using the Alberta Infant Motor Scale (AIMS), Neuro-Sensory Motor Developmental Assessment (NSMDA) and Touwen Infant Neurological Examination (TINE). The relationships between GMA at four timepoints and the 12-month outcome assessments were examined using regression models. Results: Abnormal GMA at all four timepoints were associated with worse continuous scores on the AIMS, NSMDA and four TINE clusters (p=<0.05). Abnormal GMA prior to term and at term-equivalent were associated with increased odds of mild-severe dysfunction on the NSMDA (odds ratio [OR] 4.26; 95% CI 1.55, 11.71; p=<0.01 and OR 4.16; 95% CI 1.55, 11.17; p=<0.01, respectively) and abnormal GMA prior to term with increased odds of suboptimal-abnormal motor function on the TINE (OR 2.75; 95% CI 1.10, 6.85; p=0.03). There was little evidence of an association between abnormal GMA and abnormal motor function according to the AIMS cutoff score (< 5th centile). Conclusions/Significance: Abnormal GMA prior to term and at term-equivalent age were associated with adverse neurodevel- opment at 12 months’ corrected age in children born very preterm. The inclusion of early GMA in current clinical prac- tice may facilitate earlier targeted intervention and referrals in this high-risk population. J9 Detailed scoring of general movements in high- risk preterm infants is related to early brain structure and 2 year outcomes C PEYTON 1 , M MSALL 2 , M SCHREIBER 2 , C EINSPIELER 3 , A DROBYSHEVSKY 4 1 University of Chicago, Comer Children’s Hospital, Chicago, IL, USA; 2 University of Chicago, Chicago, IL, USA; 3 Medical University of Graz, Graz, Austria; 4 Evanston Northshore University Healthsystem, Evanston, IL, USA Background and Objective(s): The detailed quality of infant gen- eral movements, may be a marker of early brain dysfuntion and may reflect white matter (WM) integrity underlying neu- rodevelopmental outcomes. Tract-based spatial statistics (TBSS) is a observer-independent tool that analyzes multi- subject diffusion data for WM microstructure. Our objective was to describe in preterm infants the relationships between detailed ratings of general movements at 10–15 weeks post- term age (PTA), WM microstructure at term-equivalent age (TEA) and 2 year neurodevelopmental outcomes on the Bay- ley Scales of Infant and Toddler Development (BSID-III). Study Design: Prospective cohort study design of very preterm children enrolled in a neuroprotective clinical trial. This is an observational analyses of the cohort masked to treatment sta- tus and primary outcome. Study Participants & Setting: 48 infants born at ≤ 31 weeks ges- tational age, with a birthweight <1500gms, and requiring sup- plemental oxygen were recruited at a university hospital. Materials/Methods: The infants received a cerebral MRI scan at TEA. The infants were next assessed at 10–15 weeks PTA. The detailed scoring of the GMA was completed with video analysis by two expert GMA certified testers using Prechtl’s methodology. Subjects returned at two years of age and were tested with the BSID-III. Fractional anisotrophy (FA) were calculated by using the Diffusion Toolbox in FMRIB. All sub- jects’ FA images were aligned to a target in a common space by using an optimized TBSS protocol for neonates. The aligned images were then used to create another mean FA map and a mean FA skeleton, which represented the centers of all tracts common to the group. Voxelwise cross-subject statistics was performed to assess the relationship between FA and detailed GMA scores. In all analyses, a p value <0.05 was considered statistically significant. Results: Of the 48 infants enrolled, 4 infants were lost to fol- low-up at 2 years of age and one infant died, leaving 43 infants (90%) for analysis. Mean age at BSID-III testing was 23.90 months. Infants with lower detailed scores on the GMA had significantly smaller FA in the corpus callosum, posterior limb of the internal capsule and left cerebral peduncle. Detailed scores on the GMA at 10–15 weeks were significantly related to outcomes on the BSID-III at 2 years of age: motor (R2=19%, p=0.004), language (R2=12%, p=0.022), and cogni- tive (R2=13%, p=0.016). Conclusions/Significance: The quality of spontaneous motor behavior in infants is related to early white matter microstruc- ture and 2 year neurodevelopmental outcome. These findings suggest early motor repertoire may be an indicator of struc- tural connectivity and a biomarker for future neurodevelop- mental performance. J10 The Specific Test of Early Infant Motor Performance (STEP) predicts Bayley outcomes P COKER-BOLT 1 , L GOWER 2 , H MOSS 2 , T BROWN 2 , V RAMAKRISHNAN 2 , D JENKINS 2 1 Academic Medical Center, Charleston, SC, USA; 2 Medical University of South Carolina, Charleston, SC, USA Background and Objective(s): The Specific Test of Early Infant Motor Performance (STEP) has been introduced as a quick developmental test to screen preterm infants who are at risk for motor delays. For validation, we must determine the STEP’s ability to predict later neurodevelopmental outcome. The specific aim was to determine the relationship of early STEP scores at 0–3 months corrected age (CA) with neuroimaging at term CA and Bayley-III scores at 1-year. Study Design: Prospective cohort study of preterm infants at- risk for developmental delays. Study Participants & Setting: Preterm infants born between 24– 34 weeks gestational age (GA); discharged by 44 weeks GA. Exclusion criteria were presence of congenital brain abnormal- ities or other major congenital abnormalities. All procedures were completed at an infant developmental research lab. Materials/Methods: Outcome measures included the STEP at 0–3 months, Bayley-III gross motor and cognitive scaled scores at 12 months CGA, and Magnetic Resonance Spec- troscopy (MRS) and Diffusion Kurtosis Imaging (DKI) at term CGA. A Receiver Operating Characteristic (ROC) Curve determined maximal sensitivity and specificity of STEP scores for Bayley III gross motor scores. STEP cutoff scores for high and low-risk groups were then used in generalized linear mod- els to predict neuroimaging abnormalities in MRS metabolites 64 Abstracts