Boning Up on Bone Drugs

Can a drug intended to prevent fractures cause them? The scoop on osteoporosis meds.

by Katharine Davis Fishman
Photograph: Photo: Coolife and Carole & Pauline

Late last fall I started walking like a penguin. The array of specialists I waddled in to see—the anesthesiologist, the physiatrist (a doctor who supervises physical therapy and rehabilitation), the rheumatologist, the sports-med doctor and finally the back surgeon—were baffled by my gait and the increasingly intense throbbing in my left buttock. The back doctor called in the hip doctor down the hall to see what he made of it. Nothing. So the back doctor made an appointment for me to see a neurologist.

At 11 o’clock the night of the back- doctor visit, I was heading for the closet to hang up my jacket when I tripped on a rug. As I slid down the wall, my upper thigh shot out to a 45-degree angle, and I felt an excruciating pain. “Joe, call 911!” I shouted to my husband.

My femur, or thigh bone, had fractured. Doctors implanted a titanium rod and two screws, and I spent three weeks, including rehab, at the nearest regional trauma center.

Two months after the accident, when I showed up (still using a walker) for follow-up care at the Hospital for Special Surgery in New York City, I learned that the fracture had probably been caused by bisphosphonates. Those were the what-a-nuisance drugs that had me getting up early, swigging down a little pill with a big mug of water and—forbidden to eat for the next 30 to 60 minutes—enviously watching my husband enjoy his muffin, all in the interest of avoiding . . . hip fractures! After an osteoporosis diagnosis, I’d swallowed Fosamax for nearly 10 years, stopped for a year after I developed an ulcer, then spent three years on Boniva, the Sally Field drug. All this pill taking helped; I moved from osteoporosis to osteopenia, a milder condition. And yet a silly at-home accident had just broken my femur.

It wasn’t as if I’d been using some kind of risky, experimental medication. In 2004 pharmacists filled more than 18 million prescriptions for Fosamax, manufactured by Merck. The first bisphosphonate available, it has been around since 1995 and was the best seller in its class until 2008, when its patent expired and generic equivalents became available. The potential market for Fosamax and its sister drugs—Actonel, Didronel, Boniva, Aredia, Reclast, Zometa and Skelid—is huge: 10 million Americans have osteoporosis, and an additional 34 million are estimated to have its precursor, low bone mass, also called osteopenia.

Bisphosphonates are known to produce side effects, and a few of those listed in the Fosamax package information for patients are “irritation, inflammation or ulcers of the esophagus, which may sometimes bleed. . . . Fosamax may cause jawbone problems in some people [which] may include infection and delayed healing after teeth are pulled. . . . Call your doctor if you develop severe bone, muscle or joint pain. . . . Some patients have experienced fracture in a specific part of the thigh bone. Call your doctor if you develop new or unusual pain in the hip or thigh.” That last warning refers to the kind of bone break I had just experienced: extremely rare (or, more formally, atypical) femur fractures, incurred when the patient is doing nothing very demanding.

So why are these drugs universally recommended (albeit with qualifications) by some of the same doctors who identified the side effects? A few months after I had thrown aside both the walker and the cane that followed it, I put on my rumpled trench coat and set off to seek some answers.
Mixed Messages
The first thing I learned was that bisphosphonates are specifically intended to protect against something more serious than what I’d had: hip fractures. My break, on the shaft of the thigh, would eventually heal. But what we commonly call a hip fracture is much worse. It occurs either at the two knobs known as trochanters that sit on the top of each femur or at the neck between them and has dire consequences. Some 24 percent of elderly women and men who fracture their hips die within a year, another quarter end up incapacitated in nursing homes, and most of the rest never walk quite right again.

Both psychological and physiological factors account for this rapid decline and high death rate. Ethel Siris, MD, director of the Toni Stabile Osteoporosis Center at Columbia University Medical Center and a past president of the National Osteoporosis Foundation, explained to me that many formerly active patients in their seventies and eighties become depressed after undergoing surgery, spending months in a rehab facility and then finding they still can’t get around.

Very low bone density, or osteoporosis, is not the only factor in fracture risk, but it is the most important and the easiest to measure. Osteoporosis most often becomes a concern around menopause. When you’re younger, your bones stay healthy through a mineral process called turnover, in which breakdown cells (osteoclasts) remove the old, calcium-deficient bone, and buildup cells (osteoblasts) replace it with new, calcium-rich bone. In the years just before and after menopause, the breakdown osteoclast cells become more active, and the buildup osteoblasts grow sluggish, partly because estrogen, now being rapidly lost, is a key element in the turnover process. The result is bone that is less dense and more fragile and at its laciest is labeled osteoporotic. Only 4 percent of women in their fifties suffer from osteoporosis, while 52 percent of women over 80 meet criteria for the disease.

Before the Food and Drug Administration approved Fosamax in 1995, the only drug available for treating osteoporosis was estrogen, prescribed to prevent bone loss and thereby fractures. Fosamax, like the bisphosphonates developed later, blocks the work of the breakdown osteoclast cells but does not stimulate the buildup osteoblasts, so it slows decline but cannot rebuild bone.

In 1997 the FDA allowed for a broader use of Fosamax, as an osteoporosis preventive. And while the agency says it conducted its own analysis of clinical data before approving that use of the drug, in September 1998 an agency statistician re—evaluated the numbers Merck had submitted. Merck had pooled data from two studies, with different time spans, populations and goals, and the statistician raised the possibility that this may have distorted the results. He also suggested that the drug’s effectiveness varied over time with the test populations’ medical histories: it worked best with patients who had either full-fledged osteoporosis or prior fractures, a finding that called into question how useful the drug would be at preventing fractures among women who simply had low bone mass and had never broken their bones.

Still, once the FDA approval was granted, the drugs, which had been given primarily to people with osteoporosis, could now be dispensed to a much larger group on the basis of bone scans produced by a machine called a DXA (or sometimes DEXA). The scan, which came into wide use in the late 1990s, generates a “T-score” that rates a patient’s bone-mineral density relative to that of healthy women at age 30, when bone mass peaks. A T-score below –2.5 equals osteoporosis; a score of –1 to –2.5 is low enough to make the patient a candidate for preventive treatment.

This ability to measure below—normal bone density fed into the grand old American notion of proactivity, of nipping a disease in the bud before it could get troublesome. Soon many doctors were prescribing bisphosphonates to patients in their midfifties in hopes of warding off osteoporosis and the grim consequences of hip fractures. The change also fed the drug manufacturers’ bottom lines. When pharmaceutical companies add osteopenic patients to those with osteoporosis, the market more than triples.

Danger Signs
In 2000 more people started taking the drugs, and reports of side effects began to dribble out, most notably indigestion, a problem solved for many when Merck changed the regimen so that patients took a pill weekly rather than daily (the less often the pills are swallowed, the less they irritate your gullet, and the less vulnerable you are to indigestion). Other emerging side effects were more alarming. Around 2004, Joseph Lane, an orthopedist who’s chief of the Metabolic Bone Disease Service at the Hospital for Special Surgery, began to notice strange events among some patients who’d been taking bisphosphonates for about six years. Two women stand out particularly in Lane’s memory. “One had been complaining of thigh pain for three months,” he remembers. “She’d had two epidural injections for back pain, and while she was in a swimming pool she broke her femur, simply by turning around. Number two is a woman who was getting on a plane to go visit her grandchildren. Similar story: She had earlier complained of sciatica, but her doctors didn’t take an X-ray. Instead, they gave her an MRI and an epidural injection. Then, the day of her flight, she climbed the stairs during boarding and broke her femur going up.”

In 2005, Lane read an article in the Journal of Clinical Endocrinology and Metabolism that jibed with what he’d been seeing. A team at the University of Texas in Dallas and at Henry Ford Hospital in Detroit had biopsied nine patients with osteoporosis or osteopenia, most of whom had been taking alendronate—Fosamax’s generic name—for three to eight years and had “spontaneous nonspinal fractures” in odd places that took an unusually long time to heal. Bone biopsies showed “minimal, or no, identifiable osteoblasts” (the buildup cells) and low breakdown activity—a syndrome that became known in lay language as frozen bone.

This was a small study with no control group, and some patients were taking drugs besides alendronate. But the authors called for more research “to determine how long bisphosphonates can safely be given.” Lane and four colleagues within the New York–Presbyterian/Weill Cornell Medical Center complex reviewed the records of all patients admitted to its trauma center from 2002 to 2007 with femur-shaft fractures that were “low energy,” meaning they had been incurred while the person was simply standing around (or splashing in a swimming pool).

The 70 trauma center patients whose cases they reviewed had all suffered from what’s technically known as an atypical subtrochanteric femur -fracture—and more than a third had been taking Fosamax, the bisphosphonate that’s been available the longest. Three quarters of these Fosamax patients shared a particular radiographic pattern: a simple horizontal or diagonal fracture with a sort of beaky overhang of bone. Their X-rays looked exactly like the one I was presented with two years later. The pattern was 98 percent specific to Fosamax users, and those who displayed it had been using the drug significantly longer than those whose breaks did not look like that. A follow-up study matching a smaller group of patients who had femur fractures with a control group of subjects who had ordinary hip fractures showed roughly similar results: Nearly a third of those with thigh fractures were on Fosamax, as opposed to one ninth of the hip patients; two thirds of the thigh patients on Fosamax showed the pattern, and they tended to have been on Fosamax longer than those with hip fractures.

Meanwhile, more reports were coming out, one from Singapore and one from Japan. Researchers hypothesized that bisphosphonates produce bone that is brittle and fracture-prone and that the thigh pain comes from little stress fractures that don’t heal but rather accumulate and build up to one big kahuna of a break. The patients Lane sees are active women seven to 10 years younger than those who break their hips (which happens on average at age 82). “Every woman I’ve seen has been out there shopping, working, running after her grandchildren, doing stuff,” he says. “These are not couch potatoes. I have never seen this kind of fracture in a nursing home patient.”

The subject of bisphosphonates and femur fractures was a point of controversy at the American Society for Bone and Mineral Research conference in the fall of 2008, when a case was made that these drugs were not the problem. Bo Abrahamsen, an endocrinologist at Copenhagen University Hospital, had studied registry data for all patients born in 1945 or earlier who entered Danish hospitals with fractures from 1996 to 2005. He fished out those who had been treated with bisphosphonates and matched them with controls who hadn’t taken those drugs but had experienced a fracture in the same place. If Lane’s hypothesis were true, the patients taking bisphosphonates would have gotten more of the atypical-femur fractures while the untreated controls would have gotten more of the old-fashioned osteoporotic fractures, yet that hadn’t happened.

Thus Abrahamsen concluded that the fractures Lane had deemed atypical were just ordinary osteoporotic fractures after all. The U.S. government, which is continuing to investigate the side effects of bisphosphonates, cited Abrahamsen’s study when it released a statement this March, saying, “At this point, the data that the FDA has reviewed have not shown a clear connection between bisphosphonate use and a risk of atypical subtrochanteric femur fractures.”

Merck, not surprisingly, concurs. While the company would not allow me to interview its researchers, a statement from its media-relations department said, “In clinical studies, Fosamax has not been associated with increased fracture risk at any skeletal site.” A recent analysis funded by Merck and another drug company also found no connection, though the investigators said their study was too small to be definitive.
On the other hand, two New York teams, one at Columbia and another (including Lane) at the Hospital for Special Surgery, recently presented small controlled studies of postmenopausal women with osteoporosis at the 2010 annual meeting of the American Academy of Orthopaedic Surgeons. The HSS team biopsied osteoporotic bone while the Columbia group analyzed the patients’ scans, and both studies buttress the theory that long-term bisphosphonate use alters bone properties so as to increase the risk of atypical femur fractures.

The risk appears quite small, however, relative to the odds of experiencing a hip fracture. This issue was explored in a large study done by orthopedist Richard Dell of -Kaiser-Permanente in Southern California. The study is important because Kaiser-Permanente is, as Dell puts it, “a closed shop,” meaning it provides comprehensive medical care and has complete medical records for large numbers of people. Dell looked at films and files of 172,000 people age 45 and older taking oral bisphos-phonates for more than three years, and he concluded that the drugs prevented 3,050 hip fractures that would have occurred in the untreated general population. At the same time, only 50 people in that group suffered bisphosphonate femur fractures like mine. (Merck is still looking into the question; it has funded a study run by Kaiser Pacific Northwest on the distribution and classification of femur fractures. Like Dell’s, this study will review X-rays and medical charts.)

Weighing the Risks
So the question is, How can doctors identify ahead of time the likely candidates for femur fractures—that is, patients who should not be given bisphosphonates, or at least not for very long? Dell, now analyzing his data before publication, has a few hints. Like Lane, he suspects that the most likely femur-fracture victims are younger (in their sixties and early seventies), more active and on the medication for a long time; some seem to be on other drugs as well, including prednisone and estrogen.

But there is an element of Russian roulette in the situation: Very few people get these odd femur fractures, and doctors can’t tell whether you’ll be one of them. So it’s important that people who take bisphosphonates be truly at risk for osteoporotic fractures. This calls into question the idea of using bisphosphonates as preventive medicine, and as a result the practice of routinely giving pills to almost every postmenopausal woman in her fifties with low bone density in order to ward off osteoporosis is becoming obsolete. However, some doctors probably haven’t gotten the memo yet. The reproductive endocrinologist—gynecologist Wulf Utian, the founder of the North American Menopause Society, points out that the U.S. is a large and diverse country, and prescribing practices are “all over the map.”

To better determine the benefit-risk ratio of taking bisphosphonates, the World Health Organization came up with FRAX, a test, still not widely used, that calculates your 10-year risk for osteoporotic fractures (see “What’s Your Bone-Break Risk?” on page 122). “In the early to mid 2000s, physicians relied almost exclusively on bone-density results and fracture history to make decisions about prescribing a bisphosphonate or other medication,” says Marjorie Luckey, medical director of the St. Barnabas Osteoporosis Center in Livingston, New Jersey. “Medication was often prescribed to prevent osteopenia from progressing to osteoporosis. But now we have tools that predict who’s at high risk for fractures so that we can target pharmacological treatment to those who need it most. The FRAX is not perfect, but it’s better than what we had before. With patients who have low bone density and are not at high risk for fractures in the near future, we try to convince them to take calcium and vitamin D, exercise regularly, stop smoking and limit alcohol. Many women can slow bone loss with just these steps once they are past the immediate menopause period.” (But if you’re taking bisphosphonates, you should continue unless your doctor advises you to stop, says the Endocrine Society.)

Some doctors now give patients, particularly those on Fosamax, a holiday of at least a year after a term of treatment. Here’s how that works, according to Lane: After you take Fosamax or a similar drug for three to five years, bone turnover, detected by a urine or blood test, begins to decrease, and the bone gets “sleepy”—that is, the collagen in the bone picks up chemicals that alter its properties, making it predisposed to fracture. When the bone markers in a blood or urine test go down, you stop the drug. Then you get the markers tested at regular intervals, and when they go up, you start using the drug again, perhaps at a smaller dose or in a weaker formulation.

How does all this research play out if you’re in your fifties and you’re told you have osteoporosis or a too-high fracture risk? A variety of effective drugs exist, and patients at high risk of fracture (like me) are now prescribed a nonbisphosphonate drug called For-teo, which offers the benefit of building new bone as well as preventing bone loss. Its sobering downsides, however, are that the patient must inject it into her stomach every day, it retails for $1,100-plus per month, and because of safety concerns, it can be taken for only two years maximum.

For other patients with osteoporosis, specialists may switch between medications. Lane may, for example, move a patient off the longest-lasting bisphosphonate, alendronate (Fosamax and its generic versions), to Actonel, a less powerful one. Another drug often used is the milder Evista, which is not a bisphosphonate. It has been shown to protect the spine, another potential site for osteoporotic fractures, but not the hip, and therefore only some patients will benefit from it. Evista has the added advantage of helping fend off breast cancer.

Luckey says the field may evolve so that different medications are prescribed at different stages of life. One scenario might involve a patient taking a low dose of estrogen close to menopause to protect bones and help with menopausal symptoms, followed by Evista to reduce the risk of breast cancer and osteoporosis, and then use of a bisphosphonate, if needed, later in life when hip-fracture prevention becomes increasingly important.

The bottom line: Experts agree that bisphosphonates prevent a lot of fractures in elderly patients with severe osteoporosis, and more of these people should be getting the drugs. If you are in your fifties and have a mother in her eighties, most likely she is a better candidate for bisphosphonate therapy than you are. Before taking these drugs, consult with your doctor to be absolutely sure you have real osteoporosis or a high FRAX score. If you do begin this course of therapy, get checked after three to five years to see if you still need it. As Dell observes, “You don’t give tumor drugs to prevent a tumor, and you shouldn’t give an oral bisphosphonate to prevent osteoporosis in someone who doesn’t have osteoporosis. This is a potent medication, not to be used by people who don’t need it.”

Originally published in the June 2010 issue of

First Published Mon, 2010-04-26 13:28

Find this story at: