<<

. 14
( 68 .)



>>

placebo) experienced adverse events, and the most common adverse events were
headache, nausea, infection, pain, and accidental injury. Lamotrigine-related rash
occurred in 8% of patients during open stabilization and in no patients in the
randomized phase. There were no serious rashes during either phase of the study,
and no patients required hospitalization for a rash. From study entry to study
point, there was no mean change in body weight for the lamotrigine group.
This study had several limitations. Because of the absence of monotherapy data
relevant to this design, no statistical justification was used in determining sample
size. The actual enrollment of fewer than 100 patients per treatment arm limited
the power of the primary outcome analysis to approximately 47%. In contrast, the
analysis of survival in study was retrospectively determined to have been powered
at approximately 83%. Additionally, the design of this study did not permit an
analysis of time to relapse into a full episode of depression, hypomania, or mania
since patients were withdrawn at the first signs of relapse. In addition, this study
enrolled a higher proportion of patients with BP-I disorder than is thought to
occur in the rapid-cycling population.
It has been noted that lamotrigine was well tolerated during the current study.
The type and frequency of adverse events after lamotrigine treatment were com-
parable to placebo.
In this study, lamotrigine demonstrated efficacy in the prevention of the
recurrence of mood symptoms over a 6-month period. The results of this study
suggest that lamotrigine may be a well-tolerated and effective mood stabilizer with
prophylactic properties when used as monotherapy in some patients with rapid-
cycling bipolar disorder. Lamotrigine may be an especially effective mood
74 O. Elhaj and J. R. Calabrese


stabilizer for patients diagnosed with BP-II disorder. Despite this study™s pre-
viously mentioned limitations, it remains the first controlled study of its kind,
especially with regard to the process of data analysis utilized, which is more
generally suitable to prophylaxis studies.
Concurrently, Frye et al. (2000) reported a placebo-controlled study of lamotrigine
and gabapentin monotherapy in refractory mood disorders. The demographic profile
of the study participants who were available for evaluation in all three treatment
phases (n ¼ 31) included 18 women and 13 men; 11 BP-I and 14 BP-II; 23 rapid-
cycling and two non-rapid-cycling; and six unipolar patients.
The design initially employed a double-blind random assignment to parallel
arms. This was followed by 6-week cross-over trials, in which patients eventually
received all three medications as monotherapy (gabapentin, lamotrigine, or
placebo). Between phases, a cross-over period of approximately 1 week was used.
Patients were treated only with the study medications except for the following:
four patients who continued receiving stable levothyroxine supplementation for
corrected primary hypothyroidism; one patient who continued receiving diuretic
therapy for essential hypertension; and two patients, who each continued receiving
stable prior medications (i.e., triiodothyronine and clonazepam). The number of
patients randomized to lamotrigine (L), gabapentin (G), and placebo (P) were as
follows: PLG: 6; PGL: 5; LPG: 6; LGP: 6; GPL: 4; and GLP: 5. Lamotrigine was
initially administered at a dose of 25 mg daily for the first week, and titrated up to
300“500 mg/day for the fifth through sixth weeks. Gabapentin was administered
initially at a dose of 900 mg/day and was titrated to 4800 mg by the fifth through
sixth weeks. After the three phases of the study were completed, the patients who
had responded to a particular phase (I, II, or III) were offered the option of
returning to that phase, still on a blinded basis, for response confirmation.
The primary outcome measure of overall improvement was the CGI scale that
was modified for bipolar illness (CGI-BP). Supplementary ratings used in the
evaluation and completion of the CGI response rating included prospective self-
and observer-rated life charting, Hamilton-D, Young Mania Rating Scale (YMRS),
the Spielberger State Anxiety Scale, and Brief Psychiatric Rating Scale (BPRS).
CGI-BP change determinations were made by a consensus of blinded research
physicians and clinicians, both in comparison with the previous phase of illness
and the worst phase of documented illness. Previous treatment exposure and
documented treatment failures, including therapeutic level with inadequate
response, clinical intolerance, or affective relapse, were the following: for lithium,
28 (90%) of 31 patients experienced prior exposure and 28 (100%) of 28 patients
experienced prior treatment failure; for valproic acid, 26 (79%) of 31 and 21
(81%) of 26; and for carbamazepine, 20 (58%) of 31 and 14 (70%) of 20,
respectively.
75 Rapid-cycling bipolar disorder


Cochran™s Q-statistic was used to compare the overall CGI response (i.e.,
considered a rating of either much or very much improved) for those patients
who completed all three phases. Any significant Cochran™s Q statistic was followed
by a post hoc test. This test allowed pairwise comparisons of the proportion of
responders in the medication phases to determine the location of a difference. The
Fisher exact test was used for differential response rates based on gender status.

Clinical response
The mean daily doses at week 6 were 274 Æ 128 mg for lamotrigine
and 3987 Æ 856 mg for gabapentin. There was no difference in lamotrigine and
gabapentin doses between responders and non-responders. The response rates based
on the overall CGI rating of much or very much improved were the following:
lamotrigine, 52% (16 / 31); gabapentin, 26% (8 / 31), and placebo, 23% (7 / 31).
Post hoc Q differences (df ¼ 1, n ¼ 31) were the following: lamotrigine versus
gabapentin (Qdiff ¼ 5.33, P ¼ 0.011), lamotrigine versus placebo (Qdiff ¼ 4.76,
P ¼ 0.022), and gabapentin versus placebo (Qdiff ¼ 0.08, P ¼ 0.700). The CGI
response rates for manic and depressive components of the illness separately
were similar to the overall response rate, but they only achieved trend levels of
significance. The response rates for mania were the following: lamotrigine, 44%
(11 / 25); gabapentin, 20% (5 / 25); and placebo, 32% (8 / 25). For depression, the
response rates were the following: lamotrigine, 45% (14 / 31); gabapentin, 26%
(8 / 31); and placebo, 19% (6 / 31). The response rate observed during phase I was
highly similar to that in the whole study using all three phases of the cross-over
trial; it was 50% (5 / 10) for lamotrigine, 33% (3 / 9) for gabapentin, and 18% (2 /
11) for placebo. In addition, when the response data were analyzed as a function of
a positive response in the preceding phase, only 23% of lamotrigine responders,
50% of gabapentin responders, and 0% of placebo responders were also partial
responders in the previous phase, and would thus have entered the next phase
somewhat improved. This finding further indicates that there was not a greater
percentage of lamotrigine-responsive patients who were responders in the preced-
ing phase. Moreover, there were no significant differences at the baseline level on
any of the supplementary ratings of severity of illness. There was no difference
between the response rates based on gender. Both agents were generally well
tolerated, with the exception of one patient, who was administered lamotrigine
and who developed a rash after the 6-week study phase was over. The rash
occurred in week 15 (during the continuation treatment), progressed to toxic
epidermal necrolysis, and required the patient to be hospitalized in an intensive
care burn unit. The patient recovered fully. A pairwise contrast (lamotrigine versus
gabapentin, F ¼ 5.884, P ¼ 0.021) showed that patients lost weight when they
received lamotrigine relative to the weight gained when they received gabapentin.
76 O. Elhaj and J. R. Calabrese


This study suggests that lamotrigine monotherapy is superior to both gabapentin
and placebo treatments in patients with refractory affective disorders. Although
cross-over designs have the potential for producing carry-over and other con-
founding effects, this did not seem to occur in this study, according to the authors,
nor did it affect the interpretation of the outcome data based on a variety of
considerations. These data suggest that the response rates in this study were not
confounded by the treatment phase, nor were they influenced by carry-over effects
on the severity of illness. Nonetheless, the 6-week treatment phases and total study
period of 18 weeks is a short period in which to assess the efficacy or persistence of
response. A very high percentage (92%) of this bipolar sample was rapid cycling,
which is substantially greater than the general population estimates of this course
specifier (DSM-IV). Both refractory unipolar and bipolar patients were included
in this preliminary study. In addition, the direct application of these preliminary
monotherapy results to community treatment guidelines is limited, given that a
combination treatment is the norm and monotherapy regimens are rarely used in
clinical practice. In contrast to the weight gain observed during treatment with
gabapentin, patients lost weight while receiving lamotrigine therapy during this
6-week trial, further increasing the possibility of a better side-effects profile on
weight than other mood stabilizers such as lithium or valproate. However, and
despite this study™s controlled condition and vigorous data analysis, its applic-
ability to the larger population of patients with bipolar disorder, especially with
the rapid-cycling course, remains limited. It is worth noting that the cross-over
design, taking into consideration the rapidly changing nature of the rapid-cycling
course, might interfere with our ability to draw solid conclusions (Maj, 2001).
Additionally, there was no clear differentiation between previous treatments™
ineffectiveness and patient intolerance thereof.
Walden et al. (2000) reported an open longitudinal investigation, where 14
patients with rapid-cycling bipolar disorder were treated for 1 year with either
lithium or lamotrigine as a mood stabilizer. Out of the seven patients with
lithium, three out of seven (43%) had fewer than four and four out of seven
(57%) had four or more episodes. In the lamotrigine group, six out of seven
(86%) had fewer than four, and one out of seven (14%) had more than four
affective episodes (depressive, manic, hypomanic, or mixed). In fact, three out of
seven (43%) of the patients who were on lamotrigine therapy were without any
further affective episodes. There was no evidence of a preferential antidepressant
versus antimanic efficacy. Although the study is limited by the small number of
patients, the results are in line with other investigations, suggesting efficacy for
lamotrigine and a suboptimal response for lithium in rapid-cycling bipolar
disorder. These preliminary data need to be confirmed with controlled double-
blind studies.
77 Rapid-cycling bipolar disorder


Despite the generally positive outcomes of the previously cited studies on the
use of lamotrigine in the treatment of rapid-cycling bipolar disorder, more con-
trolled studies with longer courses are needed to clarify the role of this promising
agent.


Lithium
Despite being the oldest among the pharmacological armamentarium in the
treatment of bipolar disorder, lithium continues to draw attention to its utility
as an effective agent in the treatment of different aspects and phases of this
disorder, and especially in the rapid-cycling course. While some researchers con-
tinue to demonstrate and advocate for lithium efficacy in the treatment of rapid-
cycling bipolar disorder (Baldessarini et al., 2000, 2002; Swann et al., 2000; Viguera
et al., 2001; Tondo et al., 2001), others have argued that lithium has a poor efficacy
in treating rapid-cycling bipolar disorder, even when supplemented by anti-
depressants and neuroleptics (Post et al., 2000; Bowden, 2001).
Baldessarini et al. (2000) concluded, in a study evaluating the factors associated
with rapid-cycling status, that this subtype of bipolar disorder was strongly
associated with type II diagnosis, higher average pre-lithium episode frequency
and percentage time ill, and weakly with female sex, but not with greater overall
morbidity during treatment. In 360 DSM-IV BP-I (n ¼ 218), and BP-II (n ¼ 142)
disorder subjects (64% women) followed over an average of 13.3 years, researchers
evaluated factors associated with rapid-cycling status with bivariate and multi-
variate techniques, and response to lithium maintenance treatment (recurrence
rates, time ill, survival analysis of time to recurrence on lithium). Their results
indicated that the rapid-cycling risk (15.6% of cases) was 5.1 times greater in BP-II
versus BP-I subjects (30.3% versus 6.0%), in minor excess in women versus men
(17.9% versus 11.5%), and associated with premorbid cyclothymia, depressive first
episodes, older onset age, and being employed or married. Before lithium, rapid-
cycling versus non-rapid-cycling cases had more mean total (3.9 / 1.2), manic, and
depressive episodes/year, and greater percent time ill (60% versus 38%). During
treatment, prior rapid-cycling status was unrelated to time to first recurrence and
other measures of morbidity and improvement, including percent time ill, although
depressive episodes were 2.7 times more frequent, and there was 13.7% less chance
of full protection from all recurrences in rapid-cycling cases. Limitations of the study
included that it was naturalistic, without random assignment or blind assessment.
However, its length and the number of subjects studied make its conclusions
noteworthy.
On the other hand, Swann et al. (2000) reported that at least four previous
depressive or 12 previous manic episodes are associated with reduced antimanic
78 O. Elhaj and J. R. Calabrese


response to lithium. They found that response to lithium, but not to divalproex or
placebo, worsened with increased depressive or manic episodes. A history of more
than 11 manic or four depressive episodes was associated with response to lithium
that did not differ from placebo. Effects of previous depressive and manic episodes
appeared independent, and could not be accounted for by increased rapid cycling
or mixed states.
These cross-sectional data cannot distinguish whether this represents progres-
sive development of lithium resistance with repeated episodes, or patients who
had frequent episodes that were lithium-resistant from the start. The reported
increased episodes could be associated with an increased number of lithium
discontinuations, which might lead to neurophysiological changes adversely
affecting response to lithium, but not to divalproex. However, previous favorable
response to lithium in the present study predicted a favorable response in the index
episode, arguing against loss of lithium response with repeated episodes of treat-
ment. Further, most patients have similar responsiveness to lithium before and
after lithium discontinuation.
Viguera et al. (2001) compared the clinical characteristics of 360 women and
men with DSM-IV BP-I or BP-II disorder before and during clinical lithium
maintenance monotherapy in a mood-disorders clinic. They utilized preliminary
bivariate comparisons, multivariate analysis, and survival analysis of time stable
during treatment. They found that women (n ¼ 229) versus men (n ¼ 131) were
more likely to have BP-II disorder (1.6 times), 3.2 years older at illness onset, more
often depressed-before-manic (1.4 times), considered unipolar depressive 1.9
years longer, and started maintenance treatment 5.5 years later. However,
women differed little from men before treatment in overall morbidity, average
episode frequency, and risk of suicide attempts. Contrary to prediction, women
showed non-significantly superior responses to lithium treatment and a significant
60% longer median time before a first recurrence during treatment, despite 7%
lower average serum lithium concentrations. Women were diagnosed as bipolar
later than men, with corresponding delay of lithium maintenance treatment that
proved to be at least as effective as in men. A possible interpretation of this study™s
results might be a confirmation of the hypothesis that the increased rate of rapid
cycling in women is probably a reflection of the increased treatment with anti-
depressants. Frequency of episodes did not seem to differ greatly before the
treatment with lithium started, taking into consideration that women were more
misdiagnosed as unipolar depressed, and that they had more depressed episodes
than men.
To clarify further the role lithium plays in the thyroid gland™s function, Kupka
et al. (2002) reported the lack of association between lithium exposure and the
high rate of autoimmune thyroiditis in bipolar disorder. The thyroid peroxidase
79 Rapid-cycling bipolar disorder


antibodies (TPO-Abs) of outpatients with DSM-IV bipolar disorder (n ¼ 226), a
population control group (n ¼ 252), and psychiatric inpatients of any diagnosis
(n ¼ 3190) were measured. The TPO-Abs were more prevalent in bipolar patients
(28%) than population and psychiatric controls (3“18%). The presence of TPO-
Abs in bipolar patients was associated with thyroid failure, but not with age,
gender, mood state, rapid cycling, or lithium exposure. Thyroid failure was present
in 17% of bipolar patients and more prevalent in women. It was associated with
lithium exposure, especially in the presence of TPO-Abs, but not with current
rapid cycling, although an association may have been masked by thyroid hormone
replacement. The authors concluded that thyroid autoimmunity was highly pre-
valent in this sample of outpatients with bipolar disorder and not associated with
lithium treatment. These variables appear to be independent risk factors for the
development of hypothyroidism, especially in women with bipolar disorder.


Divalproex sodium
Swann (2001) analyzed the prediction of treatment response in acute mania in
controlled clinical trials with divalproex. For predictive factors, 179 subjects in
three parallel groups (divalproex, lithium, and placebo) were evaluated over a
period of 21 days by using structured interviews. For the follow-on study, 372
stabilized patients were randomized to three groups: divalproex, lithium, or
placebo. The results showed that patients with manic episodes with depressive
symptoms or with rapid cycling exhibited good response to divalproex. It was
concluded that a high number of both manic and depressive prior episodes is
predictive of poor response to lithium and favorable response to divalproex. The
results of this study add to the available evidence of the utility of divalproex
sodium in the treatment of rapid-cycling bipolar disorder.

<<

. 14
( 68 .)



>>