Meeting Banner
Abstract #2966

Lactate-Edited MRS in Neonatal Hypoxic Ischemic Encephalopathy

Ruth L. O'Gorman1, Carola Sabandal1, Vera Bernet1, Cornelia Hagmann2, Martin A. Janich3, Ralph Noeske4

1University Children's Hospital, Zurich, CH, Switzerland; 2University Hospital, Zurich, CH, Switzerland; 3Diagnostics & Biomedical Tech Laboratory, GE Global Research, Mnchen, Germany; 4Applied Science Laboratory, GE Healthcare, Berlin, Germany

This study investigates the relative sensitivity of lactate editing in comparison to standard PRESS MRS for detecting lactate in neonates with suspected hypoxic ischemic encephalopathy (HIE). Measured lactate concentrations and Lac/NAA ratios were higher for the lactate-edited spectra relative to corresponding values from standard PRESS (TE=35/144 ms). The short TE (35 ms) and lactate-edited spectra show comparable sensitivity to lactate but the specificity and fit reliability was higher for the lactate-edited spectra. Therefore, lactate-edited MRS can improve the specificity and reliability of lactate detection in infants with suspected HIE, particularly in the presence of strong lipid signals.

Keywords

absent according accurate acute addition adequate affect alternate applied approximately arch arising asphyxia assessed audience bandwidth basal basing basis biomedical births bounds brain broadband cerebral channel chemical child children coil collected combined comparable concentration concentrations condition confounded cutoff detecting detection diagnostics disability displacement doublet doublets edit edited editing encephalopathy errors evaluate example experimental fetal field ganglia global glycol head hospital hypoxia improve included indicator infant infants injury laboratory lactate lancet lawn lenient lipid live long male martin masked measured meta metabolites middle mortality neonatal neonates neurological outcome overlap particularly patient peaks pediatric pediatrics perinatal presence press prognostic propylene pulses receive recent reduce referred refocusing reliability represents researchers resolved risk scaled scanner science selective sensitivity serious sets short slice specificity spectra spectral spectrum strong studies suboptimal subtracted suspected table target tech term thought version visualized water written yield