Background/Purpose: Interstitial lung disease (ILD) is one of the most severe complications in systemic sclerosis (SSc). The diagnosis and monitoring of ILD is based on instrumental examinations as spirometry and imaging techniques as high resolution CT (HRCT) and lung ultrasound (LUS). In this context, lung magnetic resonance imaging (MRI) is an emerging tool with unexplored potentialities. The aim of this work was to investigate the role of lung MRI in the diagnosis and monitoring of SSc-ILD patients, then making comparisons with other assessment methods.

Methods: We enrolled SSc patients affected by ILD which presented worsening characteristics such as to require the initiation or change of immunosuppressive therapy. Evaluations were performed at baseline and after 6 months. Epidemiological and SSc-specific data were collected for each patient. Spirometry assessed forced vital capacity (FVC), total lung capacity (TLC) and diffusing capacity of the lungs for carbon monoxide (DLCO). Gradient echo T2* standard-weighted sequences obtained before and after contrast were evaluated for lung MRI, analyzing both healthy and pathological parenchyma in each lung. HRCT images were analyzed with CALIPER software to obtain the percentages of lung parenchyma with ILD. For LUS, antero-posterior scans were obtained with quantitative evaluation of B lines and semiquantitative evaluation of pleural irregularities. A radiological comparison between the basal images and those after 6 months was then made to express a final judgment between improvement, stability or worsening.

Results: Fifteen patients with SSc-ILD were enrolled (66.6% female, median age 50 years, median disease duration 3 years). The mean values of the T2* sequences in the pathological areas of the lung were directly related to the number of B lines at the LUS (p=0.01; ρ=0.637) and inversely related to the spirometric parameters of FVC (p=0.05; ρ=-0.558), TLC (p=0.01; ρ=-0.802) and DLCO (p=0.02; ρ=-0.650). There was no significant correlation between MRI sequences values and HRCT CALIPER data. Male sex was associated with worse baseline MRI pathological values (p=0.05), whereas there were no differences regarding age, disease duration, skin subset or autoantibody. When comparing baseline with 6-month images, the agreement in the final radiological judgment between MRI and HRCT was 92.3%, whereas that between LUS and MRI was 78.5%, similar to the agreement between LUS and HRCT which was 75%.

Conclusion: Lung MRI is an emergent non-invasive investigation technique that, unlike HRCT, does not expose the subject to radiations and, unlike LUS, is not operator-dependent and is able to provide images of the entire lung parenchyma. Our study showed that T2* sequences of the pathological areas on lung MRI are related to the worsening of spirometric parameters and to the alterations highlighted on LUS. Furthermore, the agreement in radiological judgment between HRCT and lung MRI appears very high. Although more validation studies are needed and much potential remains to be explored, our data encourage the use of pulmonary MRI in the evaluation of ILD in SSc patients.

« Back to ACR Convergence 2022

ACR Meeting Abstracts - https://acrabstracts.org/abstract/lung-magnetic-resonance-imaging-in-systemic-sclerosis-associated-interstitial-lung-disease-potentiality-and-comparison-with-other-imaging-techniques/