Application of rapid on-site evaluation combined with flexible bronchoscopy in the diagnosis of lung lesions

Background: Pathology is considered the gold standard for the diagnosis of lung lesions, but the pathological result is relatively lagging and cannot provide real-time guidance for the biopsy procedure. Objective: To investigate the potential application of rapid on-site evaluation (ROSE) during flexible bronchoscopy (FB) in the evaluation and diagnosis of lung lesions. Patients and methods: Consecutive patients who underwent FB for the diagnosis of lung lesions between August 2022 and February 2023 were included in this retrospective study. 294 patients underwent FB with ROSE, while 304 patients underwent FB without ROSE. The final pathological results and the number of patients undergoing repeat biopsies were recorded in both groups. Specifically, we conducted separate statistical analysis for patients undergoing different biopsy methods, including the endobronchial biopsy (EBB), radial probe endobronchial ultrasound transbronchial lung biopsy with guide sheath (r-EBUS-GS-TBLB), and the endobronchial ultrasoundguided transbronchial needle aspiration (EBUS-TBNA) to study the detailed roles that ROSE plays under different biopsy methods. Results: The adequacy rate of biopsy specimens from the non-ROSE group was significantly lower than that of the ROSE group (259/281 = 92.17 % vs. 263/268 = 98.13 %, p = 0.001). Meanwhile, fewer patients underwent repeat biopsies in the ROSE group compared to the non-ROSE group (2/294 = 0.68 % vs. 10/304 = 3.29 %, p = 0.023). For the ROSE group, the consistency between ROSE diagnoses and final pathological diagnoses was 94.40 % (kappa = 0.886), with 95.58 % for benign diseases and 93.55 % for malignant diseases. Conclusion: The utility of ROSE during FB increases the adequacy rate of biopsy specimens and thus decreases the need for repeat biopsies in patients with lung lesions to get a definite diagnosis. Moreover, the high consistency between ROSE diagnoses and final pathological diagnoses suggests that ROSE is a reliable tool for optimizing the diagnosis of lung lesions.