Grading Airway Stenosis Down to the Segmental Level Using Virtual Bronchoscopy: Analysis Per Patient

Nine findings were true-positive, and 13 were false-positive. The 13 false-positive findings were for the left upper lobar apicoposterior segmental bronchus (B I/II) [two patients], the left upper lobar anterior segmental bronchus (B III) [two patients], the right lower lobar superior segmental bronchus (B VI), the right upper lobar apical segmental bronchus (B I) [two patients], the right upper lobar posterior segmental bronchus (B II), the right upper lobar anterior segmental bronchus (B III), the left lower lobar anteromedial segmental bronchus (B VII/VIII), the right lower lobar medial basal segmental bronchus (B VII), the middle lobar medial segmental bronchus (B IV), and middle lobar lateral segmental bronchus (B V). In seven of the 13 false-positive stenoses, mucus was noted on flexible bronchoscopy, which was removed through irrigation and suction. One false-negative finding was made in the left lower lobe superior segmental bronchus (B VI). The sensitivity (90.0%), accuracy (95.5%), and negative predictive value (99.6%) of virtual bronchoscopy for the diagnosis of segmental airway stenosis were similar to those for the central airways. The positive predictive value, however, was lower for the segmental airways (40.9%) than for the central airways (84.4%).

Virtual bronchoscopy correctly graded 286 of 302 airway regions. It correctly graded all nine true-positive stenoses. The correlation between virtual bronchoscopic and flexible bronchoscopic grading of peripheral airway stenosis was lower (r = 0.61; p