Presentations

Abstract

Introduction: The diagnostic yield of endobronchial ultrasound-guided fine needle aspiration (EBUS-FNA) can approach 72-95% in mediastinal and hilar lymph nodes. However, these percentages refer to the final diagnosis made by the pathologist after the procedure is completed. Rapid On-Site Evaluation (ROSE) is used during the procedure to obtain a preliminary pathology report. When performed on FNA samples, ROSE is frequently non-diagnostic, compelling the bronchoscopist to sample additional lymph nodes or switch modalities to navigational bronchoscopy for direct sampling of the lesion. Here, we investigate the utility of performing intranodal forceps biopsies under EBUS guidance to improve the diagnostic accuracy of ROSE.

Methods: 42 adult patients with mediastinal or hilar lymphadenopathy were included in this single-center retrospective study with procedure dates between February 2021 and January 2023. A single bronchoscopist performed all procedures. After identification of the pathologic lymph node(s), EBUS-FNA sampling was performed with samples being evaluated by ROSE. If these specimens were non-diagnostic, then intranodal forceps biopsy was attempted. When successful, forceps biopsies underwent ROSE as well. The diagnostic sensitivity for biopsy ROSE was calculated using the final pathology reports as the "gold standard" for diagnosis.

Results: In our 42 patients, 43 lymph nodes underwent forceps biopsy. 83% (36/43) of the lymph nodes sampled had ROSE performed on both FNA and the biopsy. In 34 of the 36 cases, the reason for biopsy was a non-diagnostic FNA. Of these, 17 (50%) had a positive biopsy ROSE. In one patient, there was a false positive with ROSE of the biopsy being read as "lesional cells", but the final pathology having a diagnosis of "reactive cells." When used as an add-on procedure for negative FNA ROSE, intranodal forceps biopsies had a sensitivity of 61.5%, specificity of 87.5%, positive predictive value of 94.1%, and negative predictive value of 41.2%. No complications were attributed to the intranodal forceps biopsies.

Conclusion: While EBUS-FNA is an excellent diagnostic tool, false negative preliminary results on ROSE frequently occur. Our study shows that intra-nodal forceps biopsy can be performed safely and effectively on a wide array of mediastinal lymph node stations, and that it can improve the immediate diagnostic yield of ROSE. In our experience, the use of intranodal forceps biopsy as an add-on procedure for negative FNA ROSE reduced the need for additional procedures in approximately 50% of cases. Further studies are recommended to generalize these findings as well as investigate the overall meaningful clinical benefits of intranodal forceps biopsy.