Bone Fracture Facts: Prevalence and Incidence Higher With HIV

Abstract: Bone disease has affected people with HIV since the early days of the epidemic. Over the years, low bone mineral density has proved frequent in HIV-positive people, who bear a high burden of traditional bone loss risk factors. Accumulated evidence from the United States, Canada, and Western Europe indicates that people with HIV run a higher risk of fracture than comparison groups. Five cross-sectional studies confirmed significantly higher fracture prevalence in people with HIV than in comparison groups. Ten longitudinal comparative studies all found higher fracture incidence in HIV populations, and the higher fracture rate with HIV proved significant in seven of these studies, independently of certain classical risk factors. In one U.S. study, fracture incidence rose steadily over the years in people with HIV but not in the general population.

It all started with a bad tooth. A very bad tooth. In November 1989 -- a scant 5 years after researchers discovered the AIDS virus -- specialists in Copenhagen described what appears to be the first published case of bone disease in an HIV-positive person: tooth loss and necrosis of the alveolar (tooth socket) bone in a patient with trigeminal herpes zoster.1

Alveolar necrosis (bone death) did not emerge as a prominent complication of HIV infection, but over the next several years groups in the UK and France reported avascular necrosis of the hip2 and femoral head3 in people with HIV. In 1993 researchers in Spain compared 16 HIV-positive people with 27 healthy HIV-negatives and charted significantly lower levels of osteocalcin -- a bone formation marker -- in the HIV group.4 Bone mass also proved lower in the group with HIV. Although that difference lacked statistical significance, scores of later studies would confirm significantly lower bone mineral density (BMD) in people with HIV than in HIV-negative comparison groups.

BMD offers a convenient signal of bone health, but HIV clinicians and the people they see care more about the ultimate consequence of waning bone density: fractures. And plenty of research, detailed in this article, records higher fracture rates in people with than without HIV. The published history of broken bones in HIV-positive people dates back at least to 2001, when Italian clinicians reported fractures "after trivial trauma" in two men with AIDS -- one with osteopenia, the other with osteoporosis.5 One man was 49 and the other 51, and both had central and peripheral lipodystrophy. Both men had taken indinavir plus stavudine/lamivudine, both had well-controlled HIV infection, and neither had abnormal lipids.

Twelve years later meta-analysis of seven studies comparing fracture incidence in people with and without HIV calculated a 58% higher incidence of all fractures in the HIV group (pooled incidence rate ratio [IRR] 1.58, 95% confidence interval [CI] 1.25 to 2.00) and a 35% higher incidence of fragility fractures (pooled IRR 1.35, 95% CI 1.10 to 1.65).6 (Fragility fractures are those following minimal trauma and usually affecting the hip, spine, or wrist.7) Published from 2007 through 2012, these studies confirmed several traditional fracture risk factors in people with HIV -- older age, white race, low weight, alcohol or substance use, diabetes, and liver disease.6 Six of seven studies in the meta-analysis figured that hepatitis C virus (HCV) infection inflated fracture risk.

Researchers who ran the meta-analysis rate the 58% higher all-fracture risk and the 35% higher fragility-fracture risk "modest" surges in risk with HIV.6 A review of five cross-sectional studies (Table 1) and new longitudinal studies published since the meta-analysis (Table 2) add weight to the conclusion that HIV makes broken bones more likely.

Table 1. Cross-Sectional Studies of Fracture With Versus Without HIV
First AuthorYear(s), Site(s), Study Typen HIV+/HIV-Risk Factors HIV+/HIV-Fracture Prevalence HIV+/HIV-

Higher Fx prevalence with HIV
2007, Canada, cross-sectional case-control138/402 women matched for age and region; 100/138 HIV+ on ARTAge 37.7/38.0, BMI 25.0/26.2, Aboriginal 12.5%/2%, black 16.2%/1%; HIV+ smoked, injected drugs, used steroids moreFragility Fx prevalence 26.1%/17.3%,* OR 1.7 (1.1-2.6);* BMD similar HIV+/HIV

Higher Fx prevalence with HIV
1996-2008, Boston, population-based comparison8525/2,208,792; female 34.9%/55.8%; male 65.2%/44.1%Female:
Age 27.5%/32.4% >50; 39.3%/62.5% white, 30.6%/6.4% black

Age 36.3%/25.4% >50; 55.1%/64.3% white, 17.9%/6.1% black
Vertebral, wrist, hip fracture prevalence 2.87/1.77 per 100 persons;* overall Fx 2.49/1.72 female,* 3.08/1.83 male*

Higher Fx prevalence with HIV
1998-2010, Italy, case-control160/163 men matched for age Vertebral Fx prevalence 26.9%/12.9%;* any Fx 29.6%/12.9%* on ART, 17.1%/12.9%*

Higher Fx prevalence with HIV
2002-2009; Italy, case-control2854 HIV+/8562 general population; 37% women; all HIV+ on ARTAge 46 overallAll Fx prevalence 10.8%/0.7% <40 y,* 15.2%/0.9% 41-50 y,* 14.8%/1.3% 51-60 y;* 12.5%/2.5% >60*; higher prevalence with HIV independent of age, sex, hypertension

Higher Fx prevalence with HIV
2009-2010, London, case-control222/222 matched for age; 133/44 men, 89/178 womenFemale:
Age 44.6/45.2; BMI 27.9/25.1*; smokers 12.4%/19.7%; heavy alcohol 1.1%/1.1%

Age 46.2/46.9; BMI 24.3/26.9*; smokers 35.3%/22.7%; heavy alcohol 18%/0%* Overall:48% white, 38% black
Osteoporosis prevalence 17.6%/3.6%;* Fx prevalence 20.3%/7%,* OR 3.27

* Statistically significant.

ART, antiretroviral therapy; BMD, bone mineral density; BMI, body mass index; Fx, fracture; OR, odds ratio; py, person-years.

Table 2. Longitudinal Studies of Fracture With Versus Without HIV
First AuthorYear(s), Site(s), Study Typen HIV+/HIV-Risk Factors HIV+/HIV-Fracture Prevalence HIV+/HIV-

No higher Fx risk with HIV

(Not in meta-analysis6)
1993-2007, U.S. PACTG 219/219C, prospective cohort; 4.97 y F/U1326 HIV+/649 HIV-exposed; 51% female in both groups; all HIV+ on ARTAge 7.1/5.8 y; 11%/11% white, 62%/53% black; 24%/35% Hispanic; steroids ever 2%/1%Fx incidence 1.2/1.1 per 1000 py (NS), IRR 1.1 (0.2-5.5); no difference by BMI or steroid use

11% higher incidence with HIV

No higher Fx risk with HIV
2002-2006, NYC, CHAMPS cohort328/231 men; 87% of HIV+ with ART experienceAll 49+ years, median 55; 12%/19% white, 61%/50% black, 23%/28% Hispanic; 52%/70% overweight or obese; 61%/75% smokers; 86%/94% ever drug useFx incidence 3.1/2.6 per 100 py (NS); femoral neck and lumbar BMD significantly lower in HIV+

19% higher incidence with HIV

No higher Fx risk with HIV
2002-2008, U.S. prospective WIHS cohort, 5.4 y F/U1728/663 women; 66% HIV+ on ART at index visitAge 40.4/36.1;* post-menopause 19.6%/11.2%;* 13.3%/10.7% white, 56.3%/58.4% black, 27.2%/27.0% Hispanic; BMI 28.5/30.0;* 45.3%/50.8% smokers;* 2.1%/3.9% heavy drinkers;* HCV 25.4%/14.5%;*Fx incidence 1.8/1.4 per 100 py (NS); fragility Fx incidence 0.58/0.53 per 100 py (NS)

29% higher all-Fx incidence with HIV; 9% higher fragility Fx incidence
Lo Re16

Higher Fx risk with HCV/HIV vs no infection
1999-2005; U.S. Medicaid system; F/U 5.2 y HCV/HIV+, 2.3 y HCV+ only; 3.7 y HIV+ only, 2.2-2.7 y HCV/HIV-36,950 HCV/HIV+, 276,901 HCV+ only, 95,827 HIV+ only, 3,110,904 HCV/HIV-; women 29.3% HCV/HIV+, 46.4% HCV+ only, 36.9% HIV+ only, 29.3-46.5% HCV/HIV-Age 42 HCV/HIV+, 47 HCV+ only, 39 HIV+ only, 42-48 HCV/HIV-; 27.8% HCV/HIV+, 46.1% HCV+ only, 27.3% HIV+ only, ~39% HCV/HIV- white; respective % black 39.8%, 21.3%, 44.4%, ~18%, respective % smoker 10.1%, 11.6%, 4.6%, ~3%Hip Fx incidence 3.06 per 1000 py HCV/HIV+, 2.69 HCV+ only, 1.95 HIV+ only, 1.29 HCV/HIV-; HCV/HIV+ aHR 1.38 vs HCV+ only,* 1.76 vs HIV+ females only, 1.36 vs HIV+ males only,* 2.65 vs HCV/HIV- females,* 2.20 vs HCV/HIV- males*

Higher Fx risk with HIV
1997-2009; U.S. VACS-VC prospective cohort; F/U 6.0/6.9 y*40,115/79,203 men; 75% HIV+ with ART experienceAge at enrollment 34%/34% over 50; age at fracture 54/53;* 55%/55% black/Hispanic; BMI 25/28;* alcohol abuse 16%/15%;* smoker 61%/54%;* steroid use 5%/3%*aHR for fragility Fx 1.24 (1.11-1.39);† aHR after further adjustment for BMI 1.10 (0.97-1.25) (NS)

Higher Fx risk with HIV
2000-2008; U.S. HOPS prospective cohort compared with age- and sex-matched general-population cohort; F/U 3.8 y5826 HIV+; 79% men; 73% with ART experienceAge 40 y; 51.8% white, 33.0% black, 11.7% Hispanic; BMI 24.4Standardized all-Fx incidence per 10,000 HIV+/HIV- 57.7/29.1 in 2000; 84.8/38.1 in 2002:* 81.1/26.0 in 2004;* 83.2/35.9 in 2006;* in HIV+ (but not HIV-) incidence rose significantly from 2000 to 2008

Higher Fx risk with HIV

(Not in meta-analysis6)
2007-2009; Catalonia, Spain; population-based cohort comparison; F/U 3.0 y2489/1,115,667 40 y or older; 75.3%/47.8% male*Age 50.0/61.3;* BMI 24.5/28.4;* smoker 53.3%/18.9%;* heavy alcohol 2.7%/1.8%*Fx incidence 8.03/7.93 per 1000 py; aHR 4.7 (2.4-9.5) for hip Fx,* aHR 1.8 (1.2-2.5) for osteoporotic Fx*

Higher Fx risk with HIV or HIV/HCV
1995-2009; Denmark; population-based matched cohort comparison; F/U 6.5 y HIV+, 9.6 y population controls5306/26,530; 76%/76% male; 78% of HIV+ started ART in study periodAge 36.7/36.7; 80% HIV+ white; 16% HIV+ also HCV+Fx incidence 21.0/13.5 per 1000 py;* for all Fx IRR 1.5 (1.4-1.7) for HIV+, 1.3 (1.2-1.4) for HIV+ only, 2.9 (2.5-3.4) for HIV/HCV+, 1.6 (1.4-1.8) for low-energy Fx in HIV+ and 3.8 (3.0-4.9) HIV/HCV+ (all comparisons vs population controls)

Higher Fx risk with HIV

(Not in meta-analysis6)
Jan-Dec 2000; Denmark; nationwide case-control study124,655 Fx cases/373,962 age- and gender-matched controls without FxAge 43.4 Fx cases and no-Fx controls; 48.2% cases and controls men; Charlson comorbidity index 1-2 in 16.8% of cases and 12.8% of controls,* 3-4 in 4.4% of cases and 2.4% of controls;* alcoholism 7.1% vs 2.5%;* ever use steroid 54.3% vs 50.7%*0.40 per 1000 with Fx had HIV vs 0.14 per 1000 without Fx; OR for any Fx with vs without HIV 2.89 (1.99-4.18)*, for hip Fx OR 8.99 (1.39-58.0)*, forearm Fx OR 3.5 (1.26-9.72)*, spine Fx OR 9.00 (1.39-58.1)*

Higher Fx risk with HIV/HBV vs no infection

(Not in meta-analysis6)
1997-2007; U.S. Medicaid cohort comparison; F/U 5 y4156 Medicaid patients treated for HBV/HIV, 2053 treated with HBV only, 96,253 treated with HIV only, 746,794 randomly sampled Medicaid clientsMedicaid clientsHip Fx incidence HBV/HIV vs HIV only aHR 1.37 (1.03-1.83)*; vs HBV only aHR 2.62 (0.92-7.51, NS), vs no infection aHR 1.35 (1.03-1.84)*

* Statistically significant.

† After adjustment for demographics, comorbid disease, smoking, alcohol abuse. aHR, adjusted hazard ratio; ART, antiretroviral therapy; BMD, bone mineral density; BMI, body mass index; F/U, follow-up; HOPS, HIV Outpatient Study; IRR, incidence rate ratio; NS, not significant; OR, odds ratio; PACTG, Pediatric AIDS Clinical Trials Group; py, person-years; VACS-VC, Veterans Aging Cohort Study Virtual Cohort; WIHS, Women's Interagency HIV Study.

All five cross-sectional studies, which span 1996 through 2010, found significantly higher fracture prevalence with HIV than in comparison populations.8-12 Two of these cross-sectional studies took place in Italy and one each in Canada, the UK, and the U.S. Three focused on fragility fractures8-1011,12 The Canadian study included only women,8 one Italian study included only men,10 and the remaining three studies considered both women and men. The Canadian8 and U.S.9 studies found 60% to 70% higher fragility fracture prevalence in HIV populations than in comparison groups -- rates close to the 58% higher all-fracture incidence calculated in the meta-analysis.6

These cross-sectional analyses involve almost 12,000 people with HIV and over 2 million controls. The largest review -- a population-based comparison in Boston's Partners HealthCare System -- focused on 8525 people with HIV and 2.2 million without an HIV diagnosis (Table 1).9 Women made up one third of the HIV group, and 72.5% of them were younger than 50. Among men, 64% were under 50. Whites, who run a higher fracture risk than blacks, accounted for 39% of women with HIV and 55% of men.

Prevalence of vertebral, wrist, and hip fractures (all fragility fracture sites) stood at 2.87 per 100 persons in the HIV group, 62% higher than the 1.77 per 100 rate in the HIV-negative comparison group.9 In both women and men, fracture prevalence was higher with than without HIV in every 10-year age group analyzed, starting with 20 to 29 for men and with 30 to 39 for women, though these differences were not always statistically significant. Fracture prevalence remained higher with than without HIV when the researchers looked at three nonoverlapping periods, 1997-1999, 2001-2003, and 2005-2007.

As one would expect, fracture prevalence rose with age in the Boston study.9 But the relative difference between people with and without HIV also rose with age (see Figure 2). Among 60- to 69-year-old women and men with HIV, prevalence exceeded 5.5 per 100 people, compared with rates of 2.15 per 100 for women and 1.58 for men the same age in the general population. With more HIV-positive people surviving into their 60s and beyond, these findings suggest HIV clinicians will be spending more time helping older patients recover from debilitating fractures.

The 10 studies of fracture incidence ran from 1993 through 2009 (Table 2),13-22 and four were not in the 2013 fracture incidence meta-analysis.6 Seven studies took place in the United States, two in Denmark,20,21 and one in Spain.19 One study involved only children in the Pediatric AIDS Clinical Trials Group (PACTG),13 two involved only men in the New York CHAMPS cohort14 or the U.S. Veterans Aging Cohort Study (VACS),17 one included only women in the U.S. Women's Interagency HIV Study (WIHS),15 and the rest studied both men and women.

Of these 10 fracture incidence analyses, three did not find a significantly higher fracture rate with than without HIV, five did find a significantly higher fracture rate with HIV, and three found a higher fracture rate with HIV/HCV or HIV/HBV than without infection. Of the three studies that saw no more incident fractures with than without HIV, a 1993-2007 study involved 5- to 10-year-old U.S. children with or exposed to HIV,13 a 2002-2006 U.S. study involved only men (n = 559),14 and a 2002-2008 U.S. study involved only women (n = 2391).15 In all three of these studies, fracture incidence was higher in the HIV group (from 9% to 29% higher) but not significantly higher.

In the 1997-2009 all-male VACS analysis, only 34% of veterans with or without HIV were older than 50, and 55% were black or Hispanic (whites have a higher fracture risk).17 Median body mass index (BMI) measured 25 kg/m2 in the HIV group, significantly lower than the 28 kg/m2 in the HIV-negative group but still on the lower end of the overweight spectrum. An analysis adjusted for demographics, comorbid disease, smoking, and alcohol abuse determined that men with HIV had a 24% higher fragility fracture rate than men without HIV (Table 2). After further adjustment for BMI, veterans with HIV had a 10% higher fragility fracture rate, but now the difference from the comparison group fell short of statistical significance.

HIV Outpatient Study (HOPS) investigators measured all-fracture incidence in 5826 people with HIV from 2000 through 2008 and indirectly standardized those numbers to a general-population cohort by age and sex through 2006.18 The HIV group had a median age of 40 (interquartile range [IQR] 34-46), 79% were men, and half were white. Standardized all-fracture incidence proved higher with HIV in every year analyzed from 2000 to 2006 and significantly higher with HIV in 2001, 2002, 2003, 2004, 2005, and 2006 (Figure 1). Incidence rose significantly from 2000 to 2008 in the HIV group but not in the comparison group. The HOPS team noted that the rising fracture incidence through the years could reflect improved fracture record keeping as providers became more aware of bone problems in people with HIV, it could reflect a true jump in incidence as people lived longer with HIV, or it could reflect both factors.

Fracture Incidence With vs Without HIV
Fracture Incidence With vs Without HIV Figure 1. Standardized all-fracture incidence per 10,000 population proved consistently higher in HIV-positive men and women in the HIV Outpatient Study (HOPS) than in the general population from 2000 through 2006 and significantly higher in every year after 2000.18

Among reports not included in the meta-analysis, a nationwide case-control study in Denmark compared 124,655 people with a new fracture from January through December 2000 and 373,962 people without fractures in that period.21 With matching for age and gender, age averaged 43.4 in both cases and controls and 52% were women. People with fractures included a significantly higher proportion who abused alcohol (7.1% versus 2.5%) or ever used steroids (54.3% versus 50.7%).

In this age- and gender-matched comparison, HIV prevalence stood significantly higher among people who had a fracture during the study year than among no-fracture controls (0.04% versus 0.01%, P < 0.01).21 Odds of any fracture stood almost 3 times higher with HIV, while HIV raised chances of breaks at fragility-fracture sites even higher: 9 times at the hip, 3.5 times at the forearm, and 9 times at the spine (Table 2). All of these analyses factored in fracture history, alcoholism, use of medications affecting fracture risk, and annual income. The strength of these associations held true in men and women and in younger and middle-aged populations. The Danish team concluded their overall result "is in line with other recent publications and adds to a growing body of evidence suggesting that HIV-infected patients should be assessed for fracture risk as part of their routine care."21

Research in the general population indicates that every 1 standard deviation lower bone mineral density measure approximately doubles fracture risk.23 Together, findings from these cross-sectional and longitudinal comparisons of HIV-positive and negative people confirm that lower bone density with HIV does mean a higher fracture rate. In 2015 recommendations for evaluating and managing bone disease in people with HIV (see "When to Use FRAX and DXA -- and What They Mean" in this issue), eight experts concur that "patients with HIV infection have a higher risk of low bone mineral density and fragility fracture than the general population."24


  1. Schwartz O, Pindborg JJ, Svenningsen A. Tooth exfoliation and necrosis of the alveolar bone following trigeminal herpes zoster in HIV-infected patient. Tandlaegebladet. 1989;93:623-627.
  2. Goorney BP, Lacey H, Thurairajasingam S, Brown JD. Avascular necrosis of the hip in a man with HIV infection. Genitourin Med. 1990;66:451-452.
  3. Chevalier X, Larget-Piet B, Hernigou P, Gherardi R. Avascular necrosis of the femoral head in HIV-infected patients. J Bone Joint Surg Br. 1993;75:160.
  4. Hernández Quero J, Ortego Centeno N, Muñoz-Torres M, Martinez Pérez MA, Higuera Torres-Puchol JM.. Alterations in bone turnover in HIV-positive patients. Infection. 1993;21:220-222.
  5. Guaraldi G, Ventura P, Albuzza M, et al. Pathological fractures in AIDS patients with osteopenia and osteoporosis induced by antiretroviral therapy. AIDS. 2001;15:137-138.
  6. Shiau S, Broun EC, Arpadi SM, Yin MT. Incident fractures in HIV-infected individuals: a systematic review and meta-analysis. AIDS. 2013;27:1949-1957.
  7. American Academy of Orthopaedic Surgeons. Position statement. Recommendations for enhancing the care of patients with fragility fractures. Revised December 2009.
  8. Prior J, Burdge D, Maan E, et al. Fragility fractures and bone mineral density in HIV positive women: a case-control population-based study. Osteoporos Int. 2007;18:1345-1353.
  9. 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. 2008;93:3499-3504.
  10. Torti C, Mazziotti G, Soldini PA, et al. High prevalence of radiological vertebral fractures in HIV-infected males. Endocrine. 2012;41:512-517.
  11. Guaraldi G, Orlando G, Zona S, et al. Premature age-related comorbidities among HIV-infected persons compared with the general population. Clin Infect Dis. 2011;53:1120-1126.
  12. Peters BS, Perry M, Wierzbicki AS, et al. A cross-sectional randomised study of fracture risk in people with HIV infection in the Probono 1 study. PLoS One. 2013;8:e78048.
  13. Siberry GK, Li H, Jacobson D; Pediatric AIDS Clinical Trials Group (PACTG) 219/219C Study. Fracture risk by HIV infection status in perinatally HIV-exposed children. AIDS Res Hum Retroviruses. 2012;28:247-250.
  14. Arnsten JH, Freeman R, Howard AA, Floris-Moore M, Lo Y, Klein RS. Decreased bone mineral density and increased fracture risk in aging men with or at risk for HIV infection. AIDS. 2007;21:617-623.
  15. Yin MT, Shi Q, Hoover DR, et al. Fracture incidence in HIV-infected women: results from the Women's Interagency HIV Study. AIDS. 2010;24:2679-2686.
  16. Lo Re V 3rd, Volk J, Newcomb CW, et al. Risk of hip fracture associated with hepatitis C virus infection and hepatitis C/human immunodeficiency virus coinfection. Hepatology. 2012;56:1688-1698.
  17. Womack JA, Goulet JL, Gibert C, et al. Increased risk of fragility fractures among HIV infected compared to uninfected male veterans. PLoS One. 2011;6:e17217.
  18. Young B, Dao CN, Buchacz K, et al. Increased rates of bone fracture among HIV-infected persons in the HIV Outpatient Study (HOPS) compared with the U.S. general population, 2000-2006. Clin Infect Dis. 2011;52:1061-1068.
  19. Güerri-Fernandez R, Vestergaard P, Carbonell C, et al. HIV infection is strongly associated with hip fracture risk, independently of age, gender, and comorbidities: a population-based cohort study. J Bone Miner Res. 2013;28:1259-1263.
  20. Hansen AB, Gerstoft J, Kronborg G, et al. Incidence of low and high-energy fractures in persons with and without HIV infection: a Danish population-based cohort study. AIDS. 2012;26:285-293.
  21. Prieto-Alhambra D, Güerri-Fernández R, De Vries F, et al. HIV infection and its association with an excess risk of clinical fractures: a nationwide case-control study. J Acquir Immune Defic Syndr. 2014;66:90-95.
  22. Byrne DD, Newcomb CW, Carbonari DM, et al. Increased risk of hip fracture associated with dually treated HIV/hepatitis B virus coinfection. J Viral Hepat. 2015 Mar 6. doi 10.1111/jvh.12398. Epub ahead of print.
  23. Cauley JA, Hochberg MC, Lui LY, et al. Long-term risk of incident vertebral fractures. JAMA. 2007;298:2761-2767.
  24. Brown TT, Hoy J, Borderi M, et al. Recommendations for evaluation and management of bone disease in HIV. Clin Infect Dis. 2015;60:1242-1251.