Regional neonatal brain absolute thermometry by 1 H MRS Alan Bainbridge a *, Giles S. Kendall b , Enrico De Vita c , Cornelia Hagmann b , Andrew Kapetanakis b , Ernest B. Cady a and Nicola J. Robertson b Therapeutic hypothermia is standard care for infants with moderate to severe encephalopathy. 1 H MRS thermometry (MRSt) measures regional brain absolute temperature using the temperature-dependent water chemical shift. This study evaluates the clinical feasibility of MRSt in human neonates, and correlates white matter (WM) and thalamus (Thal) MRSt with conventional rectal temperature (T rectal ) measurement. Fifty-six infants born at term underwent perinatal MRSt for suspected hypoxic–ischaemic brain injury and 33 infants born preterm had MRSt at a term-equivalent age; 56 of the 89 had T rectal measured after MRSt of either a Thal or posterior WM voxel, or both. MRSt used point-resolved spectroscopy (no water suppression; TR = 1370 ms; TE = 288 ms; 1.5  1.5  1.5 cm 3 Thal and 1.1  1.3  1.4 cm 3 WM voxels). Time domain data were phase and frequency corrected before summation and motion-corrupted data were excluded from further analysis using simple criteria [preprocessing + quality assurance (QA)]. Two published water temperature-dependence calibrations [both using cerebral creatine (Cr), choline (Cho) and N-acetylaspartate (Naa) as independent reference peaks] were compared. The temperature measurements derived from Cr, Cho and Naa were combined to give a single amplitude-weighted combination temperature (T AWC ). WM and Thal T AWC correlated linearly with T rectal (Thal slope, 0.82  0.04, R 2 = 0.85, p < 0.05; WM slope, 0.95  0.04, R 2 = 0.78, p < 0.05). Preprocessing + QA improved the correlation between WM T AWC and T rectal (R 2 increased from 0.27 to 0.78, p < 0.001). Both calibration datasets showed specific inconsistencies between the temperatures calculated using Cr, Cho and Naa reference peaks when applied to this neonatal dataset. Neonatal MRSt is clinically feasible. Preprocessing + QA improved MRSt reliability in WM. The consideration of MRSt calibration internal biases is necessary before combining MRSt temperatures from multiple reference peaks to obtain T AWC . Copyright © 2012 John Wiley & Sons, Ltd. Keywords: brain; cerebral; temperature; 1 H MRS; MRS; neonatal; thermometry; quality assurance INTRODUCTION Perinatal hypoxia–ischaemia affects 1–3 per 1000 live births in the UK and remains an important cause of death and disability (1,2). Over the last decade, several randomised clinical trials (3–7) have demonstrated that therapeutic cerebral hypothermia is a safe and effective treatment for term infants with hypoxic–ischaemic encephalopathy (HIE), and reduces mortality and neurodevelop- mental disability in survivors. Considerable work is still required to optimise cooling strategies for the newborn. Among the questions yet to be answered definitively include: what is the optimal temperature for clinical neuroprotection and what is the optimal method of delivery of therapeutic cerebral hypothermia? Published clinical results suggest that both selective head cooling combined with mild whole-body hypothermia (3) and whole-body cooling alone (4–7) are effective. Experimental models demonstrate that selective head cooling results in an increase in brain temperature between the scalp and deep brain, whereas the rectal tempera- ture (T rectal ) remains normal. This potential variation in regional brain temperature may have important implications for the efficacy of selective head cooling (8–10). Conversely, for whole- body cooling, T rectal is a useful surrogate for brain temperature (11). There is evidence that deep brain temperature is correlated with core body temperature during both normothermia and whole-body cooling. This evidence comes from direct measure- ments in experimental models (11–13) and in human adult neurosurgical patients (14,15). Currently, knowledge about regional brain temperature in healthy newborn infants and in HIE is limited, and the develop- ment of an accurate and precise method to measure regional brain temperature noninvasively may be useful in patients with acute * Correspondence to: A. Bainbridge, Department of Medical Physics and Bioengi- neering, University College London Hospitals NHS Foundation Trust, EGA Wing, Level –2, 235 Euston Road, London NW1 2BU, UK. E-mail: Alan.Bainbridge@uclh.nhs.uk a A. Bainbridge, E. B. Cady Medical Physics and Bioengineering, University College London Hospitals NHS Foundation Trust, London, UK b G. S. Kendall, C. Hagmann, A. Kapetanakis, N. J. Robertson Academic Neonatology, Institute for Women’s Health, University College London, London, UK c E. D. Vita Lysholm Department of Neuroradiology, National Hospital for Neurology and Neurosurgery, London, UK Abbreviations used: Cho, choline; CI, confidence interval; Cr, creatine; f Naa ,f Cr , f Cho ,f w , frequency of Naa, Cr, Cho and water peaks, respectively; HIE, hypoxic– ischaemic encephalopathy; LA, limits of agreement (used in Bland–Altman analysis); Lac, lactate; MRSt, MRS thermometry; Naa, N-acetylaspartate; ppm, parts per million of the main magnetic field strength; PT, preterm; QA, quality assurance; SD, standard deviation; SNR, signal-to-noise ratio; T AWC , amplitude- weighted combination temperature; Thal, thalamus/thalamic; T Naa ,T Cr ,T Cho , temperature measured using Naa, Cr and Cho peaks, respectively; T rectal , rectal temperature; WM, white matter. Research article Received: 2 July 2012, Revised: 28 August 2012, Accepted: 4 September 2012, Published online in Wiley Online Library: 16 October 2012 (wileyonlinelibrary.com) DOI: 10.1002/nbm.2879 NMR Biomed. 2013; 26: 416–423 Copyright © 2012 John Wiley & Sons, Ltd. 416