The last trimester of brain development in fetuses with complex congenital heart disease (CHD) is abnormal, with structural and metabolic abnormalities documented [1]. While resting state networks (RSN) are well characterized in preterm neonates, very little is known about the RSN in neonates with complex CHD. Importantly, healthy neonatal brain RSN has been characterized through the use of resting BOLD and Independent Component Analysis (ICA) [2]. In this study, we use ICA analysis on resting BOLD of CHD patients to characterize the neuronal activity and compare it to healthy Controls. A total of 157 term neonates were recruited and scanned from two large surgical centers for CHDs (Children’s Hospitals of Pittsburgh and Los Angeles). Of these, 117 (PCA at scan 37-46 weeks; CHD n=51, Healthy Controls n=66) were within acceptable criteria for motion in BOLD imaging, and included in the study. All were scanned using 3T Skyra (Siemens AG, Erlangen, Germany) in Pittsburgh or 3T Achieva (Philips Technologies, Hamburg, Germany) in LA. Following common sequence parameters were used: FOV=240mm, TE/TR = 32/2020 ms, with an in-plane resolution of 4x4 mm2 and slice thickness of 4mm with 4mm space. Custom pipeline in IDL[ref] for image reconstruction, brain extraction, affine motion correction, and spatial coregistration into MNI space (neonatal atlas) of BOLD images was used prior to analysis. Temporal concatenation ICA for resting analysis was conducted on CHDs and Controls as two separate groups using MELODIC on FSL (FMRIB) [3, 4].No statistically significant difference in distribution of CHD and Controls between sites (p=0.137). Total of 56 and 43 ICAs were generated for CHDs ad Controls, respectively. Of the CHD ICAs, 48% had unilateral only activation patterns compared to controls which had 23% with just unilateral activations. Default mode network, sensory, motor, cerebellum, brainstem-thalamus, and frontal executive RSNs were found in both controls and CHDs [5]. However controls exhibited additional networks such as auditory, visual eye field, fronto-parietal, and salience networks (Figure1), which were poorly delineated in the CHD patients. Notably, the areas associated with frontal executive network showed depressed activation in CHDs. In addition, complex pattern of activation involving both subcortical and cortical regions (i.e. cerebellum hemispheres and vermis, lingual gyrus, perisyvlian region, and anterior cingulate) was seen in a single component of the control group, but not in the CHD group (Figure2).Overall, the neonatal CHD group exhibited ICAs less complex patterns of activation then Controls. The CHD activation patterns were characterized by activation more limited to local regions and many of the components show patterns unilaterally constrained to one hemisphere. These ICA findings suggest CHD patients have global brain dysmaturation with disruption to callosal, intra-hemishperic and subcortical connectivity compared to healthy controls.