Semi-quantitative tracking of intra-airway fluids by computed tomography

Background: Airway secretions are a source of complications for patients with acute and chronic lung diseases, yet lack of techniques to quantitatively track secretions hampers research into clinical measures to reduce their pathologic consequences. Methods: In a preserved swine lung model, we tracked a contrasted mucus simulant (CMS) using sequential computed tomography (CT). Known drivers of secretion movement - gravity and ventilation - were tested. Ten millilitres of CMS were unilaterally introduced (1 ml min^-1) into the airways of 12 lung sets. After instillation, six lung sets were maintained prone and six were rotated 180°. Subsequently, all were mechanically ventilated for 10 min. CTs were obtained before infusion, after infusion and after ventilation ± rotation. For CT analysis, the lungs were partitioned into eight sub-cuboids using anatomic landmarks. The volumes of two CT number ranges representing CMS and poor aeration/collapse were computed in every sub-cuboid for each CT acquisition. Volume differences between study time points were used to quantify changes. Results: CMS and poor aeration/collapse volume change distributed gravitationally after infusion. After ventilation without rotation, the CMS and poor aeration/collapse volumes remained within the originally injected sub-cuboid, although the poor aeration/collapse volume expanded (27·3 ± 6·1→50·5 ± 7·4 ml, P<0·05). After ventilation + rotation, there was a reduction in the CMS and poor/aeration collapse volumes in the originally injected sub-cuboid (14·4 ± 1·7→4·4 ± 0·6 ml, P<0·05 and 18·3 ± 3·8→11·9 ± 2·7 ml, P<0·05, respectively) accompanied by increases in the gravitationally opposite sub-cuboid (1·7 ± 0·2→11·1 ± 1·1 ml, P<0·05 and 0·8 ± 0·5→40·6 ± 3·5 ml, P<0·05, respectively). Conclusion: Movement of fluids within the bronchial tree can be semi-quantitatively tracked with analysis of sequential CT acquisitions. In this isolated swine lung model, gravity had an important and brisk effect on movement of a viscous fluid, whereas ventilation tended to embed it peripherally.