Endobronchial Ultrasound for the Screening of Pulmonary Embolism in Patients with Suspected Lung Cancer: A Prospective Cohort Study.

BACKGROUND: Patients with lung cancer exhibit increased risk of pulmonary embolism (PE). While the contrast phase of computed tomography of the chest in the diagnostic work-up of suspected chest malignancy does not allow reliable detection of PE, it may be feasible to screen for present PE during endobronchial ultrasound (EBUS) examination. OBJECTIVES: The aim of this study was to establish if screening during EBUS for PE in patients with suspected lung cancer is feasible and if positive findings are predictive of PE. METHODS: Patients undergoing EBUS due to suspicion of malignancy of the chest were prospectively enrolled. The pulmonary arteries were assessed during EBUS using a standardized protocol. Patients in whom PE suspicion was raised were referred to confirmatory imaging. RESULTS: From December 2020 to August 2021, 100 patients were included. Median time for vascular assessment during EBUS was 2 min (Q1-Q3: 1-3 min). EBUS identified two suspected PEs (2%), and the number needed to scan was 50. The positive predictive value of EBUS for PE was 100%. CONCLUSION: EBUS for PE screening seems feasible and with limited time use. The PPV of positive findings for the diagnosis of PE is high, but the utility is somewhat limited by a high number needed to scan even in a high-risk population. Based on our findings, we believe that EBUS assessment of the pulmonary vasculature may have a role as a routine screening tool for PE. The assessment for PE should be implemented in EBUS training programmes, as operators should be able to recognize incidental PEs.

When Pulmonologists Are Novice to Navigational Bronchoscopy, What Predicts Diagnostic Yield?

Predicting factors of diagnostic yield in electromagnetic navigation bronchoscopy (ENB) have been explored in a number of previous studies based on data from experienced operators. However, little is known about predicting factors when the procedure is carried out by operators in the beginning of their learning curve. We here aim to identify the role of operators' experience as well as lesion- and procedure characteristics on diagnostic yield of ENB procedures in the hands of novice ENB operators. Four operators from three centers without prior ENB experience were enrolled. The outcome of consecutive ENB procedures was assessed and classified as either diagnostic or non-diagnostic and predicting factors of diagnostic yield were assessed. A total of 215 procedures were assessed. A total of 122 (57%) of the ENB procedures resulted in diagnostic biopsies. Diagnostic ENB procedures were associated with a minor yet significant difference in tumor size compared to non-diagnostic/inconclusive ENB procedures (28 mm vs. 24 mm; p = 0.03). Diagnostic ENB procedures were associated with visible lesions at either fluoroscopy (p = 0.003) or radial endobronchial ultrasound (rEBUS), (p = 0.001). In the logistic regression model, lesion visibility on fluoroscopy, but none of operator experience, the presence of a bronchus sign, lesion size, or location nor visibility on rEBUS significantly impacted the diagnostic yield. In novice ENB operators, lesion visibility on fluoroscopy was the only factor found to increase the chance of obtaining a diagnostic sample.

Learning Curves for Electromagnetic Navigation Bronchoscopy Using CUSUM Analysis.

BACKGROUND: Electromagnetic navigation bronchoscopy (ENB) is a relatively new and technically demanding procedure for the guidance of bronchoscopic biopsy to help locate small lung lesions. The results in experienced hands are well described. However, we do not know the results in unexperienced hands-in other words, we have no knowledge about how fast you can learn the procedure. AIM: The aim of this study was to draw learning curves for beginners in ENB using Cumulated Sum (CUSUM) analysis, a method for quantitative evaluation of the learning curves for clinical procedures. METHODS: Four operators from 3 centers without prior ENB experience were enrolled. The outcome of consecutive ENB procedures was assessed and classified as either diagnostic or nondiagnostic based on sample adequacy. Learning curves were drawn based on diagnostic yield. RESULTS: A total of 215 procedures were assessed. For 2 of the operators (operators 1 and 4), at least 25 to 30 procedures were necessary to obtain competency whereas operators 2 and 3 showed more horizontal learning curves indicating an overweight of diagnostic procedures from the beginning. CONCLUSION: Operators achieve competences in ENB at different paces. This must be taken in account when beginners start to learn the procedure. There is a huge need for a structured educational program and a validated test to determine competences.