Background Rheumatoid arthritis (RA) is associated with clinically relevant interstitial lung disease (ILD) in approximately 5% of patients. There are no data on articular disease activity in patients with RA and ILD. There is also no agreed treatment paradigm for such patients. Several of the drugs used to induce remission in RA may be relatively contra indicated in the presence of ILD.
Objectives We wished to assess disease activity scores in 28 joints (DAS28) in patients with RA-ILD at baseline and then after treatment, and compare results with RA controls without ILD. We also wanted to assess the relationship between severity of ILD and DAS28 scores.
Methods We collected data from six centres across the UK on patients with both RA and ILD (proven on high resolution computed tomography) diagnosed in a 25 year period from 1987 to 2012 using a standard proforma. Within this population of patients with RA-ILD, we calculated DAS28, both at baseline and following treatment, in a total of 98 patients whose data sets were complete. We compared scores with those obtained from an age and gender matched control population with RA but no ILD. We recorded all drug therapy, together with patient outcome data on survival and, where appropriate, cause of death as obtained from death certificates. We related drug therapy and DAS28 to both all cause mortality and to death from ILD using analysis of variance.
Results We followed 98 patients for a median of 5 years (range 1-12 years) from diagnosis of RA-ILD. Forty two patients died, of which 24 (57%) died from ILD. Mean group baseline DAS28 was 3.88 (SD 1.48) and fell to a mean of 3.07 (SD 1.47) [p=0.048]. Mean baseline DAS28 among RA controls was 3.78 (SD1.44), falling to a mean of 3.00 (SD 1.32). Mean DAS28 in RA patients who died from ILD was significantly lower at baseline at 3.32 (SD 1.68) [p=0.034] but equivalent at five years at 3.03 (SD 1.43). The following drugs were most frequently prescribed: sulphasalazine (59), methotrexate(54), prednisone (54), anti-TNF therapy (22) and leflunomide (19). Smaller numbers of patients received specific therapy for their ILD: mycophenolate (10), methylprednisone (9), cyclophosphamide (8). However, there was no relationship between the use of any specific drug (or drug combination) and either overall mortality or death due to ILD.
Conclusions Articular disease activity in patients with RA-ILD did not differ from RA patients without ILD. However, those patients who died from ILD did have a lower baseline DAS28, suggesting less active articular disease at diagnosis. Treatment induced a significant reduction in baseline DAS in spite of the relatively low use of MTX (possibly as result of concern about potential adverse pulmonary effects). Despite the relatively high overall mortality, we found no statistical evidence to relate the use of any specific therapeutic agent to an increased risk of death from ILD or sepsis. More information is required to assess the impact of specific therapeutic agents on disease outcome, but these data suggest that Rheumatologists should beware of ILD in those RA patients with low baseline DAS28 scores as these patients may have higher mortality from their lung involvement.
Disclosure of Interest None Declared