Abstract Background Glioma patients often experience cognitive problems, which are associated with decreased functional independence and health-related quality of life. However, the prevalence and nature of cognitive impairment in these patients is relatively underreported, particularly during stable disease. In this study we determine the prevalence of cognitive deficits and explore distinct cognitive profiles of patients with diffuse glioma at least two months after tumor treatment. Material and Methods A total of 112 patients (mean age: 43 years) without clinical or radiological progression with a grade II-IV glioma were included in this observational cohort study. Cognitive functioning was assessed in five cognitive domains (attention, information processing speed, verbal memory, working memory, and executive functioning). Cognitive impairment was defined as a z-score of 1.5 SD below that of healthy controls. Hierarchical cluster analysis was used to examine cognitive profiles. Results In total, 46 patients (41%) had cognitive impairment. More than one domain was impaired in 25 of these patients (54%). The domains were affected in decreasing order of frequency: working memory in 31%, information processing speed in 22%, attention in 19%, verbal memory in 12%, and executive functioning in 11% of the patients. A clustered heatmap revealed three prominent cognitive clusters with an increasing number of cognitive domains impaired. The first cluster displayed patients with only working memory deficits, patients in the second cluster also had deficits in information processing speed, and the last cluster showed patients with impairments across all domains. Conclusion Cognitive impairment is highly prevalent in patients with diffuse glioma during stable disease. Working memory and information processing speed are most frequently affected. Importantly, our analyses show evidence for three subgroups in cognitively impaired glioma patients. Working memory seems an important driver in cognitive impairment given the involvement in all subgroups.
Read full abstract