Background/Purpose: Rheumatoid arthritis (RA) has known extra-articular manifestations that can result in restrictive and obstructive patterns on pulmonary function measures, especially in smokers. However, investigations comparing RA and non-RA for pulmonary function change have not been performed. In this study, we compared longitudinal spirometry measures between smokers with and without RA.

Methods: We analyzed longitudinal data from COPDGene, a multicenter prospective cohort of smokers with at least 10 pack-years. We investigated longitudinal pulmonary function measures in people with RA compared to non-RA comparators and healthy non-smokers. Spirometry was conducted at baseline (2007-2011) and at a second visit 5 years later. RA cases were identified by self-reported RA and DMARD use (PPV 88%); non-RA comparators reported no RA and no DMARD use. The outcomes were the annual change in postbronchodilator percent predicted forced expiratory volume in one second (%FEV₁), percent predicted forced vital capacity (%FVC), and FEV₁/FVC ratio. We compared these outcomes between RA cases and non-RA comparators using linear regression, adjusted for age, sex, BMI, smoking status (current/past), pack-years, baseline spirometry results, and inhaled/systemic medication use for obstructive lung diseases. An additional analysis used inverse probability of censoring weighting (IPCW) to account for possible differential censoring, defined by drop-out or death before the follow-up visit was due. We also stratified the analysis by the presence of obstructive pattern (FEV₁/FVC< 0.7) at baseline.

Results: We analyzed 45 RA smoker cases and 5175 non-RA smoker comparators with available follow-up spirometry data (Table 1). The mean %FEV₁ at baseline was 77% in RA cases and 80% in non-RA comparators. The mean change in %FEV₁ from baseline to the 5-year follow-up was +1.8% in RA cases and -2.3% in non-RA comparators. Proportions of each group with improved, stable, and declined %FEV₁ are shown in Figure 1. In the multivariable linear regression models, both %FEV₁ and FEV₁/FVC showed significantly less decline in RA cases compared to non-RA comparators (%FEV₁: β 0.68, p=0.026; FEV₁/FVC: β 0.52, p=0.002; Table 2). This result was more prominent in participants with obstructive pattern at baseline (RA vs. non-RA: %FEV₁: β 1.17, p=0.0025; FEV₁/FVC: β 1.73, p< 0.001). There was no significant difference in the annual change in %FVC between the two groups. Results were similar after accounting for possible differential loss to follow-up or death.

Conclusion: In this first comparative study among smokers to examine longitudinal pulmonary function in RA, RA cases were less likely to have %FEV₁ and FEV₁/FVC declines than non-RA comparators. Results were strongest among RA cases with baseline obstructive defect and not explained by differences in smoking, suggesting that RA with obstruction may be a unique phenotype. Further studies are needed to replicate with larger sample size and to examine mechanisms and potential for reversibility related to systemic inflammation, autoimmunity, and RA treatment.

« Back to ACR Convergence 2023

ACR Meeting Abstracts - https://acrabstracts.org/abstract/rheumatoid-arthritis-and-changes-in-pulmonary-function-measures-on-spirometry-in-a-prospective-longitudinal-cohort-of-smokers/