The AGEhIV cohort study is an ongoing, prospective, cohort study in Amsterdam, with 598 HIV-1 infected individuals and a control group of 550 HIV-uninfected individuals, from the same geographic region and with similar socio-demographic and behavioral risks factors. All participants were included between 2010 and 2012, with an age ≥ 45 years. In this study, both mineral density(BMD) was evaluated by DXA scanning performed at enrolment in the cohort. Multivariable linear regression models assessed whether HIV positive status was independently associated with BMD in the lumbar spine, total hip, and femoral neck as continuous dependence variable. All models were adjusted for age, sex, menopausal status, body weight, race (black or not black), and smoking status. Final models were also adjusted for the following co-variates: physical activity, intake of dairy/fish, prescription drug and substance use, family's history of hip fracture, and HBV and HCV infection or co-infection. Finally, levels of hsCRP, D-dimer , sCD163, sCD14 and 25-hydroxy vitamin D2+D3 were also explored as potential mediators in the association between HIV and BMD. For the HIV population, multi-variable models explored as co-variates duration of HIV-infection, CDC stage, historic body weight, current and nadir CD4 cells counts, HIV-1 plasma viral load, current and prior cART use and its duration. DXA of lumbar spine, total hip, and femoral neck was performed in 581 HIV-positive (94.7% receiving cART) and 520 HIV-negative participants. The population largely consisted of men (86.7%); 79% of male were MSM. Osteoporosis ( T score ≤ -2.5 SD) or osteopenia (T score between -1 and -2.5 SD) in each of the 3 locations were significantly higher in HIV-infected individuals: prevalence of osteoporosis in ≥ 1 site 14.3% vs 6.7%; p< 0.001; total hip osteopenia: 29% vs 16%, p< 0.001. After adjustment for body weight and smoking, being HIV-positive was no longer independently associated with BMD. Low body weight was more strongly negatively associated with BMD in HIV-positive persons with a history of class B or C events. HIV co-variate such as duration of HIV infection, current or nadir CD4 cell count, CD4/CD8 cell count ratio, markers of HIV replication, pre-treatment with mono or dual NRTI were not associated with BMD. The lowest recorded historic body weight was associated significantly with BMD in the lumbar spine and femoral neck. A significant positive association was observed between current use of nevirapine and BMD in all 3 locations, and a significant negative association between the duration of exposure to high dose (≥ 400 mg/day) ritonavir and BMD in the femoral neck and total hip. Other types of ART, including current or prior TDF use or its duration, were not associated with BMD.

Conclusion: the observed lower BMD in treated HIV positive individuals was largely explained by both lower body weight and more smoking. Having experienced symptomatic HIV disease, often associated with weight loss, was another risk factor.