HIV infection increases risk of bone fractures for men, but not for younger women

This article is more than 12 years old. Click here for more recent articles on this topic

A comparison of HIV-positive and HIV-negative male veterans in the United States found that HIV infection conferred an independent but "modest" risk of fragility fractures in the hip and spine, delegates at the 17th Conference on Retroviruses and Opportunistic Infections (CROI) in San Francisco heard on Thursday.

But a second study, in a separate cohort, found that younger, pre-menopausal women with HIV did not have an elevated risk of fractures when compared to HIV-negative women of a similar age.

Loss of bone density (osteoporosis and osteopenia) is prevalent amongst people with HIV, yet relatively little is known about how HIV affects the risk of of bone fracture.


multivariate analysis

An extension of multivariable analysis that is used to model two or more outcomes at the same time.

body mass index (BMI)

Body mass index, or BMI, is a measure of body size. It combines a person's weight with their height. The BMI gives an idea of whether a person has the correct weight for their height. Below 18.5 is considered underweight; between 18.5 and 25 is normal; between 25 and 30 is overweight; and over 30 is obese. Many BMI calculators can be found on the internet.


Expresses the risk that, during one very short moment in time, a person will experience an event, given that they have not already done so.


A condition in which bone mineral density is lower than normal, but less severe than osteoporosis.


Bone disease characterised by a decrease in bone mineral density and bone mass, resulting in an increased risk of fracture (a broken bone).

Several previous studies (though not all) have found higher rates of wrist, hip and vertebral fractures in individuals with HIV, compared to age- and sex-matched HIV-negative controls.

But these studies have not always controlled for other significant risk factors, such as body mass index (BMI), alcohol intake, and other physical and mental illnesses.

In this retrospective analysis, investigators from the US Veterans Aging Cohort Study (VACS) compared the risk of fragility fractures between 40,079 HIV-positive and 79,080 HIV-negative men enrolled in the VACS observational cohort between 1997 and 2009.

Fragility fractures are those that occur with minimal impact or trauma, typically at the wrist, vertebrae, or femoral head of the hip. The two groups were racially diverse, entirely male, and had very similar demographic and health characteristics except for HIV status.

Only first fractures at the characteristic fragility fracture sites (wrist, spine and hip) were considered. Initial analysis found wrist fractures to be common amongst younger participants, with no statistically significant difference according to HIV status. The investigators therefore chose to exclude the 1233 wrist fractures from all analyses.

With this exclusion, a total of 919 fragility fractures were observed, 297 vertebral and 622 hip, during a median of eight years follow-up. Mean age at the time of fracture was 55 years.

The first, unadjusted analysis showed that fractures were 53% more likely among patients with HIV (hazard ratio [HR] = 1.53; 95% confidence interval [CI], 1.34 to 1.75).

A multivariate analysis was then done to adjust for race, low body weight, alcohol use, and other known risk factors. (However, data on smoking, family history, and bone density level were not included in the analysis.) HIV remained significantly associated with a 38% increased risk of fragility fracture (HR = 1.38; 95% CI, 1.18–1.60). The following other risk factors were also identified:

  • low body weight (cachexia): HR = 2.83; 95% CI, 2.26–3.54
  • cerebrovascular disease: HR = 1.89; 95% CI, 1.34–2.65
  • white race: HR = 1.79; 95% CI, 1.57–2.04
  • alcohol abuse: HR = 1.73; 95% CI, 1.42–2.10
  • older age: HR = 1.53 per decade; 95% CI, 1.44–1.63
  • enrolment before 1999 (protective): HR = 0.74; 95% CI, 0.64–0.87

A final multivariate analysis of the HIV-positive cohort found no significant effect due to tenofovir (Viread), stavudine (d4T, Zerit), NNRTI or protease inhibitor use at baseline; however, higher CD4 cell counts were slightly protective (HR = 0.96 per each additional 100 cells/mm3; 95% CI, 0.91–0.99).

As this cohort was composed entirely of men, no inferences could be made regarding HIV-positive women. A second study, of HIV-positive women, is discussed below.

In a related press conference, lead investigator Julie Womack commented that the level of risk purely due to HIV infection was more modest than that found in other studies, but still appeared to be an independent factor and a contributor to greater overall risk. (Jules Levin, of the National AIDS Treatment Advocacy Project, contested the term "modest", as fractures are serious and their prevalence will only increase as people with HIV age.)

Womack also noted that their observed fracture rates could include some fractures due to trauma rather than fragility and could therefore be overestimates.

Andrew Carr (of St Vincent's Hospital, Sydney) questioned why cumulative exposure to ART was not investigated instead of the snapshot of use at baseline, stating that "you can't definitively say it's HIV-related if you haven't eliminated the possibility that it's ART-related."

Womack felt that HIV infection alone did not merit DEXA bone density scans, and stated that "most" fragility fractures happened in people who did not have osteopenia or osteoporosis. DEXA, she said, "might be indicated for persons with moderate to high risk for fragility fracture."

However, Womack also recommended that "all people at risk" should be encouraged to start appropriate protective measures such as weight-bearing exercise, vitamin D supplementation and smoking cessation. Asked to clarify who should be considered "at risk", she told aidsmap that she would recommend preventive measures for people with any known risk factor or factors, including HIV infection.

Risk of bone fracture unaffected by HIV status in cohort of young US women

HIV-positive women in the US Women’s Interagency Health Study (WIHS) were not at higher risk of bone fracture than their HIV-negative counterparts, lead investigator Michael Yin told delegates at the conference in a Thursday session on cardiovascular and bone disorders.

This study analysed retrospective data from the WIHS database to compare the incidence of fracture between HIV-positive and HIV-negative women, to identify risk factors for fracture, and to determine if HIV infection was an independent risk factor.

The study information was obtained at six monthly clinic visits when the women were asked to report any fractures. This study looked at trauma fractures, in which the bone is broken by a severe impact, as well as fragility fractures (those which occur with minimal impact, typically at the wrist, spine or hip), and the two types were distinguished in the analysis. Only first fractures were considered.

The findings were reported as incidence density rates – the number of participants who had fractures, divided by the time during which they were at risk (reported in person-years). (Incidence density accounts for varying lengths of follow-up.)

The study included the 2391 women (1728 HIV-positive, 663 HIV-negative) for whom fracture data were reported and who had had at least one additional visit.

Risk factors for fracture were more prevalent amongst women with HIV than the HIV-negative women. Those with HIV were older on average (40 vs 36 years), had a lower average body mass index (28 kg/m2 vs 30 kg/m2), were more likely to be post-menopausal (20% vs 11%), and to be co-infected with hepatitis C virus (25% vs 15%). However, they were less likely to smoke (45% vs 50%) or have high alcohol consumption (2% vs 4%). Previous history of fracture did not differ by HIV status.

The HIV-positive women had a median baseline CD4 cell count of 482 cells/mm3 and a median nadir (lowest-ever) count of 233 cells/mm3; 38% had had an AIDS-defining illness. Sixty-six percent were taking antiretroviral therapy at study entry.

During a median of 5.4 years of follow-up, 148 fractures (9%) occurred in the HIV-positive group and 47 (7%) in the HIV-negative. Overall fracture rates at any body site did not significantly differ between the HIV-positive and HIV-negative women (1.79 vs 1.41 per 100 person-years, p = 0.13).

Differences by HIV status were also non-significant for fractures at each of the individual body sites.

Multivariate analysis found that increasing age, white race, hepatitis C infection, and higher serum creatinine levels were significant risk factors for bone fracture, but that HIV infection (when other factors were controlled for) was not.

Looking only at the women with HIV, the investigators found that a history of AIDS-defining illness increased the risk of fracture (hazard ratio [HR] = 1.9, 95% confidence interval [CI], 1.3-2.7, p = 0.0008), as did older age (HR = 1.2 per additional decade, 95% CI 1.0-1.5, p = 0.047). No association was found with CD4 cell count or cumulative exposure to antiretroviral therapy.

The study team concluded that traditional risk factors, but not HIV status, were predictive of bone fractures in this study group of mostly premenopausal, largely black women. Yin believed that "oestrogen has a tremendous protective effect, and we may see a shift in post-menopausal women," a population that the investigating group intends to study further.

Further information

You can view abstract 129 and abstract 130 on the official conference website.

You can also view a webcast and slides of this session on the official conference website.


Womack J et al. HIV-infection and fragility fracture risk among male veterans. Seventeenth Conference on Retroviruses and Opportunistic Infections, abstract 129, San Francisco, 2010.

Yin M et al. Fracture rates are not increased in younger HIV-positive women. Seventeenth Conference on Retroviruses and Opportunistic Infections, abstract 130, San Francisco, 2010.