Time domain estimation of infant respiratory mechanical properties and lung volume during high frequency oscillatory ventilation -

Abstract

Assessments of lung mechanics parameters and volume, and how these physiologic parameters change with treatment and during mechanical ventilation, may be of great value to clinicians caring for patients in respiratory failure and requiring mechanical respiratory support. Obtaining reliable and accurate mechanics and volume data is, however, challenging particularly in severely preterm infants whose condition requires using HFOV (High Frequency Oscillatory Ventilation) as a means of respiratory support. Additionally, these assessments generally suffer from being impractical, entail interrupting the HFOV support, and require different methods for lung mechanics and lung volume assessments. This thesis research tested the hypothesis that a model-based approach for the analysis of HFOV time-domain pressure, flow, volume ventilation data measured at multiple mean airway pressure (Paw) and pressure amplitude (ΔP) settings will provide a practical and minimally intrusive means of obtaining reliable and physiologically sensible lung mechanical-volume parameters, and can potentially guide clinical management of this vulnerable infant population. The developed method does not interrupt the mechanical support of the infant and for the first time attempts to simultaneously obtain respiratory resistance and compliance estimates, separate airway and tissue mechanical properties and estimate alveolar lung volume. This thesis research developed and compared three mathematical models that were applied in inverse-fashion (model fitting of experimental data) to available previously collected infant data in the context of physiologically driven and evidence-based simplifying assumptions. The physiologic applicability of the results (parameter estimates) was assessed through direct comparisons to available infant data in the relevant medical research literature.