Given the elevated risk of osteoporosis in people living with HIV -- especially in men over 50 and postmenopausal women -- "clinicians need to be aware of optimal screening and management of bone disease," according to Todd T. Brown, M.D., Ph.D., of Johns Hopkins University, who gave a review of the subject at IDWeek 2014 in Philadelphia.
Multiple studies show the risk of bone disease to be quite high. In a 2006 meta-analysis, the overall prevalence of osteoporosis in people living with HIV was 15%. The incidence of osteoporosis and osteopenia is likely to increase over time and as people living with HIV age, and could lead to an increased risk of fractures. For instance, a study looking at data from a large U.S. health care system (the Partners HealthCare System) from 1996-2008 found a significantly higher fracture prevalence in people living with HIV than in HIV-negative patients -- both in men and women -- with fracture risk becoming more pronounced with increasing age.
The pathogenesis of reduced bone loss in people living with HIV involves host factors, HIV disease itself and antiretroviral therapy.
Host-related factors that are common among people living with HIV include: low body weight, smoking, alcohol use, opiate use, hepatitis C (HCV) coinfection, physical inactivity, hypogonadism and low vitamin D levels.
HIV disease could also play a role in bone loss, with inflammation and viral proteins driving bone resorption on one hand, and reducing bone formation on the other.
Moreover, the initiation of antiretroviral therapy in general and toxicity related to specific drugs have been shown to be associated with a clinically significant loss in bone mineral density (BMD). At least a dozen trials using a variety of different antiretroviral therapy regimens have reported about a 2%-6% BMD loss 48-96 weeks following antiretroviral therapy initiation.
The mechanism for this is unclear, but some data suggest pre-antiretroviral-therapy immune deficiency or immune restoration disease may be to blame. A recent analysis on antiretroviral therapy initiation in treatment-naive patients found that having a lower CD4 count prior to antiretroviral therapy initiation was associated with greater decreases in BMD. Having a greater degree of immune reconstitution (rise in CD4 count) after starting antiretroviral therapy was not associated with BMD loss.
There are also data showing that some antiretroviral drugs -- namely the protease inhibitors (PIs) and tenofovir (Viread) -- have independent detrimental effects on BMD.
PI use has been associated with an increased risk of bone disease. For instance, ACTG 5224s found that there was greater BMD loss in the lumbar spine after initiation of antiretroviral therapy with a ritonavir (Norvir)-boosted atazanavir (Reyataz) than with efavirenz (Sustiva, Stocrin), though the effects on total hip BMD were similar.
In ACTG 5260s, there was a greater loss in both lumbar spine and total hip BMD on boosted atazanavir and boosted darunavir (Prezista) versus raltegravir (Isentress) -- there was no difference between atazanavir and darunavir.
A study in the Veterans Affairs (VA) Clinical Case Registry found an increased risk of osteoporotic fractures with cumulative exposure to PI-based regimens.
Finally, the SPIRAL study found that switching from a PI-containing regimen to one based on raltegravir modestly increased BMD.
Similarly, tenofovir has been linked to HIV-associated bone disease. Several studies have found that after antiretroviral therapy initiation, there appears to be a 1%-2% greater loss at the lumbar spine and hip bone on a tenofovir-containing regimen than on regimens containing other nucleoside reverse transcriptase inhibitors (NRTIs).
The VA study found significantly greater risk of osteoporotic fractures with tenofovir-containing regimens than PI-containing regimens.
Finally, two studies presented at CROI in recent years reported improvements in those with low BMD on tenofovir-based regimens in the femoral neck when switched to abacavir (Ziagen), and in both the total hip and lumbar spine when switched to raltegravir.
But when should a clinician screen for osteoporosis? According to the 2013 U.S. National Osteoporosis Foundation (NOF) guidelines, dual-energy X-ray absorptiometry (DXA) screening should be performed on: anyone with a history of fragility fracture, women over 65, men over 70, and postmenopausal women and men between 50 and 70 if there is concern based on risk factors (including HIV).
The World Health Organization (WHO) suggests the following functional definitions for bone disease as measured by DXA:
- Osteoporosis: T-score < -2.5.
- Osteopenia: T-score = -1.0 to -2.5.
- Normal: T-score > -1.0.
- A Z-score (< -2.0) is used in men below 50 years of age and premenopausal women.
For each standard deviation (SD) decrease in BMD, the risk of fracture increases 1.5-3.0 fold. However, BMD explains only about 50% of fracture risk.
According to the 2013 NOF guidelines, the following patients should be treated (applies to postmenopausal women and men over 50):
- Those with hip and vertebral fractures.
- Those with BMD T-scores ≤ -2.5 at the femoral neck, total hip or spine.
- Those with T-scores between -1 and -2.5 at above sites and 10-year hip fracture probability ≥ 3% or 10-year major osteoporosis-related fracture probability ≥ 20% based on the FRAX (WHO's fracture assessment tool) model.
The FRAX tool is available online or as an app and, based upon some basic history and DXA data, calculates 10-year fracture risk.
But clinicians should remember and check for secondary causes of low BMD in their patients, including vitamin D deficiency, hyperparathyroidism, hypogonadism and phosphate wasting.
Vitamin D deficiency and phosphate wasting may lead to osteomalacia -- bone softening resulting from impaired bone mineralization. Osteomalacia is the most important differential diagnosis for low BMD and may be accompanied by weakness, fracture, pain, anorexia and weight loss. It is treated with vitamin D and calcium with or without phosphate -- but bisphosphonates are not indicated.
Calcium and vitamin D supplementation is a general recommendation for treating bone disease. However, at CROI 2014, a randomized study reported that high-dose vitamin D3 (4,000 IU daily) plus calcium supplementation (1,000-mg calcium carbonate daily) attenuated the loss of BMD seen in people living with HIV after the initiation of antiretroviral therapy, particularly at the hip, which is the site of greatest concern for fragility fracture.
Additionally, patients should also be advised to stop smoking, reduce their alcohol intake, and to start a program of weight bearing exercise, according to Brown.
Clinicians should assess the patient's fall risk, and ask the patient whether they are worried about falling. If they are at risk, they should be encouraged to start strength and balance training (the NOF site has some examples of exercises for patients).
In persons at increased fracture risk, Brown thinks clinicians should consider avoiding initiating antiretroviral therapy with a PI- or tenofovir-containing regimen and switching off of those regimens in those who develop bone weakness on antiretroviral therapy.
Otherwise, treatment guidelines with bone specific medications should follow those established for the general population, including bisphosphonates, selective estrogen receptor modulators and PTH analogues.
Brown noted that there are some important differences in cost, effectiveness, ease of adherence and side effect profile to consider when choosing which bisphosphonate is right for which patient. Since bisphosphonates are incorporated into the skeleton, and continue to have an antiresorptive effect for a period after dosing, it may be possible to reduce the risk of some of the potential side effects (including osteonecrosis of the jaw, atypical femur fractures, atrial fibrillation and esophageal cancer) by taking a bisphosphonate drug holiday (a recent paper provides some recommendations for drug holidays).
Theo Smart is an HIV activist and medical writer with more than 20 years of experience. You can follow him on Twitter @theosmart.