Cognitive screening for impairments following stroke

Abstract

The primary motivation for this thesis is to define cognitive impairments following stroke, to assess them in a valid and practical manner, and to discriminate post-stroke cognitive profiles from other types of cerebrovascular disease (CVD). First, an overview of 33 diagnostic test accuracy systematic reviews is presented to identify the main conditions for the assessment of patients with CVD and the most sensitive screening tools recommended for them. The reviews focused on the detection of cognitive impairment, dementia, and stroke evaluated by DSM guidelines and MCI diagnosed by Petersen criteria. Analyses show that MMSE is recommended for dementia, whereas for MCI and stroke, the most frequently reported tests are MoCA and ACE-R, with OCS also recommended in the recent stroke reviews. The overview found no reports on tests in Russian. Therefore, the online survey for Russian-speaking clinicians about the preferred methods of cognitive assessment and rehabilitation was conducted. 110 participants reported using MoCA, MMSE, and Luria Neuropsychological Battery for stroke, whereas for dementia, the FAB was also mentioned. These results show some compatibility between Russian and international practices. Following the overview and survey outcomes, a Russian version of the OCS was developed. The advantages of OCS over MOCA and MMSE are minimal demands on language production, hand movement, and visuospatial processing to increase sensitivity to executive functions, number processing, memory, language, and praxis abilities. After linguistic adaptation, normative data were collected on 60 healthy and 205 post-stroke participants to determine cut-off scores, convergent validity, test-retest reliability, internal consistency, sensitivity, and specificity. Results showed the Rus-OCS had a fair level of psychometric properties allowing to recommend it for further validation and clinical use. Additional work was conducted to investigate the discriminant ability of the Rus-OCS for patients with stroke (n = 175) and small vessel disease (SVD) (n = 41). The MoCA and Rus-OCS scores were compared in these groups, including the impact of age, gender, education, and the disease duration. The results suggest the use of MoCA as a primary screen to detect cognitive impairment, as it is faster and can discriminate stroke and control groups. However, for the clinical population, the use of Rus-OCS could be more informative, as it can contrast stroke and SVD groups by numerical, visuospatial, and naming subtests together with scores on language and memory domains. The concluding chapter presents a conceptual framework for stroke education with reference to securing new resources for patient education. A case study of a patient education initiative in Moscow was reported to illustrate this framework. 24 patients participated in the Stroke School for seven weeks. They were assessed with WHODAS.2.0 before and after the intervention and showed improvements in cognition, life activities, participation, and sociability domains. The results demonstrate how an educational initiative in the inpatient clinical settings may enhance the benefits of other rehabilitative interventions, although participants’ learning motivation should be considered. Overall, the results from this thesis establish a new benchmark for the future research of cognitive assessment and related applications in stroke rehabilitation for Russian speakers.