Utility of virtual bronchoscopic navigation for peripheral pulmonary lesions

Abstract

Background and Purpose: Virtual bronchoscopic navigation (VBN) has been reported to be efficient for diagnosis of peripheral pulmonary lesions (PPLs). This study aimed to evaluate its efficacy at our institution. Methods: This was a retrospective review on patients who underwent bronchoscopy for PPLs at our hospital from April 2011 to December 2014. LungPoint® (Broncus) was used for VBN starting October 2013; by logistic regression analysis, diagnostic yields were compared before and after VBN introduction. Results: This study included a total of 220 patients with median age of 69 yrs (range 20–94) and male preponderance of 134. As of this time, 189/ 220 cases have final diagnoses. The before group (n= 129) and the after group (n = 91) were similar in terms of age, sex, and size, shape, and location of PPLs. Diagnostic yield was 90.6% (48/ 53) with both VBN and endobronchial ultrasound with a guide sheath (EBUS-GS), 77.1% (54/ 70) with EBUS-GS only, and 74.2% (49/ 66) without VBN or EBUS-GS. Adjusting for the use of EBUS-GS and PPL size, there were significant differences between groups (OR, 2.87; 95% CI, 1.09 to 7.56; p = 0.033). The diagnostic yield for PPLs ≤ 20 mm had a tendency to be higher in the after group (82.1%, 23/ 28) compared with that of the before group (60.0%, 24/ 40) (OR, 1.72; 95% CI, 0.415 to 7.12; p = 0.455). Compared with the before group, the after group had significantly longer median duration of examination (47 vs. 41 mins, p = 0.0057) and time elapsed until the start of sample collection (24 vs. 17 mins, p = 0.00035). Conclusions: Although it did not shorten the duration of examination, VBN increased the diagnostic yield of bronchoscopy for PPLs, especially for small lesions.