Evaluation of the Utility of Rapid On-site Cytology (ROSE) Performed by an Interventional Pulmonologist During EBUS-TBNA. A Prospective Observational Study
Evaluation of the Utility of Rapid On-site Cytology (ROSE) Performed by an Interventional Pulmonologist During EBUS-TBNA. A Prospective Observational Study
This study is a single-centre, prospective observational analysis conducted at the Alessandria Hospital, aimed to evaluating the role of Rapid On-Site Evaluation (ROSE) performed directly by the interventional pulmonologist during endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) procedures. EBUS-TBNA is a gold-standard minimally invasive technique for the diagnosis and staging of mediastinal diseases, particularly lung cancer, allowing the collection of cytological material from lymph nodes and lesions adjacent to the airways.
ROSE involves the immediate cytological assessment of the collected sample, with the aim of verifying its adequacy and providing an initial diagnostic indication directly in the operating theatre. Traditionally, this assessment is entrusted to the cytopathologist; however, their presence is not always guaranteed in clinical practice, due to organisational and resource constraints. For this reason, in recent years there has been growing interest in the possibility that the interventional pulmonologist, if adequately trained, could perform this assessment independently.
The aim of the study is therefore to assess the usefulness of ROSE performed by a pulmonologist during EBUS-TBNA, by comparing the results obtained with those of the cytopathologist. In particular, the study analyses the agreement between the two professionals in terms of the adequacy of the cytological sample and diagnostic accuracy.
The main question the study aims to answer is whether the on-the-spot cytological assessment performed by the pulmonologist is sufficiently reliable and consistent with that of the cytopathologist to be used in clinical practice as a valid alternative, particularly in situations where the cytopathologist is not present in the procedure room.
The hypothesis underlying the study is that the use of ROSE by the pulmonologist may help to improve the efficiency of the procedure, reducing the number of needle passes, the duration of the examination and overall costs, without compromising diagnostic quality. The expected results could therefore have significant organisational and clinical implications, promoting wider adoption of ROSE and optimising the diagnostic and therapeutic pathway for patients undergoing EBUS-TBNA.
Endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) is a well-established minimally invasive technique for the diagnosis and staging of mediastinal and hilar lymphadenopathy, particularly in patients with suspected or confirmed lung cancer. The procedure enables real-time sampling of lymph nodes and peribronchial lesions, providing cytological material for diagnostic evaluation and, when necessary, for ancillary studies such as immunohistochemistry and molecular testing. Ensuring the adequacy and representativeness of collected samples is critical, especially in the context of personalized oncology, where therapeutic decisions rely on accurate histological and molecular characterization.
Rapid On-Site Evaluation (ROSE) is an intra-procedural technique that involves the immediate cytological assessment of aspirated material to determine sample adequacy and provide a preliminary diagnostic impression. Traditionally, ROSE is performed by a cytopathologist or cytotechnologist present during the procedure. However, the limited availability of these professionals in many centers restricts the routine use of ROSE, leading to potential inefficiencies such as unnecessary needle passes, longer procedure times, and suboptimal sample collection.
This study is a prospective, observational, single-center investigation conducted within routine clinical practice, aimed at evaluating the feasibility and diagnostic performance of ROSE when performed by an interventional pulmonologist. The study is designed as a quality improvement initiative (no-profit) and focuses on assessing whether appropriately trained pulmonologists can reliably perform on-site cytological evaluation during EBUS-TBNA procedures.
During each procedure, samples are obtained according to standard clinical indications. A portion of the collected material is processed following routine cytopathological workflows, including fixation in formalin and preparation of cell blocks for definitive diagnosis. An additional portion of the specimen is smeared into glass slides, rapidly stained (e.g., Diff-Quik), and immediately evaluated on-site by the interventional pulmonologist. The pulmonologist assesses both the adequacy of the specimen and provides a preliminary classification based on cytomorphological features.
Specimens are categorized into predefined diagnostic groups, including: diagnostic (presence of clearly atypical or malignant cells), adequate negative (presence of sufficient lymphoid elements without atypia in lymph node samples), inadequate (insufficient cellular material or absence of representative elements), and indeterminate/suspicious (presence of atypical features not sufficient for a definitive diagnosis). The final cytopathological evaluation, performed by an experienced cytopathologist, serves as the reference standard. Blinding procedures may be applied to minimize bias in interpretation.
The study aims to evaluate the level of agreement between the pulmonologist's on-site assessment and the final cytopathological diagnosis, focusing on both adequacy and diagnostic interpretation. The results are expected to inform the potential role of pulmonologist-performed ROSE as an alternative model in settings where cytopathology support is limited.
Data collection is performed using a secure, web-based electronic data capture system (REDCap), which ensures structured data entry, auditability, and compliance with Good Clinical Practice (GCP) and applicable data protection regulations (including GDPR).
Inclusion Criteria:
Exclusion Criteria: