Sepsis-associated encephalopathy (SAE) refers to central nervous system dysfunction during the systemic inflammatory response to infection. In septic patients with encephalopathy MRI has indicated both gray and white matter abnormalities that were associated with worse cognitive outcome including delirium. To improve our understanding of sepsis-associated hemodynamic, metabolic, and structural changes, different MRI sequences were performed in rats that either underwent an i.p injection of saline or bacterial lipopolysaccharide (LPS) 2.5h earlier or cecal ligation and puncture (CLP) 24h earlier. After ip LPS, phase contrast MRI was performed to study anterior and middle cerebral arteries flow and Arterial Spin Labeling (ASL) to study perfusion of white and grey matter brain structures. Diffusion Weighted Imaging (DWI) sequences was used to assess structural changes. After CLP surgery, ASL was used to study microcirculation changes. T2-Weighted Imaging, Diffusion Tensor Imaging (DTI) and tract-based spatial statistics (TBSS) were performed to characterize structural events in different brain structures. After imaging, animals were sacrificed and their brains processed for histology to detect the vasoactive prostaglandin-synthesizing enzyme cyclooxygenase-2 (COX-2) and the astrocytic aquaporin-4 water channel (AQP4) the expression of which can be upregulated during inflammation, to assess the presence of perivascular immunoglobulins (Ig) indicating blood-brain barrier (BBB) leakage and to study glia cell morphology as both microglia and astrocytes are known to change their morphology in inflammatory conditions. Magnetic resonance rat brain imaging indicated no hemodynamic changes in the grey matter after ip LPS administration while an increased CBF was shown in corpus callosum white matter as indicated by ASL. DTI indicated increased water diffusion parallel to fibers of the corpus callosum white matter. These changes were accompanied by BBB breakdown in the white matter and adjacent cortical and striatal grey matter as indicated by the perivascular presence of IgG, but no major changes in vascular COX-2 or white matter glia cell morphology. CLP induced sepsis-associated CNS dysfunction resulted in higher T2-weighted contrast intensities in the cortex, striatum and base of the brain, decreased blood perfusion distribution to the cortex and increased water diffusion in the corpus callosum and ventral striatum compared to sham surgery. These changes were associated in the white matter with modifications in glia cells morphology and in the grey matter with lower expression of constitutive COX-2 expression and AQP4 in the cerebral cortex. The comparison between CLP that underwent or not MRI under isoflurane anesthesia indicated reduced inflammatory response as indicated by COX-2 expression, reduced glia activation and reduced BBB damage in CLP that underwent MRI under isoflurane anesthesia. Collectively, our results suggest that hemodynamic changes may occur in the absence of altered flow in forebrain irrigating arteries. Then, altered white matter structure is an early step in SAE pathogenesis that may result either from BBB breakdown or glial cells activation. This study underlies the deleterious effects of a single exposure to isoflurane anesthesia that may be mitigated by a second exposure in sham-operated rats and the effects of CLP-induced systemic inflammation on glial cells that can be attenuated by imaging under isoflurane anesthesia.