Agitation as a Paradoxical Effect of Divalproex Sodium: A Case Report

Mr. B. is a 39-year-old man with a 24-year history of schizoaffective disorder, cocaine and cannabinoid dependence, and noncompliance with medications. Mr. B. has multiple psychiatric hospitalizations, including commitment to the state psychiatric hospital. He was admitted to the UAB Center for Psychiatric Medicine for noncompliance with outpatient commitment, experiencing auditory hallucinations, persecutory delusions, and depressed mood without suicidal or homicidal ideation.

Physical examination was normal and his weight was 152 pounds. Urine drug screen was positive for cocaine only. Complete blood count with differential, Chem-7, liver function tests, and thyroid profile were normal except for a red blood cell count of 4.37 million/μl (range 4.40–5.80), mean corpuscular volume of 101 fl (range 83–99), and mean corpuscular hemoglobin of 34 pg (range 27–32). Valproic level was not checked on the day of admission. Since his prior admission 2 years earlier he had been on 1,250 mg of divalproex a day and had responded reasonably well when compliant. Upon admission, he was restarted on fluphenazine 5 mg bid and hs, divalproex sodium 500 mg in the morning and 750 mg at night, benztropine 1 mg bid, and clonazepam 1 mg q12 hr.

At the beginning of treatment, from time to time he was withdrawn, drowsy, preoccupied, depressed, and wanting to stay in bed; at other times he was hyperactive, irritable, agitated, pacing, admitting to racing thoughts, and talking loudly to himself. On several occasions he had to be restrained and given prn lorazepam and fluphenazine injections. Blood valproic level on day 5 of his hospitalization was 52 μg/ml. Fluphenazine dose was increased on day 6 to 5 mg bid and 10 mg at night. A repeat valproic level on day 7 was 74 μg/ml. He participated in unit activities, where he demonstrated minimal insight, poor judgment, and was withdrawn, disorganized, preoccupied with internal stimuli, and unable to focus for longer than 5 minutes. When redirected, he became agitated and irritable. Mr. B. was somatically preoccupied during his hospital stay, with complaints of constipation, increased thirst, throat pain, abdominal pain, nausea, loose stools, chest pain, shortness of breath, and tremor.

On day 12 of his hospitalization, he was argumentative and hypomanic. Considering his low valproic level of 59 μg/ml on that day, divalproex was increased to 1,000 mg in the morning and 1,000 mg at night. Despite this increased dose, the patient continued to be hypomanic, very talkative, loud, angry, loose, and easily agitated. On day 16 the valproic level was 69 μg/ml and the dose of divalproex was again increased to 1,000 mg in the morning, 500 mg in the afternoon, and 1,000 mg at night to control his ongoing hypomania, agitation, and hyperactivity. The following morning he began damaging his room, making loud noises, drinking excessive water, verbalizing with pressured speech, and seeking attention.

In the middle of the night on day 20 he rearranged furniture in his room. His valproic level taken that day was 79 μg/ml, which is low considering the total of 2,500 mg of divalproex he had been receiving. Later that day, without being provoked, he became very confused, loud, and extremely hostile and threatened the nursing staff. Attempts to redirect the patient failed and he became still more vocal. He received restraints, prn lorazepam, and fluphenazine.

It was hypothesized that this increased hyperactivity, hostility, and behavioral deterioration might be associated with increasing the dose of divalproex. Hence, divalproex was discontinued. In addition, fluphenazine was increased to 5 mg bid and 15 mg hs, clonazepam to 1 mg bid and hs. Propranolol 10 mg bid and hs was added to his medication regimen to control his hostility and aggression.

The following morning the patient was much calmer and without hallucinations. Propranolol was increased to 20 mg bid and hs. Blood ammonia level was 45 μmol/l on that day. Nursing staff also reported that the patient had become less restless, less impulsive and irritable, and more pleasant and cooperative. He agreed easily to his fluphenazine decanoate 50 mg injection that day and apologized to the staff for his earlier aggressive behavior. He mentioned that he felt better with discontinuation of divalproex. From time to time he was somewhat agitated, but those episodes were quite isolated and much less intense than earlier in the course of treatment and were easily manageable. He was compliant with his medication until the day of discharge.

Literature reporting this possible paradoxical side effect of divalproex is scant. Bellman and Ross² reported a similar effect of sodium valproate in a 14-year-old Asian male with seizure disorder who was started on 42 mg/kg (1,600 mg) per day of the medication. After 2 weeks the patient became confused, having recurrent episodes of bizarre behavior in which he stared vacantly, “shouted gibberish,” and had frightening hallucinations. The medication was stopped. Two days later he was still confused and appeared frightened, but he had steadily improved since the sodium valproate was stopped. The medication was recommenced at half the dose, and 10 days later his behavior returned to normal. Alvarez et al.³ reported the case of a 35-year-old male with seizure disorder who was assigned medication randomly as a part of a triple-blind study. At that time he was already receiving 100 mg phenytoin, 200 mg phenobarbital, and 175 mg thioridazine daily. Gradually the patient became drowsy, confused, disoriented, belligerent, and agitated, refusing to go to work. When the code was broken it was seen that he had been receiving 2,000 mg of valproic acid, which was stopped. Afterwards he was more responsive and more talkative, and 5 days later he “came back to his normal self.”

Behavioral alterations including irritability, longer and deeper sleep, hyperactivity, and aggressiveness were reported by Herranz et al.⁴ in a side-effect study where 88 pediatric patients received valproate as a monotherapy. Five out of 17 adult patients with acute mania who were treated with valproate under a placebo-controlled, double-blind condition showed no response to therapy and an increase of 3% to 13% in scores on the Young Mania Rating Scale.⁵ Among its many behavioral effects in animals, valproic acid induces hyperactivity and head shakes in rats.⁶ Among 100 children with seizure disorder who were treated with valproic acid, 8 children showed adverse effects characterized by belligerent behavior and hallucinations.⁷ The package insert for Depakote (Abbott Laboratories) mentions emotional upset, depression, psychosis, aggression, hyperactivity, hostility, and behavioral deterioration as possible side effects.⁸

At this point we are not sure of the specific cause or mechanism of the paradoxical effect of divalproex observed during the course of Mr. B.'s treatment. We are unsure whether divalproex was solely responsible for this agitation; the fact that his agitation was resolved toward the end of treatment could be explained by effects of other medications, or the episode might have resolved spontaneously. Drug interaction with other co-administered medications (fluphenazine, clonazepam, benztropine) might also have been partially responsible for the deterioration or the improvement. One interesting observation was that despite the high doses of the medication, the blood valproic level remained quite low. Considering similar reports in other literature, we urge physicians to be alert to the possibility of behavioral deterioration induced by this medication.