Automatic Processing of Nasal Pressure Recordings to Derive Continuous Side-Selective Nasal Airflow and Conductance

• Published in: Frontiers in physiology Vol. 9; p. 1814
• Main Authors: Urner, Lorenz M, Kohler, Malcolm, Bloch, Konrad E
• Format: Journal Article
• Language: English
• Published: Switzerland Frontiers Media S.A 07.01.2019
• Subjects: nasal airflow; nasal prong pressure transducer; nasal resistance; noninvasive monitoring; Physiology; rhinitis; rhinomanometry; noninvasive monitoring; nasal airflow; signal processing; nasal resistance; rhinitis; physiological monitoring; nasal prong pressure transducer; rhinomanometry
• ISSN: 1664-042X; 1664-042X
• DOI: 10.3389/fphys.2018.01814

Monitoring of nasal airflow and conductance provides crucial insights into the variable nature of the nasal resistance, nasal cycle, and ventilation. We have previously shown that tracking of pressure swings at the entrance of each nasal passage by a dedicated catheter system allows bilateral monitoring of nasal airflow over several hours but requires complex linearization and calibration procedures. Side-selective nasal conductance is derived from linearized and calibrated bilateral nasal pressure swings and corresponding driving pressure, i.e., the transnasal pressure difference derived from an epipharyngeal catheter. Manual analysis of such recordings and computation of instantaneous conductance as the ratio of flow to driving pressure over several hours is extremely tedious, time consuming, and therefore not suitable for routine practice. To address this point, we developed and validated a software for automatic processing of nasal and epipharyngeal pressure recordings as a convenient tool for studying the nasal ventilation. The software applies an eight-parameter logistic model to transform nasal pressure swings into side-selective estimates of airflow that are calibrated and further processed along with epipharyngeal pressure to compute bilateral nasal conductance over consecutive, user-selectable time-segments. Essential processing steps include (1) offset correction, (2) low-pass filtering, (3) cross-correlation, (4) cutting of signals into individual breaths, (5) normalization, (6) ensemble averaging to obtain a mean pressure signal for each nasal side, (7) derivation of airflow, conductance, and further variables. Among four evaluated algorithms for calculation of nasal conductance, the derivative of the airflow-pressure curve according to the mean value theorem agreed closest with the gold standard, i.e., the conductance derived from airflow measured by a pneumotachograph attached to an oral-nasal mask and transnasal pressure. In combination with the nasal catheter system, our novel software represents a valuable tool for use in clinical practice and research to conveniently investigate nasal ventilation and its changes occurring spontaneously or in response to various exposures and therapeutic interventions.