Rheumatoid arthritis (RA) affects approximately 0.5–1% of the population. Clinically significant interstitial lung disease (ILD) occurs in approximately 5–10% of individuals with RA. The frequency of subclinical disease is much higher, with up to 58% of persons with RA having abnormalities consistent with ILD on one or more investigations [1], [2], [3]. Histologically, a usual interstitial pneumonia (UIP) pattern is more common in RA-ILD compared with ILD associated with other connective tissue diseases. A range of other patterns are also recognised including non-specific interstitial pneumonia (NSIP), organising pneumonia (OP), lymphocytic interstitial pneumonia (LIP), diffuse alveolar damage (DAD), and desquamative interstitial pneumonia (DIP) [4,5].

The aetiology of RA-ILD is incompletely understood, and is believed to be multifactorial with a combination of environmental and genetic risk factors. The risk of developing ILD has been reported to be higher in those with an older age at RA onset [1,6], in men [1,3,7,8], and amongst smokers [9,10]. Smoking has also been associated with a UIP pattern [11]. Increased articular disease activity has also been shown to be associated with an increased risk of RA-ILD; HR 2.22, 95% CI 1.28–3.82, for moderate/high disease activity compared to group with remission/low disease activity [12]. A variant in the promoter of MUC5B has been associated with RA-ILD, and in particular a UIP pattern of disease (adjusted OR 3.1, 95% CI 1.8, 5.4, for RA-ILD vs RA without ILD in combined analysis of discovery population and multiethnic case series) [13].

Observational studies suggest potential benefit of a number of immunosuppressive therapies, but data from randomised controlled studies in RA-ILD are lacking [14]. Antifibrotic agents have been shown to slow deterioration in pulmonary function tests in some patients with RA-ILD. The INBUILD trial was a 52-week randomised double blind placebo-controlled trial of nintedanib in 663 patients with progressive fibrosing interstitial lung disease, that included 89 participants with RA-ILD, and reported a lower rate of annual decline in forced vital capacity (FVC) in patients who received nintedanib compared with placebo. Mortality was studied as a secondary endpoint, and was not found to differ significantly between the nintedanib group and the placebo group in the overall study, or in the subgroup of individuals with progressive autoimmune disease related ILDs [15,16]. TRAIL 1 was a phase 2 randomised controlled trial of pirfenidone in RA-ILD, that was stopped early due to slow recruitment and the COVID-19 pandemic. The available results did not demonstrate a statistically significant difference in the primary composite endpoint of decline from baseline in percent predicted FVC of 10% or greater or death during the 52-week treatment period; a statistically significant reduction in annual decline in FVC was reported [17].

Further research is required to better understand the natural history of RA-ILD. Duration of survival has not been a focus of previous systematic reviews. Understanding prognosis in terms of duration of survival can assist meaningful discussions with individuals with RA-ILD and may help serve as a baseline for future studies that have mortality as an outcome. Therefore, we undertook a systematic review to determine the duration of survival, from time of diagnosis of RA-ILD to death from any cause, regardless of treatments.