Phenotypical Characterization of Airway Morphology in Post-Infectious vs Post-Lung Transplantation Bronchiolitis Obliterans

Purpose Post-infectious bronchiolitis obliterans (PIBO) forms a chronic obstructive lung disease in children following a severe viral respiratory infection. Subsequent dysanaptic lung growth, with differential development of the alveolar compartment over the airway compartment, was clinically described, but remains elusive. We therefore performed a detailed 3D airway characterization in PIBO compared to bronchiolitis obliterans syndrome (BOS) after lung transplantation and discarded donor lungs. Methods Explant lungs from matched adult patients who underwent lung transplantation for end-stage PIBO (n=5), redo lung transplantation patients with end-stage BOS (n=5), and discarded donor lungs (n=5) were included. Lungs were air-inflated and frozen in fumes of liquid nitrogen. Ex vivo computed tomography (CT) scans (resolution 700 µm) were obtained and used for radiologic scoring and semi-automated 3D airway segmentation. Next, frozen lungs were sampled on matched locations (n=4/lung) and scanned with microCT (resolution 10 µm). Results Radiologic scoring indicated more airway wall thickening (p=0.079), bronchial dilatation (p=0.087), and pleural thickening (p=0.071) in PIBO vs BOS. The number of airways per generation was similar between groups (p=0.53), but airway diameters were increased in PIBO (generation 6-11) and in BOS lungs (generation 7-10) compared to controls, without differences between PIBO and BOS. Significantly more airways ended in BO lesions in BOS vs PIBO (p=0.016, mean 25.60% vs 12.60%), although the size of the BO lesions was increased in PIBO vs BOS (p=0.032, median 27.23mm³ vs 5.01mm³). Subsequent micro CT assessment revealed a significant lower number of terminal bronchioles (TB) in PIBO (p=0.0009, PIBO: mean 3131 TB/lung, BOS: 8360 TB/lung, controls: 10342 TB/lung), although diameters of TB were similar across groups (p=0.37). In addition, alveolar spaces tended to be larger in PIBO (p=0.070). 3D micro CT analysis confirmed that in BOS, BO lesions are only segmental with preserved TBs, while in PIBO, the airway is completely lost following the obstruction. Conclusion PIBO lungs displayed a more than two-fold decrease in TB compared to BOS, with fewer but larger BO lesions and non-reopening of airways distal from BO. Our findings provide potential pathophysiologic evidence of dysanaptic lung growth in PIBO.