Introduction Lactate (Lac) / N-acetyl-aspartate (Naa) peak area ratio measured at long echo time on proton magnetic resonance spectroscopy (1H MRS) in the thalamus is a robust biomarker of outcome between day 5–14 in babies presenting with neonatal encephalopathy (NE).¹ Previous studies have been done at 2.35 and 1.5 Tesla; a specific median Lac/Naa threshold of 0.29 differentiated those infants with subsequent good and adverse outcome at 2 years. The optimum threshold may differ at higher field strengths (eg 3 Tesla) as the T2 values of Lac and Naa will be different to those at lower field.

Aim To determine whether the threshold for assigning prognosis is altered at 3 Tesla.

Patients and methods 42 infants with NE were scanned using 1H MRS (PRESS: TR = 2298 ms, TE = 288 ms, 1.5cm³ voxel in thalamus) at 3 Tesla. Of these, 26 (gestational age: 39.6 +/- 0.3 weeks, birth weight: 3155 ± 425 grams, postnatal age at scan 5.6 +/- 1.7 days; mean +/- SD) had 12 month clinical follow up using Bayley III scales. 20 infants had good outcomes (normal or mild impairment) and 6 had adverse outcomes (death or severe disability).

Results Figure 1 shows the ROC curve for Lac/Naa. Using a 0.29 threshold yielded a true positive rate of 100% for adverse outcomes with 2 false positive results. (PPV: 0.75; NPV: 1.00)

• Download figure
• Open in new tab
• Download powerpoint

Abstract PC.30 Figure 1

ROC curve for Lac/Naa at 3T

Conclusion A thalamic 1H MRS Lac/Naa peak area ratio threshold of 0.29 differentiates babies with subsequent normal and adverse outcomes at 3T.

Reference

1. Thayyil S, Chandrasekaran M, Taylor AM, Bainbridge A, Cady EB, Chong WK, Murad S, Omar RZ, Robertson NJ. In-vivo cerebral magnetic resonance biomarkers for predicting long-term neurodevelopmental outcome following neonatal encephalopathy: A meta-analysis. Pediatrics. 2010;125(2):e382–e395