As highly active antiretroviral therapy (HAART) for HIV infection allows patients to live longer, many are being confronted with additional health challenges related to ageing.1 Apart from the classical risk factors, there are morbidities not classically considered to be HIV related that have been associated with ongoing HIV replication, chronic immune activation, and long-term HAART.1,2
In these lines, a number of studies have reported a decrease in bone mineral density (BMD) among HIV-infected patients.1,3,4 A meta-analysis1 of 12 cross-sectional studies in HIV-infected adults found that the probabilities of osteopenia and osteoporosis were 6.4 and 3.7 times higher, respectively, in HIV-infected than in uninfected patients. However, the consequences of low bone mass on fracture risk among patients diagnosed with HIV infection are not well known, as data are still limited and in some cases contradictory.
We have therefore used the nationwide Danish health registries to explore the existing association between HIV clinical diagnosis and fracture risk using a case–control design.
Setting and Source of Data
The extensive nature of registers in Denmark covering contacts to the health sector offers good possibilities for studies on the occurrence of fractures.5 Using the unique 10-digit civil registry number that is assigned to all Danish citizens shortly after birth, a complete hospital discharge and prescription history can be established for each individual and valid linkage between population-based registries can be obtained. The unique civil registry number is used in all registers, that is, if a person buys a drug on prescription, the drug is registered as bought by this individual and the same applies for admissions to hospitals and contacts to general practitioners for reimbursement purposes. Because of the extensive nature of the registers, only a few values were missing for socioeconomic status, such as civil status, working status, and income.
This case–control study was performed within the Danish population that constituted approximately 5.3 million individuals during the study period.
The study was subject to control by the National Board of Health and the Danish Data Protection Agency.
This study was designed as a classical case–control study. Cases were all subjects, both genders and all ages, who sustained a fracture during the year 2000. Controls were matched subjects without a fracture in the same year using the criteria below. Exposure was use of drugs and diseases before the date of fracture or a matched index date in the controls. Information on fractures and diseases before the fracture was based on hospital records of in- and outpatients.
Identification of Fracture Cases
In Denmark, The National Hospital Discharge Register covers all contacts (on in- or outpatient basis) to the hospitals.5 The register was founded in 1977, but outpatient records were first completely incorporated from 1995. The files of The National Hospital Discharge Register include information on the civil registry number of the patient, date of discharge, and discharge diagnoses, assigned exclusively by the physician at discharge according to the Danish version of the International Classification of Diseases, eighth revision (ICD-8) until the end of 1993, and to the Danish version of the ICD, 10th revision (ICD-10). The register has nationwide coverage of public hospitals with an almost 100% completeness of recordings and a high precision of diagnoses,6,7 particularly for fracture diagnoses.8 Using The National Hospital Discharge Register, we identified all subjects, who had sustained a fracture between January 1, 2000, and December 31, 2000 (n = 124,655). The following end points were assessed: any clinical fracture, hip fracture (neck and pertrochanteric), distal forearm fracture, clinical spine fracture, and/or any nontraumatic fracture (any fracture not presenting with an accident mechanism code signaling a trauma of more than a fall at the same level or less as fracture energy). Based on accident codes and admission codes (eg, hospitalized from home), incident fractures were identified and separated from readmissions.
Selection of Population-Based Controls
Using the Civil Registration System, which has electronic records on all changes in vital status, including change of address and date of death for the entire Danish population since 1968, we randomly selected up to 3 controls for each case, matched by gender, year of birth, and region. The controls were selected using the incidence density sampling technique.9
Data on HIV Infection
Patients with a diagnosis of AIDS/HIV according to ICD-8 code 07983 and ICD-10: B20, B21, B22, B23, and B24, were identified from the National Hospital Discharge Register.10 Date of HIV clinical diagnosis was accounted for in time-varying models.
Using The National Hospital Discharge Register,6 we gathered information on the number of days spent in hospital the year preceding fracture (year 1999) and history of a fracture in the period 1977–2000. Similarly, data from the National Bureau of Statistics were obtained for a more accurate patient characterization, including income, social status, working status, and educational status in 1999. The National Health Organization Register information was then used to study number of contacts to general practitioners and practicing specialists for the period 1996–2000.
Information on alcoholism was collected as appearance of a diagnosis of alcoholism in the National Hospital Discharge Register or in the Psychiatric Central Register6 or a prescription of disulfiram in the Prescriptions database. Data on use of drugs with a potential effect on bone metabolism and/or fracture risk (corticosteroids, sedatives, opioids, antidepressants, anticonvulsants, and antipsychotics) were gathered from the Prescriptions database.
Data from different registers were merged at the National Bureau of Statistics, and for each subject, the 10-digit civil registry number was substituted by a unique anonymous ID.
The analyses of the association between HIV status and fractures in the year 2000 (cases versus controls) were carried out using crude and multivariable conditional logistic regression. The latter were adjusted for the following a priori–defined potential confounders: previous fracture, alcoholism, annual income in the previous year, use of corticosteroids, and sedatives. Furthermore, this logistic model for any fracture was adjusted for use of an a priori–defined list of drugs potentially involved in the causal pathway: opioids, antidepressants, anticonvulsants, and antipsychotics.
In addition, separate analyses for hip, forearm, and spine fracture cases and matched controls were also performed using these same methods. Stratified analyses for age strata (young age <40 years, middle age 40–60, and elderly >60 years) and gender were carried out, and potential interactions with these were tested for introducing multiplicative terms into the logistic model.
Finally, we studied the effect of time from HIV diagnosis on any fracture risk using a categorical variable for HIV-infected patients (up to 2 years, 2.1–4 years, 4.1–6 years, 6.1–7.5 years, and beyond 7.5 years) and fitted a smooth spline plot for visualization of this effect.
All these analyses were performed using STATA 12.0 (STATA Corp, College Station, Tex) and SPSS 19.0 (SPSS Inc, Chicago, Ill). SPSS was used to generate the datasets from raw data and check the completeness of data, whereas STATA was used for the actual statistical analyses.
No informed consent was required for this study, as we used exclusively routinely collected data.
A total of 124,655 fracture cases were identified in 2000 in Denmark, and 373,962 controls were matched on age and gender. Hence, both cases and controls had similar age [mean (SD) = 43.4 (27.4) years], and 51.8% of both populations were women. Controls had significantly higher annual income, were more likely to be married and actively working, and less likely to suffer alcoholism and to use antiepileptic drugs, sedatives/hypnotics, and corticosteroids. In addition, controls also had less comorbidities and less commonly a fracture history (Table 1).
Prevalence of HIV infection was 0.04% (n = 50) among cases and 0.01% (n = 52) among controls (P < 0.01), equivalent to an overall unadjusted (age and gender matched) odds ratio (OR) of 2.89 (1.99–4.18). Similarly, 3 out of 10,530 (0.03%) hip fracture cases had a previous HIV infection, compared with 1 out of 31,535 (<0.01%) matched controls: OR = 8.99 (1.39–58.0). The prevalence of HIV infection among forearm and spine fracture cases was 7 out of 20,035 (0.03%) and 3 out of 3364 (0.09%), whereas the proportions of HIV-infected matched controls were 6 out of 60,030 (0.01%) and 1 out of 10,079 (0.01%), respectively, equivalent to age- and gender-matched ORs of 3.50 (1.26–9.72) and 9.00 (1.39–58.1), respectively. Adjustment for fracture history, alcoholism, use of corticosteroids, use of sedatives, and annual income attenuated the observed associations for all but for overall fracture [multivariate adjusted OR = 1.99 (1.31–3.03)] (Table 2). This latter association stood for further adjustment for use of opioids, antidepressants, anticonvulsants, and antipsychotics: OR = 1.76 (1.14–2.71).
Stratified analyses by gender demonstrated similarly increased prevalence of HIV among fracture cases in men and women: 45 (0.07%) male and 5 (0.01%) female cases had a history of HIV infection, compared with 48 (0.03%) and 4 (<0.01%) controls. Estimated ORs were therefore 2.73 (1.78–4.19) for HIV-infected men and 3.75 (1.48–9.50) for women with a clinical diagnosis of HIV. Age-stratified analyses offered comparable results in the young and middle-aged populations [OR = 2.76 (1.57–4.86) and OR = 3.12 (1.80–5.41), respectively] but no significant increase in HIV prevalence among elderly fracture cases: 1 (<0.01%) case aged 60 or older had a history of HIV infection and 2 (<0.01%) controls had an equivalent HIV infection status [OR = 1.50 (0.14–16.5)].
The excess risk associated with HIV infection increased with time from diagnosis: for the HIV-infected cases for up to 2 (median 0.84) years, adjusted OR 1.20 (0.53–2.73); for 2.1–4 (median 3.11) years, OR 2.30 (1.12–4.72); for 4.1–6 (median 4.98) years, OR 2.10 (1.06–4.16); for 6.1–7.5 (median 6.64) years, OR 2.5 (1.14–5.67); and for >7.5 (median 9.62) years, OR 3.00 (1.39–6.47). According to the smooth spline produced, the effect increases rapidly in the first 2–3 years after the diagnosis of HIV infection to then continue increasing less steeply (Fig. 1).
We report a significantly increased prevalence of HIV infection of almost 3-fold among fracture cases, compared with gender- and age-matched controls from the Danish health registries. This is, in our data, independent of potential confounders, including fracture history, alcoholism, use of potentially involved medications (sedatives, opioids, antidepressants, anticonvulsants, antipsychotics, and corticosteroids), and annual income.
Even more importantly, we demonstrate an age- and gender-adjusted 9-fold higher risk of hip fractures in HIV-infected patients. Similarly, risk of clinical spine fractures seems to be 9 times higher in patients diagnosed with HIV. A lower effect size (but still significant effect) was observed for the association between forearm fractures and HIV status, with a 3.5-fold increase among the HIV-infected patients. This last association was though attenuated after multivariable adjustment for potential confounders.
The strength of the associations observed between HIV status and fractures is similar for men and women and in younger and middle-aged populations. By contrast, no such relationship was seen among patients aged 60 years or older, possibly because of a reduced number of HIV-infected cases and controls in these strata. In a study by Bonjoch et al,11 male gender was reported as an independent factor for bone loss in HIV patients. Triant et al12 analyzed fracture prevalence among HIV and non-HIV patients. Instead, we compared the prevalence of HIV in fractured patients versus matched controls. Despite this, the general conclusions of the study are similar to our findings, where HIV infection and fracture seem to be associated.
Most previous reports on the association between HIV infection and fractures studied exclusively American patients. In contrast, our data come from a Northern European population, with different representation of ethnic groups and probably different risk factors. Nonetheless, the results highlighting an association between HIV infection and fractures, even, once adjusting for different risk factor, are similar.
Finally, we report for the first time a time-varying association between HIV infection and fracture risk: the excess risk associated with HIV infection seems to increase rapidly in the first 2–3 years after diagnosis and then continues increasing more slowly. Other studies like the one by Yin et al13 have found that fracture risk is highest in the first 2 years after HAART initiation. Again, whether bone deterioration is the consequence of HIV itself or HAART remains controversial.
Despite the evidence of an association between fractures and HIV infection was inconclusive until recently, a number of studies have now consistently pointed to an increased risk in the HIV-infected patients. A just published retrospective cohort study carried out by our group reported similar results, with age- and gender-adjusted hazard ratios of 6.2 (95% confidence interval: 3.5 to 10.9; P < 0.001) and 2.7 (2.01 to 3.5; P < 0.001) for hip and clinical major fractures, respectively, among patients from Catalonia (Spain).14 Other population-based studies support these findings.12,15,16
Although the causal pathways for such associations remain obscure, different potential factors have been described in the literature. Collin et al17 demonstrated an increased incidence of all fractures in a cohort of HIV-infected men and women on HAART for 10 years, raising concerns of undesired effects of HAART on bone metabolism. In a recent study by Womack et al,18 a similar association between fractures and HIV was described, but this was no longer significant after adjustment for body mass index and comorbidities, suggesting that these might explain the increased risk of fractures in HIV populations. Smaller studies, with more restricted samples,4,19 have suggested that HIV infection is independently associated with BMD reductions in aging men and with increased fracture risk in these patients.
There is therefore a growing body of evidence that indicates that people with HIV are at high risk for osteoporosis and fractures. However, the pathogenesis responsible for this excess risk remains unclear. Many studies show that the early bone loss seen in HIV patients can be attributed to the use of certain antiretroviral therapies, such as protease inhibitors and/or tenofovir.20–22 In a substudy of 214 patients included in the SMART trial, BMD decreased further in the group receiving continuous HAART when compared with those on intermittent therapy. More recently, Bedimo et al23 have reported a cumulative effect of continuous tenofovir and lopinavir/ritonavir use on osteoporotic fracture risk. Notwithstanding this, other studies have suggested that the chronic immune activation produced by HIV infection might also play an important role in the development of osteoporotic fractures among HIV-infected populations.24,25 Various mechanisms might explain the effects of the virus on bone metabolism, including increased production of pro-inflammatory cytokines such as tumor necrosis factor alpha26 or interleukin 6,27,28 and altered vitamin D metabolism in untreated HIV patients, which lead to excessive bone resorption.29 This narrow interaction between the virus and the patient's immunological status is indeed relevant, as an association between nadir CD4 counts and fractures has been consistently shown.15,21,23 Our study has many limitations, the main one being the scarcity of detailed information on HIV infection and antiretroviral therapies. In addition, we could not investigate other potential confounders, such as BMD, smoking, or body mass index. And, HAART medications are not available in this dataset, as these drugs are not dispensed in community pharmacies but in hospital settings. However, we have used highly validated nationwide data, and our study includes a big and representative sample of patients, with information gathered during routine clinical practice. We also expect low recall bias, as HIV infection is likely to be accurately registered because of the relevant consequences that the coding of this disease has for patients' access to appropriate therapies.
In summary, our study suggests that HIV-infected adults are at highly increased risk of hip and other bone fractures, compared with the general population. This finding 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.
1. Brown TT, Qaqish RB. Antiretroviral therapy and the prevalence of osteopenia and osteoporosis: a meta-analytic review. AIDS. 2006;20:2165–2174.
2. Mills EJ, Barnighausen T, Negin J. HIV and aging—preparing for the challenges ahead. N Engl J Med. 2012;366:1270–1273.
3. Knobel H, Guelar A, Vallecillo G, et al.. Osteopenia in HIV-infected patients: is it the disease or is it the treatment? AIDS. 2001;15:807–808.
4. Arnsten JH, Freeman R, Howard AA, et al.. Decreased bone mineral density and increased fracture risk in aging men with or at risk for HIV infection. AIDS. 2007;21:617–623.
5. Frank L. Epidemiology. When an entire country is a cohort. Science. 2000;287:2398–2399.
6. Andersen TF, Madsen M, Jorgensen J, et al.. The Danish national hospital register. A valuable source of data for modern health sciences. Dan Med Bull. 1999;46:263–268.
7. NIH Consensus Development Panel on Osteoporosis Prevention, Diagnosis, and Therapy. Osteoporosis prevention, diagnosis, and therapy. JAMA. 2001;285:785–795.
8. Vestergaard P, Mosekilde L. Fracture risk in patients with celiac disease, crohn's disease, and ulcerative colitis: a nationwide follow-up study of 16,416 patients in Denmark. Am J Epidemiol. 2002;156:1–10.
9. Wacholder S, McLaughlin JK, Silverman DT, et al.. Selection of controls in case-control studies. I. Principles. Am J Epidemiol. 1992;135:1019–1028.
10. Quan H, Sundararajan V, Halfon P, et al.. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care. 2005;43:1130–1139.
11. Bonjoch A, Figueras M, Estany C, et al.; Osteoporosis Study Group. High prevalence of and progression to low bone mineral density in HIV-infected patients: a longitudinal cohort study. AIDS. 2010;24:2827–2833.
12. Triant VA, Brown TT, Lee H, et al.. 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.
13. Yin MT, Zhang CA, McMahon DJ, et al.. Higher rates of bone loss in postmenopausal HIV-infected women: a longitudinal study. J Clin Endocrinol Metab. 2012;97:554–562.
14. Guerri-Fernandez R, Vestergaard P, Carbonell C, et al.. “HIV infection is strongly associated with hip fracture risk, independently of age, gender and co-morbidities: a population-based cohort study.” J Bone Miner Res. 2013;28:1259–1263.
15. Young B, Dao CN, Buchacz K, et al.; HIV Outpatient Study (HOPS) Investigators. Increased rates of bone fracture among HIV-infected persons in the HIV outpatient study (HOPS) compared with the US general population, 2000-2006. Clin Infect Dis. 2011;52:1061–1068.
16. 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.
17. Collin F, Duval X, Le Moing V, et al.; ANRS CO8 APROCO-COPILOTE study group. Ten-year incidence and risk factors of bone fractures in a cohort of treated HIV1-infected adults. AIDS. 2009;23:1021–1024.
18. Womack JA, Goulet JL, Gibert C, et al.; Veterans Aging Cohort Study Project Team. Increased risk of fragility fractures among HIV infected compared to uninfected male veterans. PLoS One. 2011;6:e17217.
19. Arnsten JH, Freeman R, Howard AA, et al.. HIV infection and bone mineral density in middle-aged women. Clin Infect Dis. 2006;42:1014–1020.
20. Grund B, Peng G, Gibert CL, et al.; INSIGHT SMART Body Composition Substudy Group. Continuous antiretroviral therapy decreases bone mineral density. AIDS. 2009;23:1519–1529.
21. Womack JA, Goulet JL, Gibert C, et al.; Veterans Aging Cohort Study Project Team. Physiologic frailty and fragility fracture in HIV-infected male veterans. Clin Infect Dis. 2013;56:1498–1504.
22. Walker Harris V, Brown TT. Bone loss in the HIV-infected patient: evidence, clinical implications, and treatment strategies. J Infect Dis. 2012;205(suppl 3):S391–S398.
23. Bedimo R, Maalouf NM, Zhang S, et al.. Osteoporotic fracture risk associated with cumulative exposure to tenofovir and other antiretroviral agents. AIDS. 2012;26:825–831.
24. Tebas P, Powderly WG, Claxton S, et al.. Accelerated bone mineral loss in HIV-infected patients receiving potent antiretroviral therapy. AIDS. 2000;14:F63–F67.
25. Gallant JE, Staszewski S, Pozniak AL, et al.; 903 Study Group. Efficacy and safety of tenofovir DF vs stavudine in combination therapy in antiretroviral-naive patients: a 3-year randomized trial. JAMA. 2004;292:191–201.
26. Lam J, Takeshita S, Barker JE, et al.. TNF-alpha induces osteoclastogenesis by direct stimulation of macrophages exposed to permissive levels of RANK ligand. J Clin Invest. 2000;106:1481–1488.
27. Hashizume M, Hayakawa N, Mihara M. IL-6 trans-signalling directly induces RANKL on fibroblast-like synovial cells and is involved in RANKL induction by TNF-alpha and IL-17. Rheumatology (Oxford). 2008;47:1635–1640.
28. Ishimi Y, Miyaura C, Jin CH, et al.. IL-6 is produced by osteoblasts and induces bone resorption. J Immunol. 1990;145:3297–3303.
29. Gibellini D, Borderi M, De Crignis E, et al.. RANKL/OPG/TRAIL plasma levels and bone mass loss evaluation in antiretroviral naive HIV-1-positive men. J Med Virol. 2007;79:1446–1454.