Article
Neurocognitive symptom clusters in patients with newly diagnosed glioblastoma
Neurokognitive Symptom-Cluster bei Patienten mit neu diagnostiziertem Glioblastom
Search Medline for
Authors
Published: | May 25, 2022 |
---|
Outline
Text
Objective: Many patients with glioblastoma (GB) suffer from neurocognitive deficits. This can cause severe difficulties to manage daily-living activities and reduce quality of life. It is important to identify the affected cognitive domains in order to provide specific support. A single test is usually insufficient to characterize the cognitive deficit. The aim of this study is to identify clusters of tests representing different neurocognitive domains.
Methods: 228 (146 males, 82 females, median age 63 years [24-83 years]) patients with newly diagnosed unilateral GB were tested prior to therapy with a set of 11 standard neurocognitive tests (cf. Results). A possible influence of age, gender and education level was examined by an analysis of variance. The data were normalized to percentile ranks. A hierarchical cluster analysis based on Pearson correlation ratios followed by a bootstrap procedure (using R 4.0.0, package “ClustOfVar”) was performed (i) to test the stability of the partitions and find a suitable number of clusters using the adjusted Rand index (ARI, ranging from 0 – 1) and (ii) to evaluate the similarity between grouped tests as quantified by the within-cluster sum of squares (WSS).
Results: For a partition of the data in 6 clusters, substantial stability was obtained (mean ARI=0.75; overall gain in cohesion of 51%). The symptom clusters were classified as follows: (1) lexical fluency and verbal memory (Hopkins Verbal Learning Test, Controlled Oral Word Association Test; WSS=2.97), (2) working memory and executive function (Stroop Colour & Word Test, Digit Span Test; WSS=1.65), (3) graphomotor speed and visuospatial construction ability (Digit Symbol Test, Trail Making Test, 9 Hole Peg Test, Complex Figure Test: copy trial; WSS=1.93), (4) visuospatial attention (Symbol Cancellation Test, WSS=1.00), (5) visuospatial judgement (Judgment of Line Orientation Test; WSS=1.25) and (6) visuospatial memory (Complex Figure Test: delayed recall trial; WSS=1.94).
Conclusion: A wide range of cognitive domains - including lexical fluency, working memory and executive functions, visuospatial attention, judgement and memory - can be investigated by the set of 11 neurocognitive tests, standardized for GB patients. Allocating grouped test results to stable domain clusters can not only increase the test validity compared to a single test, but also provides the caregiver with a routine interpretation strategy and help to identify the patients' need for specific support.