In a national study of older Danes who had previously had a fracture and were taking bisphosphonates, the risk of having a serious though rare atypical femoral fracture (AFF) was greater after 3-5 years of bisphosphonate use.
The risk quickly dropped after patients stopped taking a bisphosphonate, which suggests that bisphosphonate “holidays” may be useful for some patients, the researchers said. These findings support previous work.
But the study also found that 34% of the AFFs occurred in patients who had not been taking a bisphosphonate. That rate is higher than the 6%-22% that has been reported by others.
Doug Bauer, MD, from the University of California, San Francisco, presented the new study findings during the virtual American Society of Bone and Mineral Research 2020 annual meeting.
“We found no clear risk factor that accounts for this increased risk [for AFFs] among those not exposed to bisphosphonates,” he said, “but we believe this was a real finding, as our study protocol ensured that the study radiologists were completely blinded to treatments received.”
Suzanne N. Morin, MD, who was not involved in this research, pointed out that the reported AFF risks related to bisphosphonate dose and cessation are in keeping with findings of other studies, including a recent large study by Dennis M. Black, MD, and colleagues that was published in the New England Journal of Medicine.
That study found that Asians are at higher risk for AFFs than White persons. Others have reported that specific femur geometry or physique and use of glucocorticoids increase AFF risk, Dr. Morin, from the Research Institute of the McGill University Health Center, Montreal, said in an interview.
The current study suggests that rheumatoid arthritis may be a risk factor, she added.
The fact that the rate of AFFs among patients who had not been exposed to bisphosphonates was higher than previously reported “may be due to differences in the method they used to ascertain the fractures or in medication use,” she speculated.
The clinical implications of research to date are that “the risk of AFF should not dissuade patients and providers from short-term use of bisphosphonates [3-5 years],” Dr. Bauer said. He noted that most patients should not take a bisphosphonate for longer than this unless they have a very high fracture risk.
Similarly, Dr. Morin said that clinicians “should consider initiating bisphosphonate in those at high risk for fractures and reevaluate their use after 3-6 years, depending on individual’s risk profile.”
AFF is serious but rare complication of bisphosphonate use
“Since first reported over 10 years ago, it has become clear that AFFs are a rare but serious complication of bisphosphonate therapy,” Dr. Bauer explained. However, there is still uncertainty about the magnitude of this risk, including the absolute risk for AFFs among adults who take bisphosphonates and those who do not.
To study this, the researchers analyzed data from national health care and pharmacy records and a radiology image database in Denmark. They identified almost 5,000 adults who were aged 50 years or older and who experienced a subtrochanteric and femoral shaft fracture during the period from 2010 to 2015. Two expert radiologists who were blinded to the patients’ clinical history or treatment identified AFF on the basis of ASBMR 2014 criteria.
The researchers compared three patient groups: 189 patients with AFF, 2,397 patients with typical subtrochanteric and femoral shaft fractures (no AFF), and35,946 adults aged older than 50 years (control persons).
Compared with patients with typical fractures, patients with AFF were younger (aged 71 vs. 77), more likely to be women (79% vs. 69%), and more likely to have RA (12% vs. 2.5%).
Compared with patients in the other two groups, those with AFF were more likely to use corticosteroids, proton pump inhibitors, statins, and hormone replacement therapy.
They were also more likely to use bisphosphonates (58%) than patients with typical subtrochanteric and femoral shaft fractures (19%) or control patients (10%).
The bisphosphonates used in Denmark at the time were mostly alendronate (85%) and rarely ibandronate (6%), intravenous zoledronic acid (5%), etidronate (3%), or risedronate (1%).