The Derivation of Epigastric Motion to Assess Neonatal Breathing and Sleep: An Exploratory Study

Abstract Introduction New non-medical monitors are offered for respiration monitoring of neonates. Epigastric motion during sleep was investigated by means of a wearable tracker in parallel to clinical monitoring. Cohort: 23 hospitalised neonates ready for discharge. Methods A 3-axes-accelerometer and -gyroscope was placed in a standard epigastric position. Between two routine care rounds signals were recorded in parallel to monitoring of impedance pneumography (IP), ECG and pulse oximetry. Motion signals vs. time charts were evaluated using 10-min episodes and semiquantitatively assigned to breathing signal quality, regular breathing, periodic breathing and confounding artefacts. The results were compared with the impedance pneumographic data. Results 26 recordings (mean duration: 210 min/infant) were conducted without bradycardia or apnea alarm. The gestational age at birth ranged 28.9 to 41.1 and at recording from 35.6 to 42.3 postmenstrual weeks. Motion patterns of quiet sleep with regular breathing, periodic breathing and active sleep with confounding body movements were found. The longitudinal and transversal gyroscope axes resulted in best signal quality. Periodic breathing was found in up to 80% of episodes and decreased inversely with gestational age showing significantly more periodic breathing in preterm infants. Respiration signals of the gyroscope vs. IP showed a low bias and highly variating frequencies. Conclusions Standardized motion trackers may detect typical neonatal breathing and body-motion-patterns, that could help to classify neonatal sleep. Respiratory rates can only be determined during quiet sleep.