The Utility of Rapid On-Site Evaluation during Bronchoscopic Biopsy: A 2-Year Respiratory Endoscopy Central Experience

BACKGROUND: Rapid on-site evaluation (ROSE) is commonly used to evaluate the adequacy of biopsy materials in fine-needle aspiration; however, the diagnostic performance of ROSE during fiber optic bronchoscopy (FOB) biopsy under direct vision is rarely reported. Here, we evaluated the role of ROSE during FOB biopsy of visible lesion in trachea or bronchi. METHODS: The role of ROSE was prospectively evaluated in consecutive bronchoscopy specimens obtained between January 2016 and January 2018. The agreement and accuracy between ROSE and final histopathological interpretation were assessed. The frequency and possible reasons for discrepancy between ROSE and definitive histopathology results were identified. Histological and cytological classification was performed according to the International Association for the Study of Lung Cancer, the American Thoracic Society, and the European Respiratory Society (IASLC/ATS/ERS) criteria of lung ADCs classification. RESULTS: The study enrolled 651 patients, of which 33 were excluded because of insufficient cells. Final diagnosis of malignancy was achieved in 462 cases (74.8%), whereas 156 cases (25.2%) were nonmalignant. ROSE and pathology were well correlated for the diagnosis of squamous cell carcinoma (SCC) (Kappa = 0.718, p < 0.05), adenocarcinoma (AdC) (Kappa = 0.662; p < 0.05) and small cell lung cancer (SCLC) (Kappa = 0.955; p < 0.05). In 24 cases diagnosed as malignant by ROSE and nonmalignant by pathology, the lesion tissues were surgically excised and re-analyzed, and the 24 cases were finally confirmed as malignant by pathology. CONCLUSIONS: ROSE technique allows bronchoscopists to obtain viable and adequate material for the diagnosis of histopathology, and provides them with an onsite preliminary diagnosis especially in cases with inconclusive macroscopic appearance. ROSE and pathology should be used in combination to increase the accuracy of diagnosis.