Phenotyping pulmonary sarcoidosis with CT descriptors using BTS ILD registry data

Crooks, R., McCall, M., Minnis, P., Achaiah, A., Casimo, L., Hewitt, R., Hodkinson, C., Khan, F, Morris, H., Palmer, E., Stewart, I., Thomas, G., Loughenbury, M., Souto, M., Fletcher, S.V. and Chaudhuri, N. (2024) Phenotyping pulmonary sarcoidosis with CT descriptors using BTS ILD registry data. Thorax, 79 (Suppl. 2), A146-A147, [P67]. (doi:10.1136/thorax-2024-BTSabstracts.228).

Abstract

Introduction and aims: pulmonary sarcoidosis has a variable clinical course, with up to one third experiencing a chronic progression. Current phenotyping, often based on age, symptoms, and imaging, aims to predict outcomes but lacks accuracy. The Scadding staging system, reliant on chest X-rays, no longer meets modern standards. Computed tomography (CT) scans can reveal more nuanced findings. A consensus has emerged on distinct CT phenotypes, suggesting a promising avenue for improved classification. Our study aims to validate phenotypes using real-world data to enhance treatment decisions and prognostic accuracy in sarcoidosis.

Methods: we performed a retrospective cohort study using the BTS Sarcoidosis registry data stratifying patients by CXR stage and by CT descriptors. Baseline demographics were recorded and outcome measures included MRC Dyspnoea scale, lung function testing, and treatment with systemic oral therapy within 3 months.

Results: 772 individual records were available from the registry. The mean age was 51.6 (13.1), 60% were male, 48% were diagnosed within the last 12 months and 46% received systemic treatment (namely prednisolone or DMARDs) within 3 months. 497 patients had CXR stage recorded (64%), with 581 having CT descriptors (75%). Significant findings are presented in table 1. Stratifying by CXR stage revealed significant differences in terms of age, diagnosis within 12 months, MRC dyspnoea scale (CXR stage 3+4), airflow limitation (CXR stage 4), and treatment. CXR stage 1 was most common (23%). Stratifying by CT descriptors resulted in greater homogeneity of demographics and outcome measures. Pulmonary nodularity was the most common CT finding (24%). Transforming CT descriptors into discrete sarcoidosis phenotypes, especially ones with clinical implications for disease status and prognostication, has proved challenging. Limitations to using CT imaging in this way include inter observer variability, the timing of CT imaging and the inherent information reduction associated with qualitative data.

Conclusions: nearly all sarcoidosis patients have had at least one CT study prior to diagnosis. Typically they reveal abnormalities not evident on CXR, the implications of such findings in the context of Scadding stage are largely unknown, but it is likely that CT features can enrich our capacity to phenotype.

This record has no associated files available for download.

Contributors

Author: S.V. Fletcher

Download statistics