Valproate (Depakote) appears to be effective not only in treating mood
symptoms associated with bipolar disorder, but also in reducing drinking in
those patients with co-occurring alcoholism.
Data from a small, randomized, double-blind, controlled trial indicate that
dual effectiveness is the result of independent effects on the two different
disorders. The data provide support for the use of valproate to treat the more
than half of all patients with bipolar disorder who have co-occurring
substance abuse disorders. Among those with co-occurring substance abuse,
alcoholism is the most frequent.
"This has really been a population that has been previously neglected
in research," noted Ihsan Salloum, M.D., M.P.H., an associate professor
of psychiatry at the University of Pittsburgh School of Medicine. "Most
clinical trials involving patients with bipolar disorder have used `pure'
patients; they have excluded those with co-occurring disorders, and
conversely, those studies with patients who have substance abuse disorders
have commonly excluded any patients who have any co-occurring psychiatric
The result, Salloum told Psychiatric News, is a void in the
evidence base. Yet physicians dealing with a patient with bipolar disorder
know that "they are dealing with a patient who most likely also has a
substance abuse disorder. The degree of difficulty in treating the patient is
much worse; there simply are no clear guidelines that address these comorbid
Salloum and his colleagues had been encouraged by recent reports of small
trials studying the use of several anticonvulsant mood stabilizers, including
valproate and topiramate, in patients with alcoholism. Valproate, Salloum
noted, had already been granted an indication for bipolar disorder, so he and
his team set out to look at the effects of the drug in patients with both
bipolar and alcoholism.
They enrolled 59 patients with bipolar I disorder and alcoholism in a
24-week, double-blind, placebo-controlled, randomized parallel group trial.
Recognizing that it would be unethical to have patients in a trial for 24
weeks on placebo only, Salloum and his colleagues added lithium to the
protocol. All patients in the trial received "treatment as
usual"—which included lithium plus a psychosocial intervention
that included individual psychotherapy focused on the patients dealing with
both disorders as well as taking medication. In addition, patients were then
randomly assigned to have either valproate or placebo added to their treatment
The group receiving valproate was found to have a significantly lower
proportion of heavy drinking days and a trend toward fewer drinks per drinking
day compared with those who received placebo.
"When we added medication adherence as a variable, both findings
became statistically significant," Salloum noted. "That is, those
patients who had higher serum levels of valproate had significantly better
alcohol use outcomes overall—in reduced proportion of heavy drinking
days and in fewer drinks per heavy drinking day, as well as fewer drinks per
day throughout the study."
Valproate therapy also significantly prolonged the time to relapse to
sustained heavy drinking, to an average of 93 days, compared with an average
of 62 days for those on placebo.
Salloum and his colleagues also monitored mood symptoms but saw no
statistically significant differences between the two groups.
"The study was relatively small," Salloum noted. "Perhaps
if we had had more patients, we may have had more power to detect a
difference. But all patients were receiving lithium for their bipolar
symptoms, so the addition of valproate may not have made a noticeable
Nonetheless, Salloum added, "I think this is an encouraging study. It
is really the first to be done in this dually diagnosed adult population using
a double-blind, placebo-controlled paradigm."
Salloum's hope is that the trial will be replicated with larger numbers of
patients and in multiple clinical sites.
An abstract of "Efficacy of Valproate Maintenance in Patients
With Bipolar Disorder and Alcoholism" is posted online at<http://archpsyc.ama-assn.org/cgi/content/abstract/62/1/37>.▪
Arch Gen Psychiatry20056237