Russell

The main problem with bone as we get older, as I understand it, is really osteopenia, not only from classical calcium deficiency but also from problems with vitamin D deficiency and metabolism. Is it at all useful, Dr. Dawson-Huges, to think about osteopenia of the elderly in terms of a mixed medical problem of calcium and vitamin D deficiencies?

Dawson-Hughes

Traditionally, our training and thinking had been that vitamin D deficiency causes osteomalacia, softening of the bones in adults, and that a variety of other situations lead to osteoporosis or bone thinning. It has been much more widely recognized in the last few years that vitamin D insufficiency, that is, not extreme deficiency but an inadequate supply of vitamin D contributes to the osteoporosis problem.

An inadequate intake and inadequate skin synthesis of vitamin D, leading to low-normal or frankly low levels of vitamin D, is associated with decreased calcium absorption. An inadequate amount of absorbed calcium results in a subtle lowering of the ionized calcium concentration in blood. This triggers parathyroid hormone secretion and raises blood PTH levels. Parathyroid hormone is one of the most potent bone resorbing agents known, so even a subtle increase in PTH causes an increase in bone resorption, which results in measurable bone loss. A high PTH, by stimulating bone loss, increases the risk of fracture. Recent evidence indicates that the high resorption and bone turnover rates associated with higher PTH levels (independent of bone density effects) also increase fracture risk. So, in summary, bone loss and high turnover rates that result from inadequate calcium and/or vitamin D intakes contribute to osteoporotic fracture risk.

Russell

There also used to be a classification of type 1 and type 2 osteoporosis, with type 1 being the perimenopausal type and type 2 being the age-related type. Is that a useful classification to still be using?

Menopause

Dawson-Hughes

It is useful to recognize that there are two things going on after the menopause. In the first five to eight years after menopause, there is very rapid bone loss that is directly related to the loss of the gonadal hormone, estrogen. In addition to estrogen-deficiency related bone loss, there is an underlying age-related loss of bone mass. The age-related decline occurs at about 1% of the skeletal mass per year in postmenopausal women. One can use the type 1 and 2 classification but it is somewhat more descriptive to use estrogen-deficiency-related bone loss and age-related bone loss.

Russell

In the United States, exactly what is the extent of the clinical problem both in terms of spinal fractures and hip fractures? How much morbidity and cost of the medical care system are we talking about?

Dawson-Hughes

There are estimated to be about 1.3 million new osteoporotic fractures per year. Of these, about 1/2 million are vertebral fractures and about 1/4 million are hip fractures, with the remainder being fractures of the forearm and all other sites combined. A fifty-year-old Caucasian woman has a lifetime risk of hip fracture of about 14%. The economics of this problem are related to the cost of hip fractures, the acute hospital care, the rehabilitation costs, as well as to the loss of functional status of the hip fracture patient. It is estimated that $13.8 billion dollars were spent in 1995 for osteoporosis related fractures. This estimate is expected to rise dramatically over the next 50 years with the increase of the at risk population.

Russell

The clinical presentation of patients with osteoporosis is oftentimes acute back pain but this seems like a very late stage of the disease. Also, there are many new ways of measuring bone mineralization. Could you explain these different methods and give us some indication of how and when they should be used?

Diagnosis

Dawson-Hughes

With recent advances in technology, there has been a shift in the diagnosis of osteoporosis. In the past, the diagnosis was made at the time of fracture. Today, with the widespread availability of precise measurement devices, such as dual-energy x-ray absorptiometry, we now rely on indirect assessments of bone mineral density for the diagnosis. Typically measured sites are spine and hip. Less precise, but less expensive peripheral scanners, have recently become widely used. These absorptiometry and ultrasound devices measure other skeletal sites such as the heel, forearm, and finger.

There is a strong inverse association between spine and hip bone density and risk of fracture. For example, in the very large Study of Osteoporotic Fractures, an observational study of almost 10,000 women aged 65 and older, a one standard deviation loss in bone mineral density (equivalent to about 10% loss) from the hip was associated with a 2.5 fold increased risk of hip fracture.(1) It is because of this linkage that one can use bone mineral density measures as a surrogate for fracture outcome. The ability of the newer peripheral measurement devices to predict fractures is under investigation.

Russell

Can you describe the World Health Organization criteria for diagnosing osteoporosis?

Dawson-Hughes

In 1994, the World Health Organization put forward the definitions shown in Table 1.(2)

* T-score = Standard deviation of the mean for young adult white women.

From: Kanis 1994.(2)

The T-Score used refers to the number of standard deviations from the mean value of a young Caucasian reference population. For example, a woman with a T-score of -1 has bone mineral density that is one standard deviation or roughly 10% below the mean value for healthy young adult females. Using T-scores in the definition, the World Health Organization defined normal as a T-Score of -1 or above, osteopenia as a T-Score of -2.5 to -1 and osteoporosis as a T-Score of -2.5 and below. With this set of definitions, the prevalence of osteoporosis and osteopenia in the United States, based on the HANES III data, are shown in Table 2.(3)

Another 38% of women have osteopenia.

From: Looker, 1995.(3)

Who's at Risk?

Russell

Dr. Dawson-Hughes, who should be screened for osteoporosis and how? Should the entire Caucasian female population in the United States be screened, for example, by using these techniques?

Dawson-Hughes

There have been a number of positions taken on the subject of who should be evaluated and treated for osteoporosis. Most of these I would consider to be "expert opinion"guidelines. Very recently, a set of evidence-based guidelines was published by the National Osteoporosis Foundation (NOF).(4) A physician's guide that was derived from this evidence-based analysis addresses the subject in post-menopausal Caucasian females.(5) A summary of the key findings is shown in Table 3.

Specifically, post-menopausal women aged 50 to 65 who have one or more osteoporosis risk factors (from the list shown in Table 4) should have a bone density test.

Note that poor health and frailty, which may or may not be modifiable, appear under both headings. The four italicized items -- personal or family history of fracture, smoking, and low body weight -- were demonstrated in a large, ongoing, prospective US study to be key factors in determining the risk of hip fracture (independent of bone density).

From: National Osteoporosis Foundation Physicians Guide to Prevention and Treatment of Osteoporosis, 1998.(4)

If the T-Score is -1.5 or below, the woman should be treated. All women age 65 and older should have a risk factor assessment and a bone density test. Their need for a bone density test is not dependent upon the presence of risk factors but the decision to treat is somewhat contingent upon risk factor presence. Specifically, women with a T-Score of -1.5 or below and one or more risk factors should be treated. Women who have a T-Score of -2.0 and below should be treated, whether or not they have a risk factor(s) shown in Table 4. The NOF Guide also indicates that all women who have had a vertebral or hip fracture in the past or any fracture after age 40 should be treated. Most physicians will want to have a bone density test at baseline in order to be able to follow the effectiveness of treatment.

Russell

These guidelines that you have been giving us have been for Caucasian females since this is the group in the population that has the most problems with fractures due to osteoporosis. Have there been guidelines developed for women of different racial groups or, indeed, have there been guidelines developed for screening men?

Dawson-Hughes

The NOF Guidelines, as indicated earlier, are evidence based. It turns out that the evidence is restricted almost entirely to Caucasian women. This speaks to the need for studies in African-American, Hispanic, Asian and other women and to the entire population of men. Currently, we have no evidence of the effectiveness of treatment in these populations, regrettably.

Treatment

Russell

There is a plethora of drugs and nutrients that have been used in the treatment of osteoporosis, including calcium, vitamin D, estrogens, calcitonin and biphosphonates. Can you give us a practical approach -- which drugs and modalities of treatment should we use?

Dawson-Hughes

Again, I will refer to the NOF guidelines which have recently addressed the evidence of effectiveness of different treatments.(4) First, I would use only drugs that have been approved by the FDA for either treatment or prevention of osteoporosis. Drugs currently approved for treatment include hormone replacement therapy (HRT), alendronate and calcitonin. The NOF analysis takes into account the anti-fracture efficacy and cost of these drugs and ranks them as follows: hormone replacement therapy is the most cost effective, alendronate is second most cost effective and calcitonin is the least cost effective. For women who are eligible for hormone replacement therapy treatment, this would be the treatment of choice. For those who are either ineligible or do not tolerate (HRT) treatment, alendronate would be the next consideration. We have the least fracture data on calcitonin. We know it is effective in reducing bone mineral density losses from the spine but not as effective as the other agents at the hip, so it would be the last choice.

There is another drug that has received FDA approval for the prevention indication but its review for treatment is pending. This is the first of the selective estrogen receptor modulators (SERMs), raloxifene. This drug increases bone mineral density modestly, that is, to a lesser degree than does hormone replacement therapy. Preliminary evidence from ongoing trials indicates that it may lower vertebral fracture rates by about 40% but this is not yet established. The NOF guidelines recommends calcium and vitamin D as cost effective agents for all post-menopausal American women.(4) Currently, intakes of these nutrients in post-menopausal women are about 50% of the amount recommended. An interesting recent meta-analysis suggested that higher calcium intakes (1200 vs. 600 mg/day) improved the bone density responses of post-menopausal women to HRT.(6)

Russell

What is your approach for a woman who has a family history in a first degree relative with breast cancer? Do you avoid estrogens in that case or do you use estrogens but recommend frequent breast examination and yearly mammography? What is the best approach for this very large group of patients?

Dawson-Hughes

Along with many other physicians, I tend to hold back in the use of HRT in women who have a primary relative with breast cancer. On the other hand, there is a certain point, if one is unable to control frequent fractures in a patient with other drugs, at which this decision must be reconsidered. All women on HRT should have annual mammograms and frequent breast exams. In women on unopposed estrogen, which is not recommended in very many circumstances, uterine ultrasound exams and/or uterine endometrial biopsies should be performed, at least annually. More commonly, in the United States, estrogen is used with progesterone. This lowers the risk of endometrial cancer that occurs with the use of estrogen alone.

There is another condition that has recently been associated with HRT use. Approximately a four-fold increase in the incidence of phlebitis and thromboembolism occurs among estrogen users.(7) Because of the rarity of the event, this increased risk would result in one to two new cases for every 10-15,000 women treated with HRT. Estrogen has long been known to prevent heart disease in post-menopausal women. It is worth noting here the final results of the recently published HERS study.(8) This trial, in women aged 65 and older who had a personal history of heart disease, reported new heart disease events on HRT and placebo. After three years of treatment, the rates of new cardiac events were equal in the two treatment groups. This report has raised considerable concern about the value of HRT in lowering risk of cardiovascular disease events in women with pre-existing heart disease and further study of this important issue is warranted.

Russell

Biphosphonates have also had been reported to have some important and unwanted side effects. Could you elaborate on this and the frequency of these?

Dawson-Hughes

Bisphosphonates are pyrophosphate analogues. Those in which one of the side chains contains an amino group tend to cause more frequent side effects. Alendronate is an amino bisphosphonate that, in clinical trials, did not cause much, if any, side effects. However, in clinical practice, many have observed problems with esophageal irritation and gastritis-like symptoms. There have been isolated reports of esophageal perforation. Currently, we have no clear knowledge about the frequency of this problem in a clinical practice setting. In the randomized alendronate trials, women with a history of significant upper GI disease were excluded from participation and this is thought to account for the different profiles seen in the trials compared with the anecdotal reports among practicing physicians. The issue of side effects should not overshadow the fact that this highly effective drug is well tolerated by the vast majority of patients.

Etidronate, which has long been used in the treatment of Paget's disease (and off label for osteoporosis), is not an amino bisphosphonate and does not cause significant side effects. Many other bisphosphonates are currently under development and the specific side effect profiles of these drugs will be very important in decisions about their use for the treatment of osteoporosis.

Osteoporosis Among Men

Russell

What can you tell us about the general treatment of older men with osteoporosis?

Dawson-Hughes

We are seeing an increasing number of men referred for evaluation and treatment of unexpected fractures. The evaluation of men is similar to that of women. One looks for secondary causes. A common secondary cause in men is a renal leak of calcium. This can be detected by measurement of calcium content in a 24-hr urine specimen. Prior to the collection, I usually place the patient on the recommended intake of 1200 mg of calcium per day. Another common cause of osteopenia in men is hypogonadism, which can be readily assessed by measurement of a free testosterone level. However, prior to treatment with androgens, men need to be evaluated for prostate hypertrophy, for the condition can be aggravated by testosterone replacement. A positive family history of osteoporosis is also common in men with osteoporosis.

Prevention

Russell

I would like to turn, lastly, to the issue of prevention of osteoporosis, since so many dollars could be saved if we prevent fractures from occurring in the first place. What are thought to be the most effective preventive measures at this time?

Dawson-Hughes:

A number of randomized controlled trials have shown efficacy of calcium and/or vitamin D in reducing bone loss and, in several studies, fracture incidence in older individuals.(9),(10) According to the USD Continuing Survey in 1994, approximately half of women aged 50 and older and maybe 40% of men in the United States are consuming less calcium than is recommended. Similar values are likely to be true for vitamin D, although the quality of the evidence is lower for vitamin D intake. Exercise is important and weight-bearing exercises in the upright position appear to be the most effective in preserving the skeleton.

Examples of beneficial exercise include walking, running, weight lifting and any sport carried out in the full upright position. Less effective would be bicycling and still less effective for the skeleton, partial weight-bearing exercises like swimming. The usual suspects, smoking and excess alcohol use, are as bad for the skeleton as they are for many other body systems. The National Academy of Sciences recently assessed evidence on optimal calcium and vitamin D intakes and increased the recommended calcium intake from 800 to 1200 mg/day for men and women over age 50.(11) Vitamin D intake recommendations had formerly been set at 200 IU. It remains at 200 for individuals up to age 50 but rose to 400 IU/per day for men and women ages 51-70 and to 600 IU/day for men and women over age 71 (Table 5).

From: National Academy of Sciences, Dietary Reference Intakes, 1997.(11)

The Academy recommended food sources of calcium and vitamin D to the greatest extent possible.

But many will not achieve these intake levels without some supplementation. There are a variety of effective calcium supplements. Calcium should be taken in doses of 500 mg or less for optimal absorption. If one chooses calcium carbonate, then this supplement is more consistently absorbed when taken just after meals. With the other supplements, there is no known benefit from their use at mealtime.

Russell

Thanks for this discussion.