Chronic divalproex sodium to attenuate agitation and clinical progression of Alzheimer disease

Abstract

Context: Agitation and psychosis are common in Alzheimer disease and cause considerable morbidity. We attempted to delay or to prevent agitation and psychosis with the use of divalproex sodium (valproate).

Objective: To determine whether treatment with valproate could delay or prevent emergence of agitation or psychosis.

Design, setting, and patients: A multicenter, randomized, double-blind, placebo-controlled trial of flexible-dose valproate in 313 (of 513 screened) individuals with moderate Alzheimer disease who had not yet experienced agitation or psychosis. The study was conducted from November 1, 2005, through March 31, 2009, at 46 sites in the United States.

Intervention: Participants were randomly assigned to valproate treatment at a target dose of 10 to 12 mg per kilogram of body weight per day or identical-appearing placebo for 24 months followed by a 2-month period of single-blind placebo treatment.

Main outcome measure: Time to emergence of clinically significant agitation or psychosis.

Results: A total of 122 participants (59 receiving valproate and 63 receiving placebo) completed 24 months of treatment while taking study medication; 42 (27 receiving valproate and 15 receiving placebo) reached 24 months having discontinued study medication; 150 reached month 26. There was no difference between groups in time to emergence of agitation or psychosis (Cox proportional hazard ratio, 0.96; P = .88). There was no difference between groups in change on any secondary outcome. The valproate group had higher rates of somnolence, gait disturbance, tremor, diarrhea, and weakness. Eighty-eight participants underwent magnetic resonance imaging scans at baseline and 12 months; the valproate group showed greater loss in hippocampal and whole-brain volume, accompanied by greater ventricular expansion (P < .001).

Conclusion: Valproate treatment did not delay emergence of agitation or psychosis or slow cognitive or functional decline in patients with moderate Alzheimer disease and was associated with significant toxic effects.

Trial registration: ClinicalTrials.gov NCT00071721.

Figure 1.

Disposition of subjects. *Reasons for screen failures were as follows. Cognitive screen (Mini-Mental State Examination) out of range: 110; behavioral screen (Neuropsychiatric Inventory) out of range: 49; withdrawn consent/miscellaneous: 41; medically unstable: 16; excluded medication: 15; abnormal laboratory test result: 13; excluded comorbid diagnosis: 6; age out of range: 5. †Reasons for treatment discontinuation (as indicated by study site personnel) were as follows (number receiving placebo, number receiving valproate, respectively): lost to follow-up/death: 17, 13; concern about placebo condition: 8, 4; protocol violation: 2, 0; concern about length of protocol: 4, 4; concern about potential adverse effects of study medication: 12, 25; end point reached: 23, 23; investigator judgment: 8, 6; frequency of assessments: 3, 4; study partner unwilling or unable to participate: 12, 19; concern about safety risk: 4, 7; nonadherence: 7, 3; participant chose alternative treatment: 9, 4; other: 28, 30. (Note: a subject might have >1 reason for discontinuation.)

Figure 2.

Survival to end point using Cox proportional hazards model. The hazard ratio was 0.96 (P=.88).