Background/Purpose

High-resolution computed tomography of the chest (HRCT) performed for assessment of interstitial lung disease (ILD) in patients with systemic sclerosis (SSc) frequently reveals a patulous esophagus, but the significance is unclear. Assuming that the risk of aspiration is directly related to the magnitude of esophageal dilatation, we hypothesized that a greater HRCT esophageal diameter is associated with worse pulmonary outcomes, as measured by severity of ILD on HRCT and pulmonary function tests (PFT).

Methods

A cross-sectional study of Northwestern Scleroderma Registry patients with HRCT was conducted.  Subjects met the ACR 1980 SSc or three out of five CREST (Calcinosis cutis, Raynaud phenomenon, Esophageal dysmotility, Sclerodactyly, Telangiectasias) criteria. The esophageal diameter was measured at three levels: above the aortic arch, between the aortic arch and the right inferior pulmonary vein, and between the right inferior pulmonary vein and the diaphragmatic hiatus. Widest esophageal diameter (WED) was used as the predictor variable.  A modified Likert scale  (0= none, 1= <5%, 2= 6-25%, 3 for 26-50%, 4=51-75%, 5= >75%) was used to score fibrosis and ground glass opacities.  Total ILD (outcome variable) was calculated as the sum of fibrosis and ground glass scores.  Secondary outcomes were PFT including total lung capacity (TLC), forced vital capacity (FVC) and diffusing capacity for carbon monoxide adjusted for hemoglobin % predicted (DLCO). Standardized regression coefficients (β) between WED and the outcome variables were calculated adjusting for age, sex, ethnicity, disease subtype (limited or diffuse cutaneous SSc), disease duration (years since first non-Raynaud symptom), proton pump inhibitor use, C-reactive protein and modified Rodnan skin score.    

Results

Three hundred eleven subjects had HRCT.  Eight subjects without available HRCT images and one subject who had undergone esophagectomy were excluded.  Twenty-eight patients who had mixed connective tissue disease, SSc sine scleroderma, or overlap syndromes were excluded, leaving 275 subjects for analysis.  Adjusted standardized regressions demonstrated positive associations between WED and total ILD score (β=0.33, p<0.0001), fibrosis (β=0.32, p<0.0001), and ground glass opacities (β=0.28, p<0.0001) (Table). There were negative associations between WED and TLC % predicted (β=-0.24,p=0.0002), FVC % predicted (β=-0.23, p=0.0004), and adjusted DLCO % predicted (β=-0.23;p=0.005).

Conclusion

Increasing esophageal diameter on HRCT in patients with SSc is associated with more severe radiographic ILD, lower lung volumes, and lower DLCO % predicted.  Longitudinal studies should be done to elucidate the temporal relationship between esophageal disease and ILD in persons with SSc.  Future trials of aggressive management of esophageal disease to prevent ILD may be warranted in persons with SSc.