Late-onset axial spondyloarthritis: data from Reuma-check cohort.

objectivesTo estimate the prevalence of late-onset axial spondyloarthritis (lo-axSpA) and to identify clinical, laboratory, and imaging features associated with this phenotype.

methodsThis single-center, observational study included patients diagnosed with axSpA from the "Reuma-check" SpA program. Patients with a symptom onset ≥ 45 years were classified as lo-axSpA, as opposed to early-onset axSpA (eo-axSpA, onset < 45 years). The prevalence of lo-axSpA was calculated, and lo-axSpA and eo-axSpA were compared in terms of clinical, laboratory and imaging characteristics. Factors associated with lo-axSpA were analyzed with univariable followed by multivariable logistic regression.

resultsA total of 126 patients were included, 35 (28%) were lo-axSpA. Comparing lo-axSpA vs. eo-axSpA, significant differences were observed: higher female prevalence in lo-axSpA vs. eo-axSpA (51% vs. 29%), lower NSAID response (52% vs. 73%), increased skin psoriasis prevalence (42% vs. 17%,), and shorter diagnosis delay (40 vs. 93 months). In the multivariable analysis, male sex and diagnosis delay were independently and inversely associated with lo-axSpA (OR 0.2, 95% CI 0.06-0.8 and OR 0.9, 95% CI 0.96-0.99, respectively), while psoriasis was associated with a higher odds for lo-axSpA (OR 4.8, 95% CI 1.1-29).

conclusionlo-axSpA was present in more than a quarter of the patients. Although recall bias in the symptom duration cannot be excluded, the presentation with lo-axSpA seems to be associated with distinct features, being more frequent in females and more associated with psoriasis and with a shorter diagnostic delay. Key Points • Late-onset axSpA (≥ 45Y) is observed in 28% in our cohort, a higher frequency than previously reported. • Female sex and psoriasis are associated with a higher likelihood for late-onset axSpA.