# The Impact of Nonsteroidal Anti‐Inflammatory Drugs on Radiographic Spinal Progression in Patients With Axial Spondyloarthritis: 10‐Year Results From an Inception Cohort

## TL;DR

This study shows that higher NSAID use, especially COX2 inhibitors, may slow spinal damage in axial spondyloarthritis patients over 10 years.

## Contribution

The study provides 10-year evidence on how NSAID types affect spinal progression in axial spondyloarthritis subgroups.

## Key findings

- Higher NSAID intake was linked to slower radiographic spinal progression in axial spondyloarthritis.
- The effect was strongest in patients with radiographic axial spondyloarthritis.
- COX2 inhibitors showed a slightly stronger, though not statistically significant, benefit compared to nonselective NSAIDs.

## Abstract

This study aims to investigate the impact of nonsteroidal anti‐inflammatory drug (NSAID) intake on radiographic spinal progression in axial spondyloarthritis (axSpA), considering different NSAID types (COX‐2 inhibitors [COX2i] and nonselective NSAIDs [ns‐NSAIDs]) and disease subgroups (radiographic [r‐axSpA] and nonradiographic [nr‐axSpA]).

Leveraging data from the German Spondyloarthritis Inception Cohort (GESPIC), we conducted analyses on 252 patients with axSpA (139 with nr‐axSpA and 113 with r‐axSpA), who had minimum two sets of spinal radiographs. The outcome was progression in modified Stoke Ankylosing Spondylitis Spine Score (mSASSS) in two‐year intervals. We fitted sequential conditional mean models by using generalized estimating equations and adjusting for longitudinal repeated measures of exposure and time‐dependent confounders. We report β‐coefficients with 95% confidence intervals (CIs) for outcomes reflecting the progression in mSASSS per 10‐point increase in NSAID intake score.

At baseline, 201 (80.0%) patients were under NSAID treatment, with 46 (18%) taking COX2i and 156 (62%) taking ns‐NSAIDs, and mean total NSAID intake score was 38.3 ± 35.5. A 10‐point increase in NSAID intake score was associated with retardation of radiographic progression (β = −0.052, 95% CI: −0.097 to −0.007), with this effect being most pronounced in patients with r‐axSpA (β = −0.077, 95% CI: −0.152 to −0.003). COX2i showed a slightly lower point estimate (although not statistically significant) in progression compared to ns‐NSAIDs among all patients with axSpA (β = −0.061 and −0.045, respectively).

Our findings suggest a beneficial effect of higher NSAID intake, particularly COX2i, on slowing radiographic progression in axSpA. These findings may help inform therapeutic strategies, particularly in r‐axSpA, although further research is needed for nr‐axSpA.