Researchers from the University of Bologna, Italy, studied the relation of amplitude integrated EEG findings in the first 24 hrs of life to brain metabolic changes, detected by proton MR spectroscopy (H-MRS) at 7-10 days of life, in 32 term newborns with hypoxic-ischemic encephalopathy (HIE). a-EEG at 6, 12 and 24 hrs of life was significantly correlated with outcome, and showed improvement in newborns with normal H-MRS and good outcome and a deterioration in those with abnormal H-MRS and poor outcome. a-EEG time course documents the severity, as defined by H-MRS, and evolution of brain damage following an HI event in non-cooled newborns. Both a-EEG and MRS are correlated with outcome.
At follow-up till 2 years of age in 31 surviving patients, 21 (67.7%) had a normal motor outcome, 7 (22.6%) developed transitory hypertonia, and 3 (9.7%) developed CP. Newborns with normal motor outcome or transitory hypertonia had a mean General Quotient of 104 on the Griffiths Mental Development Scale. BAER was normal in all cases; a cortical visual impairment occurred in 2 of 3 patients with CP. Eight patients (25%) developed neonatal seizures; the a-EEG background was severely abnormal in 3 and moderately abnormal in 5 cases at 6 hrs of life. Newborns with a normal a-EEG background at 6 hrs developed normally, whereas 2 with severely abnormal EEG background had a poor prognosis (1 died, 1 had CP). Of the 5 with moderately abnormal EEG background pattern at 6 hrs, 2 with a deteriorating pattern at 12 hrs developed CP, whereas 3 unchanged at 12 hrs and recovered at 24 hrs showed a normal outcome. Electrographic seizures did not change prognostic accuracy of a-EEG background pattern. MRI abnormalities were observed in 5/31 patients. 
COMMENT. Seizure activity dedected at a-EEG in HIE newborns was associated with a poor outcome only in patients with abnormal a-EEG background pattern. Patients with seizures but normal EEG background at 24 hrs of life had a normal outcome. Background pattern on the a-EEG at onset of seizures or status epilepticus is the best predictor of outcome. 
H-MRS shows a better correlation with outcome of newborns with HIE compared to conventional MRI. a-EEG has a high sensitivity for alterations in the H-MRS performed in the subacute period after perinatal asphyxia, and both a-EEG and H-MRS have a similar sensitivity and specificity for the prediction of outcome of HIE in term newborn. These a-EEG data obtained at 24 hrs of life may be useful in prediction of subsequent development of seizures in newborns with HIE.