Presented at the Neonatal Society 2016 Summer Meeting.
da Costa CS1, Placek MM2, Czosnyka M3, Smielewski P3, Cabella B3, Kasprowicz M2, Austin T1
1 The Rosie Hospital, Cambridge University Hospitals NHS Foundation Trust, UK
2 Department of Biomedical Engineering, Wroclaw University of Technology, Wroclaw, Poland
3 Department of Clinical Neurosciences, Academic Neurosurgical Unit, University of Cambridge
Background: A healthy biological system is one that has the ability to react and adapt to physiological changes in its environment. When complexity is reduced or lost, a pathological status is potentially present. Multiscale entropy (MSE) analysis has been used to assess the complexity of biological systems (1). In neonates, near-infrared spectroscopy (NIRS) has been used to assess cerebral oxygenation and fluctuations in cerebral blood flow. Our aim was to apply MSE analysis to assess the complexity of NIRS signals and its correlation with outcome.
Methods: Prospective observational study of 61 preterm infants with a median (range) gestation age (GA) of 26+6 weeks (23+3-31+0) with an indwelling arterial catheter. All infants were studied before 24 hours of age, following parental consent. A NIRS sensor was placed on the infant’s temporoparietal area of the head. NIRS signals as oxygenated haemoglobin (HbO2), deoxygenated haemoglobin (Hb), tissue oxygenation index (TOI) and tissue haemoglobin index (THI) and systemic signals as continuous mean arterial blood pressure (MABP), heart rate (HR) and arterial saturation (SaO2) were collected using ICM+ software (2).
Results: 20 infants had intraventricular haemorrhage (IVH) during the admission on the unit and 8 infants died before corrected GA at term. MSE was lower in all scale factors in those infants who had IVH and those who died. Lower complex index (CoI) of HbO2, CoI-Hb and CoI-TOI were observed in those infants who developed IVH compared to those who did not (P=0.002, P=0.010 and P=0.003 respectively). Mean CoI-HbO2, CoI-Hb and CoI-THI were lower in those infants who died compared to those who survived (P=0.002, P=0.004 and P=0.003, respectively). Binary logistical regression identified CRIB II (P=0.031), sepsis (P=0.025) and CoIHbO2 (P=0.004) as predictors of IVH when they were analyzed independently. When these variables were used in a multiple logistical regression model, only CoI-HbO2 remained independently associated with IVH. The only Col of systemic signals associated with clinical outcome was Col-MABP, which was associated with mortality.
Conclusion: Decreased complexity of brain signals was associated with IVH and mortality, supporting previous studies in adults and children, showing that decreased complexity of biological signals is associated with poor outcome (3). CoI-HbO2 decreased complexity was independently associated with an increased likelihood of having IVH.
Corresponding author: firstname.lastname@example.org
1. Costa M et al. Multiscale entropy analysis of complex physiologic time series. Phys Rev Lett. 2002;89(6):068102.
2. Smielewski P et al ICM+: software for on-line analysis of bedside monitoring after severe head trauma. Acta neurochirurgica Supplement. 2005;95:43-49.
3. Norris PR, Anderson SM, Jenkins JM, Williams AE, Morris JA, Jr. Heart rate multiscale entropy at three hours predicts hospital mortality in 3,154 trauma patients. Shock. 2008;30(1):17-22.