ROBOTIC BRONCHOSCOPY: OUR FIRST 100 NODULES

Abstract

TOPIC: Procedures TYPE: Original Investigations PURPOSE: Diagnosis of peripheral pulmonary lung nodules is a clinical challenge requiring a multi-disciplinary approach. Robotic bronchoscopy (RB) has come into the forefront of novel diagnostic procedures allowing for visualization and stable access to distal airways providing better access to nodules. However, limited data currently exist as to the diagnostic yield of RB and the learning curve required to become proficient. METHODS: This was a retrospective, single center study utilizing the robotic bronchoscope (Ion, Intuitive Medical) to sample peripheral pulmonary nodules. Data collected include nodule characteristics, length of procedure, use of linear endobronchial ultrasound (EBUS), and overall yield for malignant and benign nodules. Results from the first 6 months of use were compared to results from the last 8 months. Samples which were atypical or did not show a definite benign or malignant diagnosis were classified as non-diagnostic. Nodules were classified into true positive (TP), true negative (TN), or false negative (FN). TP were nodules with biopsy positive carcinoma or a defined benign diagnosis. TN nodules were non-diagnostic that had a benign diagnosis confirmed by resection, or follow-up imaging demonstrating decreased size of the nodule or a minimum of 6 months of stability. FN nodules were non-diagnostic and subsequently found to be cancer or treated as such with empiric radiation. RESULTS: 95 patients with 103 nodules were biopsied between February 2020 and April 2021. Median lesion size was 19 mm (range 7- 69mm). Based on RB 9(8.7%) of biopsies were benign, 60(58%) malignant and 34(33%) non-diagnostic. Benign RB diagnosis included hamartoma, fungal elements, organizing pneumonia and granulomas. In the non-diagnostic group 23/33 had complete follow up data and 13 (57%) were confirmed as benign.When comparing the first 6 months to the last 8 months the TP rate improved from 20/33 (61%) to 49/60 (82%) and the FN rate decreased from 4/33(12.1%) to 6/60 (10%) (p=0.0415). A total of 70 nodules were ultimately defined as malignant. RB confirmed diagnosis in 60(86%) of those. In the first 6 months malignancy was confirmed in 19/23 (83%) and this improved to 41/47 (87%; p=0.72).The average procedure time for RB did not differ significantly between the first 6 months and thereafter (81.9min vs 70.9min, p =0.1300). The procedure time for RB with EBUS decreased significantly from 138.9min to 104.5min (p = 0.0081). CONCLUSIONS: In this study, a learning curve effect was seen with significant improvements in TP/FN rates, and shorter procedure times when performed in conjunction with EBUS. In patients with cancer, RB is even more effective with yield as high as 87%. CLINICAL IMPLICATIONS: Our initial experience shows excellent results that may be superior to electromagnetic navigation bronchoscopy. A learning curve was demonstrated. DISCLOSURES: No relevant relationships by Stephanie Baltaji, source=Web Response No relevant relationships by Hiran Fernando, source=Web Response Speaker/Speaker's Bureau relationship with AstraZeneca Please note: $5001 - $20000 Added 11/30/2020 by Sohini Ghosh, source=Web Response, value=Consulting fee Consultant relationship with Intuitiv Please note: $1001 - $5000 Added 11/30/2020 by Sohini Ghosh, source=Web Response, value=Consulting fee Speaker/Speaker's Bureau relationship with AstraZeneca Please note: 2013-2021 Added 04/30/2021 by Benny Weksler, source=Web Response, value=Consulting fee No relevant relationships by Ping Zheng, source=Web Response