Session S73.3

Time Frequency Relationships between Heart Rate and Respiration: A Diagnosis Tool for Late Onset Sepsis in Sick Premature Infants

G Carrault*, A Beuchee, P Pladys, L Senhadji, A Hernandez

Université de Rennes
Rennes, France

The diagnosis of late onset sepsis in premature infants remains difficult because clinical signs are subtle and non-specific and none of the laboratory tests, including CRP and blood culture, have high predictive accuracy. This lack of reliability of laboratory tests often results in anticipatory antimicrobial treatment. Heart rate variability (HRV) analysis emerges as a promising diagnostic tool. Entropy and long-range fractal correlation are decreased in premature infants with proven sepsis and presenting with frequent and severe bradycardias. Besides this, respiration and its relations to HRV appear to be less explored in the perspective of a multivariate monitoring system for the diagnosis of neonatal sepsis. The objective of this study was to determine if the heart rate behavior, respiratory amplitude and analysis of their relationships may help the diagnosis of infection in premature infants with cardiac decelerations. During the last decades, numerous works have been dedicated to the development of signal processing methods aimed at measuring the degree of association between signals. We use here a novel estimator of the linear relationship between nonstationary signals based on the cross correlation of narrow band filtered signals. The tests were performed on a cohort study of 60 premature infants (post-menstrual age <33 weeks and chronological age >72 hours), hospitalized in the neonatal intensive care unit at the university Hospital of Rennes over two periods June 2003 to June 2004 and January to July 2007, who presented severe (need of bag and mask to resuscitate) or frequent (more than 1 per hour) bradycardias. Results show that the non stationary estimator may enhance the readability of the time-frequency representation of relationship and, thus, can improve the interpretation of nonstationary interdependencies in signals. The results show that the correlation in the very low frequency band (VLF) tended to be higher in the sepsis group (r²VLF was significant >0.5). The highest correlation in higher frequencies (>0.02 Hz) was not associated with sepsis.

(Abstract Control Number: 176)