TREATMENT OF TAUNTING BEHAVIOR IN VASCULAR DEMENTIA WITH VALPROIC ACID

Abstract

To the Editor: Up to 90% of people with dementia exhibit significant behavioral problems at some time during the course of their illness.1 Physical aggression is an unfortunately common and reasonably well-studied problem in this population, but not as much attention has been given to less-urgent behaviors. A patient with vascular dementia and behavioral disinhibition is reported whose nonaggressive taunting of other patients responded well to valproic acid, an agent commonly used for physical aggression in this population. A 91-year-old man had resided in the Community Living Center (formerly Nursing Home Care Unit) for approximately 15 months. Pertinent medical problems included asbestosis, coronary artery disease, hypertension, benign prostatic hypertrophy, and hypothyroidism. Psychiatrically, there was a remote history of mild generalized anxiety disorder in complete remission with sertraline and an approximately 6-year history of mild dementia with amnestic and executive features felt to be consistent with vascular dementia. There were initially no behavioral problems aside from some mild tactlessness and Witzelsucht (inappropriate humor and jocularity), but after approximately 1 year, he began to taunt other residents. These behaviors were invariably directed toward several severely cognitively impaired residents and generally occurred when he thought that he was not being watched. Behaviors included teasing one resident about his sex life, gently pulling on another resident's ear, stroking another sleeping resident's face with a piece of paper, and gently kicking another resident in the leg. On several other occasions, he was reported to have rattled a bird cage on the unit. None of these behaviors resulted in any injury, and he never gave the appearance of being angry or of trying to inflict harm. Medications and laboratory levels were noncontributory, and there was no evidence of intercurrent illness. Extensive inquiry revealed that these behaviors were completely out of character for him. The behavioral problems did not respond at all to behavioral interventions, and a trial of beta-blockers was abandoned because of bradycardia. He was eventually started on valproic acid, which was titrated to a maximum dose of 750 mg twice a day. Upon attaining therapeutic steady-state serum levels (60–80 μg/mL), the taunting behavior ceased almost completely and had not recurred over the following 3 months. Although there is no scientific literature specifically addressing taunting behavior in dementia, it is likely that it is common and that it represents a manifestation of the more-general behavioral disinhibition seen in patients with frontal or subcortical deficits (e.g., in vascular dementia).2,3 Typical behaviors may range from easily tolerated ones (tactlessness, flirtatiousness) to more problematic ones (taunting, stealing) to frankly urgent ones (physical or sexual aggression).3 This patient's behaviors were not dangerous and mostly a nuisance but affected other residents to the extent that they threatened his ability to remain in the facility. Valproic acid seems to have a role in the management of physical aggression in dementia,4 although there is some conflicting evidence.5 As for nonphysical behaviors, there is only a single report of its effectiveness in nonphysical aggression,6 a patient with angry, aggressive behaviors (shouting vulgarities, threats of violence) that seemed quantitatively and qualitatively different from taunting. Valproic acid has not yet been suggested to be specifically helpful for disinhibited behaviors, although this has been suggested for carbamazepine,7 another anticonvulsant and mood stabilizer. It seems reasonable to speculate that the disinhibited nature of a behavior, rather than its level of potential physical harm, might predict response to an agent of this class. It is gratifying to find that valproic acid may have some role in the management of less overtly dangerous behaviors. Conflict of Interest: Neither of the authors have any actual or potential financial, personal, or other conflict of interest pertinent to this work. Author Contributions: Both authors contributed to the conceptualization of this work, literature review, manuscript preparation, and final approval. Sponsor's Role: There was no sponsorship of this work and no source of funding.