The Neuropsychiatry of Aging

Many questions face our field as we anticipate the impending exponential growth of our elder population. How do we draw distinctions between normal aging and age-related diseases on the moving landscape of physiologic change over time? How do we become experts in the neurobiology of aging to ensure a better quality of life for the oldest old? In a world of tertiary and quaternary care centers, do we risk losing the primary care perspective on the circle of life? Is it possible that some conditions at the end of life reflect developmental stages that cannot respond to interventions any more than an edentulous neonate could respond to dentures? On the more positive side, how can we capitalize on the beneficial attributes of aging, i.e., the wisdom and philosophical perspective that only years of experience can bestow?

Three articles in this issue of the Journal address diverse topics in the trajectory of aging, ranging from the mechanisms of neural activity to the estimation of decisional capacity. In this issue, we find a report on the age-related differences in modulation of prefrontal cortical cerebral blood flow, a report on the functional neuroanatomy of paired associate learning among patients with Alzheimer’s disease, and a report addressing one of the highest levels of cognitive functioning in the real-world setting, i.e., the capacity to vote among persons with Alzheimer’s disease. The scope of these articles reflects the range of topics on aging that will assume great importance in the next several decades.

In the first of these articles, Freo and colleagues used [¹⁵O]H₂O positron emission tomography to assess regional cerebral blood flow (rCBF) during a cholinergic challenge to understand the relative contribution of disrupted cholinergic function to late-life cognitive decline. Regional blood flow during a working memory task in younger and older adults was compared after an intravenous dose of physostigmine. The physostigmine was observed to significantly improve working memory performance in both age groups. It was further noted that the prefrontal regions demonstrating increased blood flow during the placebo condition showed significantly lower rCBF during the physostigmine condition in both groups. Notably, there were differences in the younger versus older groups in the location of rCBF changes, i.e., increased rCBF during placebo was observed in the right middle and inferior frontal cortex in the younger subjects but the older individuals demonstrated increased blood flow in the anterior and ventral prefrontal regions. The authors noted that the two groups showed brain regions that were commonly modulated, as well as areas that were differentially affected by cholinergic potentiation. This work raises important questions as to whether recruitment of more dispersed areas of the prefrontal cortex with aging may represent an adaptive response compensating for functional losses. Furthermore, this adaptation may then have implications for regions of interest in pharmacological interventions that do not necessarily fall within the anticipated structural regions associated with a given disease. The implications of this work highlight the importance of distinguishing neural adaptations from pathological changes in the pursuit of appropriate interventions that may seek either to support beneficial functional adaptations or reduce any deviations from the optimal function observed in younger adults.

It is precisely this type of research that challenges the field to define “disease” within an aging population. Temple et al. (1) commented that

As we look toward advances in genetic research, this distinction becomes ever more complex. For example, the study of polymorphic variations in genes may identify specific genotypes that represent a risk for morbidity but may or may not ultimately lead to the outcome of disease. In the development of interventions that seek to prophylactically treat high-risk conditions based on genetic information, we are now blurring the boundaries of disease and risk factors. When the focus of intervention may be simple age-related functional changes, we are further blurring the distinctions between normal physiological variation and disease.

The article by Gould et al. in this issue addresses the problem of how to estimate differences in functional imaging among persons with Alzheimer’s disease when task performance and effort during a functional imaging procedure may differ markedly from healthy subjects. This study used functional magnetic resonance imaging to assess performance on a visuospatial paired associate learning task after controlling the estimated task difficulty such that patients and comparison subjects performed at the same relative levels of effort. These investigators observed blood-oxygen-level-dependent (BOLD) responses in frontal-parietal and occipital regions during successful associative learning in both groups. With increasing task difficulty, linear increases in occipital-parietal regions were noted during encoding and retrieval, but these did not differ markedly between the patients and the comparison subjects. The authors concluded that similar brain activations occur during successful paired associate learning in patients with Alzheimer’s disease and comparison subjects, providing that the subjects were able to conduct the tasks with approximated levels of task difficulty. This article reveals another example of the complexities associated with estimating compromised neural activity and how task performance and task difficulty may significantly contribute to variance in these estimations. It was observed that lateral and medial parietal regions displayed greater BOLD activation in the patients during encoding and retrieval, which may have reflected additional recruitment of regions that were noted in healthy subjects to be activated during visuospatial paired associate learning. Again, this was thought to reflect a possible adaptive mechanism employed by patients to compensate for disease-related neuropathology. Functional changes that represent adaptation to disease may be difficult to distinguish from direct disease-related pathology. However, this distinction may be critically important as we seek to take future steps in mapping the treatment effects of various interventions.

In a third article on age-related conditions, Appelbaum and colleagues examined arguably the highest level of executive functioning in public life: the act of voting for political leaders and policy. The authors used a competence assessment tool for voting among persons with Alzheimer’s disease based on the elements of understanding the nature of voting, the ability to make a choice, and measures of appreciation and reasoning. It was observed that voting capacity performance correlated strongly with Mini-Mental State Examination scores. Overall, this study found that patients with milder conditions generally displayed adequate performance on capacity measures, whereas those with moderate disease were more variable, and persons with severe dementia generally did not demonstrate an adequate capacity to vote.

Of interest, the competence assessment tool for voting (CAT-V) was derived in part from the Doe criteria, the result of a federal court decision in Maine—Doe v. Rowe—in which persons under guardianship due to chronic mental illness contested their automatic exclusion from voting. The court proposed criteria that permitted voting, provided that the individual did not lack the ability to understand the nature of voting or lack the ability to make an individual choice. Hence, the result of this decision was to facilitate voting in a class of citizens who were previously excluded by a provision in Maine’s constitution. In contrast, as suggested here, the use of tools such as the CAT-V may effect the opposite result, i.e., they may serve to reduce voting within a group of citizens who are no longer capable. As mentioned by the authors, the increasing number of elderly persons with dementia, combined with a high rate of voting among older groups, suggests that “further refinement of approaches to identifying potential voters with inadequate capacity will become increasingly important to our electoral system.” Without question, this area begs for much additional research and careful thought, particularly in view of potential misuse of such tests in populations of any age. In taking the brave step of raising this issue, the authors opened the door to a host of questions with significant ethical implications.

Whether the issue is voting, neural activation, or cerebral blood flow, the challenges ahead lie in separating the aspects of aging that require only simple monitoring from aspects that require intervention or other decisive actions to protect the patient or society. With our increasing awareness of the multiple factors that converge in the development of dementia (2), it is clear that the continuum of neurodegeneration will only become more of a continuum as we parse smaller and smaller measures of change. Progress in mapping brain function combined with an awareness that the experience of dementia is closely tied to sociocultural influences (3) lends itself to an unlimited number of important questions and challenges for future research as we try, possibly to no avail, to discern the lines between aging and age-related disease. Perhaps the best guidance in this endeavor will come from retaining the bigger picture as a backdrop, i.e., the circle of life.

Address correspondence and reprint requests to Dr. Schultz, Roy and Lucille Carver College of Medicine, University of Iowa, Iowa City, IA 52242; [email protected] (e-mail).