Quantification of tissue-engineered trachea performance with computational fluid dynamics.

Objectives/Hypothesis: Current techniques for airway characterization include endoscopic or radiographic measurements that produce static, two-dimensional descriptions. As pathology can be multilevel, irregularly shaped, and dynamic, minimal luminal area (MLA) may not provide the most comprehensive description or diagnostic metric. Our aim was to examine the utilization of computational fluid dynamics (CFD) for the purpose of defining airway stenosis using an ovine model of tissue-engineered tracheal graft (TETG) implantation. Study Design: Animal research model. Methods: TETGs were implanted into sheep, and MLA was quantified with imaging and endoscopic measurements. Graft stenosis was managed with endoscopic dilation and stenting when indicated. Geometries of the TETG were reconstructed from three-dimensional fluoroscopic images. CFD simulations were used to calculate peak flow velocity (PFV) and peak wall shear stress (PWSS). These metrics were compared to values derived from a quantitative respiratory symptom score. Results: Elevated PFV and PWSS derived from CFD modeling correlated with increased respiratory symptoms. Immediate pre- and postimplantation CFD metrics were similar, and implanted sheep were asymptomatic. Respiratory symptoms improved with stenting, which maintained graft architecture similar to dilation procedures. With stenting, baseline PFV (0.33 m/s) and PWSS (0.006 Pa) were sustained for the remainder of the study. MLA measurements collected via bronchoscopy were also correlated with respiratory symptoms. PFV and PWSS found via CFD were correlated (R-2 = 0.92 and 0.99, respectively) with respiratory symptoms compared to MLA (R-2 = 0.61). Conclusions: CFD is valid for informed interventions based on multilevel, complex airflow and airway characteristics. Furthermore, CFD may be used to evaluate TETG functionality.