CROI 2015 Program and Abstracts

Abstract Listing

Poster Abstracts

Results: Compared to HIV negative group, those with HIV were younger, more likely to be male and less likely to be Caucasian. Those with HIV had lower baseline BMD at FN, TH and LS (all p<0.05, Table 1). Although BMD declined at all three sites in both groups (Table 1), there was no signficant between-group difference in rate of absolute change in BMD at LS (+0.002 g/cm 2 /year, p=0.51), TH (-0.001 g/cm 2 /year, p=0.69) and FN (-0.004 g/cm 2 /year, p=0.08) after adjustment for age, gender, ethnicity, smoking status and body mass index. In a HIV-specific sub-analyses (table 1), one IU/L increase in baseline ALP was independently associated with a greater subsequent BMD loss at FN (0.00016 g/cm 2 /year loss, p=0.016). Additional adjustment for CTX-1, P1NP or OC did not change this association and longitudinal changes in BMD were not associated with current or cumulative exposure to tenofovir disoproxil fumarate (TDF), nadir CD4+T-cell count, or traditional risk factors of age, gender, ethnicity, PTH and 25(OH) D. Characteristics of HIV-positive and HIV negative participants

Data are median (IQR) unless specified. ART = antiretroviral therapy, MSM = men who have sex with men. IDU = intravenous drug user. TDF = tenofovir disoproxil fumarate. BMD = bone mineral density. TH = total hip. FN = femoral neck. LS = lumbar spine. BL = baseline. Conclusions: Although those with HIV have lower BMD, we observed no difference in rate of BMD loss between groups over time. That higher ALP was associated with greater decreases in FN BMD in those with HIV may reflect altered bone turnover, although exposure to TDF was not implicated in progressive BMD loss. 775 Fracture Incidence Is Increased in Aging HIV-InfectedWomen Anjali Sharma 1 ; Qiuhu Shi 2 ; Donald R. Hoover 3 ; Kathryn Anastos 1 ; Phyllis C.Tien 4 ; Mary A.Young 5 ; Mardge Cohen 6 ; Deborah Gustafson 7 ; MichaelTYin 8 1 Albert Einstein College of Medicine, Bronx, NY, US; 2 New York Medical College, Valhalla, NY, US; 3 Rutgers University, Piscataway, NJ, US; 4 University of California San Francisco, San Francisco, CA, US; 5 Georgetown University, Washington D.C., DC, US; 6 John H. Stroger Jr. Hospital of Cook County, Chicago, IL, US; 7 State University of New York Downstate Medical Center, Brooklyn, NY, US; 8 College of Physicians and Surgeons, Columbia University, New York, NY, US Background: Low bone mineral density (BMD) is common among HIV-infected individuals. Some studies find higher fracture incidence in HIV+ than HIV- individuals. There is particular concern for increased rates of fracture in older HIV-infected women as they continue to age. Methods: We analyzed a cohort of 2375 (1713 HIV+, 662 HIV-) women enrolled in the Women’s Interagency HIV Study (WIHS) to evaluate the effects of HIV infection on time to first self-reported fragility and non-fragility fracture, and determine risks for incident fracture, between 2003-2013. Incident fractures were assessed at semiannual visits. Associations of traditional fracture risk factors and HIV disease characteristics measured at index visit with subsequent incident fracture were estimated using proportional hazards models. Results: HIV+ women were older than HIV- (median 40 vs 35 yrs) and had lower body mass index (28 vs 30kg/m 2 ), but were racially similar (59% vs 62% black). HIV+ women were more likely to be post-menopausal (19% vs 11%), HCV infected at study entry (24% vs 15%), and use vitamin D at index visit (42% vs 28%), but less likely to smoke (45% vs 51%). Among HIV+, mean CD4+ count was 480 cells/ μ L and 63%were taking HAART at index visit. New fractures occurred in 300 (17.5%) HIV+ women and 90 (13.6%) HIV- women, including new fragility fractures in 82 (4.8%) HIV+ and 24 (3.6%) HIV- women. Unadjusted fracture incidence rates were higher in HIV+ versus HIV- women (2.19 vs 1.54/100py, p=0.002); however, unadjusted incidence rates of fragility fractures did not statistically differ (0.56/100py vs. 0.39/100py, p=0.13). Adjusted for age, race, prior fracture, and history of cocaine and injection drug use, HIV+ women had significantly higher incidence rate of any fracture compared with HIV- women (aHR1.30; 95% CI 1.02-1.66). Among HIV+ women, older age, white race, prior fracture, smoking, and prior AIDS defining illness (aHR=1.56; 95% CI: 1.22-1.98) significantly predicted new fracture, while CD4+ count and antiretroviral exposure did not. Conclusions: Fracture rates were higher among HIV+ than HIV- women. Cocaine and injection drug use were also significant predictors of incident fracture. Further research is needed to understand whether the risk of fracture associated with cocaine use relates to increased rate of falls, or direct effects on bone metabolism. Our data support optimizing musculoskeletal health in older HIV-infected women.

Poster Abstracts

474

CROI 2015