Aims: Cognitive decline in SCD may manifest despite the absence of overt stroke and adds considerable burden to patients, including limitations in academic performance and day-to-day life. Data on objective screening of cognitive function in SCD is scarce, especially in the age group>50 years. Our aim was to evaluate the performance of SCD adult patients in the Montreal Cognitive Assessment (MoCA) questionnaire, and to identify possible predictive factors for test performance, including anatomical findings of brain imaging. Methods: This retrospective study included 61 SCD patients in which MoCA was applied as part of routine care. This test can detect mild cognitive alterations and was validated to the Portuguese language and in SCD. Seven domains of cognitive function were assessed: visual-spatial/executive, naming, attention, language, abstraction, delayed recall and orientation, in a maximum of 30 points. Demographic, clinical and laboratory data acquired from patients’medical records were evaluated for associations with the MoCA score. Magnetic Resonance Imaging with voxel-based morphometric analysis (VBM) was used to assess macro-structural changes in the brain tissue of 28 of these patients. Results: The median age of patients was 46 years (21-66), with 33 females and 28 males, distributed as: 40 homozygous SS, 13 SC, 5 Sβ0 and 3 Sβ+. Hydroxyureia (HU) was used by 45 patients (35 SS, 2 SC, 5 Sβ0 and 3 Sβ+). The median MoCA score was 22 points (8-30), with 50 (82%) of the patients scoring<27, and therefore considered to have some level of cognitive decline. As expected, MoCA score declines with aging (Pearson correlation: r=-0.4; p=0.002). Interestingly, there was no association with income but we observed a strict relation with educational status of the patients (r=0.62; p<0.01) and of their parents (mother: r=0.46, p<0.01; father: r=0.52, p<0.01). We found no relationship with hemoglobin levels, cell counts, hemolysis/inflammation markers, genotype, HU use, clinical complications (stroke, retinopathy, bone necrosis, leg ulcers, and chronic kidney disease) or transfusion load throughout life. On the other hand, there was an inverse correlation with proteinuria levels (r=-0.34, p=0.01). Logistic regression confirmed age, lower education and proteinuria as independent predictors of worse performance in MoCA. The longitudinal morphometric comparison of images of the gray matter of patients in relation to 50 healthy controls showed areas of significant atrophy in patients, mainly in cerebellum, left frontal and occipital lobes and bilateral temporal lobe. Interestingly, 8 of these patients had a new MRI after 1 year, and despite the short time interval, progressive atrophy was observed in right temporal and left frontal lobes and basal ganglia. Discussion: Our results demonstrate that cognitive decline is highly prevalent in SCD and progresses with age. The significant correlation with proteinuria may indicate an association of impaired cognition with disease severity, and the common MRI findings may reflect the significant changes in domains such as memory, learning and executive function. Conclusion: The relationship of cognitive decline with poor education demonstrates that this complication is even more expressive in situations of low intellectual stimulation (the reality of the majority of SCD patients), and should encourage the development of health policies to address this issue.