Aim: The aim of this study was to describe lesion patterns, distribution, and evolution in posterior reversible encephalopathy syndrome (PRES) in a larger single-center population.
Methods: Scans and follow-up, if available, of 50 patients with PRES between 2002 and 2011 were reviewed retrospectively. Lesion patterns, extent, and signal intensity changes were identified and graded on fluid-attenuated inversion recovery (FLAIR) and diffusion-weighted images. Hemorrhagic changes were identified on T2* or susceptibility-weighted images, and gadolinium enhancement on T1-weighted images was identified if available.
Results: The most frequently affected regions on FLAIR were the frontal lobes in 54 %, occipital lobes in 34.3 %, and parietal lobes in 31.0 % of cases, thus 65.3 % in the posterior regions. Temporal lobes were affected in 10.6 %, the cerebellum in 6.5 %, and basal ganglia in 1.6 %. Division into vascular supply showed involvement in the anterior circulation in 66.5 % and in the posterior circulation in 33.5 % of cases. On diffusion-weighted imaging (DWI), vasogenic edema was observed in 6.9 %, cytotoxic edema in 9.1 %, and both in 2 % of cases. In 31.9 %, there was shine through, and in 15.9 %, there was shine through as well as cytotoxic or vasogenic edema. Topologic distribution on DWI showed affection of the frontal lobes in 43.5 %, occipital lobes in 25.8 %, parietal lobes in 17.7 %, temporal lobes in 11.3 %, and cerebellum in 1.6 %. T2* or susceptibility-weighted images showed spot-like hemosiderin accumulation in 17.2 % of cases. In 23.1 %, enhancement was seen. Follow-up magnetic resonance imaging showed complete resolution in 66.6 % of patients.
Conclusion: The spectrum of imaging findings in PRES is wide. Almost always subcortical and cortical structures are involved. Although posterior changes are prominent in this syndrome, frontal involvement is more frequent than posterior on FLAIR imaging and DWI. On DWI, mixed patterns are not uncommon. Reversibility generally takes place independent of DWI pathology. Hypertension was not a prognostic factor.