B98-08 Asthma’s Kinematic Fingerprints Remain Detectable When Standard-of-Care Physiomarkers Are Silent

Abstract Background We validated ARK (Analysis of Respiratory Kinematics), a novel method which quantifies thoraco-abdominal wall motion with breathing. This approach confirmed what seasoned clinician-educators have long taught: that breaths encode more physiological information than the respiratory rate alone, and that this additional information is prognostically valuable. Pediatric asthma is a uniquely compelling application for ARK: early detection is notoriously difficult; the pathophysiology is rooted in thoraco-abdominal wall mechanics; with 7% of U. S. children affected and 50 billion in annual direct costs, advances in predictive monitoring could improve outcomes while reducing costs on an enormous scale. We sought to assess the technical feasibility and discriminative utility of respiratory kinematic measurements in assessing pediatric asthma. Methods We recorded 5 minutes of upper rib, lower rib, and abdominal motion using inertial sensors in 23 healthy and 22 asthmatic awake children lying supine with a 30° head elevation. Recordings with ≥2. 5 minutes of artifact-free signal were analyzed. We extracted 34 clinically meaningful kinematic features. We used two-sample t-tests to compare the kinematics of healthy and asthmatic children, applying Bonferroni correction for the 34 comparisons. For features that differed by diagnosis, we assessed interactions with age, sex, and race using linear regression and evaluated correlations with pulmonary function tests (PFT) in the asthma group. Results The participants’ age (median 10 years; range 4-17), sex (65% male), and race (65% White, 12% Black) did not vary by diagnosis (healthy vs asthmatic). Standard physiomarkers (median; interquartile range) were largely normal in asthmatics, including respiratory rate (22 breaths/minute; 18-26), oxygen saturation (98%; 97-99), and FEV1/FVC ratio (82; 75-86). The acceptability criterion (≥2. 5 minutes of clean signal) was met in 91% of the recordings. The most discriminative feature (See Figure: “Rib-Abdomen Dominance Index”) quantified whether breathing was rib-dominant (positive values), abdomen-dominant (negative values), or balanced (near zero). Asthmatic children had significantly more abdomen-dominant breathing than healthy children (-1. 2 vs -0. 02; corrected p 0. 01) ; abdomen-dominance was associated with lower FEV1/FVC ratios. Sex and race interaction trends were observed but not significant due to low power. Conclusions Respiratory kinematic features were feasible to acquire and revealed meaningful differences – even in stable asthma with near-normal standard-of-care markers. These kinematic fingerprints – indelible even in clinically quiescent disease – offer a new window into asthma phenotyping. Larger studies of kinematic phenotypes could deliver high-performing early warning systems – bringing order to the chaotic lives of children with asthma, their families, and physicians This abstract is funded by: UVA Wallace H. Coulter Center for Translational Research