Objectives To establish the level of observer variation for the current ATS/ERS/JRS/ALAT criteria for a diagnosis of usual interstitial pneumonia (UIP) on CT among a large group of thoracic radiologists of varying levels of experience. Materials and methods 112 observers (96 of whom were thoracic radiologists) categorised CTs of 150 consecutive patients with fibrotic lung disease using the ATS/ERS/JRS/ALAT CT criteria for a UIP pattern (3 categories—UIP, possibly UIP and inconsistent with UIP). The presence of honeycombing, traction bronchiectasis and emphysema was also scored using a 3-point scale (definitely present, possibly present, absent). Observer agreement for the UIP categorisation and for the 3 CT patterns in the entire observer group and in subgroups stratified by observer experience, were evaluated. Results Interobserver agreement across the diagnosis category scores among the 112 observers was moderate, ranging from 0.48 (IQR 0.18) for general radiologists to 0.52 (IQR 0.20) for thoracic radiologists of 10–20 years’ experience. A binary score for UIP versus possible or inconsistent with UIP was examined. Observer agreement for this binary score was only moderate. No significant differences in agreement levels were identified when the CTs were stratified according to multidisciplinary team (MDT) diagnosis or patient age or when observers were categorised according to experience. Observer agreement for each of honeycombing, traction bronchiectasis and emphysema were 0.59±0.12, 0.42±0.15 and 0.43±0.18, respectively. Conclusions Interobserver agreement for the current ATS/ERS/JRS/ALAT CT criteria for UIP is only moderate among thoracic radiologists, irrespective of their experience, and did not vary with patient age or the MDT diagnosis.

Mortality in pulmonary sarcoidosis is highly variable and a reliable prognostic algorithm for disease staging and for guiding management decisions is needed. The objective of this study is to derive and test a staging system for determining prognosis in pulmonary sarcoidosis. We identified the prognostic value of high-resolution computed tomography (HRCT) patterns and pulmonary function tests, including the composite physiological index (CPI) in patients with pulmonary sarcoidosis. We integrated prognostic physiological and HRCT variables to form a clinical staging algorithm predictive of mortality in a test cohort. The staging system was externally validated in a separate cohort by the same methods of discrimination used in the primary analysis and tested for clinical applicability by four test observers. The test cohort included 251 patients with pulmonary sarcoidosis in the study referred to the Sarcoidosis clinic at the Royal Brompton Hospital, UK, between Jan 1, 2000, and June 30, 2010. The CPI was the strongest predictor of mortality (HR 1·04, 95% CI 1·02-1·06, p<0·0001) in the test cohort. An optimal CPI threshold of 40 units was identified (HR 4·24, 2·84-6·33, p<0·0001). The CPI40, main pulmonary artery diameter to ascending aorta diameter ratio (MPAD/AAD), and an extent of fibrosis threshold of 20% were combined to form a staging algorithm. When assessed in the validation cohort (n=252), this staging system was strikingly more predictive of mortality than any individual variable alone (HR 5·89, 2·68-10·08, p<0·0001). The staging system was successfully applied to the test and validation cohorts combined, by two radiologists and two physicians. A clear prognostic separation of patients with pulmonary sarcoidosis is provided by a simple staging system integrating the CPI and two HRCT variables.

The incidence of indeterminate pulmonary nodules has risen constantly over the past few years. Determination of lung nodule malignancy is pivotal, because the early diagnosis of lung cancer could lead to a definitive intervention. According to the current international guidelines, size and growth rate represent the main indicators to determine the nature of a pulmonary nodule. However, there are some limitations in evaluating and characterising nodules when only their dimensions are taken into account. There is no single method for measuring nodules, and intrinsic errors, which can determine variations in nodule measurement and in growth assessment, do exist when performing measurements either manually or with automated or semi-automated methods. When considering subsolid nodules the presence and size of a solid component is the major determinant of malignancy and nodule management, as reported in the latest guidelines. Nevertheless, other nodule morphological characteristics have been associated with an increased risk of malignancy. In addition, the clinical context should not be overlooked in determining the probability of malignancy. Predictive models have been proposed as a potential means to overcome the limitations of a sized-based assessment of the malignancy risk for indeterminate pulmonary nodules.

Diagnostic and Interventional Radiology

Rationale: Reliable outcome prediction in patients with fibrotic lung disease using baseline high-resolution computed tomography (HRCT) data remains challenging. Objectives: To evaluate the prognostic accuracy of a deep learning algorithm (SOFIA [Systematic Objective Fibrotic Imaging Analysis Algorithm]), trained and validated in the identification of usual interstitial pneumonia (UIP)-like features on HRCT (UIP probability), in a large cohort of well-characterized patients with progressive fibrotic lung disease drawn from a national registry. Methods: SOFIA and radiologist UIP probabilities were converted to Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED)-based UIP probability categories (UIP not included in the differential, 0–4%; low probability of UIP, 5–29%; intermediate probability of UIP, 30–69%; high probability of UIP, 70–94%; and pathognomonic for UIP, 95–100%), and their prognostic utility was assessed using Cox proportional hazards modeling. Measurements and Main Results: In multivariable analysis adjusting for age, sex, guideline-based radiologic diagnosis, anddisease severity (using total interstitial lung disease [ILD] extent on HRCT, percent predicted FVC, Dl_CO, or the composite physiologic index), only SOFIA UIP probability PIOPED categories predicted survival. SOFIA-PIOPED UIP probability categories remained prognostically significant in patients considered indeterminate (n = 83) by expert radiologist consensus (hazard ratio, 1.73; P < 0.0001; 95% confidence interval, 1.40–2.14). In patients undergoing surgical lung biopsy (n = 86), after adjusting for guideline-based histologic pattern and total ILD extent on HRCT, only SOFIA-PIOPED probabilities were predictive of mortality (hazard ratio, 1.75; P < 0.0001; 95% confidence interval, 1.37–2.25). Conclusions: Deep learning–based UIP probability on HRCT provides enhanced outcome prediction in patients with progressive fibrotic lung disease when compared with expert radiologist evaluation or guideline-based histologic pattern. In principle, this tool may be useful in multidisciplinary characterization of fibrotic lung disease. The utility of this technology as a decision support system when ILD expertise is unavailable requires further investigation.