Localised proton magnetic resonance spectroscopy of the brain after perinatal hypoxia: a preliminary report

Objectives. Perinatal hypoxic ischaemic injury is a significant cause of ne urodevelopmental impairment. The aim of this study was to evaluate localise d proton magnetic resonance spectroscopy (H-1-MRS) after birth asphyxia. Materials and methods. Thirty newborn infants suspected of having perinatal asphyxia (Apgar score < 3) were studied. The mean gestational age was 37 w eeks; mean age at the MR examination was 18 days and mean weight was 2.9 kg . A 1.5-T unit was used for imaging and spectroscopy. None of the babies ha d mechanically assisted ventilation, No sedation was used. Axial T1-weighte d and T2-weighted images were obtained. H-1-MRS was recorded in a single vo xel, localised in white matter, using a STEAM sequence. Results. Image quality was good in 25 of 30 babies. H-1-MRS was performed i n 19 of 30 subjects, with adequate quality in 16. Choline, creatine/phospho creatine and N-acetylaspartate peaks and peak-area ratios were analysed. La ctate was detected in four infants. The N-acetylaspartate/choline ratio was lower in infants with an impaired neurological outcome, but the difference was not statistically significant. Conclusions. This study suggests that H-1-MRS may be useful for assessing c erebral metabolism in the neonate. A raised lactate level and decreased N-a cetylaspartate/choline ratio may be predictive of a poor outcome. However, in our experience this method is limited by the difficulty in performing th e examination during the first hours after birth in critically ill babies, the problems related to use of a monovoxel sequence, the dispersion of the ratios and the lack of determination of the absolute concentration of the m etabolites.