Fracture Risk Quantified for Older Spondyloarthritis Patients

Medicare beneficiaries with ankylosing spondylitis (AS) incur frequent bone fractures, a new study indicated, with the risk boosted by some unsurprising factors but also including one that was not expected.

Fee-for-service Medicare claims data from 2016-2018 showed that AS patients suffered fractures at a rate of 76.7 per 1,000 person-years, according to Rachael Stovall, MD, MAS, of the University of California San Francisco, and colleagues.

Older age and a record of previous fractures were among the risk factors for fractures, the group reported in Arthritis Care & Research, with odds ratios of 2.80 (95% CI 1.39-5.65) for ages 85 and up versus those younger than 65, and 5.24 (95% CI 3.44-7.99) for patients with at least one previous fracture versus those with none.

But one other factor significantly linked to fracture risk stood out: use of opioids. AS patients taking morphine-equivalent doses greater than 30 mg/day had nearly doubled risk of fracture (OR 1.86, 95% CI 1.08-3.19) compared with non-users of opioids. Low doses (up to 10 mg/day) were also a significant risk factor, with an odds ratio of 1.56 (95% CI 1.03-2.38); patients on mid-range opioid doses, i.e., between 10 and 30 mg/day in morphine-equivalents, had numerically greater fracture rates that fell just short of statistical significance (OR 1.59, 95% CI 0.95-2.66).

These latter findings carry an important clinical message, Stovall and colleagues suggested. Insofar as opioid use is potentially modifiable, “this high-risk population should be considered for interventions to mitigate risk,” the authors wrote.

As well, in light of the risk associated with previous fractures, Stovall and colleagues argued that AS patients with a fracture history should be “prioritized for fracture prevention interventions, such as fall avoidance programs and osteoporosis treatment.”

While AS is similar to rheumatoid arthritis in many ways, one unique aspect is that bone physiology is frequently altered. In particular, Stovall’s group explained, “osteoproliferation, new bone formation, is a key feature. … It includes the development of syndesmophytes, ankylosis, and ligamentous ossification of the spine.” This can result in osteoporosis but can also “alter an individual’s center of gravity and put them at an increased risk of falling,” the researchers noted.

Yet while these risks are known qualitatively to rheumatologists, “we do not have recent or comprehensive U.S. population level data on fracture risk among adults with [AS],” the group continued. Consequently, they undertook the current study using Medicare claims data collected through the Rheumatology Informatics System for Effectiveness (RISE) registry.

This database included 1,426 Medicare enrollees with AS with 2 years of follow-up beginning in 2016, of whom 197 experienced fractures. Three-quarters of these individuals had just one fracture during the study period. The most common fracture location was the spine, at a rate of 21.1 per 1,000 person-years, followed by the forearm/wrist/hand at 17.4 and the ankle-foot at 12.8 per 1,000 person-years.

Besides age, fracture history, and opioid use, Stovall and colleagues examined a host of other potential risk factors, including glucocorticoid use, body mass index, smoking history, sex, race/ethnicity, and comorbidity burden. None had significant associations with incident fractures nor even numerical differences suggestive of an effect. (Interestingly, very obese individuals incurred fracture at relatively low rates, with an odds ratio of 0.60 versus normal-weight people [95% CI 0.31-1.16] but no trend was apparent.)

The researchers wrote that they were unaware of any previous finding of increased fracture risk associated with opioid use in the AS population. This finding in the current study, they suggested, probably comes from falls, although earlier studies had found decreased bone mineral density with opioids.

Limitations to the study included the relatively small patient sample, the short (2-year) follow-up, the reliance on administrative data, and the potential for unmeasured factors to influence the results.

Please enable JavaScript to view the

comments powered by Disqus.