background   The ante mortem diagnosis of Alzheimer's disease is based on the patients clinical symptoms and on various criteria. The National Institute of Neurological and Communicative Disorder and Stroke-Alzheimers's Disease and Related Disorder Association criteria (nincds-adrda) are considered to be the gold standard. This method, however, is very labour-intensive and is simply not feasible in many hospitals. In our clinic we use the Mini Mental State Examination (mmse) or the cognitive part of the Cambridge Examination for Mental Disorders of the Elderly cognitive (camcog) for diagnosing possible cases of Alzheimer's disease.
aim   To investigate whether the predictive value of the mmse/camcog could be made to exceed that of the nincds/adrda if we genotype the following Alzheimers-related genes: apolipoprotein E, α1-antichymotrypsin, presenilin-1 and tumour necrosis factor-α. method The study population consisted of 223 patients and 100 controls. On the basis of the nincds-adrda criteria 84 patients were suffering from Alzheimer's disease. Besides subjecting all patients to psychometric tests we identified in each patient the genotype of possible markers for Alzheimer's disease.
results   Age, mmse and camcog together had a predictive value of 76%, to which mmse contributed the most. Genotyping did not raise the predictive value of mmse/camcog above that of nincds/adrda. conclusion   There is no clear relation between the genetic markers studied here and clinical diagnosis of Alzheimers disease based on the nincds-adrda criteria. For the time being, therefore, there seems to be little or no justification for genotyping in a clinical setting.