[Epidemiologic and clinico-pathologic study on senile dementia in a Japanese community, Hisayama]

To elucidate a definite prevalence of dementia and its risk factors in the senescence, 887 subjects aged 65 years or over of Hisayama residents (94.6% of the whole population in the same age-range) were examined from May to December, 1985, as to whether he or she has suffered dementia. Karasawa's "Clinical criteria for the severity of dementia", Hasegawa's Dementia Scale and Hachinski's ischemic score were used as a procedure to identify demented persons. Out of 887 examined, 59 subjects were diagnosed to have had dementia (6.7%) with female to male ratio being 1.2. The prevalence rates of dementia increased with advanced age decades and sharply grew up over tha age of 80 years. Brain morphology in 50 of the 59 dementias was re-evaluated by autopsy and/or CT scan, during a 55-month follow-up period from January 1986 to July 1990. Coincidence rate between clinical evaluation and morphological diagnosis was 90.5% for vascular dementia (VD), 62.5% for senile dementia of Alzheimer type (SDAT), and 28.6% for other or ill-defined dementia (Others). Misdiagnosis of clinical evaluation for VD was mainly due to uncertain history of head trauma, while the subjects suffering from atypical stroke tended to be diagnosed as SDAT. Out of the 21 cases who were initially diagnosed as having other or ill-defined dementia, 8 were re-evaluated as VD, and 7 as SDAT later. Most cases who were finally diagnosed as having VD had multi-infarcts without an episode of stroke attack, while those who were classified to SDAT had less severely undergone dementia. The final type-specific prevalence of dementia was 56% for VD, 26% for SDAT and 18% for Others, respectively. The ratio of VD to SDAT was 2.2, indicating more frequent VD than SDAT. Retrospective case-control study on risk factors of VD was performed using selected parameters obtained at examinations in 1973 or 1978, comparing 27 VD cases and 789 non-demented cases. As a consequence, aging, hypertension, electro-cardiographic abnormalities (Minnesota code 3-1 and/or 4-1, 2, 3) and high hematocrit were taken as important risk factors for VD. These risk factors were very similar to those for lacunar infarcts among Hisayama residents. The difficulty of epidemiologic study on dementia in general population and the accuracy of diagnosis for type of dementia were discussed.