Session S22.6

Interaction between Heart Rate Variability and Respiration under Vibro-Tactile Stimulation in Premature Infants with Apnea

P Indic*, EB Salisbury, D Paydarfar, EN Brown, R Barbieri

Massachusetts General Hospital
Boston, MA, USA

Apnea often occurs in premature infants due to immaturity of developing systems and has been observed in more than 70% of infants born under 35 weeks post conception. Although it is well documented that bradycardia and hypoxia often follows apnea, the impact of stimuli on cardio-respiratory interaction has not been sufficiently investigated. In this work we propose to study this interaction by multivariate autoregressive analysis of heart rate variability (HRV) and respiration (RESP). We have considered 10 subjects under a reactive experimental protocol. In this protocol, a vibro-tactile stimulus is applied through a mattress when a pause in breathing greater than 5 seconds is detected by a computerized automatic system. This innovative non-invasive procedure has been suggested to reduce the occurrence of apnea episodes in premature infants.
For our analysis, the ECG and the abdominal respiratory recordings are divided into different sub epochs based on the condition of the mattress (OFF vs. ON). The RR interval is extracted from the raw ECG signal using a peak detection algorithm and then visually inspected for erroneous or missing detection of the R peaks. Both the RR and RESP series are sampled at 3Hz prior to application of a bivariate autoregressive model. Classification of traditional frequency ranges for studying the interaction of HRV and RESP is not applicable for the premature infant, as the normal breathing frequency is around 1 Hz. Hence in addition to the low frequency (LF: 0.01-0.15Hz) and high frequency (HF: 0.15-0.45 Hz) ranges, we introduce two new frequency regions: the Super High Frequency (SHF: 0.45-0.7Hz) and Ultra High Frequency (UHF: 0.7-1.5Hz). Using the bivariate model, we have computed the maximum coherence in all these ranges. Our analysis revealed significantly higher coherence in the UHF range compared to other ranges during both OFF and ON conditions; however, the magnitude of coherence averaged for the group was less than 0.5. The coherence is increased during ON compared to OFF condition in the LF range, whereas there is no significant increase in coherence in the other regions. These observations indicate a milder, but present, respiratory sinus arrhythmia in premature infants, as a relatively higher coherence is maintained in the UHF range. A significant influence of the vibro-tactile stimulus on the interaction between HRV and RESP at LF ranges is also observed, whereas not much influence is observed in the other frequency regions.

(Abstract Control Number: 171)