Is the Montreal Cognitive Assessment superior to the Mini-Mental State Examination to detect poststroke cognitive impairment? A study with neuropsychological evaluation.

BACKGROUND AND PURPOSE A screening test is required to improve the diagnosis of poststroke cognitive impairment. The Montreal Cognitive Assessment (MoCA), a newly designed screening test, has been found to be more sensitive than Mini-Mental State Examination (MMSE), but its clinical value has not been established by means of a comprehensive neuropsychological battery. This study was designed to assess the value of MoCA and MMSE to detect poststroke cognitive impairment determined by a neuropsychological battery. METHODS Both screening tests and a neuropsychological battery were administered during the acute phase in 95 patients referred for recent infarct or hemorrhage. Raw MMSE and MoCA scores were used with published cutoffs and new cutoff scores for MMSE and MoCA were also computed after adjustment for age and education. RESULTS Using raw scores, MoCA was more frequently impaired (P=0.0001) than MMSE. MoCA showed good sensitivity (sensitivity, 0.94) but moderate specificity (specificity, 0.42; positive predictive value, 0.77; negative predictive value, 0.76), whereas an inverse profile was observed for MMSE (sensitivity, 0.66; specificity, 0.97; positive predictive value, 0.98; negative predictive value, 0.58). Adjusted scores with new cutoffs (MMSE(adj) ≤24, MoCA(adj) ≤20) provided good sensitivity and very good specificity for both tests (MMSE(adj): sensitivity, 0.7, specificity, 0.97, positive predictive value, 0.98, negative predictive value, 0.61; MoCA(adj): sensitivity, 0.67, specificity, 0.9, positive predictive value, 0.93, negative predictive value, 0.57). On receiver operating characteristic curve analysis, areas under the curve of all scores were >0.88. CONCLUSIONS The previously reported high sensitivity of MoCA is associated with low specificity. Both screening tests are moderately sensitive to acute poststroke cognitive impairment. This study provides indications for the diagnosis of poststroke cognitive impairment.