In a neuropathological study of 81 brains of prospectively studied subjects of 80 years of age or older at the time of death, 13 cases (16%), including 4 men and 9 women, had hippocampal sclerosis (HpScl) affecting the vulnerable region of the hippocampus. In demented subjects of 80 years of age or older, the frequency of HpScl was even higher, 26%. Cases with HpScl had significantly fewer hippocampal senile plaques (SP) and neurofibrillary tangles (NFT) and parahippocampal NFT than cases without HpScl, but did not differ significantly in any of the other measured pathological parameters. Enzyme-linked analysis of synaptic protein immunoreactivity in a subset of 33 cases demonstrated significant decreases in the hippocampus, but not in frontal, temporal, parietal or parahippocampal cortices. All but 1 of the cases with HpScl had Blessed information, memory and concentration scores (BIMC) of 8 or more, and all were considered to be demented. In some patients memory disturbance was disproportionate to deficits in other cognitive areas. All but 4 of the cases with HpScl had many non-neuritic, amyloid plaques in the neocortex meeting NIA criteria for Alzheimer's disease (AD); however, given the advanced age of the subjects, amyloid plaques were considered to represent age-related cerebral amyloid deposition ("pathological aging") in most cases. Only 3 cases had both many SP and NFT in multiple cortical regions consistent with AD. Another case had brain stem and cortical Lewy bodies consistent with diffuse Lewy body disease (DLBD). A few ballooned neurons were present in the limbic cortices in 3 cases, including one case of dementia with argyrophilic grains (DAG) in limbic and orbital frontal and temporal cortices. The 8 cases without AD, DLBD or DAG included 4 cases in which no other obvious cause of dementia was detected and 4 cases in which HpScl was accompanied by either multiple cerebral infarcts or leukoencephalopathy, or both, that could have contributed to dementia. Patients with HpScl had risk factors, clinical signs and post-mortem pathological findings of cardiovascular disease, but due to the high prevalence of these conditions in very old humans, no significant correlation with HpScl was detected. This study demonstrates that HpScl is a common post-mortem finding in demented, but not normal, elderly subjects. It may contribute to. or be a marker for, the increased risk of dementia in subjects with documented cardiovascular disease or a history of myocardial infarction.

The hearts obtained at autopsy of 67 patients with hypertension, diabetes mellitus, or both were examined microscopically and histochemically, and the amount of fibrosis was determined. Significant differences in heart weight, interstitial fibrosis, replacement fibrosis, and perivascular fibrosis were found among the groups. The mean heart weight of the hypertensive-diabetic patients was significantly greater than that of the hypertensive patients and the diabetic patients. The amount of microscopic fibrosis increased between the groups, the lowest in hypertensive hearts, midrange in diabetic hearts, and highest in hypertensive-diabetic hearts. Total fibrosis correlated with heart weight among diabetic and hypertensive-diabetic patients and was significantly greater among patients with congestive heart failure, most of whom had histories of both hypertension and diabetes. The microscopic grade of fibrosis correlated significantly (p less than 0.01) with a quantitative, histochemical determination of the amount of collagen per milligram of total noncollagenous protein in the heart tissue. Myocardial fibrosis may contribute to the diastolic dysfunction typical of hypertensive-diabetic cardiomyopathy, in which congestive heart failure is a common sequela. The importance of hypertension in the pathogenesis of severe diabetic heart disease is discussed.

No abstract