Accuracy of Rapid On-Site Evaluation in Robotic-Assisted Bronchoscopy Fine Needle Aspirations of Lung Nodules

Background: Shape-sensing robotic-assisted bronchoscopy (ssRAB) is an emerging technique for sampling lung nodules. When evaluating the value of pairing ssRAB with rapid on-site evaluation (ROSE), other studies have only looked at diagnostic yield. This study aims to assess the diagnostic concordance of ROSE and histopathologic biopsies in ssRAB sampling. Methods: We analyzed lung nodules sampled by ssRAB fine needle aspiration (FNA) and had ROSE. The following parameters were recorded: preliminary and final cytology diagnoses, concurrent histopathology biopsy diagnosis, tool type, number of passes for ROSE, lesion size, and location. A χ2 test was performed for categorical variables, and one-way analysis of variance was used to compare means for continuous variables. A P -value <0.05 was considered statistically significant. Results: A total of 249 FNA cytology specimens from 234 patients were obtained, with 202 concurrent histopathologic biopsies. The lesion size, location, and proceduralist experience did not affect cytology diagnosis; however, the diagnostic yield improved with increasing lesion size. Diagnostic yield plateaued after 3 passes ( P =0.01). Forceps biopsies had a lower diagnostic yield (74%) than cryoprobe biopsies (87%), and forceps were more likely to yield a nondiagnostic sample than cryobiopsies ( P =0.04). A positive ROSE preliminary diagnosis for malignancy correlated highly with a positive histopathology biopsy (86% correlation). Conclusion: A positive result for malignancy on ROSE correlated highly with a positive histopathology biopsy. Three fine needle passes were sufficient for diagnostic results, with a plateau in subsequent biopsies. Histopathologic tissue diagnosis was superior with the cryoprobe than with tissue obtained using conventional forceps.