The hypothesis to be tested is whether children below age thirteen with the same symptom
cluster targeted in that protocol (explosive temper outbursts and severe mood swings,
Explosive Mood Disorder) will show the same response to open label Depakote that the
adolescents showed. An addition in this protocol is to see if this hypothesis holds under
double blind placebo controlled conditions.
The investigators have now completed open treatment with six children and have found that the
same criteria that selected a treatment responsive group of adolescents and adults also
worked for five of these six children. The investigators now propose to test these findings
under double blind conditions. There is no change in the overall safety profile of the
treatment. The only change is the addition of a placebo phase. All children will receive
active treatment at some point.
In addition, the investigators hypothesize that the clinical syndrome of temper and
irritability will tend to resolve with development and that families will not exhibit
bipolarity in first degree relatives but will exhibit temper outbursts. This hypothesis will
be studied in a follow-up study.
Data Summary
OPEN TRIAL
Patient Age Dose Acute Outcome Long Term Outcome
age 10 1000mgs Good Excellent
12 1000mgs Fair-good Poor (non-compliant
10 1000mgs Excellent Good
9 1000mgs Excellent Very good
12 1000mgs Poor Poor
7 500mgs Excellent Good-excellent 7 11 500 mgs Dropped out after 3 weeks; never took
meds Poor
BACKGROUND The disruptive disorders of childhood and adolescence are conduct,
oppositional-defiant, and attention deficit hyperactivity disorders. Of these, only the
latter has a specific pharmacologic treatment. The other two reliably describe problem
behaviors of special education classes, chemical dependency units and family courts but have
few treatment implications. In a previous communications, the investigators described
adolescent subjects with temper outburst and severe mood swings. The investigators labeled
the syndrome Explosive Mood Disorder (EMD) and drafted tentative criteria for temper outburst
and irritable mood swings (see Inclusion Criteria for EMD). The goal of this study is to
demonstrate that Depakote can help a certain type of problem child, one whose disruptiveness
is associated with mood dysregulation.
The investigators hypothesized the subjects' mood dysregulation and aggressive disinhibition
to be "limbic" pathology. The limbic system is heavily implicated in behavior and emotion,
and work by previous investigators suggested episodic dyscontrol can result from limbic
irritability. Since carbamazapine and valproic acid are used in limbic seizures and bipolar
spectrum disorders, the investigators predicted subjects would improve on divalproex sodium
(Depakote), the enteric coated form of valproic acid. The investigators treated thirteen
patients meeting the above criteria criteria with open label divalproex sodium. In all cases,
there were strikingly beneficial results, including a sharp decline in the frequency and
severity of temper outbursts and the severity of mood swings.
The typical patient would progress from irritation to screaming, door-slamming, property
destruction or fighting explosively, i.e., without intermediate steps. When not exploding,
the patient's affect and frustration tolerance would fluctuate throughout the day. The mood
could be irritable, normal or silly but always unpredictable. The patients had chronic
interpersonal difficulties, especially with authority figures, thus serious problems with
family, school and work were common. They tended to end up with deviant peers, since others
could not tolerate the unpredictable interactions. Though some of patients met criteria for
Conduct or Oppositional Defiant Disorder, these diagnoses were too heterogeneous to describe
the Depakote responders. Cyclothymia and Bipolar II Disorder did not capture the chronic,
maladaptive interactions with peers and authority figures. Intermittent Explosive Disorder
missed the above plus the connection to mood dysregulation. Borderline Personality implied a
specific type of interpersonal pathology (idealization, devaluation, manipulation) that was
not necessarily present.
The closest match the investigators found in the literature to our patients was "Emotionally
Unstable Character Disorder" (EUCD). This diagnosis described young adults with extreme
irritability, mood swings, chronic maladaptive behaviors, and a clear response to a mood
stabilizing agent (lithium carbonate). However, even this diagnosis does not capture the
explosiveness which is prominent in the sample of Depakote responders. The investigators
therefore decided to use the term Explosive Mood Disorder (EMD) to capture the temper
outbursts and the mood lability.
Campbell reported hospitalized conduct disordered children with affective (as opposed to
predatory) aggression improvement on lithium and Findling reported similar findings in
similar outpatients treated with risperidone. The key feature of irritable mood swings is not
directly addressed in these studies. In addition, other studies of childhood aggression in
Autism, Tourette Syndrome, Bipolar Disorder and mental deficiency suggest a role for
risperidone. Nonetheless, it is still not clear what alternative pharmacological treatments
for temper driven by irritable mood swings are available.
The investigators treated ten subjects who met screening criteria for EMD with open label
Depakote 1000mgs. Baseline evaluation included a full medical and psychiatric interview, the
Structured Clinical Interview for Diagnostic and Statistical Manual of Psychiatry-III-R and a
record of the total number of explosions during the previous week and thirty days. The latter
estimate was obtained from as many sources as possible. At minimum, the patient and one
parent had to each give an estimate. As many other sources as possible were also asked to
estimate the lability of the patient's mood during the previous week and thirty days. Each
week the same information would be obtained about the previous week and a global assessment
made as to whether the patient was unimproved, minimally improved, much improved or very much
improved.
Twelve adolescents agreed to participate, and two dropped out before receiving an adequate
trial. The other ten adolescents showed clear improvement in mood and number of explosions,
with the latter typically declining by 90-100%. All ten adolescent patients were judged at
least much improved by the psychiatrist, parents and patient. These results were later
replicated under double-blind, placebo controlled conditions.
Teachers and parents approached the investigators with children under the age of 13 and asked
us to treat them. The investigators obtained permission to collect open label data on those
youngsters. The completion of open-label data has led to the current study.