A comparison between the MoCA and the MMSE visuoexecutive sub-tests in detecting abnormalities in TIA/stroke patients

Background Executive dysfunction predicts stroke risk, dementia, and mortality. The Montreal cognitive assessment detects more visuoexecutive dysfunction than the mini-mental state examination but it is unclear which of the individual Montreal cognitive assessment visuoexecutive items contribute to the better performance of the Montreal cognitive assessment. We therefore determined the relative performance of the Montreal cognitive assessment visuoexecutive sub-tests versus the mini-mental state examination pentagon copying in patients with stroke and transient ischemic attack. Methods Mini-mental state examination and Montreal cognitive assessment were administered to a prospective, population-based cohort of stroke, and transient ischemic attack patients from the Oxford Vascular Study at six month or five-year follow-up between November 2007 and June 2009. We compared the proportion of participants with incorrect Montreal cognitive assessment visuoexecutive tasks and sub-tasks but correct mini-mental state examination pentagon copying versus the proportion with incorrect MMSE pentagon copying but correct visuoexecutive Montreal cognitive assessment sub-test and individual sub-test items. Results Among 412 patients assessed with the mini-mental state examination and Montreal cognitive assessment, the Montreal cognitive assessment detected more visuoexecutive dysfunction than the mini-mental state examination (OR 11.4, 95% CI 8.2-15.8, p<0.001). The likelihood of incorrect mini-mental state examination pentagon copying increased as the numbers of correct MoCA visuoexecutive responses decreased: 2/106 (1.9%) and 9/10 (90.0%) incorrect mini-mental state examination pentagon copying for 5/5 and 0/5 correct Montreal cognitive assessment visuoexecutive tasks, respectively (p for trend 0.005). Each Montreal cognitive assessment visuoexecutive sub-task, including trails (39.6%), cube copying (49.5%), and clock drawing (59.0%), detected more patients with visuoexecutive dysfunction than the mini-mental state examination pentagon copying (20.6%, p<0.001). Conclusion All three of the Montreal cognitive assessment visuoexecutive sub-tests detected more abnormalities than the mini-mental state examination pentagon copying and thus contributed to the over 10-fold superiority of Montreal cognitive assessment over the mini-mental state examination for detection of visuoexecutive dysfunction.