsource Medical Group; and Vivian Yeh: Andrew Escajeda Clinic, City or Pasadena Public Health.
1. Brown TT, Qaqish RB. Antiretroviral therapy and the prevalence of osteopenia and osteoporosis: a meta-analytic review. AIDS
2. Klein RS, Lo Y, Santoro N, et al. Androgen levels in older men who have or who are at risk of acquiring HIV infection. Clin Infect Dis
3. Saves M, Chene G, Ducimetiere P, et al. Risk factors for coronary heart disease in patients treated for human immunodeficiency virus infection compared with the general population. Clin Infect Dis
4. Bolland MJ, Grey AB, Gamble GD, et al. Review #: low body weight mediates the relationship between HIV infection and low bone mineral density: a meta-analysis. J Clin Endocrinol Metab
5. Zenilman JM, Hook EW III, Shepherd M, et al. Alcohol and other substance use in STD clinic patients: relationships with STDs and prevalent HIV infection. Sex Transm Dis
6. Kwan TS, Padrines M, Theoleyre S, et al. RANKL, TNF-alpha/IL-1: interrelations in bone resorption pathophysiology. Cytokine Growth Factor Rev
7. Fakruddin JM, Laurence J. HIV-1 Vpr enhances production of receptor of activated NF-kappaB ligand (RANKL) via potentiation of glucocorticoid receptor activity. Arch Virol
8. Fakruddin JM, Laurence J. HIV envelope gp120-mediated regulation of osteoclastogenesis via receptor activator of nuclear factor kappa B ligand (RANKL) secretion and its modulation by certain HIV protease inhibitors through interferon-gamma/RANKL cross-talk. J Biol Chem
9. Pan G, Kilby M, McDonald JM. Modulation of osteoclastogenesis induced by nucleoside reverse transcriptase inhibitors. AIDS Res Hum Retroviruses
10. Pan G, Wu X, McKenna MA, et al. AZT enhances osteoclastogenesis and bone loss. AIDS Res Hum Retroviruses
11. Castillo AB, Tarantal AF, Watnik MR, et al. Tenofovir treatment at 30 mg/kg/day can inhibit cortical bone mineralization in growing rhesus monkeys (Macaca mulatta
). J Orthop Res
12. Jain RG, Lenhard JM. Select HIV protease inhibitors alter bone and fat metabolism ex vivo. J Biol Chem
13. Wang MW, Wei S, Faccio R, et al. The HIV protease inhibitor ritonavir blocks osteoclastogenesis and function by impairing RANKL-induced signaling. J Clin Invest
14. Gallant JE, Staszewski S, Pozniak AL, et al. Efficacy and safety of tenofovir DF vs stavudine in combination therapy in antiretroviral-naive patients: a 3-year randomized trial. JAMA
15. Bonnet E, Mabile L, Ruidavets J-B, et al. Important changes in bone metabolism soon after commencing HAART. Presented at: 9th Workshop on Adverse Drug Reactions and Lipodystrophy in HIV; July 19-21, 2007; Sydney, Australia.
16. Duvivier C, Kolta S, Assoumou L, et al. First line PI-containing regimens cause a greater decrease in bone mineral density compared to NNRTI-containing regimens in HIV-1 infected patients: a substudy of HIPPOCAMPE-ANRS 121 trial . Presented at: 15th Conference on Retroviruses and Opportunistic Infections; February 3-8, 2008; Boston, MA.
17. Tebas P, Umbleja T, Dube MP, et al. Initiation of ART is associated with bone loss independent of the specific ART regimen: results of ACTG A5005s. Presented at: 14th Conference on Retroviruses and Opportunistic Infections; February 25-28, 2007; Los Angeles, CA.
18. Cameron DW, da Silva B, Arribas JR, et al. Significant sparing of peripheral lipoatrophy by HIV treatment with LPV/r + ZDV/3TC induction followed by LPV/r monotherapy compared with EFV + ZDV/3TC [44LB]. Presented at: 14th Conference of Retroviruses and Opportunistic Infections; February 25-28, 2007; Los Angeles, CA.
19. Cameron DW, da Silva B, Arribas JR, et al. A two-year randomised controlled clinical trial in antiretroviral-naive patients using lopinavir/ritonavir monotherapy after initial induction treatment compared to an efavirenz 3-drug regimen: 96 week results (Study M03-613) [THLB0201]. Presented at: XVIth International AIDS Conference; August 13-18, 2006; Toronto, Ontario.
20. Campo RE, da Silva B, Gathe J, et al. Predictors of loss of virologic response in subjects who deintensified to lopinavir/ritonavir monotherapy after acheiving plasma HIV-1 RNA < 50 copies/mL on lopinavir/ritonavir plus zidovudine/lamivudine. Presented at: 14th Conference of Retroviruses and Opportunistic Infections; February 25-28, 2007; Los Angeles, CA.
21. Mynarcik DC, McNurlan MA, Steigbigel RT, et al. Association of severe insulin resistance with both loss of limb fat and elevated serum tumor necrosis factor receptor levels in HIV lipodystrophy. J Acquir Immune Defic Syndr
22. Recker R, Lappe J, Davies K, et al. Characterization of perimenopausal bone loss: a prospective study. J Bone Miner Res
23. Matthews DR, Hosker JP, Rudenski AS, et al. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia
24. Mondy K, Yarasheski K, Powderly WG, et al. Longitudinal evolution of bone mineral density and bone markers in human immunodeficiency virus-infected individuals. Clin Infect Dis
25. Nolan D, Upton R, McKinnon E, et al. Stable or increasing bone mineral density in HIV-infected patients treated with nelfinavir or indinavir. AIDS
26. Dolan SE, Kanter JR, Grinspoon S. Longitudinal analysis of bone density in human immunodeficiency virus-infected women. J Clin Endocrinol Metab
27. Lenhard JM, Furfine ES, Jain RG, et al. HIV protease inhibitors block adipogenesis and increase lipolysis in vitro. Antiviral Res
28. Tebas P, Powderly WG, Claxton S, et al. Accelerated bone mineral loss in HIV-infected patients receiving potent antiretroviral therapy. AIDS
29. Rivas P, Gorgolas M, Garcia-Delgado R, et al. Evolution of bone mineral density in AIDS patients on treatment with zidovudine/lamivudine plus abacavir or lopinavir/ritonavir. HIV Med
30. Martin A, Smith DE, Carr A, et al. Reversibility of lipoatrophy in HIV-infected patients 2 years after switching from a thymidine analogue to abacavir: the MITOX Extension Study. AIDS
31. Ding C, Parameswaran V, Udayan R, et al. Circulating levels of inflammatory markers predict change in bone mineral density and resorption in older adults: a longitudinal study. J Clin Endocrinol Metab
32. Aukrust P, Haug CJ, Ueland T, et al. Decreased bone formative and enhanced resorptive markers in human immunodeficiency virus infection: indication of normalization of the bone-remodeling process during highly active antiretroviral therapy. J Clin Endocrinol Metab
33. Kearns AE, Khosla S, Kostenuik PJ. Receptor activator of nuclear factor kappaB ligand and osteoprotegerin regulation of bone remodeling in health and disease. Endocr Rev
34. Warming L, Hassager C, Christiansen C. Changes in bone mineral density with age in men and women: a longitudinal study. Osteoporos Int
35. Knoke JD, Barrett-Connor E. Weight loss: a determinant of hip bone loss in older men and women. The Rancho Bernardo Study. Am J Epidemiol
36. Pothiwala P, Evans EM, Chapman-Novakofski KM. Ethnic variation in risk for osteoporosis among women: a review of biological and behavioral factors. J Womens Health (Larchmt)
37. Cazanave C, Dupon M, Lavignolle-Aurillac V, et al. Reduced bone mineral density in HIV-infected patients: prevalence and associated factors. AIDS
38. Madeddu G, Spanu A, Solinas P, et al. Bone mass loss and vitamin D metabolism impairment in HIV patients receiving highly active antiretroviral therapy. Q J Nucl Med Mol Imaging
39. Schwartz AV, Sellmeyer DE, Strotmeyer ES, et al. Diabetes and bone loss at the hip in older black and white adults. J Bone Miner Res
40. Brown TT, Ruppe MD, Kassner R, et al. Reduced bone mineral density in human immunodeficiency virus-infected patients and its association with increased central adiposity and postload hyperglycemia. J Clin Endocrinol Metab
41. Guaraldi G, Orlando G, Squillace N, et al. Prevalence of secondary causes of osteoporosis among HIV-infected individuals. Presented at: 8th International Workshop on Adverse Drug Reactions and Lipodystrophy in HIV; September 24-26, 2006; San Francisco, CA.
42. 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