Coinfection

IDWeek 2014: Study Shows HIV/HCV Coinfected Women Have Lower Bone Density

ART-treated women with HIV/HCV coinfection have greater deficits in some structural bone parameters compared to women with HIV only, HCV only, or neither virus, according to the results of a cross-sectional study presented at IDWeek 2014 last week in Philadelphia. Among women with HCV, bone loss was most profound in those with stage 3-4 liver fibrosis or cirrhosis, which adversely effects bone health.

"Compared to healthy reference participants, the women with HIV/HCV had decreased tibial trabecular volumetric bone mineral and cortical thinning, and significant endocortical bone loss, a pattern observed in other inflammatory diseases," according to Vincent Lo Re of the University of Pennsylvania, who presented the study findings.

Background

People with HIV and hepatitis C virus (HCV) coinfection -- particularly women -- have an increased risk of low bone mineral density (BMD) and hip fractures than those living with HIV alone, and these bone changes may precede liver cirrhosis. This has also been seen in people with HIV and hepatitis B virus (HBV) coinfection.

The mechanisms underlying BMD loss and fractures in coinfected women are unclear, though inflammation -- evidenced by elevated levels of inflammatory cytokines such as tumor necrosis factor-alfa (TNF-alfa), interleukin-1 (IL-1), and IL-6 -- appears to play a role. Furthermore, the structural underpinnings for skeletal fragility in HIV/HCV coinfection are unknown. Better characterizing these bone deficits and mechanisms for bone loss could help guide studies of appropriate therapies.

Dual-energy X-ray absorptiometry (DEXA) is widely used to screen for osteoporosis. This is a 2-dimensional projection technique that summarizes total BMD in trabecular bone (spongy bone) and cortical bone (the dense outer shell of bone), but it cannot distinguish between trabecular and cortical bone to reveal which aspect of the bone’s structure is changing.

Peripheral quantitative computed tomography (pQCT), however, can address the limitations of DEXA by providing a 3-dimensional image of the bone, making it possible to describe trabecular volumetric BMD (vBMD) and cortical bone dimensions, including cortical thickness, cortical area, periosteal circumference, and endosteal circumference.

Lo Re and colleagues hypothesized that if they used pQCT of the tibia (shin bone) and whole-body DEXA to evaluate musculoskeletal parameters in women with HIV/HCV coinfection, they would find significant differences when compared to uninfected women or women living with HIV or HCV alone.

The Study

The team performed a cross-sectional study of women attending 3 HIV and hepatitis C clinics in Philadelphia, including 50 ART-treated (stable for 3 or more months) women with HIV/HCV coinfection, 50 ART-treated women with HIV alone, and 51 women with HCV alone. Participants were excluded if they were pregnant or breastfeeding or if they had other conditions known to affect nutrition or bone, such as opportunistic infections, malignancies, active substance abuse, or liver decompensation.

Tibial trabecular and cortical volumetric BMD and cortical dimensions were determined by pQCT. Whole-body fat mass and appendicular lean mass in the limbs were measured by DEXA. Age- and race-specific Z-scores (standardized scores within the normal distribution for age and race) were generated using data from a reference group of 263 healthy women without HIV or liver disease or on any medications that could affect bone or nutrition. Linear regression was used to evaluate differences in mean pQCT Z-scores between the groups, adjusted for fat and lean mass. Data were also gathered on vitamin D metabolism, parathyroid hormone (PTH), and TNF-alfa, IL-1 beta, and IL-6 levels.

The median age ranged from 47 to 55 years in the HIV/HCV, HIV-only, and HCV-only groups (highest in the HCV-only arm), and was 47 in the reference group. The majority of women (78%-84%) in the HIV/HCV, HIV, and HCV groups were black, compared with 42% in the reference group. Women in the HIV/HCV group and the reference group had significantly lower body mass index (BMI) than the HIV-only or HCV-only groups. 80% of coinfected women, 32% in the HIV-only group, and 49% in the HCV-only group -- but just 13% in the reference group -- were smokers.

Most women with HIV alone (74%) or HIV/HCV coinfection (86%) had undetectable HIV viral load and the median CD4 T-cell count was 605-635 cells/mm³. About 75% of women with HIV had used tenofovir (Viread, also in Atripla and other coformulations). Nearly one-quarter of women in the HIV/HCV and HCV groups had advanced fibrosis (stage 3) or cirrhosis (stage 4). Women who had received prior hepatitis C treatment were excluded.

Results

• HIV/HCV coinfected women had lower mean Z-scores for trabecular vBMD (-0.85 vs -0.01), cortical thickness (-0.77 vs 0.00), and cortical area (-0.61 vs +0.03) compared to the reference group.
• These differences were statistically significant, and remained so after adjustment for smoking, physical activity, fat and lean mass.
• Decreased cortical dimensions were due to greater mean Z-scores for endosteal circumference (+0.67); periosteal circumference was slightly greater (+0.04), but this did not reach statistical significance.  
• HIV/HCV coinfected women had a significantly lower mean trabecular vBMD Z-score than women with HIV alone (-0.85 vs 0.06) or those with HCV alone (-0.85 vs -0.21), but there were no differences in cortical dimensions across these groups.
• Among participants with chronic HCV infection, those with stage 3 or 4 liver fibrosis or cirrhosis had significantly lower trabecular vBMD (-0.90 vs -0.44) and cortical thickness (-1.1 vs -0.36) Z-scores than those with stage 0-2 (absent to moderate) fibrosis.
• There were no differences in vitamin D metabolism or levels of most inflammatory cytokines between women with HIV/HCV, HIV alone, or HCV alone.
• The exception was TNF-alfa, which was significantly elevated in coinfected women compared to women with either HIV alone or HCV alone.

There are a number of limitations to this study, one being its cross-sectional design (looking at bone parameters at a single point, rather than changes over time). Also, the team had no information about the duration of HIV and/or HCV infection, and was unable to confirm HIV or HCV negative status in the presumed healthy reference group. Finally, it is not clear whether the study’s findings are generalizable to men.

Nevertheless, the study does suggest that future research should evaluate the mechanisms for these structural abnormalities and explore whether successful hepatitis C treatment can reverse bone loss among people living with HIV/HCV coinfection.

10/22/14

Reference

V Lo Re, K Lynn, E Stumm, et al. Structural Bone Deficits in HIV/HCV, HCV-Monoinfected, and HIV-Monoinfected Women. IDWeek 2014. Philadelphia, October 8-12, 2014. Abstract 645.