Multi-domain Cognitive Testing: A Biomarker for Classifying the Cognitive Status of Mild Cognitive Impairment and Alzheimer's Disease

CONTEXT

Cognition is impaired in Alzheimer's disease (AD) and patients with mild cognitive impairment (MCI) with varying levels of magnitude.

AIM

The present study aimed to identify a biomarker for classifying MCI and AD using multi-domain cognitive testing.

SETTINGS AND DESIGN

This was a cross-sectional study.

METHODS AND MATERIALS

26 AD patients, 28 MCI patients and 25 controls were recruited. Cognitive assessment of different domains was done using standard questionnaires and cognitive function tests.

STATISTICAL ANALYSIS USED

Cognitive task scores were compared between the groups using multivariate analysis of variance (MANOVA).

RESULTS

Patients with AD had significantly lower MMSE, CDR, cognitive task scores compared to controls and MCI. Cognitive scores of all tasks for MCI were significantly less than controls, except MMSE and digits forward score. ROC analysis showed that picture memory had 100% sensitivity, 91.6% specificity for AD and 88.4% sensitivity, 92.5% specificity for MCI. Word memory had 92.3% specificity, 100% specificity for AD and 80.7% specificity, 84.6% specificity for MCI.

CONCLUSIONS

The global cognitive tools are less specific in bringing out the differences especially between MCI and control. Limitation of MMSE, heterogeneity of MCI and differential impairment of various domains of cognition, demands the inclusion of multi-domain cognitive evaluation especially picture and word memory tasks with high sensitivity and specificity into the existing diagnostic protocol. ROC results also suggested the continuum of cognitive impairment and MCI as a transitional stage leaving more scope on the quantum of research required for intervention to halt the structural and functional decline.