P83 Early experience of multimodally directed slim/ultraslim bronchoscopy at a UK centre

Background We set up a multimodally directed slim and ultraslim bronchoscopy service at our institution in December 2018 to aid in the diagnosis of hard to reach peripheral lung lesions. This service uses a combination of virtual bronchoscopic navigation (VBN) software, radial probe endobronchial ultrasound (RP-EBUS) with or without guide sheath (GS), 2D fluoroscopy and either a slim (2.0 mm working channel) or ultraslim (1.7 mm) bronchoscope. Methods We performed a retrospective analysis of the first 20 consecutive cases referred to this service. In all patients VBN, RP-EBUS and 2D fluoroscopy were used in combination. The choice of a slim or ultraslim bronchoscope was at the discretion of the operator and usually guided by the location of the lesion. All procedures were performed under conscious sedation. Tissue sampling included bronchial lavage, brushings using either a 1.5 mm or 1.9 mm cytology brush, and transbronchial biopsy using a 1.5 mm or ≥1.9 mm forceps. Size selection was based on the use of GS or the bronchoscope used. Patients with a non-malignant diagnosis were referred on for further investigations or surveillance for at least 3 months. Results The mean age was 71 ±8 years. 4 out of 9 patients with non-malignant cells were true negative for malignancy on follow up CT; 5 were deemed to be false negative. The overall diagnostic accuracy was 75% in our cohort. The sensitivity for diagnosing malignancy was 73% with a cancer prevalence of 80%. All patients tolerated the procedure well. Other than one small post-procedure pneumothorax (treated conservatively) there were no significant complications. The use of VBN significantly reduced procedural time, with directed navigation to the area of abnormality which was confirmed on RP-EBUS in all cases. Whilst use of larger biopsy forceps or brush to improve diagnostic yield necessitates the removal of the GS for sampling (and may preclude use of the ultraslim scope), concurrent use of fluoroscopy helped to confirm the same location was being consistently sampled. Conclusions Our initial experience of multimodally directed slim/ultraslim bronchoscopy is very promising with a high diagnostic accuracy in the sampling of peripheral lung lesions.