PHILADELPHIA— Neuropsychiatric symptoms are nearly universal in all phases of Alzheimer’s disease, and a better understanding of their cause and treatment is of “great importance” clinically, said Constantine G. Lyketsos, MD, MHS. Despite the public’s view that the disease is primarily cognitive, these symptoms “have serious adverse consequences for the patients, are probably right now some of the most treatable symptoms of dementia, and are likely part of the dementia prodrome,” he said.

Speaking at the Ninth International Conference on Alzheimer’s Disease and Related Disorders, Dr. Lyketsos, Professor of Psychiatry and Behavioral Sciences and Codirector, Division of Geriatric Psychiatry and Neuropsychiatry at the Johns Hopkins University and Hospital in Baltimore, centered his discussion around these questions: How common are the disturbances? How are we now approaching their classification? What do we know about their impact and cause? What do we know about treating them? Where is this area of the dementia field headed?

PREVALENCE AND CLASSIFICATION

The population-based Cache County, Utah, and Cardiovascular Health studies have found that over the course of dementia, more than 90% of patients develop one or more psychiatric and related behavioral features, with apathy, depression, agitation, and irritability being the most common, Dr. Lyketsos said. “One of the things the field has struggled with for quite awhile is how to classify these. We have not arrived to the point that we can approach them by understanding their cause—though there is some progress in that area—so most of the classificatory schemes have focused on their phenomenology. There have been three approaches: looking at them as individual symptoms, applying categories that were developed by the American Psychiatric Association in the DSM-IV, or empirically, trying to see how they naturally aggregate.”

Taking an individual symptom approach “is risky,” Dr. Lyketsos asserted. “In my view, it’s not totally parsimonious; in the clinical world that makes the treatment focus very narrow, so many times taking this approach has patients ending up on several medicines at once and, in fact, goes against data that these symptoms co-occur on a regular basis.” A cross-tabulation of Neuropsychiatric Inventory– defined symptoms found that of 45 two-way comparisons, “38 have significant odds ratios of association, suggesting that the relationship between these symptoms is fairly complicated.”

Dr. Lyketsos said although it would be “heuristically useful” to apply DSM-IV categories to the symptoms observed in Alzheimer’s disease, the phenotypic disturbances are in fact very different from what the categories intended. “For example, the psychotic symptoms of people with dementia that are sometimes referred to are very different from what you see in schizophrenia or mood disorders. Visual hallucinations predominate, and the delusions the patients exhibit are quite different. They are much less likely to be well formed.” For example, belief of misidentification or a spouse having an affair—which is rarely seen in schizophrenia—is “actually fairly common in this setting,” he remarked.

“Similarly, the mood disorders—or the term I prefer, affective disorders—associated with dementia are less characterized by crying, guilty feelings, and the like and are more likely to be associated with anhedonia, irritability, nonspecific worry, and in fact to be frequently associated with the co-occurrence of delusions. And suicidality, believe it or not, in people with Alzheimer’s disease, is very, very rare.”

In lieu of these methods, Dr. Lyketsos favors an empirical classification. Long used in psychiatry, the approach looks at the natural aggregation of symptoms and applies statistical methods “to see whether there are distinct subgroups of patients, based on their neuropsychiatric symptom profile,” he said. Studies suggest three groups of disturbances: affective, psychotic, and monosymptomatic—which relates to disparate behaviors in the later stages of dementia—“that may simply reflect executive dysfunction when there’s advanced brain damage … such as disinhibition, calling out, aggression, wandering.”

Dr. Lyketsos noted that the field has recognized these three groups, and proposed criteria are now being validated in population and clinical investigations and as targets for therapeutic intervention.

IMPACT AND ETIOLOGY

The impact of these disturbances can be quite significant, Dr. Lyketsos said. Depression worsens twofold the severity of impairment in activities of daily living, even after adjustment for Mini-Mental State Examination (MMSE) score or medical comorbidity. Agitation and related behaviors in later stages are associated with lower ratings on quality-of-life scales, “and there’s a fairly consistent literature suggesting that disease progression is accelerated in the presence of these symptoms. Institutionalization happens earlier and, in fact, functional decline is more rapid.” In addition, caregivers may suffer if physical aggression is exhibited, and cost of care can be increased by an estimated $10,000 annually.

Two schools of thought—which are likely interrelated—exist as to cause of symptoms, Dr. Lyketsos said. “They may well be reactions of cognitively impaired individuals who are attempting to understand/interact with their environment— if you will, more of a psychological understanding—or these may be directly caused by the underlying brain damage.”

Although evidence that such symptoms are linked to brain damage is fairly extensive, a point of “clear coherence” has not yet been reached. Examples of the link include correlations between plaque and tangle burdens with severity of Alzheimer’s disease pathology, neurotransmitter impairments—in which depression has fairly consistently been associated with loss of locus ceruleus and presumably noradrenergic loss to the brain—and “suggestions that psychosis is related to relative preservation of the dopamine system, with relative loss within the cholinergic system,” Dr. Lyketsos elaborated. Genetic associations have also been reported—particularly, those involving dopamine or serotonin receptor gene polymorphisms, he noted.

Clinical factors likely related to the occurrence of symptoms are “quite woefully underresearched,” he added. One consistent association observed—and not just in dementia—“is that older people who are visually impaired are much more likely to exhibit visual hallucinations. But in addition, especially in long-term care environments, there seem to be specific environmental triggers that make these behaviors emerge; for example, during personal care, poor communication, over- and under-stimulation in the environment, boredom, and related features.”

PRODROMES AND PROSPECTS

One area of convergence relates to how neuropsychiatric symptoms fit within various dementia prodromes. According to Dr. Lyketsos, a question actively being investigated—but for which there are few answers—is whether the presence of these symptoms affects the risk of dementia. In one study, patients with a Clinical Dementia Rating of 0.5 converted to dementia at a higher rate if they had neuropsychiatric symptoms; two studies have reported that late-life depression appears to increase the relative risk of dementia threefold to fourfold.

Several prospective studies have found “that there is an increased risk of developing dementia with depression whether or not there are cognitive symptoms at the same time,” Dr. Lyketsos said. “This raises the question as to whether we want to be thinking of something like mild affective impairment, because the converging evidence would suggest that a constellation of affective symptoms, including dysphoria, anhedonia, apathy, irritability—what seems to be a nonclassical mood state or affective state—may well be part of the dementia prodrome, whether or not cognitive symptoms are present. If it occurs with cognitive symptoms, there is a synergistic affect on dementia risk, but even in the absence of cognitive symptoms, late-life depression seems in and of itself to predict an increased risk of dementia.”

Dr. Lyketsos cited two studies from Cache County. The first found that, while all individuals with CIND (cognitive impairment, no dementia) at baseline declined on the MMSE, those with concomitant neuropsychiatric symptoms experienced “substantially faster” declines. The second found that late-life depression, with or without cognitive symptoms, is in fact a risk factor for incident dementia but that there is an apparent interaction with apolipoprotein (APOE) genotype, “suggesting that the association with late-life depression is primarily focused on individuals who are not carriers of an APOE ε4 gene.”

PHARMACOTHERAPY AND PSYCHOSOCIAL INTERVENTIONS

The majority of treatment studies have focused on Alzheimer-type dementia, “and on balance, they suggest modest effects. If you compare these effects to our ability to improve cognitive or functional declines of dementia, actually we do better in these areas. We can generally take care of or reduce most of the symptoms, particularly the milder symptoms in dementia patients. But the problem, in part, with the treatment studies has been that there’s not been focus on specific disturbances. That is an approach that’s only started to happen in the last several years,” Dr. Lyketsos said.

The most rigorously studied approaches have been pharmacologic. “Antidepressants, antipsychotics, anticonvulsants, and cholinesterase inhibitors have all been found to have some efficacy for the treatment of individual aspects,” he said.

Results of studies of antidepressants targeting various definitions of depressive symptoms in the context of dementia have been equivocal, with half suggesting efficacy over placebo to be nonexistent, due to substantial placebo responses. One 12-week study in patients with fairly severe types of depression and concurrent probable Alzheimer’s disease found that full response approached 40% in patients receiving a median dose of 100 mg sertraline and delayed a decline in activities of daily living. Those who remained depressed, however, reached a point at which admission to assisted living would be indicated.

Antipsychotics at moderate or even higher doses than those generally used in the geriatric setting are more effective than lower doses when targeted at psychosis or agitation, particularly in the later stages. However, although the newer class of atypical antipsychotics—risperidone, olanzapine, aripiprazole, quetiapine—all have modest efficacy, “unfortunately, there have been some concerns raised about whether these drugs in this population increase the risk of cerebrovascular incidents, so this is starting to affect their clinical use, to some extent,” he said. Anticonvulsants seem to have modest effects for agitation aggression; cholinesterase inhibitors most consistently seem to improve apathy by a modest extent, he noted.

Most studies of psychosocial interventions in people with Alzheimer’s disease and neuropsychiatric symptoms have had methodological limitations (eg, small numbers of subjects, uncontrolled, not targeted at specific disturbances, high attrition rates), Dr. Lyketsos said. However, pooled data from controlled studies suggest “some evidence that the activity programs—especially ones that are targeted at helping patients do things that they enjoy—behavior therapy, light therapy, and environmental modifications all might reduce the level of neuropsychiatric disturbance.” In addition, providing caregivers with specific intervention skills can help reduce levels of agitation in patients.

THE FUTURE IN THEORY AND PRACTICE

In the future, the field needs “to reach agreement on the definition and differentiation of these conditions,” Dr. Lyketsos said. “The differentiation is going to be important to understanding their cause and to targeting their treatment.” Studies needed are those relating to etiology, prevention (ie, the link between affective change in Alzheimer’s disease and the predementia phase), risk factors (environmental and caregiver), better brain imaging; correlations between clinical pathology and regional changes in brain pathology, further genetic associations, and treatment, including efficacy studies targeted to specific conditions and genetic predictors of treatment response.

“But in the real clinical world these days, what is most needed are studies that will carefully look at how psychosocial therapies and pharmacological therapies compare to each other or how their combined use impacts on the outcomes of interest. There’s also an important point, which is that targeting these symptoms often has secondary benefit. By treating these, we might be attenuating the course of the disease in some ways or improving and delaying functional declines and institutionalization, and obviously we need to be looking more carefully on quality-of-life aspects and benefit to caregivers,” he concluded.

—Debra Hughes

Suggested Reading
Fitzpatrick AL, Kuller LH, Ives DG, et al. Incidence and prevalence of dementia in the Cardiovascular Health Study. J Am Geriatr Soc. 2004;52:195-204.
Steinberg M, Tschanz JT, Corcoran C, et al. The persistence of neuropsychiatric symptoms in dementia: the Cache County Study. Int J Geriatr Psychiatry. 2004;19:19-26.

Return to table of contents