THE CAMCOG - A USEFUL SCREENING INSTRUMENT FOR DEMENTIA IN STROKE PATIENTS

Background and Purpose-Most mental screening tests focus on the detect ion of cognitive deficits compatible with Alzheimer's disease. Stroke patients who develop a dementia syndrome, however, constitute a more h eterogeneous group with bath cortical and subcortical disturbances. We assessed the diagnostic accuracy of the CAMCOG (the cognitive and sel f-contained part of the Cambridge Examination for Mental Disorders of the Elderly) and the Mini-Mental State Examination (MMSE) for dementia in patients with a recent stroke. Methods-In patients aged 55 and old er who were admitted in the Rotterdam Stroke Databank, cognitive funct ioning was assessed between 3 and 9 months after the most recent strok e. The ''gold standard'' diagnosis of dementia was compatible with the criteria of the Diagnostic and Statistical Manual of Mental Disorders , Third Edition, Revised. The CAMCOG and MMSE scores were obtained ind ependent of the diagnostic procedure. Results-Of 300 consecutive patie nts, 71 (23.7%) were demented. Sixteen severely demented patients coul d not be tested and were excluded. The CAMCOG and MMSE scores were sig nificantly related to dementia (both P<0.0001) in a logistic regressio n model. Receiver operating characteristic analysis showed that the CA MCOG was a more accurate screening instrument (area under the curve fo r CAMCOG, 0.95; for MMSE, 0.90). Two other clinical variables independ ently improved the diagnostic accuracy of the MMSE and CAMCOG: patient s with a left hemispheric lesion had a lower (odds ratio, 0.3; 95% con fidence interval, 0.1 to 0.7), and patients with hemorrhagic stroke ha d a greater chance of being demented (odds ratio, 3; 95% confidence in terval, 1 to 10). The effect of left hemispheric lesion as an independ ent diagnostic factor could not be explained by selection or its assoc iation with aphasia alone. Conclusions-The CAMCOG is a feasible instru ment for use in patients with a recent transient ischemic attack or st roke. It is a more accurate screening tool for dementia than the MMSE, especially when type and site of stroke are taken into account.