AUTHOR=Niérat Marie-Cécile , Dubé Bruno-Pierre , Llontop Claudia , Bellocq Agnès , Layachi Ben Mohamed Lila , Rivals Isabelle , Straus Christian , Similowski Thomas , Laveneziana Pierantonio 

TITLE=Measuring Ventilatory Activity with Structured Light Plethysmography (SLP) Reduces Instrumental Observer Effect and Preserves Tidal Breathing Variability in Healthy and COPD

JOURNAL=Frontiers in Physiology

VOLUME=8

YEAR=2017

URL=https://www.frontiersin.org/journals/physiology/articles/10.3389/fphys.2017.00316

DOI=10.3389/fphys.2017.00316

ISSN=1664-042X

ABSTRACT=<p>The use of a mouthpiece to measure ventilatory flow with a pneumotachograph (PNT) introduces a major perturbation to breathing (“instrumental/observer effect”) and suffices to modify the respiratory behavior. Structured light plethysmography (SLP) is a non-contact method of assessment of breathing pattern during tidal breathing. Firstly, we validated the SLP measurements by comparing timing components of the ventilatory pattern obtained by SLP vs. PNT under the same condition; secondly, we compared SLP to SLP+PNT measurements of breathing pattern to evaluate the disruption of breathing pattern and breathing variability in healthy and COPD subjects. Measurements were taken during tidal breathing with SLP alone and SLP+PNT recording in 30 COPD and healthy subjects. Measurements included: respiratory frequency (R<italic>f</italic>), inspiratory, expiratory, and total breath time/duration (T<sc>i</sc>, T<sc>e</sc>, and T<sc>t</sc>). Passing-Bablok regression analysis was used to evaluate the interchangeability of timing components of the ventilatory pattern (R<italic>f</italic>, T<sc>i</sc>, T<sc>e</sc>, and T<sc>t</sc>) between measurements performed under the following experimental conditions: SLP vs. PNT, SLP+PNT vs. SLP, and SLP+PNT vs. PNT. The variability of different ventilatory variables was assessed through their coefficients of variation (CVs). In healthy: according to Passing-Bablok regression, Rf, TI, TE and TT were interchangeable between measurements obtained under the three experimental conditions (SLP vs. PNT, SLP+PNT vs. SLP, and SLP+PNT vs. PNT). All the CVs describing “traditional” ventilatory variables (R<italic>f</italic>, T<sc>i</sc>, T<sc>e</sc>, T<sc>i</sc>/T<sc>e</sc>, and T<sc>i</sc>/T<sc>t</sc>) were significantly smaller in SLP+PNT condition. This was not the case for more “specific” SLP-derived variables. In COPD: according to Passing-Bablok regression, Rf, TI, TE, and TT were interchangeable between measurements obtained under SLP vs. PNT and SLP+PNT vs. PNT, whereas only Rf, TE, and TT were interchangeable between measurements obtained under SLP+PNT vs. SLP. However, most discrete variables were significantly different between the SLP and SLP+PNT conditions and CVs were significantly lower when COPD patients were assessed in the SLP+PNT condition. Measuring ventilatory activity with SLP preserves resting tidal breathing variability, reduces instrumental observer effect and avoids any disruptions in breathing pattern induced by the use of PNT-mouthpiece-nose-clip combination.</p>