CHICAGO—Neuropsychiatric disturbances are common in dementia. Prevalence estimates range from 60% to 80% at any one time, and the lifetime risk is almost 100%, according to Anton P. Porsteinsson, MD. For all of that, however, they are notoriously difficult to treat effectively. “If you have dementia, during the course [of the disease] some kind of behavioral disturbance will surface. But they are not well defined. There’s a high rate of co-occurrence. If, at day one, you do not have agitation, aggression, or psychosis, over the next year the incidence range will be somewhere between 20% and 40%, and over the next two years, up to 50% to 60%. So if you don’t have it now, you’re likely to have it in the relatively near future,” said Dr. Porsteinsson, an Assistant Professor in the Department of Psychiatry at the University of Rochester, New York.

Behavioral disturbances also have a high persistence rate, he added in his address at the American Geriatrics Society 2001 Annual Scientific Meeting. “For agitation especially, it’s 60% to 80%, and for psychosis it can range from 40% to 60% per year. It’s the frequency and the persistence of the problem that predicts the magnitude, the impact on the environment. We think that behavioral disturbances exacerbate functional and cognitive deficits. These individuals seem to have more rapid progression, and they’re hard to care for. Even getting them to comply with the treatment is difficult. If the disturbances are happening infrequently but are very disruptive, or if some milder forms of behavioral disturbances are happening daily, you may need to become a little more aggressive in terms of intervention,” said Dr. Porsteinsson.

A RUBRIC

Because these are not syndromic pictures, the general approach to treating these behavioral disturbances can be complicated. “Teasing out target symptoms is helpful. You start to see that these symptoms often resemble some sort of known drug-responsive syndrome,” he explained. “But before you apply a label, it’s important to get the input from multiple sources. If there is a medical disorder, you treat it. If you think there’s a psychiatric syndrome, you treat that appropriately. Monitor the target symptoms. One of the things that happen is that some of the standard diseases may present a little differently in someone with a superimposed cognitive disorder. The patients may not be able to communicate their symptoms or exhibit their symptoms in the typical fashion.”

Dr. Porsteinsson recommended non-pharmacologic intervention as a first option but also supported the use of psychotropics “where they’re appropriate.” Pharmacologic therapy for behavioral disturbances in someone with dementia is administered contingent upon two results: response of target symptoms versus toxicity, he said. “You slowly increase the dose until you see one or the other. If you do get benefit you try to hold that nontoxic, efficacious dose. See whether continuous use is needed. If we look at agitation, remember how persistent it may be. As you go through the moderate to early severe stages of disease, you may have to persist with treatment. But often as you hit the later stages, so much apathy has set in that you may be able to back away from the psychotropic.”

A NOVEL ANGLE

Reports on the use of anticonvulsants for aggressive and impulsive behaviors prompted Dr. Porsteinsson and his colleagues to consider trying those agents in patients with dementia and behavioral disturbance, in whom some of the target symptoms are similar. “Our initial work focused on carbamazepine. We recruited 26 subjects and found that carbamazepine was statistically significant in terms of improvement over placebo. It was also relatively well tolerated, which was a little surprising.”

“We did a second, more rigorously controlled study, recruiting 51 agitated dementia patients in nursing homes. At the end of the study, there was about a 75% global improvement in the drug group versus only about 20% in the placebo group.” However, he added, “carbamazepine is not by far the optimal drug for this population. There was concern about prominent side effects: ataxia, sedation, gastrointestinal distress, and hematologic adverse effects, as well as drug-drug interactions, because every patient in this age-group seems to be on several medications.”

VALPROATE—THE ANTICONVULSANT ANSWER?

Valproate was also much less prone to drug interactions. “It is metabolized primarily via glucuronidation and to a much lesser extent through cytochrome P-450, so it’s not as susceptible to hepatic impairment or influence from other drugs that might have an upgrading or downgrading effect on cytochrome P-450,” explained Dr. Porsteinsson. Aspirin may increase valproate level because it increases the free fraction of valproate. However, “the usual [cardioprotective] doses of aspirin don’t seem to have much of an effect at all,” he added. Valproate itself may cause minor increases in the levels of warfarin, amitriptyline, diazepam, and some other anticonvulsants. In particular, Dr. Porsteinsson cautioned practitioners to “be especially careful about the combination of valproate and lamotrigine.”

NEUROPROTECTIVE POSSIBILITIES

He elaborated by considering some of the underlying pathophysiology of Alzheimer’s disease. “There are two main insults that happen in Alzheimer’s disease: amyloid plaques and neurofibrillary tangles. There are three main ways we theorize that valproate may influence and protect neurologic pathways. One is through a family of proteins called bcl2, which will block cytochrome c, which through its release induces several caspases and participates in the apoptosis pathways. You may see upregulated apoptosis in the proximity of amyloid plaques. Also, you have the glycogen synthase kinase 3 beta (GSK3 beta) protein, which participates in the hyperphosphorylation of tau, one of the culprits in neurofibrillary tangles. Finally, GSK3 beta is related to beta-catenin and if you up-regulate beta-catenin, it also has an impact on transcription factors in the nucleus and leads to improved cell survival,” he explained.

A CASE FOR EXTENDED-RELEASE FORMULATIONS

Dr. Porsteinsson went on to document the case of a 67-year-old patient with bipolar disorder who was chronically hypomanic and had a slowly progressive cognitive disorder. “When he gets manic, there’s no one more sexually disinhibited. You have to restrain him to give him care because his hands are all over the place,” Dr. Porsteinsson said. “He doesn’t tolerate any antipsychotics. Doesn’t tolerate lithium. When we tried Depakote, he needed about 500 mg. We could give him 250 mg twice daily, but he would accelerate in the evening and you could not get medications into him reliably. If you gave 500 mg once in the morning he would have peak-related side effects—you just couldn’t do it.

“He was the first patient I put on the extended-release divalproex. It was tolerated very well at 500 mg: improved patient compliance and improved treatment. Clinically, it’s effective. I haven’t been called to see him in a year,” he added.

NR