Leveraging Central Sleep Apnea Events to Validate the Measurement of Lung Volume Changes Using Thoracic Bio-Impedance.

Sleep disordered breathing can cause serious health issues, yet current diagnostic methods are cumbersome and prone to error. Thoracic electrical bio-impedance (BioZ) is a promising alternative, but it remains unclear whether the measured BioZ variations reflect lung volume changes. We leverage linear reductions in lung volume during central sleep apnea (CSA) events to assess whether BioZ measurements capture changes in lungs. BioZ signals from 92 sleep studies were analyzed using linear regression to quantify their slope and linearity (R2). Group differences were assessed, and a linear mixed-effects model was used to evaluate the impact of the body mass index (BMI), gender, and sleeping position. Welch's ANOVA showed significant differences between CSA and breathing segments. A chi-squared analysis showed that CSA events were more likely to exhibit negative slopes. The mixed-effects model found no BMI or gender effects, but the supine posture was significantly associated with more negative linear trends. These findings indicate that BioZ captures lung volume changes and that the sleeping position significantly modulates how clearly these changes appear in the signal.