WASHINGTON, DC—Contrary to widely held beliefs about aging, cognitive decline is not normal in old age, and it is possible to experience healthy brain aging into the 10th decade of life, according to presenters at the World Alzheimer's Congress 2000. Possible, that is, but admittedly rare.

"It is commonly assumed that as we age, we experience cognitive decline. I will put forth evidence that this may not be the case," said John C. Morris, MD, Co-Director of the Alzheimer's Disease Research Center at Washington University School of Medicine in St. Louis. He referred to a study that documented nondemented status and absence of neuropathologic evidence of Alzheimer's disease in those age 90 and older.

"We had a strict definition of what is acceptable for 'normal' aging," said Dr. Morris. The study was not population based but the volunteer subjects were very carefully assessed. Although at baseline "we demanded that healthy means healthy, without even a mild decline in cognitive function," the controls developed Alzheimer's disease at 2% to 3% a year. They weren't "supernormal" nor were they protected against developing age-related illness, Dr. Morris said.

Clinical, psychometric, and neuropathologic results were reviewed in the 69 study subjects (52 female; mean age at entry, 88.5; mean age at death, 94.1 with a range of 90 to 107). At baseline, 24 subjects were cognitively normal (Clinical Dementia Rating [CDR], 0) while the remainder had very mild dementia (CDR, 0.5), mild dementia (CDR, 1), or moderate dementia (CDR, 2). For the study purposes, said Dr. Morris, "Healthy brain aging is CDR 0. Alzheimer's disease is represented by cognitive deterioration at the 0.5 or greater level, with Alzheimer's disease confirmed at death." Seven of the 24 individuals who had a CDR of 0 at baseline did not develop dementia, but the others developed dementia.

DECLINE, OR DISEASE?

Annual psychometric evaluations in those who became demented showed cognitive decline, "but if you get strict about it, you see that the decline occurs not as a function of age, but with the onset of disease," reported Dr. Morris. The controls who remained healthy did not have plaques or tangles outside the medial temporal lobe; but subjects demonstrating cognitive decline, even at early stage, already had neuropathologic evidence of early Alzheimer's disease. Dr. Morris and his colleagues believe that these findings reflect a long preclinical stage in which neuropathic lesions accumulate before the clinical expression of disease.

"Some data suggest that the rate of developing Alzheimer's disease may taper off after age 90, but the number of cases studied is small. "What about people past age 90? Is cognitive stability still seen, and what does the neuropathology look like?" asked Dr. Morris. Clinicopathologic analyses were performed in subjects 90 or older, mostly women with high school education, who were followed for four years. At baseline, 44 of the 69 individuals were demented. Only seven of the original 24 healthy subjects remained CDR 0.

Individuals who later developed dementia performed identically at baseline as those who remained nondemented. Among individuals who were not impaired at time of entry but became demented, the psychometric data supported the clinical detection of impairment even at the CDR 0.5 level, explained Dr. Morris. "So, if you develop to CDR 0.5 even after age 90," said Dr. Morris, "it can be detected and it is substantiated by declining psychometric performance."

As for neuropathology, Alzheimer's disease was confirmed in 55 of 61 individuals with dementia of the Alzheimer type at time of death. Of the seven subjects who remained nondemented, one had sufficient plaques and tangles to be classified as having clinical Alzheimer's disease.

AMYLOIDS AND AGING

Even in individuals with healthy brain aging, all have neurofibrillary change in the medial temporal cortex, said Dr. Morris. "This finding is ubiquitous in our series, no matter whether dementia is present or not," he explained. In contrast, all subjects who died in the 0.5 or higher stage had at least moderate to severe levels of amyloid deposition, whereas the nondemented individuals had little or no amyloid pathology. Hence, amyloid deposition appears to demarcate healthy brain aging from disease.

Alzheimer's disease is very prevalent after age 90, as 90% of the study subjects were demented by the time of death. Nonetheless, the researchers' findings of at least some nondemented individuals without neuropathologic Alzheimer's disease led them to conclude that healthy brain aging can occur past age 90. "We have six people who, I think, by any standard—clinically, psychometrically, neuropathologically—have healthy brains. It is possible to live into the 10th decade with a healthy brain. I think we should look at such individuals to find out what is unique about them that allows them to escape the very common, age-associated problem with Alzheimer's disease," Dr. Morris said.

DEMENTIA MAY NOT BE DESTINY

Related research reported at the meeting suggested that cognitive decline may be the result of social factors, preventable illness, and genetic vulnerability. "Because cognitive decline is so common and frequently signals the development of dementia, it is important to understand what predicts that type of change and to distinguish aging-related and age-related changes in cognitive function and dementia," said Mary N. Haan, MPH, DrPH. Dr. Haan is an Associate Professor in the Department of Epidemiology at the University of Michigan's School of Public Health, Ann Arbor.

"There are very few population-based studies of cognitive change and dementia, particularly those that are based upon a randomly selected, representative sample," she said. Dr. Haan and colleagues designed the Cardiovascular Health Study, a long-term epidemiologic cohort study in which 70% of people who were followed over a seven-year period did not change more than one standard deviation on a test of cognitive functioning, the modified Mini Mental State Examination (MMSE).

The Cardiovascular Health Study was a prospective, population-based study of 5,201 men and women age 65 and older recruited in 1989 and 1990. A second cohort of African Americans was added in 1992, bringing the total study population to nearly 6,000. Subjects were followed through 1999 via an interview of clinical measures; intense follow-up will continue through 2004. Subjects were annually administered the modified MMSE and the Digital Symbol Substitution Test.

The first step in analysis was simply to look at overall change in cognition, as reflected in the MMSE scores, Dr. Haan explained. The average annual rate of change was –0.58, and the annual median change was 0. The range of change was –42 to +9, with the average person declining about four points during the seven-year follow-up. "So, looking at it overall, there's relatively little change in the population [of] average cognitive functioning in this particular study," she reported.

CARDIOVASCULAR DISEASE THE CULPRIT?

Measures of cardiovascular disease risk factors included oral-glucose tolerance test, apolipoprotein E (APOE) genotype, ankle-to-arm blood pressure ratio (<0.9 was considered low), wall thickness of the common and internal carotid arteries, systolic blood pressure, history of stroke at baseline, congestive heart failure, and major electrocardiographic abnormalities. After adjustment for age, race, education, and depression, all of these risk factors were significantly associated with decline in cognitive function during the seven-year follow-up.

The greatest change occurred in those patients who had stroke at baseline; they declined by nearly 5 points during the follow-up in association with a 1-point change in the modified MMSE. Because the APOE*E4 allele is associated with elevated low-density lipoprotein (LDL), the researchers examined APOE genotypes in connection with LDL, high-density lipoprotein (HDL), and total serum cholesterol.

For any genotype including an APOE*E4 allele, there was a significantly higher LDL (HDL was not affected by the APOE genotype), and subjects who had an APOE*E2/APOE*E2 genotype had a lower level of LDL. Subjects with an APOE*E2/APOE*E2 genotype had significantly less atherosclerosis, as measured by average wall thickness in the carotid artery, than did subjects with other genotypes. Those with APOE*E3/APOE*E4 or APOE*E4/APOE*E4 genotypes had significantly more atherosclerosis.

Subjects who were in the lowest quartile of carotid wall thickness and had no APOE*E4 allele had slight cognitive decline over time but did not differ significantly from others in the bottom three quartiles of carotid atherosclerosis. Among APOE*E4-negative subjects, those in the top quartile of carotid wall thickness declined significantly more than subjects in the other quartiles. However, there was no significant difference in the rate of cognitive decline between APOE*E4-negative subjects in the top quartile for carotid wall thickness and the APOE*E4-positive subjects in the bottom three quartiles.

"So, the most dramatic change occurred in the people who were APOE*E4 positive and were also in the highest quartile of atherosclerosis," Dr. Haan reported. "This is a significant interaction and also a multiplicative cross-product interaction—which means that it is not simply a matter of adding together the risk associated with APOE*E4 and the risk associated with a high level of atherosclerosis. It is a synergism that is occurring in that particular analysis." When the same data are expressed as a ratio of decline in cognition, the greatest difference in rate of decline between subjects with and without APOE*E4 occurred in subjects in the fourth quartile of carotid atherosclerosis.

PERIPHERAL VASCULAR DISEASE AND…

The association between peripheral vascular disease and cognitive decline was also examined during the study. An ankle-to-arm blood pressure ratio of less than 0.9 is considered indicative of peripheral vascular disease. Using the GEE regression model, the researchers found that subjects who had neither APOE*E4 nor peripheral vascular disease declined less than 2 points over a seven-year follow-up period. "People who had an APOE*E4 and had peripheral vascular disease declined so much more rapidly that by the end of the follow-up period they had lost an average of 12 points on the modified MMSE. This group did not differ from each other statistically, so it's a synergism between APOE*E4 and another form of subclinical cardiovascular disease that predicts cognitive decline," believes Dr. Haan.

DIABETES, TOO

The study also examined the influence of diabetes on the risk of cognitive decline and possible interactions with APOE*E4. The cognitive decline of subjects who had APOE*E4 and a baseline diagnosis of diabetes was five times faster than that of nondiabetic subjects without APOE*E4 and three times faster than that of subjects who had APOE*E4 but did not have diabetes. "Again, evidence for synergism for another subclinical cardiovascular disease and APOE*E4 in relation to cognitive decline—this combination is strongly predictive of cognitive decline."

"NO CHANGE" IS THE NORM

"This study shows that in a population-based study of people over 65, no change in cognitive functioning is the norm," reported Dr. Haan. Cognitive function in this study was influenced by modifiable exposures (such as cardiovascular diseases in the high-risk group), genetic vulnerability, and subclinical cardiovascular disease (which is primarily due to environmental influences such as diet and lack of exercise). Therefore, stressed Dr. Haan, there are implications for interventions that can identify a high-risk group for treatment and prevention of dementia and cognitive decline.

"This is not merely an artifact of the influence of APOE*E4 on coronary heart disease risk but is actually a multiplicative interaction that suggests the synergy between APOE*E4 and subclinical cardiovascular disease may be an important predictor of cognitive decline. Certainly, those people who have those combinations of effects are not normal," said Dr. Haan. "Cognitive decline is not normal in aging but is a result of a synergy between genetic factors and, in this case, subclinical cardiovascular disease."

NR