Background/Purpose: Interstitial pneumonia with autoimmune features (IPAF) is a subset of interstitial lung disease (ILD) with clinical features suggestive of, but not sufficient for, a diagnosis of a connective tissue disease (CTD). Previous reports suggest it has a prognosis intermediate between that of idiopathic pulmonary fibrosis (IPF) and that of CTD related ILD (CTD-ILD). The epidemiology and progression of IPAF in a population based cohort has not been defined.

Methods: All residents of a geographically-defined area who first met the Fischer criteria for IPAF (Eur Respir J 2015; 46; 976-987) in 2000-2015 were identified. Manual chart review of all patients with ILD with abnormal serologies listed in the IPAF criteria, or with at least one rheumatology evaluation was performed. Those with a diagnosis of a CTD prior to or within 6 months after ILD diagnosis were excluded. Baseline serologies, clinical characteristics, comorbidities, ILD subtype, pulmonary function tests, and echocardiograms were collected. Incidence rates were calculated and adjusted to the US white 2010 population. Mortality was estimated by Kaplan-Meier methods. The performance of GAP and ILD-GAP prognostic models was assessed using standardized incidence ratios (SIR).

Results: We identified 17 cases meeting classification criteria for IPAF (7 [41%] female; mean age: 72.5 years; range 35-89), with an estimated incidence of 1.2 per 100,000 (95% CI 0.6 – 1.8). The incidence increased over time (p=0.024), perhaps reflecting increasing surveillance for CTD-ILD. Based on high resolution chest CT, there were 10 (59%) with nonspecific interstitial pneumonia (NSIP), 6 (35%) with usual interstitial pneumonia (UIP), and 1 with organizing pneumonia. Seven patients (41%) were never seen by a rheumatologist but met criteria based on serologies. Antinuclear antibodies were positive in 8 patients (73%). Six patients (40%) had a DLCO < 40% or were too ill to perform DLCO at the time of meeting IPAF criteria, and these were over-represented in the UIP group (p = 0.036). Only 4 patients (24%) had pulmonary hypertension. None of the patients progressed to a diagnosis of a CTD. During a median of 2.6 years of follow-up, 10 patients died (5 year survival rate: 36%; 95% CI 18- 71; median: 1.5 years). Both the GAP and ILD-GAP models underestimated mortality (SIR for GAP: 2.16; 95% CI: 1.12-4.15; SIR for ILD-GAP: 4.79; 95% CI: 2.49-9.21). In subgroup analysis, this appeared to be due to underestimation for patients with UIP.

Conclusion: The incidence of patients with ILD meeting criteria for IPAF in this first ever population based study was 1.2 per 100,000, although it may be increasing over time. Patients are generally older, with mean age at the time of meeting IPAF criteria of 72.5 years. Mortality in this population-based cohort was higher than previously published estimates for IPF or CTD-ILD. The GAP and ILD-GAP prognostic models underestimated mortality, particularly in patients with a UIP pattern. Overall, outcomes in this population-based cohort were significantly worse than those previously reported for IPAF cohorts at tertiary care centers.

« Back to 2017 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/epidemiology-of-interstitial-pneumonia-with-autoimmune-features/