Exploring visuospatial abilities and their contribution to constructional abilities and nonverbal intelligence

Which Test Is the Best to Assess Visuo-Cognitive Impairment in Patients with Parkinson's Disease with Mild Cognitive Impairment and Dementia? A Systematic Review and Meta-Analysis

BACKGROUND

Visuo-cognitive impairment is common in patients with Parkinson's disease with mild cognitive impairment (PD-MCI) and constitutes a prognostic factor for the conversion to Parkinson's disease dementia (PDD). However, systematic analyses on which neuropsychological tests are most suitable to assess visuo-cognition in PD-MCI and PDD and to differentiate these cognitive stages are lacking.

OBJECTIVE

To review neuropsychological tests used to assess visuo-cognition including visuo-perceptual and visuo-spatial processing, visuo-constructive copying and drawing on command abilities; and to identify the visuo-cognitive subdomain as well as tests most suitable to discriminate between PD-MCI and PDD.

METHODS

MEDLINE, PsycINFO, Web of Science Core Collection, and CENTRAL were systematically searched for relevant studies assessing visuo-cognitive outcomes in patients with PD-MCI and PDD. Risk of bias was assessed using a customized form based on well-established tools. Random-effect meta-analyses were conducted.

RESULTS

33 studies were included in the systematic review. Data of 19 studies were entered in meta-analyses. Considerable heterogeneity regarding applied tests, test versions, and scoring systems exists. Data indicate that visuo-constructive command tasks are the subdomain best suited to discriminate between PD-MCI and PDD. Furthermore, they indicate that the Rey-Osterrieth-Complex-Figure Test (ROCF), Corsi Block-Tapping Test, Judgment of Line Orientation (JLO), and Clock Drawing Test (CDT) are tests able to differentiate between the two stages.

CONCLUSION

We provide suggestions for suitable visuo-cognitive tests (Corsi Block-Tapping Test, or JLO, ROCF, CDT) to improve diagnostic accuracy. Methodological challenges (e.g., heterogeneity of definitions, tests) are discussed and suggestions for future research are provided.

REGISTRATION

https://www.crd.york.ac.uk/prospero/, ID: CRD42018088244.

Representational drawing following brain injury

Research has shown that damage to either the left or right hemisphere can lead to deficits in visuoconstructional skills including drawing and figure copying. Nevertheless, research would suggest that the nature of the deficits arising from left and right brain injury are distinct in nature if not severity, with the right hemisphere, and parietal cortex specifically, seen as critical for obtaining accurate spatial relations and the left hemisphere important for effective organisation (i.e., executive function). Much of this work on drawing and figure copying following brain damage has rested on qualitative assessments or crude marking scales with descriptive anchors for what constitutes good or poor performance. We employed quantitative analyses of drawings developed to assess accuracy in novice and expert artists. We analyzed drawings of a cube and a star in 50 patients (23, left brain damaged: LBD; 27 right brain damaged: RBD) who had suffered strokes. Our analysis was sensitive to the presence of neglect on the cube (i.e., missing left sided details) with voxel-wise lesion symptom mapping (VLSM) highlighting involvement of expected brain regions (superior temporal and supramarginal gyri). With left-sided omissions removed from analyses, we failed to find any difference between LBD and RBD patients. While the presence of left neglect appeared to exaggerate errors, this was only significant for errors of scale and proportion for the star drawing. VLSM of the distinct error domains demonstrated white matter involvement (and a minor contribution from the right insula) with respect to scale errors of the cube only. Finally, blinded judgements of hemisphere of lesion based on qualitative assessment of the drawings were no better than chance. These results suggest that figure copying is a complex task relying on large scale neural networks involving both hemispheres. Clearly, models of visuoconstructional capacity that emphasise right hemisphere dominance are not entirely accurate.