Longitudinal evaluation of cognition after stroke - A systematic scoping review

Abstract

Background: Cognitive impairment affects up to 80 percent of the stroke population, however, both the available evidence about post-stroke cognition and the measures used to evaluate it longitudinally have not been well described. The aims of this systematic scoping review were: to identify and characterize studies evaluating cognition longitudinally after stroke; to summarize the cognitive instruments used and the domains they target; and to organize cognitive domains assessed using the International Classification of Functioning, Disability and Health (ICF).

Methods: We used a systematic scoping approach to search for peer-reviewed articles involving adults with stroke that evaluated cognition longitudinally. Screening of titles, abstracts, and full reports was completed independently by two reviewers, across six electronic databases (PubMed, PsycInfo, Medline, Cinahl Plus, Embase, and Web of Science). Cognitive domains were mapped to an ICF function independently by the same two reviewers, using a previously tested, standardized approach.

Results: A total of 5,540 records were found; 257 were included, representing a total pooled sample of 120,860 stroke survivors. Of these studies, 200 (78%) provided specific cognitive outcomes from the longitudinal evaluations, 57 (22%) reported model predictions, and 77 (30%) included interventions. Cognition was evaluated with 356 unique instruments, targeting 95 distinct cognitive domains, and 17 mental functions from the ICF. The Mini-Mental State Examination was the most frequently used instrument (117 reports, 46%). Other tools used longitudinally were the Trail Making Test (17% of reports), tests of verbal fluency (14%), the Functional Independence Measure (14%), the Montreal Cognitive Assessment (13%), the Digit Span (11%), and the Stroop test (10%). Global cognition was evaluated in 170 reports (66%), followed by higher-level cognitive functioning (29%), memory (28%), language (21%), attention (21%), and perceptual skills (14%). Studies using functional (or performance-based) cognitive assessments over time were scarce (< 1%).

Conclusion: Our findings indicate that whilst there is a substantial number of studies available that report longitudinal evaluations of cognition after stroke, there is large variability in the measures used and the cognitive domains they target. Nonetheless, the available data for evaluation of cognition over time after stroke can be organized and described systematically.

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Fig 1. Included studies and reasons for exclusion.

Fig 2. Sample size and origin of included studies (N = 257).

Fig 3. Twenty-five most frequently used cognitive tests and ICF domains across included studies.

ADAS-cog, Alzheimer's Disease Assessment Scale; AMT, Abbreviated Mental Test; BIT, Behavioral Inattention Test; Digit sp, Wechsler's Digit Span; FAB, Frontal Assessment Battery; FIM, Functional Independence Measure; HLCF, Higher-Level Cognitive Functioning; Log Mem, Wechsler's Logical Memory; MDRS, Mattis-Dementia Rating Scale; MMSE, Mini-Mental State Examination; MoCA, Montreal Cognitive Assessment; RAVLT, Rey Auditory Verbal Learning Test; ROCF, Rey-Osterrieth Complex Figure; TMT, Trail Making Test.

Fig 4. Cognitive tools most commonly used and the ICF functions they evaluate.

The top 7 instruments were used in 209 of 257 articles, representing 34% of the evaluations completed across all the studies. The 'Not reported' category was assigned to studies that did not specify what cognitive domain was being evaluated when using a specific assessment; The 'Other' category was assigned to indicate a cognitive domain (or group of cognitive domains) that is different from the ones listed for the specific tool. DS, Wechsler's Digit Span; FIM, Functional Independence; HLCF, Higher-Level Cognitive Functioning; ICF-ch1, International Classification of Functioning and Disability, chapter 1; MMSE, Mini-Mental State Examination; MoCA, Montreal Cognitive Assessment; TMT, Trail Making Test; VFT, Verbal Fluency Test.

Fig 5. Mental functions from the ICF evaluated by each instrument.

The 43 instruments described in this figure were used in 228 of 257 studies, and represent 74% of all the evaluations performed. Only common tools, used more than 3 times, across all papers were used to build this figure; adas, alzheimers disease assessment scale; amt, abbreviated mental test; avlt, auditory verbal learning test; bdt, block design test; bit, behavioral inattention test; camcog, cambridge cognition examination; cbt, corsi blocks test; cdt, clock drawing test; cowat, controlled oral word association test; dct, digit cancel task; dr, delayed recall test; ds, digit span test; dsym, digit symbol test; fab, frontal assessment battery; fct, figure copying test; fim, functional independence measure; gpt, grooved peg test; iq-code, informant questionnaire on cognitive decline in the elderly; lct, letter cancellation test; lm, logical memory test; mdrs, mattis-dementia rating scale; mmse, mini mental examination; moca, montreal cognitive assessment; nam, naming test; ravlt, rey auditory verbal learning test; rbans, repeatable battery for the assessment of neuropsychological status; rocf, rey-osterrieth complex figure test; sdmt, symbol digit modalities test; sim, similarities test; sis, stroke impact scale; sr, story recal test; stroop, stroop test (also called color word interference test); tmt, trail making test; token, token test; vft, verbal fluency test; vrt, visual reproduction test; wcst, wisconsin card sort test; wll, word list learning; wlr, word list recall & recognition.