Decreasing incidence of cervical spine fractures in patients with ankylosing spondylitis: A population-based study in Southeast Norway.

BACKGROUND CONTEXT: Individuals diagnosed with ankylosing spondylitis (AS) face an increased risk of spine fractures, specifically cervical spine fractures (CS-Fxs). In the past two decades, biological disease-modifying antirheumatic drugs (bDMARDs) have provided considerable relief from pain and an enhanced sense of wellbeing for a large segment of AS patients. Despite these improvements, it remains unclear whether extended use of bDMARDs can indeed reduce the risk of spine fractures.

PURPOSE: In this study, we aimed to investigate the evolving patterns and epidemiology of traumatic CS-Fxs in both AS and non-AS populations. We hypothesized that the risk of CS-Fxs among AS patients would show a decreasing trend over time, while the risk among non-AS patients would remain constant.

STUDY DESIGN/SETTING: Retrospective cohort study based on a prospective database.

PATIENT SAMPLE: A total of 3598 consecutive patients with CS-Fxs were treated at Oslo University Hospital over an 8-year period.

OUTCOME MEASURES: CS-Fxs in AS patients were contrasted with non-AS-related CS-Fxs in terms of temporal trends, age, sex, injury mechanism, associated cervical spinal cord injury (cSCI), need for surgical fixation, and 30-day mortality.

METHODS: Data regarding all CS-Fxs diagnosed between 2015 and 2022 were extracted from the Southeast Norway population-based quality control database for traumatic CS-Fxs. Categorical data were summarized using frequencies, and continuous data were summarized using medians. The Wilcoxon rank-sum test was used to compare continuous variables, and the chi-squared test and Fischer exact test were used to compare categorical variables. To investigate the trend in the incidence of fractures, two different Poisson models were fitted with the number of non-AS and AS fractures as dependent variables and the year as the explanatory variable.

RESULTS: Over an eight-year period, we registered 3622 CS-Fxs in 3598 patients, with AS patients accounting for 125 of these fractures. Relative to their non-AS counterparts, AS patients presented a 9-fold and 8-fold higher risk of initial and subsequent CS-Fxs, respectively. We observed a declining trend in AS-related CS-Fxs with an annual linear decrease of 8.4% (p=0.026), whereas non-AS-related CS-Fxs showed an annual linear increase of 3.7% (p<0.001). AS patients sustaining CS-Fxs were typically older (median age 70 versus 63 years), predominantly male (89% versus 67%), and more frequently experienced injuries due to falls (82% versus 57%). They also exhibited a higher prevalence of subaxial CS-Fxs (91% versus 62%), fewer C0-C2 CS-Fxs (14% versus 44%), a higher rate of associated cSCI (21% versus 11%), and a greater tendency for surgical fixation (66% versus 21%). We observed a 30-day mortality rate of 11% in AS patients and 5.4% in non-AS patients (p=0.005).

CONCLUSIONS: The results of this study confirm the elevated risk of CS-Fxs among AS patients, although this risk appears to show a decreasing trend. The most plausible explanation for this risk reduction is the widespread application of bDMARDs.