Which cutoff value of the Montreal Cognitive Assessment should be used for post-stroke cognitive impairment? A systematic review and meta-analysis on diagnostic test accuracy

Abstract

Background and Purpose: Post-stroke cognitive impairment (PSCI) is one of the serious complications of stroke. The Montreal Cognitive Assessment (MoCA), as a brief cognitive impairment screening tool, is widely used in stroke survivors. However, some studies have suggested that the use of the universal cut-off value of 26 may be inappropriate for detecting cognitive impairments in stroke settings. We conducted this study to identify the optimal cut-off value of the MoCA in screening for PSCI. Methods: PubMed, CINAHL, Embase, the Cochrane Library, and Web of Science were searched for eligible studies until March 23, 2023. All studies were screened by two independent researchers. The quality of each article was evaluated by the Quality Assessment of Diagnostic Accuracy Studies-2 tool. A bivariate mixed effects model was used to pool sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, diagnostic odds ratio, and the summary receiver operating characteristic curve. Results: 24 studies with a total of 4,231 patients were included in this review. Despite the lack of evidence of publication bias, a high degree of heterogeneity was observed. Meta-analysis revealed that a cut-off value of 21/22 yielded the best diagnostic accuracy. The optimal cut-off varied in different regions, stroke types, and stroke phases as well. Conclusion: The optimal cut-off of MoCA was 21/22 for stroke populations rather than the initially recommended cut-off of 26. A revised (lower) cut-off should be considered for stroke survivors.

What does cognitive screening reveal about early cognitive performance following endovascular clot retrieval and intravenous thrombolysis in acute ischaemic stroke?

“…3). Alternatively, using the strokespecific cut-off score of <22 suggested by Wei et al (2023), patients who were managed conservatively were significantly more likely (P = 0.009) to be classified as cognitively impaired compared to those treated with ECR and t-PA.…”

“…The MoCA is a quick and easy-to-administer 30-point cognitive screening tool covering several cognitive domains, including visuospatial and executive skills, memory, attention, language, abstract reasoning, and orientation (Nasreddine et al 2005). The MoCA was originally developed to detect mild cognitive impairment and dementia; however, it has since been widely validated in the stroke population and is a feasible screening tool for the detection of cognitive impairment in the acute stages after stroke (Chiti and Pantoni 2014;Potocnik et al 2020;Munthe-Kaas et al 2021;Wei et al 2023). There is, however, some debate regarding the optimal cut-off score for the detection of post-stroke cognitive impairment, with a recent meta-analysis finding the optimal cut-off score is 21/ 22 and not the original recommended cut-off score of 26 (Wei et al 2023).…”

“…The MoCA was originally developed to detect mild cognitive impairment and dementia; however, it has since been widely validated in the stroke population and is a feasible screening tool for the detection of cognitive impairment in the acute stages after stroke (Chiti and Pantoni 2014;Potocnik et al 2020;Munthe-Kaas et al 2021;Wei et al 2023). There is, however, some debate regarding the optimal cut-off score for the detection of post-stroke cognitive impairment, with a recent meta-analysis finding the optimal cut-off score is 21/ 22 and not the original recommended cut-off score of 26 (Wei et al 2023). The MoCA was adopted for this retrospective study, as it was commonly administered to stroke patients at Monash Medical Centre as part of standard care during the data collection period.…”

Cognition after stroke

“…To address this, Wei et al. 15 performed a systematic review and meta-analysis of studies on MOCA diagnostic test accuracy, which is published in this month’s IJS. They conclude that the optimal cutoff of MoCA is 21/22 for stroke populations rather than the initially recommended cutoff of 26 and suggest we should use this revised (lower) cutoff.…”