Response to Letter by Blackburn et al Regarding Article, “Is the Montreal Cognitive Assessment Superior to the Mini-Mental State Examination to Detect Poststroke Cognitive Impairment? A Study With Neuropsychological Evaluation”

Abstract

HomeStrokeVol. 42, No. 11Response to Letter by Blackburn et al Regarding Article, “Is the Montreal Cognitive Assessment Superior to the Mini-Mental State Examination to Detect Poststroke Cognitive Impairment? A Study With Neuropsychological Evaluation” Free AccessReplyPDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessReplyPDF/EPUBResponse to Letter by Blackburn et al Regarding Article, “Is the Montreal Cognitive Assessment Superior to the Mini-Mental State Examination to Detect Poststroke Cognitive Impairment? A Study With Neuropsychological Evaluation” Olivier Godefroy, MD, PhD, Jean Marc Bugnicourt, MD and Andreas Fickl, MD Olivier GodefroyOlivier Godefroy Service de Neurologie CHU Nord Amiens, France (Godefroy,Bugnicourt,Fickl) Search for more papers by this author , Jean Marc BugnicourtJean Marc Bugnicourt Service de Neurologie CHU Nord Amiens, France (Godefroy,Bugnicourt,Fickl) Search for more papers by this author and Andreas FicklAndreas Fickl Service de Neurologie CHU Nord Amiens, France (Godefroy,Bugnicourt,Fickl) Search for more papers by this author Originally published22 Sep 2011https://doi.org/10.1161/STROKEAHA.111.630368Stroke. 2011;42:e583Other version(s) of this articleYou are viewing the most recent version of this article. Previous versions: January 1, 2011: Previous Version 1 Response:We thank Blackburn et al for their interesting comments on screening tests for poststroke cognitive impairment. Given the frequency and impact of poststroke cognitive impairment, choosing the best possible screening test is important. In view of the MMSE score's poor sensitivity to vascular cognitive impairment and our recent results,1 Blackburn et al suggest that the Montreal Cognitive Assessment (MoCA) should be used instead of the MMSE, with a cutoff ≤23. Although we agree with some of Blackburn et al's points, the currently available evidence (including our own study) does not fully support their position. First, our study supported the moderate sensitivity (0.66) of MMSE when recommended cutoffs are used.1 However, sensitivity improved using cut-off after adjustment of the score for education level (Table 21). Second, Blackburn et al indicate that use of a cutoff ≤23 for the MoCA yielded a sensitivity of 0.84 and a specificity of 0.81 as indicated in Table 3.1 We wish to emphasize that this cut-off does not concern the raw MoCA score, only a specific adjustment of the score for age and education level (Table 31). Third, Blackburn et al indicate that a test with sensitivity and specificity >0.8 meets the criterion for being a good screening tool.The criterion for being an optimal screening test is a critical point. In the article, we provided 2 cut-off values for each test, either to optimize the trade-off between sensitivity and specificity (adjusted MMSE, ≤24; adjusted MoCA, ≤20) or to minimize the false-negative rate (MMSE adjusted, ≤29; MoCA adjusted, ≤26). This allows the reader to choose the test and cut-off according to his/her objective. We agree with the idea that cognitive impairment should be systematically screened for within 6 months of a stroke. Given that the primary objective of stroke care is not the systematic analysis of cognitive and behavioral complaints, the screening tool should minimize the false-negative rate and the likelihood ratio of a negative test (LRN).2 This is achieved after adjustment of the score for age and educational level using our proposed cut-offs for the MMSE (MMSE adjusted, ≤29; false-negative rate, 1 of 62 [ie, 1.6%]; LRN, 0.097) and the MoCA (MoCA adjusted, ≤26; false-negative rate, 3.2%; LRN, 0.06). Conversely, other cut-offs provide higher false-negative rates (MMSE adjusted, ≤24; false-negative rate, 31%; LRN, 0.31; MoCA adjusted, ≤20; false-negative rate, 31%; LRN, 0.36). A high false-negative rate is also observed for the cut-off proposed by Blackburn et al (MoCA adjusted, ≤23; false-negative, 13%; LRN, 0.19). Lastly, the use of a cut-off that minimizes false-negatives would result in cognitive assessment of half the screened patients1 (and not in 80% as suggested by Blackburn et al). The ongoing GRECOG-VASC study (ClinicalTrials.gov identifier: NCT01339195) will provide norms for the MoCA in French-speaking subjects and will examine the ability of these 2 screening tests to predict cognitive impairment at 6 months poststroke.Olivier Godefroy, MD, PhDJean Marc Bugnicourt, MDAndreas Fickl, MD Service de Neurologie CHU Nord Amiens, FranceDisclosuresNone.FootnotesStroke welcomes Letters to the Editor and will publish them, if suitable, as space permits. Letters must reference a Stroke published-ahead-of-print article or an article printed within the past 3 weeks. The maximum length is 750 words including no more than 5 references and 3 authors. Please submit letters typed double-spaced. Letters may be shortened or edited. Include a completed copyright transfer agreement form (available online at http://stroke.ahajournals.org and http://submit-stroke.ahajournals.org).