Intrauterine growth restriction (IUGR) due to placental insufficiency affects 5–10% of all pregnancies, and is associated with fetal undernutrition and chronic hypoxia (1,2). It is difficult to address the differential effects of "early-onset" IUGR and prematurity (birth before 37 weeks of gestation), as they represent two different problems occurring simultaneously, and can both contribute to impaired neurodevelopment. To assess brain metabolite profile differences in 1-year-old IUGR infants born prematurely (P-IUGR), compared to premature and term adequate for gestational age infants (P-AGA and T-AGA, respectively), and their association with brain structural and biophysical parameters, and neurodevelopmental outcome at two years of age.26 P-IUGR infants (birth weight <10th cent. GA; umbilical artery Doppler > 95 th centile), 22 P-AGA and 26 T-AGA infants (birth weight ³10th cent. GA) underwent brain MRI and proton-MRS at one year of age during natural sleep, on a 3 Tesla scanner (TIM TRIO, Siemens DH, Germany). All brain T1- (MPRAGE: TR/TE/TI = 2050/2.41/1050 ms) and diffusion-weighted images (DWI, SE-EPI: TR/TE = 9300/94 ms, b = 0 and 1000 s/mm2, 30 directions) were acquired, as before (3). Single-voxel spectra (1H-PRESS) were also acquired from the frontal lobe, with TR/TE = 2500/30 ms, as before (4). T1-weighted MRI data were segmented to determine regional brain volumes, white matter (WM), gray matter (GM), and cerebro-spinal fluid (CSF) content (SPM v8.0, University College London, UK) (Fig. 1). Diffusion-tensor images were estimated from DWI data (MedInria, France) to compute regional fractional anisotropy (FA) and mean diffusivity (MD). Brain parcellation based on AAL atlas was performed as in (3). Measurements were averaged in the MRS voxel and in the frontal lobe, defined as a combination of the AAL frontal regions. Brain MRS data were quantified using linear fitting of individual metabolites (LC Model, S. Provencher, CA) (Fig. 2). Neurodevelopment was evaluated at two years using the Bayley Scales 3rd Edition (BSID-III), assessing cognitive, language, motor, social-emotional and adaptative behavior.P-IUGR infants had slightly smaller brain volumes and increased MD in frontal lobe WM (MDWM) compared to P-AGA and T-AGA controls. N-acetylaspartate to creatine ratios (NAA/Cr) were significantly lower in P-IUGR than in P-AGA infants, but increased in P-AGA compared to T-AGA infants. The P-IUGR group also showed slightly lower choline compounds/Cr (Cho/Cr), borderline decrements of estimated glutathione/Cr (GSH/Cr) levels, and increased myo-inositol (+glycine) to choline ratios (MI/Cho), compared to P-AGA controls (Fig. 3). These specific metabolite changes were correlated to brain tissue volumes - lower fractions of GM in the frontal lobe, and WM in the all brain - and to diffusion properties within the MRS voxel – higher MD in GM, and lower FA in WM (Table 1). P-IUGR infants also showed a tendency for poorer neurodevelopmental outcome at two years, associated with lower levels of NAA/Cr at one year within the preterm subset (Fig. 4).The structural and biophysical changes detected in the P-IUGR infant brain are consistent with previous findings (5). Lower brain NAA and increased MI/Cho have previously been associated with poorer neurodevelopment in infants (6,7), while apparent GSH decreases have also been reported in 1-year-old term-IUGR infants (4). The structural and metabolic changes detected in this work suggest decreased fiber myelination in the frontal lobe of P-IUGR infants, and potentially increased susceptibility to oxidative damage. Preterm IUGR infants show altered brain metabolite profiles during a critical stage of brain maturation, which correlate with brain structural and biophysical parameters, and neurodevelopmental outcome. Our results suggest altered neurodevelopmental trajectories in preterm IUGR and AGA infants, compared to term AGA infants, which require further characterization.