As combined antiretroviral therapy (cART) must currently be continued for life, treatment complications have become a major concern. Osteoporosis is a growing problem in the aging HIV-infected population, being associated with an increased risk of fractures [1,2]. In a meta-analytic review by Brown and Qaqish , the pooled odds ratios for reduced bone mineral density (BMD) and osteoporosis in HIV-infected patients compared with uninfected controls were 6.4 and 3.7, respectively. Cross-sectional studies [2,4–8] of HIV-infected men on cART have reported prevalences of osteopenia ranging from 22 to 65% and of osteoporosis from 3 to 33%. First-line antiretroviral therapy has been linked to a 2–6% decline in BMD during the first 12 months [9–13], particularly with tenofovir (TDF) [10,14]. Longitudinal studies have shown that BMD remains relatively stable in treatment-experienced patients [15–17]. Here we studied the course of BMD in HIV-infected men with osteopenia, and sought risk factors for abnormal BMD decline.
Patients and methods
Eligible patients were HIV-1-infected adult men with −2.5 ≤T-score less than −1 at the hip or lumbar spine; no secondary cause of osteoporosis; a known duration of HIV-1 infection of at least 5 years or a CD4+ cell nadir less than 200/μl; and having been screened for the ANRS 120 Fosivir trial . Exclusion criteria included current or prior use of any antiosteoporotic agent . The Institutional Review Board of the Pitié-Salpêtrière Hospital approved the study protocol, and all patients provided their written informed consent.
The primary endpoint was the change in BMD at the lumbar spine and hip 24–36 months after the diagnosis of osteopenia.
Secondary objective was to identify factors associated with a decline in BMD larger than the smallest detectable difference (SDD). The SDD was defined as a decrease in BMD of −0.034 g/cm2 for the lumbar spine site and −0.027 g/cm2 for the total hip site . This is the smallest change that exceeds the variability inherent in BMD measurements .
BMD (total hip and lumbar spine) was evaluated at baseline and between months 24 and 36 by means of dual-energy-X ray-absorptiometry (DXA) with central reading as previously described [10,18,19]. As GE-Lunar (Madison, Wisconsin, USA) and Hologic (Bedford, Massachusetts, USA) DXA devices were used (the same device was always used for a given patient), the standardized BMD (sBMD, mg/cm2) was calculated . For the GE-Lunar device the sBMD was 952.2 × BMD at the lumbar spine and 979 × BMD – 31 at the hip. For the Hologic device the sBMD was 1075.5 × BMD at the lumbar spine and 1008 × BMD + 6 at the hip .
The changes between baseline and follow-up were compared using the Wilcoxon paired test for continuous variables, and the Mann–Whitney test to study the association of one continuous variable and one categorical variable. Univariate and multivariable logistic models were used to identify factors associated with a decline in BMD greater than the SDD at each site (total hip and lumbar spine).
As some parameters with missing data are known to influence BMD, and because it is better to impute missing data than to ignore them , we created 20 datasets in which missing data were imputed from each patient's other covariables, including demographic, biological and therapeutic data; risk factors for osteoporosis; and the decrease in BMD greater than the SDD or not. In these imputations, missing values were randomly sampled from their predicted distributions [25,26]. Analyses were run on each of the 20 data sets, including the imputed values, and the results were combined with Rubin's rules . We assessed whether continuous variables were better modeled as continuous variables or as tertiles or for exposure to TDF either as no exposure, current exposure and past exposure or as none, 0–2.5 years, and more than 2.5 years, based on the lowest value of Akaike's information criterion for the corresponding univariate logistic regression model, and grouped together the closest values in order to obtain two classes for certain variables. Variables with univariate P values below 0.15 were then entered in multivariable logistic models.
The SPSS software package version 18.0 for Windows (SPSS Inc., Chicago, Illinois, USA) was used for all analyses.
Among a total of 128 men screened, 97 (76%) accepted to participate. Three men were subsequently excluded because they had received antiosteoporotic agents leaving 94 men for evaluation. Of them, 46% were active smokers or had stopped smoking for less than 3 years. Median age was 46 years [interquartile range (IQR), 41–53], median BMI was 22.4 kg/m2 (20.8–25.0), median CD4+ cell count was 495 cells/μl (345–676), median CD4+ cell nadir was 164 cells/μl (69–261) and 71% had a plasma viral load below 500 copies/ml at baseline (at the time of the first DXA measurement). At baseline, the median T-score was −1.40 (−1.76 to −1.04) at the lumbar spine and −1.08 (−1.51 to −0.60) at hip. Median z-score were, respectively, −1.35 (−1.80 to −1.00) and −0.85 (−1.20 to −0.30). The median time between the two DXA measurements was 2.6 years (2.3–2.9). At the second DXA measurement, the median BMI was 22.5 kg/m2 (20.8–25.1) and the median change in BMI was 0.2 kg/m2 (IQR: −0.4–0.7, P = 0.086). Median CD4+ cell count was 587 cells/μl (446–771) and plasma HIV-RNA value was below 50 copies/ml in 87% of patients. All patients were receiving cART for a median of 7.5 years (5.8–9.5). Forty-four patients (47%) were receiving TDF, 31 (33%) had never taken TDF and 19 (20%) had stopped taking TDF for a median of 2.5 years. The median durations of protease inhibitor exposure and TDF exposure were, respectively, 6.7 years (2.7–9.9) and 1.4 years (0.0–3.6). Twenty-five OH Vitamin D, parathyroid hormone and calcium levels were available for, respectively, 42, 37 and 74 patients, with median values of 16.5 ng/ml (10.4–27.9), 48.0 pg/ml (30.6–71.9) and 2.30 mmol/l (2.25–2.37), respectively.
Evolution of bone mineral density
After a median time of 2.6 years, BMD had fallen significantly at both sites, with a mean percentage change from baseline of −1.3 ± 4.1% (P = 0.010) at the lumbar spine and −0.9 ± 4.5% (P = 0.033) at the hip (Fig. 1). The proportion of patients with a decrease in BMD larger than the SDD was 25.5% at the lumbar spine and 27.7% at the hip, with a mean decrease of −6.6 ± 2.7% at the spine and −6.0 ± 3.7% at hip. Two patients had a T-score less than −2.5 (osteoporosis) in at least one site on the second DXA. The osteoporosis incidence rate was 8.4 (1.0–30.9) per 1000 person-years. Only one fracture (traumatic) occurred during the study.
Factors predictive of a larger decline in bone mineral density
Table 1 shows the factors associated with a greater than 1 SDD decline in BMD at the lumbar spine and hip by using univariate and multivariable logistic models. At the lumbar spine, patients currently exposed to TDF (OR: 2.4, P = 0.016) were more likely than patients who had never been exposed to TDF to have a decline in BMD exceeding the SDD (Table 1a). The mean decrease in BMD was −2.2 ± 4.2% in patients currently exposed to TDF and −0.4 ± 3.9% in patients on other regimen. At the hip, patients who had been exposed to TDF for a cumulative period of less than 2.5 years had a higher risk (OR: 2.8, P = 0.010) of experiencing a decline in BMD exceeding the SDD relative to never exposed patients (Table 1b). The mean decrease in BMD at the hip were −0.7 ± 3.1% in patients who had never been exposed to TDF, −2.7 ± 4.2% in those exposed for less than 2.5 years and 0.4 ± 5.4% in those exposed for more than 2.5 years.
The evolution of bone density demineralization in HIV-infected patients with osteopenia is a key issue in the context of aging. In this prospective study on 94 middle aged men with a median age of 46 years, cART treated for a median duration of 7.5 years, we showed a significant decline in BMD at the spine and hip over a two and half time period, representing −0.5 and −0.4% per year, respectively. About one-quarter of the patients had a decline in BMD exceeding the SDD, with a mean of −2.6% per year at the spine and −2.3% per year at the hip. The incidence of osteoporosis observed here was 8.4 per 1000 person-year and was similar to that reported at 2 years in the Aquitaine Cohort .
With regards to the risks factors for osteoporosis , after accounting for the traditional factors such as age and BMI, TDF exposure was associated with a BMD decline exceeding the SDD at both the lumbar spine and the hip. Given the relatively small size of the study, we only studied exposure (current and cumulative) to drugs known to potentially influence BMD (protease inhibitors and TDF) [10,14,29]. Vitamin D, parathyroid hormone and calcium levels were not significantly associated with such a decline, possibly because of the high frequency of missing data that needed to be imputed. The high level of missing values for these parameters is explained by the fact that they were collected if they had been measured between the two DXA measurements and were not mandated per protocol.
The average decrease in BMD observed here is similar to that found in longitudinal studies of HIV-infected patients on long-term cART [16,17], but lower than that observed after 1 year of first-line antiretroviral therapy [9,10,14]. In the months following treatment initiation, immune restoration could contribute to bone loss through cytokine activation . Two recent studies [14,29] showed a larger decrease in BMD (lumbar spine and hip) in previously untreated patients receiving TDF/FTC than in those receiving ABC/3TC, after 1 year of treatment. Gallant et al.  observed a significantly larger decrease in BMD at the lumbar spine (but not at the hip), 1 year after first-line treatment initiation with TDF compared with stavudine. However, the decrease in BMD appeared between weeks 24 and 48 and then remained stable until week 144. The present study shows that TDF exposure is a risk factor for bone loss, and that bone loss continues, albeit at a slower pace, well beyond the first year of treatment. TDF might interfere with bone metabolism through its renal tubule toxicity and by increasing bone turnover . It may also induce secondary hyperparathyroidism, particularly in patients who are vitamin D-deficient , which was the case in approximately 75% of our cohort for which vitamin-D level was measured. TDF is currently far the most widely used nucleotide analogue in cART. Whether this decline could be reversible or at least slowed down after switch to therapies that avoid drugs with an impact on BMD is a question of growing importance for managing aging HIV-infected population.
In conclusion, the loss of BMD observed here in osteopenic HIV-infected patients with no identified cause of secondary osteoporosis was similar to that reported in the general population of similar age at total hip and slightly higher at lumbar spine . These data suggest that osteopenic patients with controlled viral load, receiving TDF, that is the majority of patients under care in resource rich settings, should regularly be monitored for BMD.
D.C., C.K., S.K. and S.R. were involved in the conception and design of the study.
C.K., A.S., J-P.V., M.B., C.R., S.K. and S.R. were responsible for the provision of study materials or patients. S.K. was a member of DXA-centralized reading centre.
L.A. and D.C. were responsible for statistical analysis.
L.A., C.K., J-P.V., M.B., A.S., C.R., S.K., D.C. and S.R. were involved in interpretation of the data.
L.A., D.C. and S.R. were responsible for drafting of the article.
L.A., C.K., J-P.V., M.B., A.S., C.R., S.K., D.C. and S.R. did the critical revision of the article for important intellectual content and also final approval of the article.
Appendix: members of the Osteovir study team.
Clinical Centers (in order of the number of patients screened): Hopital Pitie-Salpetriere, Paris–C.K., M.A. Valantin, L. Schneider and H. Schoen; Hopital Pitie-Salpetriere, Paris–A.S, M. Bonmarchand and M. Irguetsira; Hopital Necker, Paris–J-P.V., G. Obenga and A. Maignan; Hopital Avicenne, Bobigny–M.B., P. Honore, F. Rouges and L. Tuna; Hopital Cochin, Paris–D. Salmon-Ceron, N.Benammar, F. Almasi and V. Le Baut.
DXA Centers: Hopital Pitie-Salpetriere, Paris–S. Rozenberg and R. Inaoui; Hopital Avicenne, Bobigny–P. Weinmann; Hopital Cochin, Paris–C. Roux, S. Kolta and N Vautier.
Source of funding: the Agence Nationale de Recherche contre le SIDA et les hépatites virales (ANRS, France).
Conflicts of interest
No author has any financial or personal relationships with people or organizations that could inappropriately influence this work, although most authors have, at some stage in the past, received funding from a variety of pharmaceutical companies for research, travel grants, speaking engagements or consultancy fees.
1. Kanis JA. Diagnosis of osteoporosis and assessment of fracture risk
2. Arnsten JH, Freeman R, Howard AA, Floris-Moore M, Lo Y, Klein RS. Decreased bone mineral density and increased fracture risk in aging men with or at risk for HIV infection
3. Brown TT, Qaqish RB. Antiretroviral therapy and the prevalence of osteopenia and osteoporosis: a meta-analytic review
4. Amiel C, Ostertag A, Slama L, Baudoin C, N’Guyen T, Lajeunie E, et al. BMD is reduced in HIV-infected men irrespective of treatment
. J Bone Miner Res
5. Bonjoch A, Figueras M, Estany C, Perez-Alvarez N, Rosales J, del Rio L, et al. High prevalence of and progression to low bone mineral density in HIV-infected patients: a longitudinal cohort study
6. Bruera D, Luna N, David DO, Bergoglio LM, Zamudio J. Decreased bone mineral density in HIV-infected patients is independent of antiretroviral therapy
7. Cazanave C, Dupon M, Lavignolle-Aurillac V, Barthe N, Lawson-Ayayi S, Mehsen N, et al. Reduced bone mineral density in HIV-infected patients: prevalence and associated factors
8. Tebas P, Powderly WG, Claxton S, Marin D, Tantisiriwat W, Teitelbaum SL, Yarasheski KE. Accelerated bone mineral loss in HIV-infected patients receiving potent antiretroviral therapy
9. Briot K, Kolta S, Flandre P, Boue F, Ngo Van P, Cohen-Codar I, et al. Prospective one-year bone loss in treatment-naive HIV+ men and women on single or multiple drug HIV therapies
10. Duvivier C, Kolta S, Assoumou L, Ghosn J, Rozenberg S, Murphy RL, et al. Greater decrease in bone mineral density with protease inhibitor regimens compared with nonnucleoside reverse transcriptase inhibitor regimens in HIV-1 infected naive patients
11. Gallant JE, Staszewski S, Pozniak AL, DeJesus E, Suleiman JM, Miller MD, et al. Efficacy and safety of tenofovir DF vs stavudine in combination therapy in antiretroviral-naive patients: a 3-year randomized trial
12. Rivas P, Gorgolas M, Garcia-Delgado R, Diaz-Curie M, Goyenechea A, Fernandez-Guerrero M. Evolution of bone mineral density in AIDS patients on treatment with zidovudine/lamivudiine plus abacavir or lopinavir/ritonavir
. HIV Med
13. van Vonderen MG, Lips P, van Agtmael MA, Hassink EA, Brinkman K, Geerlings SE, et al. First line zidovudine/lamivudine/lopinavir/ritonavir leads to greater bone loss compared to nevirapine/lopinavir/ritonavir
14. 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
15. Bolland MJ, Grey AB, Horne AM, Briggs SE, Thomas MG, Ellis-Pegler RB, et al. Bone mineral density remains stable in HAART-treated HIV-infected men over 2 years
. Clin Endocrinol
16. Mondy K, Yarasheski K, Powderly WG, Whyte M, Claxton S, DeMarco D, et al. Longitudinal evolution of bone mineral density and bone markers in human immunodeficiency virus-infected individuals
. Clin Infect Dis
17. Nolan D, Upton R, McKinnon E, John M, James I, Adler B, et al. Stable or increasing bone mineral density in HIV-infected patients treated with nelfinavir or indinavir
18. Mary-Krause M, Viard JP, Ename-Mkoumazok B, Bentata M, Valantin MA, Missy P, et al. Prevalence of low bone mineral density in men and women infected with human immunodeficiency virus 1 and a proposal for screening strategy
. J Clin Densitom
19. Rozenberg S, Lanoy E, Bentata M, Viard JP, Valantin MA, Missy P, et al. Effect of alendronate on HIV-associated osteoporosis: a randomized, double-blind, placebo-controlled, 96-week trial (ANRS 120)
. AIDS Res Hum Retroviruses
20. Kolta S, Ravaud P, Fechtenbaum J, Dougados M, Roux C. Follow-up of individual patients on two DXA scanners of the same manufacturer
. Osteoporos Int
21. Ravaud P, Reny JL, Giraudeau B, Porcher R, Dougados M, Roux C. Individual smallest detectable difference in bone mineral density measurements
. J Bone Miner Res
22. Faulkner KG. Bone densitometry: choosing the proper skeletal site to measure
. J Clin Densitom
23. Genant HK, Grampp S, Gluer CC, Faulkner KG, Jergas M, Engelke K, et al. Universal standardization for dual x-ray absorptiometry: patient and phantom cross-calibration results
. J Bone Miner Res
24. Janssen KJM, Donders ART, Harrell FE, Vergouwe Y, Chen QX, Grobbee DE, Moons KGM. Missing covariate data in medical research: to impute is better than to ignore
. J Clin Epidemiol
25. Van Buuren S, Boshuizen HC, Knook DL. Multiple imputation of missing blood pressure covariates in survival analysis
. Stat Med
26. Rubin D. Multiple imputation for nonresponse in surveys
. New York:Wiley; 1987.
27. Cazanave C, Lawson-Ayayi S, Barthe N, Uwamaliya-Nziyumvira B, Kpozehouen A, Mehsen N, et al. Changes in bone mineral density: 2 year follow-up of the ANRS CO3 Aquitaine cohort
. 17th Conference on Retroviruses and Opportunistic infections. San Francisco 2010.
28. Mallon PW. HIV and bone mineral density
. Curr Opin Infect Dis
29. Stellbrink HJ, 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
30. Ofotokun I, Weitzmann N, Vunnava A, Sheth A, Villinger F, Zhou J, et al. HAART-induced Immune Reconstitution: A Driving Force Behind Bone Resorption in HIV/AIDS
. 18th Conference on Retroviruses and Opportunistic Infections. Boston 2011.
31. Masia M, Padilla S, Robledano C, Lopez N, Ramos JM, Gutierrez F. Early changes in parathyroid hormone concentrations in HIV-infected patients initiating antiretroviral therapy with tenofovir
. AIDS Res Hum Retroviruses
32. Warming L, Hassager C, Christiansen C. Changes in bone mineral density with age in men and women: a longitudinal study
. Osteoporos Int
Keywords:© 2013 Lippincott Williams & Wilkins, Inc.
antiretroviral therapy; bone mineral density; HIV; osteopenia; osteoporosis; smallest detectable difference; tenofovir