Authors (including presenting author) :
Ma KC, Kwok CL, Yee KS, Tseng CZ
Affiliation :
Department of Tuberculosis and Chest, TWGHs Wong Tai Sin Hospital
Introduction :
Bronchoscopy is a vital diagnostic procedure for lung cancer, yet distinguishing malignancies from collapse-consolidation tissue remains challenging, particularly with ultrasound alone. Rapid Onsite Evaluation (ROSE) offers real-time feedback during bronchoscopy, potentially improving diagnostic precision. This report illustrates the role of ROSE in addressing the limitations of radial endobronchial ultrasound (R-EBUS).
Objectives :
To demonstrate the effectiveness of ROSE in enhancing diagnostic accuracy and procedural efficiency during bronchoscopy for suspected lung cancer.
Methodology :
A 63-year-old male patient, Mr. Leung, presented with blood-stained sputum and pleural effusion. High-resolution computed tomography (HRCT) revealed a suspicious 2.2 cm lobulated mass in the right lower lobe. Initial bronchoscopy with R-EBUS failed to confirm malignancy due to the inherent limitation of R-EBUS in distinguishing the echogenic signal of lung cancer from surrounding collapse-consolidation tissue. A repeat bronchoscopy incorporating ROSE was performed. The biopsy position was iteratively adjusted until pleomorphic cells suggestive of malignancy were identified. Molecular testing and surgical confirmation followed.
Result & Outcome :
Results: ROSE facilitated real-time detection of pleomorphic cells suggestive of malignancy. Subsequent pathological evaluation confirmed adenocarcinoma with EGFR exon 19 deletion. Surgical intervention confirmed a 2.4 cm tumor, staged as IIIA (T1cN2M0). ROSE ensures adequate sampling for molecular testing, reduces procedural time, and minimizes false negatives without increasing complication rates. Conclusion: ROSE is an effective adjunct in bronchoscopy, overcoming R-EBUS limitations by providing instant feedback for sample adequacy and enhancing diagnostic precision. Literature review highlights that ROSE improves diagnostic yield by 14% and sensitivity for malignancy by 10%. Integrating artificial intelligence (AI) into ROSE workflows may further optimize outcomes by reducing interobserver variability and procedural inefficiencies.
