Association of Plasma Homovanillic Acid with Behavioral Symptoms in Patients Diagnosed with Dementia: A Preliminary Report Robert A. Sweet, Bruce G. Pollock, Benoit H. Mulsant, Jules Rosen, Kim Hung Lo, Jeffrey K. Yao, Ruth A. Henteleff, and Sati Mazumdar Neuroleptic treatment of psychotic symptoms or agitated behavior in elderly patients diagnosed with dementia is associated with reduced efficacy and increased rates of neuroleptic-induced parkinsonism in comparison to younger patients with schizophrenia. We report the first study to examine the relationship between an in vivo measure of dopaminergic function, plasma ho- movanillic acid (pHVA), and ratings of psychosis, agitation, and parkinsonism before and after neuroleptic treatment in dementia patients. Pretreatment pHVA was significantly correlated with parkinsonian rigidity, with a trend observed with agitation and hostility. Though mean pHVA did not change during perphenazine treatment, intraindividual change in pHVA at day 15 was correlated with improvement in hostility, with a similar trend for improvement in agitation. These preliminary findings are consistent with reports associating dopaminergic function with agitated, but not psychotic, symptoms in patients diagnosed with dementia, and with a reduced responsivity of dopaminergic systems to neuroleptic treatment in these patients. © 1997 Society of Biological Psychiatry Key Words: Homovanillic acid, dementia, Alzheimer’s disease, neuroleptic, psychosis, dopamine BIOL PSYCHIATRY 1997;42;1016 –1023 Introduction The rationale for the use of neuroleptic medication in the treatment of psychosis and agitation complicating demen- tia is largely derived by extrapolation from the observed efficacy of neuroleptics in the treatment of younger patients with psychoses. When neuroleptics are used in the acute treatment of schizophrenia, approximately 75% of patients demonstrate a moderate to marked response, in comparison to less than 25% of placebo-treated patients (Davis 1985). Estimates of neuroleptic efficacy in the treatment of behavioral syndromes complicating dementia, however, are much more modest. A total of eight random- ized, parallel group, placebo-controlled studies of neuro- leptics in the treatment of behavioral symptoms associated with dementia, comprising 285 patients, have been re- ported (Abse et al 1960; Hamilton and Bennett 1962a, 1962b; Sugerman et al 1994; Barnes et al 1982; Petrie et From the Geriatric Psychopharmacology, University of Pittsburgh School of Medicine (RAS, BGP, BHM, JR, JKY, RAH), Department of Veterans Affairs Medical Center (JKY) and University of Pittsburgh Graduate School of Public Health (KHL, SM), Pittsburgh, Pennsylvania. Address reprint requests to Robert A. Sweet, MD, 3811 O’Hara Street, Pittsburgh, PA 15213. Received August 6, 1996; revised December 9, 1996. © 1997 Society of Biological Psychiatry 0006-3223/97/$17.00 PII S0006-3223(97)00146-7