CROI 2015 Program and Abstracts

Abstract Listing

Poster Abstracts

TUESDAY, FEBRUARY 24, 2015 Session P-Q4 Poster Session

Poster Hall

2:30 pm– 4:00 pm Measuring Bone Density 776 Heel Quantitative Ultrasound to Cut Down on DXA Costs in HIV-infected Patients Marilia R. Pinzone ; Maria Gussio; Daniela Castronuovo; Adriana Di Gregorio; Benedetto M. Celesia; Bruno Cacopardo; Giuseppe Nunnari University of Catania, Catania, Italy

Background: HIV infection has been associated with increased risk of osteopenia/osteoporosis and fragility fractures. DXA is the reference standard to assess bone mass density (BMD), however it is a radiation-based and expensive technique, not easily accessible in several clinical settings. Quantitative ultrasound (QUS) of the heel is a radiation-free, easy- to-perform technique, which may help reducing the need for DXA scan. Methods: In this cross-sectional study, we evaluated the prevalence of and risk factors for low BMD by QUS (Hologic Sahara) in a cohort of HIV-infected patients. We assessed the correlation between BMD and demographic, viro-immunological and biochemical parameters. Osteopenia/osteoporosis rates were compared with those of 36 sex- and age- matched HIV-negative controls. Mann-Whitney test was used to compare continuous variables, Chi-squared test for categorical variables. Spearman’s correlation and multivariate analysis were used to assess factors associated with low BMD. Results: 152 HIV-positive patients were enrolled, 29% female, median age 47(IQR 39-54) years. 19% of patients were HCV-positive, 60% smokers, 7% had diabetes. Median BMI was 24(IQR 22-26) kg/m 2 . Median time since HIV diagnosis was 122(IQR 47-216) months, median time from HAART initiation 119(IQR 39-204) months. 91%were on HAART, 53% were receiving a PI-based regimen, 44% a NNRTI-based one, 11%were taking RAL, 60% TDF, 80% had undetectable viral load, median CD4 count was 548 (IQR 407-695) cells/ μ l. The number of subjects with Sahara t-score <-1 was significantly higher among HIV-positive patients in comparison with controls (64 vs. 33%, p=0.012). 55% of HIV-positive patients had Sahara t-score <-1, 9%<-2.5). In the univariate but not in the multivariate analysis, factors associated with lower BMD were older age (p=0.002), lower CD4 at diagnosis (r=0.35, p=0.003), duration of tenofovir exposure (r=-0.19, p=0.02) and HCV coinfection (p=0.01). According to guidelines, 41% of patients had risk factors for osteoporosis who make them eligible for DXA. By using QUS, we may avoid/delay DXA in around 30% of them. On the other hand, 59% of patients ineligible for DXA according to US HIV guidelines, had a low QUS t-score. Conclusions: In our cohort, low BMD was highly prevalent. Heel QUS is a quick, easy-to-perform, relatively inexpensive and radiation-free technique which may help reducing the 1 St. Michael’s Hospital, Toronto, Canada; 2 Toronto General Research Institute, Toronto, Canada; 3 University Health Network, Toronto, Canada; 4 Centre for Excellence in Skeletal Health Assessment, Toronto, Canada Background: Bone mineral density as assessed by dual energy x-ray absorptiometry (DXA) is often used to assess fracture risk, but is limited by the lack of a clear fracture threshold and of validation in HIV. We conducted a pilot case-control study to determine whether novel radiologic methods such as total volumetric BMD (tvBMD) at the distal radius by high resolution peripheral quantitative computed tomography (HRpQCT) and buckling ratio (BR) at the narrow neck by hip structural assessment (HSA) correlate with fracture in HIV patients. Methods: Adults with a history of low-trauma fractures after their HIV diagnosis (cases) were matched 1:1 with HIV-infected adults with no prior fractures (controls) based on age, sex, race and smoking history. Participants underwent DXA, HRpQCT and HSA once. Conditional logistic regression was used to model the relationship between fracture and DXA- derived T-scores, adjusting for the duration of HIV infection at the time of the fracture. After further adjustment for each novel radiologic parameter, nested models were compared using the likelihood ratio (LR) test to determine whether any offered additional information independent of DXA. Results: 23 matched pairs were included. Median (interquartile range, IQR) age was 50 (46,56) years, 78%were male, 78%were white and 57%were smokers. Median (IQR) duration of HIV at the time of bone measurements was 19 (11,23) years for cases and 10 (7,18) years for controls; nadir CD4 was 234 (123,370) and 166 (36,410) cells/mm 3 respectively. 4 cases and 0 controls had ever used osteoporosis medications. DXA-derived T-scores showed trends towards association with fracture, with OR=0.87 per SD of T-score (95%CI=0.53,1.43) for the L-spine and OR=0.43 (95%CI=0.17,1.12) for the hip. When tvBMD was included in the hip model, the tvBMD estimate was not significant (OR=1.29 per 10mg/cm 3 , 95%CI=0.93,1.81), but the strength of the DXA-derived T-score increased (OR=0.11, 95%CI=0.01,1.08) as did model fit (LR p =0.06). When BR was included in the hip model, the BR estimate was not significant (OR=0.69, 95%CI=0.40,1.18), and the strength of the DXA-derived T-score again increased (OR=0.20, 95%CI=0.03,1.12), though model fit was not improved (LR p=0.14). Conclusions: Novel radiologic parameters such as HRpQCT-derived tvBMD and HSA-derived BR may aid in identifying correlates of low-trauma fracture in HIV patients, and warrant further study as supplementary markers of fracture risk in HIV clinical trials. need for unnecessary DXA as well as identifying high-risk subjects, requiring a thorough workup. 777 Novel Radiographic Measures HRpQCT and HSA as Correlates of HIV-Associated Fractures Darrell H. Tan 1 ; Janet Raboud 2 ; Leah Szadkowski 2 ; Eva Szabo 4 ; Hanxian Hu 4 ; QueenieWong 4 ; Angela Cheung 4 ; SharonWalmsley 3

Poster Abstracts

475

CROI 2015