Assessing fracture risk and managing low bone mass in menopausal women

Consultant Gynaecologist, Panorama Mediclinic, Department of Gynaecology, Faculty of Health Sciences, Stellenbosch University, Cape Town, South Africa

Prior to 1994, the diagnosis of osteoporosis required the presence of a clinical fracture. Clearly, a measurable risk factor for fracture was needed. The measurement of bone mass, or bone mineral density (BMD), became a reality with the advent of techniques such as dual-energy x-ray absorptiometry (DXA). Bone mineral density is a strong indicator of fracture risk, as it accounts for 70% of bone strength. In 1994, the World Health Organization (WHO) developed a definition of osteoporosis based on BMD expressed as standard deviations of the mean young adult reference value; this measurement is known as a T-score (TABLE 1).¹

World Health Organization classification of bone-density disorders

The WHO definition provided guidance but also raised questions. The most important benefit was the identification of a cutoff point (T-score ≤-2.5) for the definition of osteoporosis that could be measured by DXA before a fracture occurred. This was a notable advance, as many novel drugs were being developed for the prevention and treatment of osteoporosis, and the new wave of evidence-based medicine demanded that investigational drugs prove fracture prevention efficacy in randomized controlled trials. As a result, almost all new randomized trials with fracture end points used a T-score of -2.5, with or without fractures, as the main inclusion criteria, in order to have sufficient statistical power to prove fracture prevention efficacy.

As osteoporosis drugs passed the test of statistical significance and became available for clinical use, the use of a T-score of -2.5 logically became the intervention threshold for therapy. While this was acceptable in terms of proven drug efficacy, the WHO definition of osteoporosis based on BMD was never designed or intended to serve as an intervention threshold. It is, in fact, a statistical tool well suited to compare the incidence of osteoporosis across nations.

Further, this single risk factor for fracture became elevated to diagnostic status. It took several years, until 2001, for the first study to show that about 50% of all fractures actually occur in patients diagnosed with low bone mass, or osteopenia (T-score between -1.0 and -2.5).^2,3 This is not surprising. Although the relative risk of fracture is higher in osteoporotic patients compared with osteopenic patients, the total group of patients with low bone mass is much larger than the pool of patients with osteoporosis. Treating only patients with osteoporosis (based on a T-score of -2.5) will target less than half of preventable fractures.

Many postmenopausal women are at risk for low bone mass, osteoporosis, and bone fracture. This review discusses strategies for assessing risk for fracture and managing low bone mass in postmenopausal women.

There is consensus that BMD screening in postmenopausal women should not be generalized, but be case specific. Because age is a strong predictive factor for fracture, both the National Osteoporosis Foundation (NOF) and the International Society of Clinical Densitometry (ISCD) recommend that all women age 65 and older should be tested.^4,5 For postmenopausal women who are younger than age 65, the need for BMD assessment should be driven by the presence of risk factors, as discussed below.

Measurement of BMD

For diagnostic and therapeutic purposes, BMD should be determined by central DXA, performed according to the principles recommended by the ISCD. The lowest T-score of the spine (L1-4), total hip, or femur neck is used for diagnosis. BMD of the distal forearm can be included if either the spine or the hip value is invalid (for example, in patients with bilateral hip replacement).

Therapeutic threshold

In accordance with the therapeutic effectiveness of available drugs, it is recommended that patients be considered for bone-specific therapy if they have either:

• A T-score <-2.5 or

• A proven fragility fracture.

Patients with a T-score >-1.0 and no fractures should not be considered for bone-specific therapy.

In patients with a T-score between -1.0 and -2.5 (low bone mass), only those at sufficient risk of fracture should be treated with bone-specific drugs. The sensitivity of BMD for identifying fracture risk can be increased by lowering the BMD threshold from -2.5; however, this would decrease the specificity, and many patients who are not at risk of fracture would be treated unnecessarily and at great cost.

The sensitivity and specificity of BMD to predict fracture risk can be increased by combining the BMD value with other clinical risk factors.

A multitude of risk factors are associated with fracture (TABLE 2). A simple clinical approach for identifying the patient with low bone mass who is at significant risk of fracture is to target the patient with osteopenia plus any 2 or more additional risk factors. Although this approach is attractive in its simplicity, it is weak in that the individual risk factors vary in weight regarding their contribution to fracture risk.

Risk factors associated with fracture (not in order of strength of correlation with fracture risk)

An alternative and preferred approach is to use a model that integrates and weighs individual risk factors with BMD measurement. Various models are available, including the FRAX calculation tool, the Study of Osteoporosis Fractures (SOF) assessment index, and the QFracture algorithm. The FRAX model was developed by the WHO and is based on more than 60,000 subjects recruited from large national cohorts and the placebo arms of large randomized controlled trials. It can thus be regarded as a validated model of fracture risk prediction. The FRAX tool is available for use free online at www.shef.ac.uk/FRAX.

Risk factors used in the FRAX model include the BMD of the femoral neck or total femur (DXA measurement), age, gender, body mass index (BMI), prior fragility fracture, parental history of hip fracture, long-term (>3 months) exposure to systemic glucocorticoids, high alcohol intake (>3 units/day), smoking, rheumatoid arthritis, and other causes of secondary osteoporosis. Risk is estimated as the 10-year probability of either a hip fracture or any major osteoporotic fracture. The FRAX tool has been modified for specific populations based on the local epidemiology of fractures.

Limitations of the FRAX model include the limited choice of clinical risk factors, restriction of DXA measurement of BMD to the femur, and the absence of any risk factors for falls or mobility. Although clinicians can use any validated model, the FRAX tool is attractive because it is constantly evaluated and updated for specific nationalities.

The FRAX tool does not provide therapeutic intervention thresholds. The decision at which level to treat must be based on local guidelines, taking into account local health priorities and cost. In the United States, the National Osteoporosis Foundation used the WHO fracture prediction algorithm along with an updated economic analysis to evaluate its existing practice recommendations.⁶

The NOF determined that treatment was cost-effective when the 10-year probability of hip fracture reached 3%, and when probability for any major osteoporotic fracture reached 20%. In a study by the National Osteoporosis Guideline Group in the United Kingdom, treatment was determined to be cost-effective at all ages if the 10-year probability of any major osteoporotic fracture exceeded 7%.⁷

No patient with osteopenia or osteoporosis should be targeted with bone-specific medication before nonpharmacologic measures have been considered. Lifestyle measures should be part of the management plan; for example, patients should be advised to avoid bone-toxic substances (eg, nicotine, excessive alcohol) and drugs such as glucocorticoids. Patients should be encouraged to engage in weight-bearing exercise and to implement fall-prevention measures.

Nutritional guidance should also be provided. Dietary management involves consuming a balanced diet rich enough in protein to maintain good muscle strength and a total daily intake of elemental calcium of 1000 to 1200 mg. Preferably, dietary sources should provide these nutrients, with supplementation reserved to make up any shortfall. Ensure vitamin D supplementation of 600 to 800 IU daily to maintain serum 25-hydroxyvitamin D levels above 20 to 30 ng/mL.

As stated, therapeutic intervention with bone-specific agents for patients with osteoporosis (T-score at or below -2.5 or fragility fracture) is supported by large randomized controlled trials. This is not the case, however, for patients with low bone mass who are at risk for fracture.

Estrogen–progestin hormone therapy

The Women’s Health Initiative (WHI) randomized controlled trial was the first study to prove that estrogen–progestin hormone therapy (HT) reduces all osteoporosis-related fractures in postmenopausal women, as well as in osteopenic women; the hip fracture hazard ratio (HR) was 0.66 (95% confidence interval [CI], 0.45 to 0.98), and the combined fractures HR was 0.76 (CI, 0.69 to 0.85).^8-11 The clinical utility of estrogen–progestin HT was limited by initial conclusions of the WHI investigators that the benefit of HT was outweighed by possible risk, even in patients at risk of fracture.

These conclusions relied heavily on the global index, a controversial, nonvalidated instrument designed by the WHI investigators.¹² The global index included earliest occurrence of coronary heart disease, invasive breast cancer, stroke, pulmonary embolism, deep venous thrombosis, endometrial cancer, colorectal cancer, hip fracture, or death due to other causes.

From a bone perspective, the most important objection to the global index is the fact that it did not include x-ray assessed vertebral fractures (as was done in all the randomized controlled trials for other bone drugs). X-ray diagnosed asymptomatic vertebral fractures are the most common fractures expected in patients with low bone mass. Further, the WHI investigators, in later publications, found the benefit/risk profile to be different and generally favorable in the 50- to 60-year-old age group, or the period of less than 10 years from menopause.¹³

Many patients with low bone mass can be expected to fall into this age group and period since menopause. The International Menopause Society and other national societies, including The North American Menopause Society, have acknowledged this period as a possible window of opportunity where estrogen–progestin HT can be considered per indication.^14,15 It is generally agreed that, based on the original WHI results, estrogen–progestin HT should not be initiated after the age of 60 years, but hormone continuation after the age of 60 years should be individualized.

In 2003, the FDA stated that if estrogen– progestin HT was considered for the sole purpose of preventing fractures, other available therapies should also be considered. Unfortunately, when based on the principle of evidence-based medicine, the alternative choices are limited, as will be seen in the following discussion.

Strontium ranelate

According to data derived from predetermined end points in 2 large randomized controlled trials, compared with placebo, strontium ranelate significantly reduced the risk of spine and hip fractures in patients with osteoporosis and osteopenia.^16,17 It is

postulated that strontium ranelate inhibits bone resorption and stimulates bone formation. Such a mode of action may explain why strontium ranelate works well in low bone mass, a condition in which bone turnover may be increased to a lesser extent compared with that in osteoporosis.

Although available in Europe and many other countries, strontium ranelate is not available in the United States.

Raloxifene

Raloxifene, a selective estrogen receptor modulator (SERM), is effective in reducing the risk of vertebral fractures (not hip fractures) in patients with osteoporosis, according to the conclusions of the Multiple Outcomes of Raloxifene Evaluation (MORE) randomized controlled trial.¹⁸ A reanalysis of the MORE data further showed that raloxifene was also effective in significantly reducing the risk of vertebral fractures in subjects with osteopenia.¹⁹ Since this was not a predetermined end point of the MORE trial, the evidence should be regarded as weak.

Raloxifene has an added advantage of offering protection against estrogen receptor–positive invasive breast cancer, but it increases the risk of venous thrombosis to the same extent as estrogen–progestin HT. Raloxifene increases hot flushes.

Risedronate

Randomized controlled trials have shown that risedronate, a bisphosphonate, prevents fractures in women with osteoporosis. In elderly postmenopausal women with osteopenia, however, risedronate has been shown to be ineffective in protecting against hip fracture in women with a BMD T-score above -2.5.²⁰

Although a post hoc subgroup analysis of pooled data from 4 different trials concluded that risedronate provided vertebral fracture protection in women with hip osteopenia,²¹ the evidence for antifracture efficacy of risedronate in women with low bone mass remains weak.

Alendronate

Another bisphosphonate, and a potent inhibitor of bone resorption, alendronate was very effective in preventing spinal fractures in postmenopausal women with osteoporosis, according to the results of the Fracture Intervention Trial (FIT 1).²² In FIT 2, a randomized controlled trial that included patients with low bone mass without vertebral fractures, efficacy in fracture prevention at the spine and hip could be shown only in patients who had a T-score of -2.5.²³ There is no evidence of efficacy when alendronate is used in patients with low bone mass.

Other drugs

Ibandronate, zoledronate, parathyroid hormone, and denosumab all have proven anti-fracture efficacy in large randomized controlled trials of women with osteoporosis. However, none of these agents have been studied specifically in women with low bone mass.

Treating only women with osteoporosis based on a T-score of -2.5 will target less than half of preventable fractures. Women with low bone mass who are at sufficient risk of fracture to warrant treatment should be identified by applying clinical risk factors in an integrated validated model of fracture prediction, such as the FRAX tool.

Treatment thresholds should be determined at a national level to ensure cost effectiveness based on local prevalence of fractures and the cost of medication.

As discussed, most drugs used to prevent fractures in women with osteoporosis lack data of efficacy in women with low bone mass. For women with low bone mass, the best evidence of drug efficacy is available for estrogen–progestin HT and strontium ranelate; weaker evidence exists for raloxifene and risedronate. Alendronate has been shown to not be effective in women with low bone mass.

Sexuality, Reproduction & Menopause ©2012 Quadrant HealthCom Inc.