Montreal cognitive assessment for cognitive detection in brain trauma patients with normal mini-mental state examination scores

Abstract

Objective To evaluate the Montreal cognitive assessment (MoCA) for detecting the mild cognitive impairment (MCI) in brain trauma patients with normal mini-mental state examination (MMSE) scores. Methods Fifty brain trauma patients with normal MMSE scores hospitalized from January 2013 to June 2014 were subjected to the MoCA test. The patients were classified as cognitive impairment group scored less than 26 on the MoCA and cognitive normal group scored 26 or above on the MoCA. Differences in MMSE and MoCA scores of the two groups were compared. Receiver operative characteristic (ROC) curve was used to determine the optimal cut-off scores in screening for MCI. Results Overall MMSE and MoCA scores were (27.84±0.89)points and (23.24±2.90)points. There was a positive correlation between MoCA and MMSE total scores (r=0.355 2, P<0.05). MCI was found in 79% of the brain trauma patients using the MoCA. MMSE total score and subscores were all similar between the two groups. MoCA total score and subscores of attention, language, abstraction and delayed recall were much higher in cognitive normal group than in cognitive impairment group (P<0.05), but there were no significant differences in visuospatial, naming and oritention domains. Area under the ROC curve for MoCA(0.871±0.038) was larger compared with MMSE (0.796±0.054) (Z=3.592, P<0.05). The optimal cut-off scores of MoCA and MMSE for the identification of MCI were 25.5 and 28.5 respectively. Conclusions MoCA and MMSE total scores are positively correlated. MoCA is a better detector for the identification of MCI in brain trauma patients than the MMSE. Key words: Craniocerebral trauma; Cognition disorders; Montreal cognitive assessment/mini-mental state examination