Age raises fracture risk more in HIV-positive people

Mark Mascolini

Age had a greater impact on fracture risk in people with HIV than in matched controls without HIV, according to a retrospective cohort study involving
well over 200,000 individuals. [1]

Among 30-to-59-year-olds, HIV with or without AIDS independently raised the fracture risk.

The study involved 238,336 adults continuously enrolled in the Ingenix Impact National Benchmark Database for more than 12 months from January 1997
through March 2008. Researchers studied 59,584 HIV-infected people matched 3-to-1 with HIV-negative people according to gender, month and year of
enrollment in the Ingenix cohort, and duration of cohort membership. They defined fracture as a low-impact nontraumatic fracture. AIDS criteria were a CD4
count <200 cells/mm3 or AIDS-defining conditions.

Nearly three quarters of HIV-positive cases and HIV-negative controls (71.5%) were men. Prevalence of several fracture risk factors were higher in the
HIV-positive group: excess alcohol use (2.5% versus 1.3% in controls), low weight (7.9% versus 1.6%), lipodystrophy (3.4% versus 0.04%), hepatitis B virus
(4.1% versus 0.2%), hepatitis C virus (6.5% versus 0.6%), and excess steroid use (5.5% versus 3.6%). But prior fracture rates were similar (2.0% among
people with HIV and 1.9% in controls), as was use of proton pump inhibitors (14.5% versus 10.5%). A slightly larger proportion of controls took
bisphosphonates (2.0% versus 1.3% of people with HIV). Only 51% in the HIV group were taking antiretrovirals.

Through 13,757 person-years of follow-up, 9027 people (3.8%) fractured a bone, including 4.2% with HIV and 3.7% without HIV. Fracture incidence was 14%
higher in the HIV group (incidence rate ratio 1.14, 95% confidence interval [CI] 1.09 to 1.20).

A statistical model considering numerous fracture risk factors identified five independent predictors of fracture in the entire cohort (HR, 95%CI):

  • Prior fracture: hazard ratio (HR) 4.49 (3.89 to 5.18)
  • Excess alcohol: HR 1.90 (1.65 to 2.20)
  • Low physical activity: HR 1.77 (1.73
    to 1.82)
  • Anti-osteoporosis bisphosphonate use: HR 1.49 (1.29 to 1.72)
  • Low weight: HR 1.32 (1.18 to 1.48)

The researchers
then identified independent fracture risk factors in three age groups:

People under 30

  • Prior fracture: HR 7.77 (3.23 to 18.67)
  • Excess alcohol: HR 2.24 (1.06 to 4.74)

People 30 to 59 years old

  • Prior fracture: HR 3.81 (3.14 to 4.63)
  • Low physical activity: HR 2.24 (1.90 to 2.65)
  • Excess alcohol: HR 1.86 (1.55 to 2.23)
  • Anti-osteoporosis bisphosphonate use: HR 1.36 (1.20 to 1.68)
  • Low weight: HR 1.30 (1.12 to
  • HIV without AIDS: HR 1.18 (1.09 to 1.28)
  • HIV with AIDS: HR 1.15 (1.06 to 1.26)
  • Vitamin D deficiency or D or calcium
    supplementation: HR 0.72 (0.54 to 0.98)

People older than 59

  • Prior fracture: HR 2.79 (1.68 to 4.64)
  • Low
    physical activity: HR 2.65 (1.67 to 4.21)

The researchers could not explain why vitamin D deficiency or vitamin D or calcium supplementation
was protective in their analysis.

Linear trend analysis showed that, compared with HIV-negative people, fracture risk rose significantly more with advancing age among HIV-infected people
without AIDS (p=0.012) or with AIDS (p=0.001). The increased fracture trend with age in HIV-infected people was higher in those with AIDS than in those
without AIDS.

The investigators noted that their analysis suffered from lack of data on bone mineral density, race, and over-the-counter drugs. Only 6% of people
analysed reported smoking, so the researchers did not including smoking in the analysis.

An earlier comparison of 8,525 people with HIV and 2,208,792 people without HIV in a Boston healthcare system found significantly higher fracture
prevalence in both men and women with HIV than in uninfected men and women. [2]


  1. Mundy LM, St. Laurent S. Bowlin S, et al. Age-stratified risk assessment for fracture among adults with and without HIV infection.
    1st International Workshop on HIV and Aging. 4-5 October 2010. Abstract O_07.
  2. Triant VA, Brown TT, Lee H, Grinspoon SK. Fracture prevalence
    among human immunodeficiency virus (HIV)-infected versus non-HIV-infected patients in a large U.S. healthcare system. J Clin Endocrinol Metab.

Links to other websites are current at date of posting but not maintained.