Nonsteroidal anti-inflammatory drugs (NSAID) are the cornerstone of treatment in axial spondyloarthritis (axSpA). We aimed to describe the real life NSAID use in early axSpA according to the disease characteristics.

The DESIR cohort is an observational study of patients presenting with inflammatory back pain suggestive of axSpA for less than 3 years. Patients who attended all 6 visits (every 6 months during the first 2 years and yearly at 3 years) were analysed. Data collected at every visit over 3 years was NSAID intake and the ASAS-NSAID score for the 6 months preceding the visit, where 100 reflects the full daily intake of an NSAID.

In all, 627 patients were assessed: 46.1% males, mean age: 33.7 (standard deviation: 8.7) years, mean symptom duration: 1.5 (0.9) year and mean BASDAI: 44.3 (25.3). Over the 6 months preceding inclusion, 92.8% had received NSAIDs; this proportion significantly decreased over time, to 73% patients after 3 years (P<0.001). This decrease was not different with regard to the fulfillment (or not) of the ASAS criteria at baseline, nor with regard to the “arm” of the ASAS criteria fulfilled (i.e. imaging vs. clinical). The median ASAS-NSAID score was 36 [interquartile range: 13–72] at inclusion, and substantially decreased to reach 7 [0–44] after 3 years (P<0.001), although BASDAI at 3 years was 33.8 (21.7).

NSAID prescription was very frequent in this early axSpA cohort, but the intake was low, and decreased significantly over time, even though symptoms remained present. Early axSpA patients may not be sufficiently treated with NSAIDs.