[Diagnostic value of amplitude-integrated electroencephalography in predicting outcome of newborn patients in neonatal intensive care unit].

Objective: To assess the diagnostic value of amplitude-integrated electroencephalography (aEEG) in predicting outcome of newborns who were at high risk for central nervous system without severe hypoxic-ischemic encephalopathy.
Methods: Forty-two consecutive patients at risks for neurological disorders referred to our level-III NICU were prospectively enrolled in the study over a period of 3 years. They were classified on the basis of their primary diagnoses including hypoglycemic brain damage, meningoencephalitis, bilirubin encephalopathy, and metabolic disease. Clinical data were collected. Amplitude-integrated and raw EEG tracings were assessed for background pattern, sleep-wake cycling, and epileptiform activity. The neuromotor development of survivors was assessed by using the Infant Neurological International Battery (INFANIB).
Result: The characteristic of aEEG tracings in 42 infants showed continuous normal voltage (CNV)(n = 15), discontinuous voltage (DC)(n = 9), burst-suppression (BS) BS(+) (n = 6), BS(-)(n = 7), flat (FT, n = 5); mature sleep-wake cycling (SWC, n = 4), immature SWC (n = 14), no SWC (n = 24); 30 infants (71.4%) had electrical seizures: single seizure (n = 6); repetitive seizures (n = 7), and status epilepticus (SE) (n = 17).aEEG of 20 infants who had poor outcome showed FT (n = 5), BS(-)/SE (n = 6), BS(-)/ repetitive seizures (n = 1) , BS(+)/SE (n = 1), BS(+)/repetitive seizures (n = 1), DC/SE(n = 6). Chi-square analysis and Spearman rank correlation analysis showed the classification of aEEG background pattern, SWC and comprehensive score (score system was developed by evaluation of the above 3 variables) were correlated with the outcome of these infants at high neurological risks.
Conclusion: Amplitude-integrated electroencephalography can provide important information of the status of cerebral function in neonates at high neurological risk and help to predict their outcome.