Stanley Center for the Innovative Treatment of Bipolar Disorder

SECOND INTERNATIONAL CONFERENCE 
ON BIPOLAR DISORDER


Introduction

Poster Sessions

Audio Recordings

Bipolar Conferences Home

Poster Sessions

Table of Contents:

The Nature of Activity-Energy Among Unipolar Patients Who Switch To Bipolar and Those With a Stable Diagnosis

Authors: Hagop S. Akiskal, M.D., Jack D. Maser, Ph.D.
and Pamela Zeller, Ph.D.

Previous research (Akiskal et al., 1995) demonstrated four personality factors important to the switching process in Bipolar Disorder. One of these factors was Activity-Energy, with the unipolars who switched to Bipolar II Disorder showing significantly higher values on this factor than those unipolars who remained unipolar. Here we present data taken from the NIMH Collaborative Depression Study on 16 Bipolar I patients when depressed, 44 Bipolar II patients when depressed, and 460 unipolar patients who were never bipolar and remained unipolar over the five years of follow-up. Items on which the Bipolar II patients (now depressed) scored significantly (p < .05) higher than the other two groups were: 1) You are happiest when involved in some project that calls for rapid action, 2) You are often so much "on the go" that sooner or later you may wear yourself out, 3) You are able to work for unusually long hours without feeling tired, and 4) Others regard you as a lively individual. The data are interpreted to mean that Bipolar II patients, even when depressed, have elevated Activity-Energy compared to Bipolar I or unipolar patients.

The Altman Self-Rating Mania Scale (ASRM)

Authors: Edward Altman, Donald Hedeker, James L. Peterson, John M. Davis

Objective: We report on the development, reliability and validity of the Altman Self-Rating Mania Scale (ASRM).

Method: The ASRM was completed during medication washout and after treatment by 22 schizophrenic, 13 schizoaffective, 36 depressed, and 34 manic patients. The CARS-M and MRS were completed at the same time to measure concurrent validity. Test-retest reliability was assessed separately on 20 depressed and 10 manic patients who completed the ASRM twice during medication washout.

Results: Principal components analysis of ASRM items revealed three factors: mania, psychotic symptoms, and irritability. Baseline mania subscale scores were significantly higher for manic patients compared to all other diagnostic groups. Manic patients had significantly decreased post-treatment scores for all three subscales. ASRM mania subscale scores were significantly correlated with MRS total scores (r = 0.718) and CARS-M mania subscale scores (r = 0.766). Test-retest reliability for the ASRM was significant for all three subscales (r = 0.86, p<.001; r = 0.89, p<.001; r = 0.88, p<.001). Cronbach's alpha resulted in values of 0.79 for subscale 1 and 0.65 for subscales 2 and 3 respectively. Significant differences in severity levels were found for some symptoms between patient ratings on the ASRM and clinician ratings on the CARS-M. The presence or absence of insight was not significantly related to patients' responses on the ASRM. Mania subscale scores of greater than 5 on the ASRM resulted in values of 85.5% for sensitivity and 87.3% for specificity.

Conclusions: The ASRM is a brief, reliable, and valid self-rating scale for assessing the presence and/or severity of manic symptomatology. Advantages of the ASRM over other self-rating mania scales are discussed. Differences between patient and clinician ratings for some items (elevated mood, grandiosity) suggest some denial or under-reporting of severity levels in manic patients with mild to moderate symptomatology.

Increased Amygdala Volume on mri is Specific for bipolar Disorder

Authors: Altshuler LL, Bartzokis GB, Grieder T, Curran J, Mintz J

Brain MRI's (SPGR-3D acquisition in the coronal plane) were obtained on 12 bipolar, 14 schizophrenic and 18 control subjects to assess for differences in limbic system structures. 3-D volume reconstruction software (ISG Technologies) was used to reorient brain images in all three planes to decrease variance of head position across subjects. The hippocampus, parahippocampus, amygdala and temporal lobes were manually traced by a rater blinded to diagnosis. Limbic structures were delineated using internal brain landmarks. Data were analyzed using repeated measures factorial ANCOVA with age and height as covariates for temporal lobes, and age and temporal lobe as covariates for other limbic structures. Design factors of diagnosis, hemisphere and their interaction were included (general linear mixed model with unstructured covariates matrix). A significant main effect of diagnosis was seen (F2,40) = 5.2, p = .01. Pairwise contrasts revealed bipolar subjects had significantly larger amygdala compared to both schizophrenic (t = 3.2, df = 40, p = .003) and control (t = 2.1, df = 40, p = .04) subjects. No significant differences in temporal lobe volumes were found across groups and no significant interaction by hemisphere was observed. Course of illness variables (duration ill, number of manias, number of depressions) and their relationship to amygdala volume were assessed (general linear mixed model using data from both hemispheres, covaried for age, cortisol, and height). Illness variables were log-transformed as they were non-normally distributed. The number of episodes of mania, not total duration ill nor number of episodes of depressions, was significantly positively correlated with amygdala size (df = 19, F = 6.16, p = .02). Implications of these findings will be discussed.

Reduced Endogenous ADP-Ribosylation of G? s in Postmortem
Bipolar Disorder Brain

Authors: S. Andreopoulos, K.P. Siu, P.P. Li, and J.J.Warsh

Recent observations support altered signal transduction processes in bipolar disorder (BD). Findings of increased levels of the stimulatory guanine nucleotide (G) protein ? subunit, ? s, elevated forskolin-stimulated cAMP production, and alterations in cytosolic protein kinase A (a downstream target of cAMP action) in autopsied cerebral cortical regions from BD postmortem brain, support the notion that ? s-mediated hyperfunctionality occurs in this disorder. Lack of alterations in ? s mRNA levels, and negative evidence of linkage between the gene encoding ? s and BD, suggest hyperfunctional ? s may occur consequent to changes in posttranslational mechanisms governing ? s turnover, such as adenosine-diphosphate (ADP)-ribosylation. To test this hypothesis, endogenous and cholera toxin (CTX)-catalyzed ADP-ribosylation products of ? s were measured in postmortem temporal (n=9), occipital (n=10) and cerebellar cortex (n=7) of BD, and age and postmortem delay matched controls. ANOVA revealed significant main effects of diagnostic group (F=5.41, df=l ,51, p=0.025) and brain region (F=8.24, df=2,51, p=0.001) for endogenous ADP-ribosylation of ? s-s (short form of ? s). On posthoc analysis, endogenous ADP-ribosylated ? s-s was significantly lower (30%) in BD temporal cortex compared to controls (ROD = 0.62 ? 0.18, mean ? SD, vs 0.89 ? 0.26; t=2.55, df=16, p<0.05) but only showed a nonsignificant trend towards a decrement in occipital and cerebellar regions. Endogenously ADP-ribosylated ? s-L (long form) was only weakly detectable at the protein concentrations (250 ? g) used. CTX-catalyzed ADP-ribosylation of ? s-s was also somewhat, but not statistically significantly lower, in BD temporal cortex compared to controls. CTX-catalyzed ADP-ribosyled ? s-L was significantly higher (F=8.31, df=2,51, p=0.001) in temporal and occipital cortex compared with cerebellum but did not differ between groups (F=0.48, df=1,51, p=0.49). These preliminary observations suggest that disturbances in the mechanisms regulating ADP-ribosylation of ? s may occur in BD brain and raise the possibility that enhanced levels and hyperfunctionality of ? s may occur secondary to alterations in these processes .

Review of Data From a Newly Established Bipolar Unit at a University Hospital in Turkey

Authors: _ zerdem AyÕ egh l, M.D., Tunca Zeliha, M.D., Kaya Nezaket, M.D.

We report data from our bipolar outpatient unit collected by using life charts of 61 registered patients (45 hospitalized during last 5 years, 16 referred from the outpatient unit). Sex distribution: Female: 42, Male:19; F/M=2.2. Age range: 19-75, (mean: 38.3? 15.1). Status of employment: 36.1% housewife, 16.4% student, 14.8% professional, 8.2% retired, 8.1% unemployed, 14.7% other. Most of our patients had either university (41%) or high school level (26.2%) education. 62% came from inner town. 41% were married, 41% unmarried and only 6.6% were divorced for reasons other than bipolarity. 45 patients were bipolar I and 16 bipolar II. Age of onset: 27.3? 11.6 (range: 13-72). 6.8% of bipolar I and 37.5 % of bipolar II patients had multiple episodes which could not be counted reliably. Total number of documented episodes were 5.9? 5.1(1-28) and 7.0? 5.3 (3-21) for bipolar I and bipolar II, respectively. There was no correlation between the age of onset and number of episodes in both groups. Mean number of episodes before lithium treatment was 4.0? 3.2. Thirty-eight patients (63%) received neuroleptics at some time during the course of illness, mostly for the acute phase for 2-4 weeks, and 10 (16.4%) were given electroconvulsive therapy. Five patients (8.2%) received neuroleptics for 57.6? 44.2 months because of delayed diagnosis. Twenty-one of 34 patients (61.8%) received lithium alone, 11 (32.4%), a combination of lithium with carbamazepine and/or valproate, and two, valproate and carbamazepine alone. Mean duration of lithium treatment was 39.1? 42.6 months. Thirteen patients stopped lithium for various reasons and developed a new episode within a year. Clinical status of the 34 patients when last seen were euthymic (27), depressed (5), hypomanic (1) and manic (1). We would like to emphasize the importance of life chart in following and documenting bipolar patients.

Lithium Regulation of Brain MyoInositol in Bipolar Affective Disorder

Authors: Joseph M. Bebchuk, M.D., Gregory Moore, Ph.D.,
Husseini K. Manji, M.D.

The discovery of lithium's efficacy as a mood-stabilizing agent revolutionized the treatment of patients with Bipolar Affective Disorder. Lithium continues to be the mainstay of treatment for both the acute and prophylactic treatment of Bipolar Affective Disorder. However, despite its unquestioned efficacy, the biochemical basis for lithium's mood-stabilizing actions remains to be fully elucidated. Furthermore, increasing evidence suggests that a significant number of patients respond poorly to lithium therapy. Studies such as these indicate two important and highly clinically relevant directions for future research: firstly, the need to better identify patients likely to respond to lithium treatment, and secondly, the necessity to develop more effective treatment regimens.

The most widely accepted hypothesis underlying lithium' s therapeutic efficacy is the inositol depletion hypothesis. This hypothesis posits that lithium produces a relative depletion of myoinositol in critical areas of the brain and it is this depletion of a major precursor of the phosphoinositide second messenger system which results in its therapeutic effects. Despite the attractiveness of the inositol depletion hypothesis, it has never been investigated in manic-depressive patients. Thus, there is a clear need to determine if lithium reduces the levels of myoinositol in critical brain regions of individuals with manic-depressive illness, and if individual differences in susceptibility to lithium-induced CNS myoinositol reductions represent major factors determining resistance or sensitivity to lithium's therapeutic effects.

We are therefore conducting resistance or sensitivity to lithium's spectroscopy (MRS) to measure levels of myo-inositol (proton spectroscopy) and inositol-1-phosphate (phosphorus spectroscopy) in the following brain areas of manic patients and healthy volunteers: i) frontal cortex; ii) hippocampus; iii) occipital cortex; iv) cerebellum. Myoinositol and inositol-l-phosphate levels are being examined at 3 times points: i) at baseline; ii) after acute (5-7 days) lithium administration; and iii) after chronic (4 week) lithium administration. To date, we have found that similar to the effects observed in rodent brain, lithium reduces the myoinositol levels in critical brain regions in patients suffering from Bipolar Affective Disorder. The hypothesis that lithium-induced alterations in brain myoinositol levels are associated with responsiveness to its therapeutic effects are currently under investigation.

ADHD and Bipolar Disorder as Predictors of Clinical Response to Lamotrigine

Author: Jeffrey L. Berlant, M.D., Ph.D.

Objective: The presenter explored predictors of clinical outcome associated with treatment of severe mood disorders with lamotrigine.

Method: Fourteen adults with unsatisfactory response to conventional medication treatment of severe mood disorders received lamotrigine in a naturalistic trial. Outcome variables included lamotrigine discontinuation and retrospective Clinical Global Impression scores for six variables: overall outcome, depression, hyperactivity, concentration, mood lability, and general functioning. The investigation examined associations of lamotrigine effects with age, sex, mood disorder diagnosis, ADHD diagnosis, and Wender Utah Rating Scale score.

Results: Ten patients met DSM IV criteria for Bipolar Disorder, ten for ADHD, and seven (50%) for both. Mean group WURS score was 41.6, mean age 37.1 years, with a 1:1 sex ratio. Mean duration of treatment was 16.9 weeks, with 57% of cases discontinuing lamotrigine. Mean CGI scores for the six variables suggested mild improvement. Age and sex had no significant effect on any outcome variables. WURS score significantly predicted medication continuation (mean 52 versus 34 for discontinuers), and five CGI scores correlated with the WURS (r = 0.50 to 0.62). Diagnosis of Bipolar Disorder had no significant effect on discontinuation rates or CGI scores, although all four non-bipolar patients discontinued lamotrigine use. Four CGI scores were significantly higher for those with ADHD diagnoses (overall outcome, depression, hyperactivity, and functioning). The lowest discontinuation rates and highest CGI outcomes were for the group with both Bipolar Disorder and ADHD. WURS scores were mildly correlated with five CGI outcomes for bipolar cases (r = 0.59 to 0.73). For five CGI outcomes the most robust correlation with WURS scores occurred with ADHD diagnosis (r = 0.70 to 0.85), and no further predictive power appeared when bipolar and ADHD groups were combined.

Conclusions: Lamotrigine may improve Bipolar Disorder. Elevated WURS scores and ADHD diagnosis may further predict treatment response. Further studies are definitely warranted.

Case Report of Synergistic Response of Rapid Cycling Bipolar Disorder to
Lamotrigine, Risperidone, and Verapamil

Author: Jeffrey L. Berlant, M.D., Ph.D.

Objective: The poster presents several months of daily mood chart recordings from a patient with bipolar disorder to demonstrate the effect of a series of medication combinations on cycle frequency.

Method: A professional male, now 33, with Bipolar Disorder, Mixed Type, Rapid Cycling Subtype, and Attention Deficit Hyperactivity Disorder, kept daily mood chart recordings for 14 days in February 1995 and thereafter from July 1995 until December 1996, except for an eight-week hiatus in early 1996. Using a 10 point scale (presented on the poster), he operationalized his mood states to facilitate internal reliability of scoring. He also developed methods for recording diurnal mood lability as an outcome variable. Records of changes in medication and dosage associate therapeutic changes with control of mood cycling frequency and diurnal lability.

Results: Following unsatisfactory trials of valproate and verapamil, carbamazepine and levothyroxine diminished mood cycling and lability but left a chronically depressed state. Decreased carbamazepine lifted mood level but released mood cycling. A verapamil/carbamazepine trial reduced mixed state symptoms and decreased mood reactivity but left significant residual depressed mood. An empirical trial of B12 injections was unhelpful. When mood cycling and chronic depression worsened, addition of risperidone in low doses mildly improved mood level and cycling. Trial discontinuation of verapamil resulted in a major depressive dip, with recovery within a day after reinstitution of verapamil. Beginning in April 1996, a combination trial of lamotrigine, verapamil, and risperidone elicited general mood enhancement, but decreasing the dose of risperidone released mood cycling and increased mood lability. The patient is currently doing well on lamotrigine 250 mg/d, verapamil 320 mg/d, and risperidone 1 mg/d.

Conclusions: Combination psychopharmacotherapy may be essential for some rapid cycling patients. Lamotrigine, but only in combination with other mood stabilizing agents, was beneficial for this individual.

Diagnostic Criteria for Mixed Manic States

Authors: Frederick Cassidy, M.D., Eileen Ahearn, M.D., Ph.D.,
Elizabeth Murry, Kara Forest, M.D.
and Bernard J. Carroll, M.B., Ph.D.

Although mixed states of bipolar disorder have been long recognized, no consensus of how best to define them has developed. Numerous researchers have suggested that the definitions adopted by DSM-III-R and DSM-IV are too rigid. Two hundred thirty-seven subjects meeting criteria for Bipolar Disorder, manic or mixed, were evaluated both by DSM-III-R criteria and the Scale for Manic States. In a previous factor analysis of this scale, a factor was identified which represented dysphoric mood in manic patients. The distribution of subject scores on that factor was bimodal, with one mode comprising patients with high scores for dysphoric mood, guilt, lability, anxiety, and suicidality.

That distribution was used to divide the cohort into two groups. Sensitivities, specificities, positive and negative predictive values and diagnostic efficiencies for the dysphoric mode of that factor were calculated for various signs and symptoms suggested to be relevant to mixed states. Six symptoms: depressed mood, anhedonia, fatigue, feelings of guilt or worthlessness, recurrent thoughts of death or recurrent suicidal thoughts, and anxiety had adequate positive predictive values for inclusion in a definition of mixed states. Various definitions of mixed states were tested against this empirical subgroup of dysphoric mania, and receiver operating curves were constructed for these definitions. A definition comprising all six symptoms performed better than the DSM-III-R definition for Bipolar Disorder, mixed. These data suggest that DSM-III-R criteria for Bipolar Disorder, mixed, are too rigid. Alternate definitions of mixed states are less restrictive and provide a better compromise between sensitivity and specificity. Further studies are needed to test the validity of current definitions of mixed bipolar disorder.

Neurometabolic and Neuropsychologic Functioning in Bipolar Children Pre- and Post- ECT

Authors: H.E. Courvoisie, M.D, S.R. Hooper, Ph.D., L. Kwock, M.D.

Electroconvulsive Therapy (ECT) is a well-established treatment for depression. Although reported effective in depressed children as young as five years old, questions remain about the developmental impact of ECT in children. To date, there are no systematic reviews of treatment impact on neurometabolic and neuropsychological outcome in children.

This presentation describes three prepubescent male children who received ECT, for a primary BPAD with intractable mania. Each had failed multiple courses of mono- and polytherapy of mood stabilizers, and other appropriate medications.

Subjects were recruited into this study as part of a larger study examining children with BPAD. The parent/caregiver participated in a structured interview, and the mania rating scale was completed by designated staff. All subjects underwent a drug washout prior to ECT. BPAD probands received a comprehensive neuropsychological evaluation and Magnetic Resonance Spectroscopy (MRS). The MRS targeted the neurometabolic activity in bilateral frontal regions based on preliminary hypotheses from our ongoing BPAD study. Two subjects received 6-12 unilateral (left-sided) treatments. One child started with unilateral and then had bilateral. All policies formulated by the American Psychiatric Association for ECT were followed.

Two bipolar patients underwent single volume localized proton MRS emphasizing the frontal brain region. Post-ECT, a 2-4 fold increase in the lipid/lactate peak area resonances was found in comparison with the pre-ECT MRS study. No significant changes were found in peak areas for N-acetylaspartate, choline, creatine, and myoinositol. These findings correlate with proton MRS studies performed by Woods and Chiu (Ann. Neuol. 1990; 28:745-749) on ECT-treated adult patients.

Concurrent neuropsychological testing revealed average intellectual abilities but significant interprofile variability in all three patients. Short-term memory impairments were noted in two patients. Follow-up MRS and neuropsychological testing will be conducted six months post-treatment.

Bipolar Disorder as Manifested Within Borderline Personality Disorder

Authors: Deltito JA, Martin LY, Riefkohl J, Kissilenko A, Halligan P,
Austria B, Morse C, Corless P

Several lines of evidence suggest that many patients classified by DSM-IV as suffering from Borderline Personality Disorder may fundamentally suffer from a Bipolar Spectrum Disorder. These findings may also extend to other Axis II Disorders marked by affective instability such as Histrionic and Sociopathic Personality Disorders. Evidence further suggests family members of patients with Borderline Personality oftentimes suffer from psychiatric syndromes which present with increased irritability and affective instability. In addition, preliminary investigation suggests that medications used to control Bipolar Disorder, such as Divalproex Sodium may be quite helpful in the overall treatment of borderline patients.

Noting the above-mentioned factors, we have embarked on a psychopathologic study of the Borderline Personality Disorder Syndrome with the intent to document and analyze the contribution of bipolar spectrum pathology to the patients’ overall psychopathological syndromes.

Data collection is now half completed and should be concluded by the early Spring of 1997. Through the use of a modified version of the SCID, the Cornell Personality Organization Questionnaire, the Family History-Research Diagnostic Criteria (FHRDC) and other tools, we have attempted to quantify the contribution of Bipolarity to the borderline state. Preliminary analysis of the data demonstrated this contribution to be significant. Through the use of logistic regression models we plan to make some determination regarding whether bipolar disorder should be conceptualized as a frequently encountered co-morbid entity, part of a larger syndrome with Borderline Personality Disorder or truly part of its causality.

This psychopathological study is meant to inform decisions regarding the use of psychopharmacological treatment regimens in the treatment of Borderline Personality Disorder and is a precursor to a larger controlled clinical trial of such agents.

The Usefulness of Divalproex Sodium in the Treatment of Bipolar and Behavioral Disorders on an Adolescent Inpatient Unit

Authors: Deltito J, Levitan J, Damore J, Zambenedetti M, Hajal F

Controlled clinical trials provide necessary information regarding the safety, tolerability and efficacy of any psychopharmacologic agent. Yet constraints inherent in such methodology leave many issues related to the usefulness of these preparations in the naturalistic setting unanswered. Therefore, in order to receive a full picture of a given agent's usefulness, data from controlled clinical trials needs to be complemented with data from the naturalistic setting.

With this in mind, we have completed the data collection relevant to the use of Divalproex Sodium in all patients admitted to an adolescent inpatient unit of New York Hospital-Cornell Medical Center-Westchester Division over a one-year period. In all, data was collected on 200 patients, 30 percent of whom were treated with Divalproex Sodium.

Data collection is completed but is in the process of analysis.

Preliminary analysis would suggest Divalproex Sodium to be a useful (safe, well-tolerated, and effective) medicine to use in this patient population for treating bipolar spectrum disorders and other commonly encountered behavioral disturbances in adolescents.

Prophylactic efficacy of lithium, carbamazepine, valproate and their combination in bipolar disorder

Authors: Kirk D. Denicoff, M.D., Earlian Smith-Jackson, R.N.,
Ann L. Bryan, B.A., S. Omar Ali, B.S., and Robert M. Post, M.D.

Objective: To study the prophylactic efficacy of lithium, carbamazepine, valproate and their combination in bipolar illness.

Method: Fifty-two outpatients who met DSM III-R criteria for bipolar illness (29 BPI, 23 BPII) were randomized in a double-blind design for an intended one year of treatment with lithium or carbamazepine, a crossover to the opposite drug in the second year, and a third year on the combination. Antidepressants and antimanic agents were used acutely as necessary. Patients entered a fourth phase of valproate (? ) lithium because of inadequate response and/or significant side effects; inadequate responders were offered triple mood stabilizer therapy (lithium, valproate, and carbamazepine). Patients received detailed evaluations monthly, and daily life chart ratings of the degree of functional incapacity associated with mania or depression.

Results: Evaluable patients who had marked or moderate improvement on the Clinical Global Impressions (CGI) scale for each phase were: lithium 14 of 42 (33.3%); carbamazepine 11 of 35 (31.4%); lithium + carbamazepine 16 of 29 (55.2%); valproate (? ) lithium 6 of 18 (33.3%); and triple therapy 3 of 7 (42.9%). The cumulative response rate was only 61.9% (19.0% marked,42.9% moderate in any phase) from the original randomized evaluable cohort. At baseline year (worst year of retrospective illness) patients were manic 26.0% of the year and depressed 33.0% of the year; this improved to 5% and 16% respectively during the patients' best treatment phase. Thus while 62% showed a noticeable improvement in at least one of the prospective phases, 38% did not, and substantial morbidity remained.

Conclusion: These data on the one hand highlight that persistence in sequentially trying and adding different mood stabilizing treatments can achieve good results for many bipolar patients. On the other hand, the lack of adequate response in a large proportion of our patients highlights the substantial need for new treatment options in the long-term prophylaxis of patients with bipolar illness.

Lamotrigine treatment in rapid cycling bipolar disorder (BPD): clinical and biological correlates

Authors: I.N. Ferrier, D. Potkins, D. Eccleston

Recent evidence suggests that the outcome in BPD is less good than previously thought with approximately 30% of patients showing an inadequate response to lithium. There are a variety of poor outcomes in BPD, one of which is rapid or ultra rapid short cycling. We compared 15 patients with BPD with poor outcome with an age sex matched population of BPD who had a good response to lithium. The degree of family history, age of onset and length of illness was similar between the two groups. The poor outcome group exhibited a greater frequency of sub-cortical patchy white matter lesions (PWML) on MRI and increased frequency of temporal lobe slow wave activity on the EEG.

The more severe forms of rapid cycling disorder respond to anticonvulsants, and in the more severe cases, combinations of anticonvulsants and lithium are required. Seven rapid cycling BPD patients (mean age 50) were put on lamotrigine in addition to lithium and sodium valproate or carbamazepine. The patients had been ill for a mean duration of 15 years and all had shown failure to respond to lithium with a partial response to anticonvulsants. Three of the patients did well over 2-3 years when lamotrigine was added in a dose of 150 to 200 mg, with a marked reduction in the frequency and severity of episodes. Two patients showed no change over one year (maximum dose was 100 mg). Two patients exhibited a worsening at the onset of the study with increased severity of over-valued ideas. Medication was discontinued immediately. All of the patients who showed a good response had an abnormal EEG before the introduction of lamotrigine but no correlation between response to medication and PWMLs has been found.

LONGITUDINAL ASSESSMENT OF THYROID FUNCTION AND MOOD STABILITY IN MANIC DEPRESSIVE ILLNESS

Authors: M.A. Frye, K.D. Denicoff, A. Bryan, E. Smith-Jackson, S. Omar Ali, D. Luckenbaugh, G.S. Leverich, and R.M. Post

There is an emerging consensus that a decrease within the normal range of thyroid indices occurs in association with an acute response to a number of antidepressant, mood stabilizing, and cognitive therapy treatments (Whybrow 1981, Joffe 1994, 1996). Longer term studies of thyroid indices, however, have shown a number of predictors of relative mood instability, including low T3 levels associated with relapse in bipolar patients maintained on lithium (Hatterer 1988, Baumgartner 1995) and increased incidence of concurrent panic disorder and relative antidepressant inefficacy with subclinical hypothyroidism (Joffe 1992). This study was conducted to evaluate the incidence of de novo hypothyroidism (grades I & II) and to assess whether these thyroid changes were associated with clinical response or mood instability.

52 bipolar outpatients participated in a randomized double blind study comparing efficacy of lithium, carbamazepine, and lithium/carbamazepine combination for up to 3 years of prospective evaluation (Denicoff 1996). Patients on thyroid supplementation prior to prospective study were not included in this analysis.

The incidence of de novo hypothyroidism grade I or II was 40 % (12/30) during lithium monotherapy, 0% during carbamazepine monotherapy, and 17.6% (3/17) for the lithium/carbamazepine combination. This group was 67% (8/12) women, 33% (4/12) men, 67% (8/12) rapid cyclers, and 33% (4/12) nonrapid cyclers. The time course to develop an elevated TSH was 76.6 days +/- 49.7 into lithium monotherapy (mean dose 1247.1 mg +/- 261.8) and 119.3 days +/- 56.3 into the combination (mean dose lithium 1178.6 mg +/- 103.5 and carbamazepine 614.3 mg +/- 186.4). Pearson correlation of elevated TSH with LCM, Hamilton, Young, and Spielberger ratings were not significant. There was no significant difference between responders and nonresponders based on subclinical hypothyroid state (Fisher's exact test: one tail p=0.53 for monotherapy, p=0.36 for combination).

Further data will be presented evaluating free T4, T3, and TSH with mood at monthly intervals over the course of prospective treatment. Preliminary analysis reveals a protective effect by carbamazepine for lithium induced subclinical hypothyroidism.

Comorbidity of Axis I Bipolar Disorder and Axis II Personality Disorder:
Prevalence and Clinical Determinants

Authors: Elizabeth L. George, M.A. and David J. Miklowitz, Ph.D.

Many studies have examined the prevalence and predictive validity of Axis II disorders among unipolar depressed patients, but few have examined these issues among bipolar patients. The few studies that do exist suggest that Axis II pathology does indeed complicate the diagnosis of bipolar disorder, but is not prevalent as frequently as one might expect.

We examined the prevalence of Axis II disorder in 47 bipolar patients who had achieved remission of symptoms, using the Personality Disorder Examination (PDE). We present data on the prevalence of personality disorders in this sample, as well as mean PDE dimensional scores. Finally, we offer preliminary findings on the associations between Axis II comorbidity and the prospective one-year course of bipolar disorder in the domains of social-occupational functioning and medication compliance. Initial findings suggest that Axis II disorders can be rated reliably among bipolar patients in remission.

Is Bipolar Disorder Underdiagnosed?Are Antidepressants Overutilized?

Authors: S. Nassir Ghaemi, M.D., Gary S. Sachs, M.D., Alice M. Chiou, Ananda K. Pandurangi, M.D., and Frederick K. Goodwin, M.D.

Objectives: Clinical experience suggests that bipolar disorder (BP) may be underdiagnosed, mood stabilizers may be underutilized, and antidepressants may be overutilized in BP. This study examined these hypotheses.

Methods: In one year, all patients in an affective disorders unit of a university hospital with discharge diagnoses of BP (n=50) or schizoaffective disorder (n=5, all manic type) were diagnosed prospectively by a psychiatrist with expertise in affective disorders based on a semistructured clinical interview using DSM-IV criteria. Clinical Global Impression of Improvement (CGI-I) scores were assigned retrospectively, blind to admission or discharge diagnoses.

Results: 21/50 (42%) of bipolar patients carried other diagnoses in the community, mainly unipolar disorder (UP) (n=19/21; 90.5%). The semi-structured clinical interview identified 50% (25/50) of BP patients with either no previous diagnosis or other diagnoses. 21 patients not diagnosed as bipolar before admission remained undiagnosed longer (7.9+8.8 years after first professional contact) than 25 who were diagnosed (0.88+8.8 years). On admission, only 1/3 used mood stabilizers, yet 1/3 used antidepressants. All except 2 (3 .7 % ) were tapered off antidepressants, with overall improvement upon discharge. Valproate was more frequently used than lithium. Response to valproate alone was lower (40%) than response to valproate plus lithium or other adjuncts (antipsychotic or clonazepam, 67-70%). With aggressive anticonvulsant treatment, response rates were similar (50-57%) for pure mania, mixed episodes, and rapid-cycling episodes. On admission, 8/10 with acute depression used antidepressants, and only 3 received mood-stabilizing cotherapy. At discharge, only 2 remained on antidepressants, and all received mood stabilizers, with mild to moderate improvement in 7/10.

Conclusion: Bipolar disorder (BP) is underdiagnosed. Systematic application of DSM-IV criteria identified 50% as previously misdiagnosed or undiagnosed. Mood stabilizers were underutilized and antidepressants overutilized. Good treatment response was achieved acutely with aggressive use of anticonvulsants, especially with lithium or adjuncts.

RELAPSE AND QUALITY OF LIFE IN BIPOLAR DISORDER

Authors: Joseph F. Goldberg, M.D. and Martin Harrow, Ph.D.

Chronic affective illness has been shown to impair overall quality of life (QOL) even after acute symptoms remit. Impaired role performance, and the economic, occupational and social disability caused by depression have been well-described; however less is known about the impact of bipolar illness on these areas. We assessed relapse,outcome, and key areas of QOL in bipolar and unipolar-depressed patients, followed up over an 8-year period. Components of QOL were examined in relation to affective relapse and objective measures of functional outcome.

A sample of 206 RDC-bipolar I and nonpsychotic unipolar-depressed inpatients from the Chicago Follow-up Study were interviewed at index and again after 2, 4.5 and 8 years. Affective relapse, rehospitalization, and global outcome were rated using standardized instruments. These factors were then compared with a 5-point QOL index based on patients' satisfaction with work, social life, economic status, living circumstances, and self-perceived overall mental health.

Results indicated that during each follow-up period, 20-30% of bipolar patients were dissatisfied with at least one aspect of QOL. Functional outcome measures, including nonrehospitalization and the percentage of time spent in occupational roles, were correlated with greater work and social satisfaction among unipolar patients, but not bipolar patients (p<.05). Affective relapse was strongly linked to work dissatisfaction among bipolar patients (p<.05), but not unipolar patients. In addition, over one-quarter of all patients who had no affective syndrome in the year preceding each follow-up still viewed their overall mental health as impaired.

Among unipolar depressed patients, QOL appears less related to affective relapse than to objective signs of role functioning and adjustment. In contrast, among bipolar patients, poor QOL appears closely tied to recurrent manic or depressive episodes, but is less clearly linked to disrupted work and social performance. Diminished QOL is a persistent feature for one-quarter or more of affectively disordered patients, even in the absence of a recent affective relapse.

Suicidality Predicts Nonremission From Acute Bipolar Episodes

Authors: Joseph F. Goldberg, M.D., Jessica L. Garno, B.S.,
Andrew C. Leon, Ph.D., and James H. Kocsis, M.D.

Suicidality is common in bipolar illness, yet little is known about the relative suicide risk for subtypes of bipolar patients. The relationship between suicide attempts and overall course of illness in mania is also poorly understood. Previous studies have identified comorbid substance abuse, inadequate treatment, rapid cycling, and psychosis as being associated with an increased risk for suicidal behavior in bipolar disorder. This pilot study examined the frequency and lethality of pre-index suicide attempts among a large cohort of bipolar patients hospitalized between 1991-1995. Previous suicidality was examined in relation to 1) subdiagnoses of "mixed" or "pure" mania; 2) overall severity of illness, and 3) likelihood for acute remission during hospitalization.

Records were reviewed for 182 DSM-III-R Bipolar I inpatients. The number, method, and outcome of previous suicide attempts were rated. Patients were classified as having mixed-dysphoric or pure-manic episodes along standardized guidelines. Demographic data were obtained from clinical records, along with information regarding current suicidality, affective and psychotic symptoms, rapid cycling, medication use, previous substance abuse and weekly clinical improvement rated by Clinical Global Impressions scores. Factors hypothesized to predict nonremission from the index episode were analyzed by logistic regression.

Results indicated: 1) prior suicide attempts were more common in patients admitted for mixed mania (35%) than pure mania (9%) (p<.05); 2) both prior and current suicidality were strongly correlated with the number of depressive symptoms among mixed-state bipolar patients (r =.23, p <.01); 3) the likelihood of acute remission was reduced by 66% for every suicide attempt made prior to index (OR=0.44, 95% CI=0.23 to 0.84).

The findings suggest that mixed-state bipolar patients have a higher suicide risk than pure- manic patients. Past suicidality may be a marker for both future mixed mania and recurrent suicidality. Mixed-manic patients may represent a subpopulation at higher risk for demoralization and suicide, as a result of increased unpredictability and chaos of internal mood states.

Adjunctive Cognitive-Behavioral Group Treatment for Bipolar Disorder

Authors: Robert A. Gould, Ph.D., Dina Hirshfeld, Ph.D.,
Noreen A. Reilly-Harrington, M.S., and Gary Sachs, M.D.

Bipolar disorder is costly in terms of psychiatric costs, medication costs, rehabilitation services, lost wages and productivity, and, in some cases, loss of life. Recent studies indicate that even patients who have good acute responses to medication and adequate maintenance treatment have a five-year relapse rate of 73%. Even after symptoms have resolved, social and functional impairment continue. Clearly, interventions which reduce the frequency and severity of episodes while improving patients' overall functioning are needed. Although numerous reports suggest that cognitive-behavioral therapy (CBT) may represent a promising adjunctive treatment, to date no controlled clinical trial of group CBT for bipolar disorder has been published.

We are conducting an ongoing controlled clinical trial comparing a cognitive-behavioral group treatment (CBGT) to standard pharmacotherapy (SPT) alone in reducing the frequency and severity of episodes, and social and occupational dysfunction over 3-month treatment and follow-up periods. Participants were evaluated at baseline via a structured clinical interview and self-report measures and were assessed monthly thereafter by an independent clinical interviewer blind to treatment condition. The CBGT is a 12-week manualized group treatment which targets: 1) medication compliance, 2) recognition of acute symptoms and personal triggers for episodes of mania and depression, 3) development of individualized plans for coping with symptoms, 4) cognitive-behavioral therapy for depression, and 5) improving stress management, conflict resolution and family functioning.

We will present preliminary data on this effectiveness of CBGT relative to SPT for approximately 20 patients who have undergone these interventions in our clinic during the past 7 months. Results will be discussed in terms of augmenting pharmacotherapy and mitigating the impact of this devastating disorder.

Clinical Subtypes of Soft Bipolar Disorders in a French
Multicenter Study: EPIDEP

Authors: Hantouche EG, Fraud JP, Allilaire JF, Sechter D, Akiskal HS

This paper presents the preliminary results of a French multi-center study in progress on 600 out and inpatients with major depressive episodes (EPIDEP). The aim of EPIDEP is to show the feasibility of validating new clinical bipolar subtypes such the spectrum of Soft Bipolar Disorders: BP type II (major depressives associated with Hypomania, Cyclothymia or Hyperthymia).

Methodology: It involves 1) training French psychiatrists in 25 sites; 2) construction of a protocol based on criteria of DSM-IV and Akiskal, as well as instruments modified from the work of Angst (Hypomania Checklist), Ahrean-Carroll (manic scales), HAM-D28 + Rosenthal (Depression scale), semistructrured interview for evaluation of affective temperaments (Akiskal et al), family history and comorbidity (Winokur's user's friendly criteria); 3) prospective follow-up (3 months).

Results: Preliminary results are presented on 250 patients with MDE. The global rate of Soft Bipolarity (BP-II disorder) was about 40%. By comparison to unipolar depressives, BP-II was significantly different on the following parameters: higher frequency of suicidal thoughts and hypersomnia during the current depressive episode; younger age of onset of first depression; higher rate of recurrence; higher scores on Hypomania checklist and Cyclothymia questionnaire; higher switching rate under current treatment by 35-40% (vs 5% in UP group) which was more correlated to level of Cyclothymia than to Hypomania score.

Clinical Subtypes of Acute Mania in a French Multicenter Study: EPIMAN

Authors: Bourgeois ML , Hantouche EG, Azorin M, JP Fraud, Akiskal, HS

This paper presents the preliminary results of a French multi-center study in progress on 100 hospitalized manic (EPIMAN). The aim of EPIMAN is to show the feasibility of validating a new clinical form of mania, such "Dysphoric Mania".

Methodology: It involves 1) training French psychiatrists in 5 sites; 2) construction of a protocol based on criteria of DSM-IV, Akiskal, McElroy et al, Swann et al, as well as instruments like Beigel-Murphy and Ahrean-Carroll (manic scales), modified HAM-D13 / 17 (Depression scale), semi-structrured interview for evaluation of affective temperaments (Akiskal et al), family history and comorbidity (Winokur’s user's friendly criteria); 3) prospective follow-up during a period of 12 months.

Results: Preliminary results are presented on 77 hospitalized manic patients. The rate of "Dysphoric Mania" or MD (defined by the presence of 2 depressive symptoms for "Probable DM" and > 3 for "Definite DM") is 38% of hospitalized manic. DM didn't represent an extreme form of mania (lower score on Beigel-Murphy scale). Significant differences versus "Pure Mania" were obtained on female over-representation (83%); lower frequency of typical manic symptomatology (grandiosity, elation, hyperactivity); longer latency before correct diagnosis; and higher rate of mixed states in the first episodes (25% vs 2%). Finally, higher level of "Irritable Temperament" was observed in the "Probable DM".

Gender differences in bipolar disorder

Authors: Victoria Hendrick, M.D., Lori Altshuler, M.D., Michael Gitlin, M.D.

Psychiatric interviews were obtained on the first 88 consecutive patients presenting for treatment at UCLA's Affective Disorders Clinic. At the time of the interview, clinicians completed database forms that were subsequently used to establish a computer database on the clinic's mood disorder patients. The databases included questions on patients' Axis I and Axis II diagnoses, demographic variables, age of onset of Axis I disorders, number of previous mood episodes and hospitalizations, history of rapid mood cycling, history of substance abuse, medical histories, and family histories of psychiatric illness.

A gender difference that emerged from the database was the greater prevalence of bipolar II disorder in female compared to male patients. While 50% of the 64 bipolar I patients were female, the prevalence rose among the 24 bipolar II patients, of whom 67% were female.

We will also present data on gender differences in the percentage of mood episodes that were depressions vs manias, in the age of onset of bipolar I or bipolar II disorders, in total number of hospitalizations for mood episodes, in family histories of psychiatric illness, and in comorbidity with other Axis I disorders. Gender differences were also identified in patterns and types of medication use.

The Psychosocial Correlates of the Recurrence of Bipolar 1 Disorder from the National Comorbidity Survey

Author: Carolyn A. Holmes, Ph.D., R.N.

Bipolar 1 (BP1) disorder is a recurrent affective psychiatric disorder with a disruptive and debilitating course. Clinical evidence indicates that the frequency of episode recurrence is positively associated with a more severe course and progressive functional deterioration. This study examined the psychosocial risk factors associated with increased recurrence in persons meeting DSM-III-R criteria for a diagnosis of BP1 disorder, using data from the National Comorbidity Survey (NCS) with a representative population sample of noninstitutionalized Americans between the ages of 15 and 54.

A diathesis-stress model was tested to ascertain whether adverse experiences during childhood (e.g., parental psychopathology, relationship difficulties, and violence during childhood) create a vulnerability that predisposes the bipolar respondent to increased risk of recurrence, especially in the presence of certain adult characteristics and circumstances (e.g., comorbidity, traumatic events, chronic stressors, and personality characteristics). The main effects of each childhood and adult predictor and the moderating and mediating effects of adult experiences on childhood adversities to predict recurrence were measured.

Two childhood adversities showed robust direct effects on recurrence, parental psychopathology and childhood abuse. Significant adult risk factors were marital difficulties, interpersonal conflicts, and current traumatic events of an aggressive or financial nature. When all significant childhood and adult predictors were placed in one multivariate model, three factors remained strongly predictive of recurrence. Childhood abuse increased the risk of recurrence over five times, and parental psychopathology and ongoing marital difficulties each doubled the risk. Remarkably, when the effects of multiple risk factors were aggregated, a strong addictive effect was seen. With each additional risk factor the risk of recurrence more than doubled. None of the childhood effects were mediated through adult adversities, but traumatic events reduced the effect size of other adult predictors, suggesting the need to further examine the possibility of a triggering effect for traumatic events.

Clinical Effectiveness of lamoTrigine in Affective Dis0rders

Authors: Scott Hoopes, M.D. Chtd., Mark Snow, Ph.D.

Fourteen males and thirty-three females, 12 to 88 years of age (Bipolar I = 5, Bipolar II = 13, Bipolar N0S = 13, Cyclothymia = 2, Major Depressive Disorders = 10, Mood Disorder = 2, Conduct Disorder = 1) were treated with Lamotrigine. Approximately 83 percent had comorbid diagnoses including substance abuse, anxiety, attention-deficit, eating and cognitive disorders. Seven patients were diagnosed with personality disorders or personality disorder traits.

Lamotrigine was used for patients who could not tolerate other thymoleptics or with disinclination to submit to blood draws, risk of weight gain, or other potentially adverse effects of alternative medications. Twelve patients failed to complete an adequate therapeutic trial because of noncompliance, failure to follow-up, or a move from the area . Follow-up assessments were available for twenty-eight patients and pending for two others. Doses from 25 to 200 mg per day taken once-at-night or twice-a-day were used. Fifteen of twenty-eight were treated with less than 100 mg per day and three with 200 mg per day. Nine patients took lamotrigine only. Combinations of lamotrigine with other medications were generally well-tolerated. Two patients discontinued because of rashes, one because of headaches, and two because of agitation. Twenty patients showed significant improvement and eight did not. Five of the nonresponders had Major Depression with or without a psychosis. Five of eight patients with no comorbid diagnases improved. Patients often reported improvement as early as the first week with doses from 25 to 100 mg per day. Patients with Bipolar I responded as well as Bipolar II patients (four of four and six of eight respectively).

Lamotrigine used for bipolar disorders often produced antidepressant effects, showed rapid response at low doses, was well-tolerated, and generally combined well with other psychotropics. Starting at 25 mg and advancing slowly minimized side effects. Many responders were younger women with rapid cycling Bipolar II who improved within the first two weeks of treatment at doses from 25 to 50 mg per day.

Guanfacine and Juvenile Bipolar IllnesS

Authors: J.P. Horrigan, M.D. and L.J. Barnhill, M.D.

Guanfacine hydrochloride (Tenex) is an alpha-2 adrenergic agonist which has received recent attention in the field of child and adolescent psychiatry due to its apparent benefits in managing attention-deficit/ hyperactivity disorder (ADHD), tic disorders, and posttraumatic stress disorder. The initial reports noted minimal side effects. This poster details six cases of adverse responses to guanfacine, drawn from an initial clinic sample of 95 outpatient boys and girls aged 8 to 15 years who were seen in a university-based developmental neuropsychiatric clinic. In each case, the patient met formal DSM-IV criteria for ADHD while four out of six also met criteria for Tourette's Disorder. Within 72 hours of initiation of guanfacine therapy, drastic changes in mood and behavior occurred in each of these individuals, culminating in states that resembled hypomania and mania, including elevated mood, poor sleep hygiene, and hypersexuality. The dose of guanfacine ranged from l to 2 mg/day. Later investigation revealed that all of the youngsters had clinical and/or familial risk factors for bipolar disorder. The authors speculate about the possible mechanisms behind these side effects, and suggest that bipolar disorder may be a relative contraindication to guanfacine therapy.

Suicide and Other Causes of Death in Patients with Bipolar and Unipolar Illness

Author: Eyd Hansen Hr yer

Several studies have found an increase in mortality in patients with bipolar and unipolar illness, compared with the general population. The single most important cause of this is suicide. Furthermore, the studies have revealed an increase in mortality from "natural" causes, especially due to cardiovascular diseases. In studies comparing the mortality of unipolar and bipolar patients, a tendency towards an increase in mortality due to "natural" causes, has been found in patients with bipolar illness. When suicide rates in bipolar and unipolar illness have been compared, the results have been contradictory. Generally, these studies have limited statistical power, especially when dividing groups according to age, gender, duration of illness, and bipolar/unipolar illness. In this project we study the mortality in patients with bipolar/unipolar illness in a large population-based sample of first-admitted patients compared with the general population in relation to age, gender, duration of illness, cause of death, and the time elapsed between discharge from a psychiatric hospital and time of death.

Methods: This project includes a total nationwide sample of approximately 20,000 patients with bipolar and unipolar manic-depressive illness, followed up to 25 years. Mortality is compared with the general population using the person-years method, calculating the SMR. Bipolar and unipolar patients are compared using survival analysis. The project is based on data from a record linkage between two nationwide Danish registers, available in the Danish Database for Psychiatric Epidemiological Research.

Results: Are not yet available. Data is currently being analyzed.

A Standard Education Programme for Patients at Lithium Maintenance Treatment*

Authors: E.A.M. Knoppert-van der Klein, C.A.L. Hoogduin, A.S. van Peski-Oosterbaan, P. K` lling, and J.R. Beck-Lie A Fat.

Research Questions: Do knowledge and attitudes about lithium treatment improve by an education programme (a videotape and a written hand-out) and does compliance improve?

Patients and Methods: Forty-six patients on lithium in remission were at random divided into two groups; five times (every six weeks) the Lithium Knowledge Test (LKT) and the Lithium Attitudes Questionnarie (LAQ) (both translated in Dutch), a list of side-effects and lithium levels were assessed: Group 1 received the programme at second visit, group 2, six weeks later.

Results and Conclusions: The educational programme showed a positive effect on patient knowledge and attitudes about the lithium treatment. Both effects decline slowly in time. No increase in reported side-effect was found.

*This study is a replication with permission of the study of Peet & Harvey, British Journal of Psychiatry 1991, 158, 197-204

A Case of Ultra-rapid Cycling Bipolar Disorder With Frontal Epilepsy in a 13 Year Old Boy

Authors: Kochman, F, Ducrocq, F, Parquet, PJ

We report the case of a 13 year old adolescent, hospitalized in our department (Child and Adolescent Psychiatry - Professor PARQUET) because of a major depressive disorder with suicidal ideas. After three days, his behaviour changed, with disappearance of depressive mood and appearance of pleasurable activities with a high potential for painful consequences (combined with irritability and aggressiveness), sexual indiscretions, grandiosity, decreased need for sleep, flight of ideas, marked impairment in social functioning. Using Kiddie-SADS-R (according to DSM-IV classification), the young patient fulfilled either criteria for Major Depressive Disorder or for Hypomanic Episodes (twice for Manic Episode), with a period of approximately one week for each episode. Plus, the adolescent had brief and sudden crisis characterized by aggressiveness, sexual concerns and sexual acts mimics, atypical upper limbs movements, and postcritical confusion.

Electroencephalography revealed frontal seizures. This patient has been dramatically improved after a Valproate treatment.

We hypothesize that bipolar disorder in children and adolescents is not a rare disease but is just massively underdiagnosed. Besides, this disease should often occur at this age by ultra-rapid cycling. This fact could also explain its misdiagnosis.

Anyway, what is the relation between this bipolar disorder and frontal seizures ?

We propose different hypotheses according to the literature.

Efficacy of Valproate/Valpromide in Ultra-rapid Cycling Bipolar Disorders n Children and Adolescents

Authors: Kochman, F, Ducrocq, F, Parquet, PJ

Bipolar disorder is a diagnosis rarely given in childhood and adolescence. A regularly biphasic disorder is described in 4 children and adolescents (9, 11, 12 and 13 years old). It was characterized by several hours to several days (period ranging from 4 hours to 10 days) of manic, hypomanic episodes, or major depressive disorders. According to Kiddie-SADS-R semi-structured interview (DSM-IV), these young patients presented mixed episodes included in a bipolar disorder.

Anyway, we think that these young patients present either (hypo)manic episodes or major depressive disorders during a very short period of time (sometimes lasting a few hours only).

Nevertheless, should we consider these children and adolescents as patients suffering from ultra-rapid cycling bipolar disorders, which could be considered as a new entity, or a form of early-onset bipolar disorder ?

We have been prescribing Valpromide for 3 patients, Valproate for one. We observed a dramatic improvement within 2 weeks (CGI score at 5.75 before, and at 1.25 after treatment). According to a second Kiddie-SADS-R assessment, none of them still met criteria for a Mood Disorder after treatment. They are still asymptomatic after at least 4 months of treatment.

This case report should be followed by further studies to validate this new clinical entity, and by a double-blind placebo-controlled trial.

Total Sleep Deprivation and Consecutive Sleep-phase-advance in bipolar Versus Unipolar Depression: Effects on Psychopathology

Authors: K` nig, A. Riemann D., Hohagen F., Kiemen A., Hornyak M.,
Steffes P., Voderholzer U., Berger, M.

Total sleep deprivation (TSD) has an immediate antidepressive effect in 60 % of depressed patients. In a pilot study we were able to show that the usual relapse into depression after successful TSD could be prevented in approximately 60 % of depressed patients by a succeeding sleep phase advance therapy (SPA). This strategy was based on studies which showed that a phase advance of the sleep period alone acts antidepressive, that naps in the morning after successful TSD have stronger depressiogenic effect than in the afternoon and that sleep deprivation in the second half of the night improves mood, but not sleep deprivation in the first half of the night. Additionally, we tested the effect of phase advance in bipolar depressed patients.

Methods: 33 inpatients (45,2? 13,4 yrs) with MDD (subtype melancholia; DSM-III-R), all responders to TSD, have participated in the study. 22 of the patients suffered from uni-, 11 from bipolar MDD. Both groups did not differ concerning age, severity of depression, number of episodes and duration of current episode. Depressed mood was measured by the 21- and 6-HAMD. Patients were considered as responders to TSD, SPA or SPD, if their 6-HAMD showed an at least 30 % reduction compared to the baseline value. Phase advance: bedtime after TSD from 5.00 p.m. until midnight and then a daily one hour delay of the sleep phase finished with the conventional bedtime from 11.00 p.m. till 6.00 a.m.

Results: 18 out of 22 patients with unipolar depression and 8 out of 11 patients with bipolar depression finished the SPA. The 6-HAMD of the unipolar MDD group improved by 53,1 ? 39,1 %, the improvement in the bipolar MDD group was 62,7 ? 34,5 %. Both groups did not show any significant difference by t-test in the response to neither TSD (p = 0,909) nor to SPA (t-test: p = 0,422).

Conclusion: TSD followed by SPA seems to present an effective method for mood stabilization. Results show similar improvement of unipolar versus bipolar MDD. This useful strategy for the treatment of depression should be used as often in bipolar as in unipolar depressed patients.

Prodromes, coping strategies, insight and social functioning in bipolar affective disorders

Author: Dominic Lam, Ph.D.

Forty patients suffering from bipolar affective disorder were interviewed for their prodromes of depression and mania, their coping strategies for these prodromes, their levels of insight and their levels of social functioning. A quarter of subjects reported that they could not detect any early warnings of depression compared with only 7.5% of subjects who reported that they could not detect prodromes of mania. There were significantly more high functioning subjects in the good coping group for prodromes of mania. More high functioning subjects were also present in the good coping group for prodromes of depression but the difference just failed to reach statistical significance. More subjects in the good coping group for prodromes of mania reported the spontaneous use of behavioural techniques, for example restraining themselves from excessive behaviour, engaging in calming activities and taking extra time to rest or sleep when they detected prodromes of mania. Similarly, subjects in the good coping group for prodromes of depression used behavioural techniques such as keeping busy. However some subjects also reported cognitive techniques of distraction from negative thoughts and recognising unrealistic thoughts and evaluating if these thoughts were worth worrying about. Subjects' current levels of depression, coping with prodromes of mania, insight and ability to recognise early warnings for depression contributed significantly to their levels of social functioning.

Cognitive Therapy for Manic Depression: a pilot study - end of therapy outcome

Author: Dominic Lam, Ph.D.

Aim of the Project: The pilot study aims at using cognitive behavioural strategies for the treatment of manic depressive illness in conjunction with pharmacological approach. We have recruited bipolar patients who are on prophylactic medication and yet are still at risk of relapsing.

Design: The pilot study is a randomised controlled design. Twenty four subjects suffering from manic depression were randomly allocated to an experimental and a control group. Subjects in the control group have usual outpatient treatment. Twelve to twenty sessions of cognitive therapy with be given to subjects in the experimental group. Therapy is based on Beck et al.'s (1979) cognitive model for affective disorder as well as specific techniques developed for bipolar patients. Specifically subjects are taught cognitive behavioural skills to: 1. deal with mood swings, 2. monitor early warning signs and manage them,

3. promote insight and compliance of medication, 4. promote a routine and good self management: 5. promote social functioning, 6. increase sense of control, 7. tackle any sense of stigma.

Inclusion criteria: 1. DSM-IV Bipolar I or Bipolar II Disorder, 2. at least two episodes of mania/hypomania in the last two years or three or more past episodes, 3. no history of non-affective non-organic psychosis or schizo-affective disorder, 4. no periods of alcoholism or drug abuse in the past year, 5. on regular medication, 6.age 18 to 65,

7. currently not in a manic episode or deeply depressed (BDI<29; MAS <9),

8. currently not in any psychological therapy, 9. no previous CBT experience.

Methods: Instruments: 1. S.A.D.S. covering the period of interest, BDI, MAS, Internal State Scale; 2. MRC Social Performance Schedule for interviewer rating (Harry et al. 1983); 3. Social adjustment scale (Cooper et al. 1982); 4. Insight interview adapted from David et al. (1992); 5. Views of Manic Depression (Hayward et al. unpublished); 6. Self-control behavior schedule (Rosenbaum, 1980); 7. Early warning and coping interview (Lam and Wong, unpublished); 8. Coping questionnaire (Wong and Lam, unpublished); Self-concept questionnaire (Robson et al., 1989); 10. Significant Others Scale (Power et al. 1988); 11. Mill Hill Verbal Scale (only on recruitment). Subjects were assessed at recruitment, and at six and twelve months. The above measures at recruitment are repeated with the S.A.D.S. covering the relevant period.

Outcome: Monthly mood rating as well as numbers of bipolar episodes during the first six months produced promising results.

Factors Associated with Poor Psychosocial Functioning among Children of Parents with Bipolar Disorder

Authors: Lapalme, M. and Hodgins, S.

The present study was designed to evaluate the psychosocial functioning of children of parents with bipolar disorder as compared to that of children of parents with no mental disorders and to identify associated factors. Forty-nine parents with a confirmed diagnosis of bipolar disorder, their spouses (biological co-parents of the child), and their 67 children were compared to 40 couples with no mental disorders and their 57 children. The children, aged five to twelve years old, were rated independently by both parents on the Child Behavior Checklist. Comorbid disorders in the bipolar parent (SCID I and II), mental disorders in the other biological parents (SCID I and II), marital adjustment (Dyadic Adjustment Scale), and parenting practices (Parenting Dimensions Inventory) were assessed. Proportionately, more of the children of parents with bipolar disorder than the children of parents with no mental disorders were rated as having psychosocial problems within the clinical range. Among both groups of families, parenting practices, but not marital adjustment, were found to be related to the presence of difficulties among the children. Within bipolar families, the presence of a personality disorder in the bipolar parents was associated with impaired psychosocial functioning in the children, particularly when the bipolar parent is the mother. Neither the presence of a comorbid axis I disorder in the bipolar parent, nor the presence of a disorder (Axis I or II) in the co-parent were found to be related to problems among the children. Results suggest that in childhood, impaired psychosocial functioning among children of parents with bipolar disorder is more strongly related to comorbid personality disorder and to poor parenting practices than to parental bipolar disorder per se.

Comparison of Functioning in Children of Parents with Bipolar Disorder (BPD) and Parents with no Serious Mental Disorder (NMD)

Authors: LaRoche, Catherine; Hodgins, Sheilagh; Marrache, Myriam
and Lapalme, Micheline

In addition to recent advances in the etiology and treatment of adult BPD, there remains a need to increase understanding and develop preventive interventions for the children of these adults. Many of these children develop different forms of psychopathology at younger and younger ages. These childhood disorders may represent antecedents of adulthood disorders and/or reactions to living with a mentally disordered parent.

This poster presents initial findings from a prospective longitudinal study comparing the development of children of parents with BPD and children of parents with no mental disorder (NMD). The experimental group includes 57 adults with a confirmed diagnosis of BPD, their spouses and 79 children. The comparison group includes 51 couples with NMD and their 77 children.

Measures for parents include a Diagnostic Interview (SCID), a Parental History of Mental Disorder, and measures of parental personality traits, parenting, social support, coping skills and a measure for family violence (CTS). Childrens' Diagnostic and Functional Measures include the Dominic (Valla et al.), Child and Parent versions of the Child Assessment Scale (CAS) (Hodges) and the Child and Adolescent Functional Assessment Scale (CAFAS) (Hodges). These interview protocols were chosen because of their appropriate fit with the developmental limitations of young children (ages 5 to 12) and for their good psychometric properties.

Initial findings show a consistent pattern of parental impairment rates ranging from the highest rate among parents with BPD, medium rates in their spouses, and the lowest rates among parents with NMD. Few diagnoses were assigned to the total child sample. However, children of BPD parents showed more impairment in both symptomatic and functional areas than children with NMD. Considerable disagreement existed between informants regarding childrens' functioning.

Genetic evidence for a bipolar disorder subtype

Authors: MacKinnon DF, Xu J, McMahon FJ, Simpson SG, Stine OC,
McInnis MG, DePaulo JR

Panic disorder frequently cosegregates with bipolar disorder in some families. In these families, we have proposed that a high risk for panic disorder may be a marker for a genetically distinct subtype of bipolar disorder. We now test this hypothesis on a sample of 28 families ascertained, psychiatrically interviewed, genetically analyzed, and reported on elsewhere as part of a genome-wide screen for loci linked with bipolar disorder. In our initial study, we found evidence of linkage using 31 markers along chromosome 18. Here, we have reevaluated these linkage results using multipoint lod score and nonparametric linkage (NPL) analyses, stratifying the sample into three groups: 1) five families in which the bipolar proband ofthe family was diagnosed as having panic disorder (RDC inclusion criteria); 2) six families in which the proband had panic attacks, but did not meet criteria for panic disorder; 3) families in which probands did not have panic disorder or any history of panic attacks. Only family members with BPI or BPII were considered affected, and included in the analysis. Multipoint NPL Z-scores were in the range of 4.0-4.8 (p=0.003-0.001) for region from D18S42 to D18S61 on 18q, only for the group of families in which the proband had panic disorder. Scores for the second group were intermediate, ranging from 1.5-2.0 (p=0.3-0.04) in the same region, while scores for the third group in this region were all negative (-0.5 to -2.0). The maximum multipoint lod score for group 1 was 2.93, at Dl8S61. This study provides evidence for genetically distinct subtypes of bipolar disorder, distinguished clinically by a difference in the risk of comorbid panic disorder in probands and affected family members. Under-standing this finding in light of other phenotypic divisions (e.g., paternal/maternal pedigrees) requires further study.

Initial Definitive Treatment of Mania in a Psychiatric Emergency Service

Authors: Lucian Manu, M. D. and Michael H. Allen, M. D.

The previously available antimanic agents, lithium and the neuroleptics, have significant limitations in the Psychiatric Emergency Service. More recently, divalproex sodium (DVX) has been shown to be superior to placebo and comparable to lithium. DVX is better tolerated and appears to have a rapid onset after achieving therapeutic blood levels. A loading strategy has been developed, which seeks to take advantage of the favorable side effect profile and rapid onset. Three published reports suggest the effectiveness of this approach.

The Bellevue Comprehensive Psychiatric Emergency Program (CPEP) treats patients intensively for up to three days in an Extended Observation Unit (EOU). Patients are evaluated psychiatrically and medically on EOU Day 1. If no contraindications emerge, consenting patients receive DVX 20 mg/kg in divided doses by the end of Day 1. Patients are under close medical and nursing supervision at all times. Patients are reevaluated on Day 2, a 12 hour VPA trough level is obtained, and the dosage sometimes adjusted for Days 2 and following. Lorazopam 2 mg is available on an as needed basis. Patients are again evaluated on Day 3 for final disposition. If sufficiently improved, patients are discharged to the community with aftercare, provided by CPEP, if necessary. If insufficiently improved, patients are admitted to a Bellevue inpatient unit.

Using routine hospital data sources, all patients who received DVX in the EOU from September 1995 to July 1996 were identified. In order to measure response to this strategy, changes in mental status, day, time, dosage, and route of adjunctive medications and inpatient admission rate were assessed. For patients discharged from CPEP to the community, recidivism at 30 days and community survival analysis are also reported. Urine toxicology, other significant laboratory results, and side effects are described.

Bipolar Disorder in the Latter Half of Life: Symptom Presentation, Global functioning, stability, and age of Onset in a Community Sample

Author: Suzanne Meeks, Ph.D.

Relatively little is known about the manifestations of bipolar disorder in late life. Many of the reports on late-life bipolar disorder are clinical case reports, usually drawn from inpatient expenences. The majority of empirical studies also have focused on a hospital-based population, and many have focused on late-onset mania or bipolar disorder. By contrast, the present paper focuses on a community-based sample of middle-aged and older adults diagnosed with bipolar disorder according to Research Diagnostic Criteria (RDC). Participants were 87 individuals who took part in a larger eight-month prospective study of severe mental illness in later life. All participants had received services in the five previous years from one of two community mental health centers in a 12 county region surrounding the Louisville, KY metropolitan area: they were selected randomly from current and former client lists of the centers. All were over the age of 40, were designated as having a severe mental illness by state criteria, and did not have primary diagnoses of substance abuse disorders. Psychiatric history and current functioning were assessed using the SADS lifetime version.

The average age of the sample was 53.60 (SD=9.78), with a range from 40 to 78. They were 73.6% women and 26.4% men, had a mean education of 12.41 years (SD=3.56) and median income under $10,000 per year. Participants were interviewed three times at four-month intervals. At the time of the first interview, 1/4 were not in an episode of illness; 1/4 were in a new episode of illness of less than five years duration, and the remainder were divided among more chronic or cycling conditions. Approximately one-third remained stable across study intervals in either an illness-free state or in a residual state with minimal symptoms; 35% were stably ill, and about a third were unstable across the eight months of the study. In spite of the fact that the majority were in acute or chronic episodes, only 34.5% were receiving lithium and 37.9% were receiving antidepressants. A similar number (36.8%) were receiving neuroleptic medications. One-third were receiving some form of psychotherapy. Normative presentation was more depressive than manic symptoms, with few psychotic symptoms. Age was unrelated to symptom presentation. Treatments received were unrelated to short-term stability.

Participants had experienced on average 14.49 prior manic episodes, and 17.77 depressive episodes. The mean duration of their longest manic episode was 120 weeks and for depressive episodes 141 weeks, although there was tremendous variation and the modal episode length was much shorter. The average age of onset of any symptoms was 24.84; the mean age at the first identifiable manic episode was 34.95, with the first depressive episode occurring approximately 5 years earlier. Age of onset was unrelated to short-term stability, which was related only to the prior number of depressive episodes experienced. However, age of onset was strongly related to global functioning, accounting for 13% of the variance in GAS scores with age controlled. No other aspects chronicity of illness were related to GAS once age of onset was entered.

This study of community-dwelling middle-aged and older adults with bipolar disorder is unique in its focus on community-dwelling individuals with early-onset bipolar disorder and long-term episodic or chronic illness. In this group of individuals, there is great variability in functioning and symptom presentation, although the norm is significant impairment and limited socioeconomic resources. Consistent with previous literature, age of onset appears to be an important predictor of global functioning. Despite the fact that the sample was identified through mental health agencies, up to two-thirds may not have been benefitting from appropriate pharmacological or psychosocial treatments.

*This study was supported by grant #R29 MH44787 from the National Inst. of Mental Health

The psychoeducational approach on the treatment of bipolar patients

Authors: Ricardo Alberto Moreno & Ana Claudia Fontes de Andrade

One of the main difficulties faced on the clinical practice with bipolar patients is the high rate of noncompliance. Recent studies have shown the efficacy of the psychoeducational approach, in addition to other treatment modalities on the improvement of compliance levels and quality of life of these patients and their families. This research aims to study the effect of this approach combined with lithium therapy in a 10 weekly session group of outpatient clients in a university hospital in Brasil. There is a control group, assisted individually in medical appointments for outcome comparison purposes. Outcome will be presented through the assessment of symptomatology, social adjustment, compliance and level of information. The assessment is being done eight weeks prior to the group sessions, at the end of the intervention and at the third, sixth and twelfth month after the end of the program for maintenance checking purposes.

Gender Differences Among Late-Onset vs. Early-Onset Bipolars

Authors: Myers, Diane S., Stabb, Sally D., & Rubin, Linda

The present study was undertaken to evaluate gender differences in symptom presentation, family history, and role of stressful life events in late-onset vs. early-onset bipolar disorder. Post (1992) has identified 10 characteristics of affective illness that may parallel the longitudinal course of bipolar disorder. Using a qualitative patternmatching design, medical records of late-onset bipolar patients were compared to early-onset bipolars using Post's 10 characteristics. Archival medical records were evaluated on 156 bipolar I patients at a large northeastern psychiatric hospital. Retrospective review of medical records on a subgroup of 35 late-onset bipolar I patients, including 15 late-onset females and 20 late-onset males, were compared with records of 121 earlyonset bipolar I patients, including 84 early-onset females and 37 early-onset males. Results partially supported the theoretical hypotheses of the study. Late-onset bipolar patients were found to have a less frequent family history of affective illness, and male late-onset bipolar patients had a more frequent association of stressful life events with all bipolar episodes than early-onset or female late-onset bipolars. In addition, late-onset bipolar patients reported a higher percentage of paranoid delusions, irritability/anger, and mixed mania symptoms as part of the symptom picture than early-onset bipolars. Using Post's 10 characteristics, results revealed that females more frequently had early experiences that may have predisposed them to later bipolar episodes than males. Late-onset females demonstrated other differences in bipolar history, including heightened vulnerability to recurrent episodes, more conditioned compensatory reactions, and more positive response to carbamazepine administration than late-onset males or early-onset bipolars. Thus, findings suggest that there are gender differences among late-onset vs. early-onset bipolar patients. More complete findings of the study and implications for further research will be presented.

Coping Resources and Life Functioning of Hospitalized People with Bipolar Disorder

Authors: L.E. Pollack, PhD, K. Kouzekanani, PhD, S. Harvin, MSN,
and R. V. Varner,MD

Purpose: The coping resources and life functioning of hospitalized people with bipolar disorder were studied with respect to race, chronicity, and gender.

Methods: Seventy-one subjects (65% female, 35% male; 63% Euro-American, 37% African-American; average age 35.6 years), diagnosed using DSM-IV criteria, were recruited from a state- and county-funded acute psychiatric hospital. The Coping Resources Inventory (CRI) was used to measure subjects' self-reported coping resources in five domains (cognitive, social, emotional, spiritual/philosophical, physical), and a total score. The Behavior and Symptom Identification Scale (BASIS-32) was used to measure subjects' self-reported difficulty in symptoms (depression/anxiety, impulsive/addictive behavior, psychosis) and functioning (relation to self/others, daily living/role functioning), and a grand mean score. Data were collected during the most recent episode, and prior to beginning an inpatient bipolar group therapy program.

Results and Conclusions: The Euro-American group reported significantly higher degrees of difficulty in the major areas of relation to self/others, depression/anxiety, daily living/role functioning, impulsive behavior, and on the BASIS-32 grand mean score, than did the AfricanAmerican group. Euro-Americans also reported significantly lower levels of coping resources in the cognitive and emotional domains, and on the CRI total score, when compared to the African-American group. These findings indicate that the Euro-American group perceived greater impairment and fewer coping resources than the African-American group. Subjects who had three or fewer hospitalizations (acute group) had significantly: (a) higher scores on their perceived degree of difficulty in relation to self/others, daily living, and on the grand mean BASIS-32 score; and (b) lower scores on the cognitive and emotional coping resource dimensions, and on the total CRI score, than did those subjects who had more than three hospitalizations (chronic group). In comparison with the chronic group, the acute group may have been more aware of, or affected by, their deterioration in life functioning, and/or were not hospitalized until deterioration was more pronounced. Significant gender differences on coping resources, as well as behavior and symptom identification, were not evident, indicating one area of homogeneity in this sample.

Neuropsychological Profiles in Bipolar Affective Disorder

Authors: Kristin B. Powell, David J. Miklowitz, & Jeffrey A. Richards

Brain dysfunction and cognitive deficits have long been correlated with clinical outcome and course of illness in schizophrenia. However, research in the neuropsychology of bipolar disorder has been limited. The present study will present neuropsychological data on a group of bipolar patients varying in severity of illness and subtype. Based upon a previous review of the literature (Powell & Miklowitz, 1994), these assessments focus on tasks related to frontal lobe functioning, and take place during periods of relative symptom remission, so as to minimize purely state-related cognitive deficits.

Subjects received an assessment battery consisting of the following measures: the Wisconsin Card Sorting Test; the California Verbal Learning Test; the Behavioral Dyscontrol Scale; Halstead-Reitan Finger Tapping; Digit Span, Vocabulary and Block Design (WAIS-R). Subjects have also been administered two experimental measures proposed to tap into more subtle aspects of prefrontal functioning: the Delayed Alternation Response Task, a measure of spatial working memory (Gold, Berman, Randolph, Goldberg, & Weinberger, 1996), and the "gambling task" developed by Antonio Damasio's research team (Bechara, Damasio, Damasio, & Anderson, 1996) as a potential indicator of orbitofrontal functioning.

Within-group analyses (N = 20) will examine the relationships between subjects' performance on neuropsychological measures and patient characteristics including diagnostic subtype (Bipolar I versus II), presence or absence of psychotic states during episodes, medication regimen (treated versus not treated with anticonvulsants) and neuropsychological risk factors (severe substance abuse, head injury, and other neurological insults). Results will be discussed in terms of their implications for diagnostic subtyping, prognosis and treatment of the "cognitive" correlates of bipolar disorder.

Cognitive Diathesis-Stress Interactions as Predictors of Bipolar and Unipolar Symptomatology

Authors: Noreen A. Reilly-Harrington, Lauren B. Alloy, and David M. Fresco

While extensive research has investigated the role of cognitive processes and life events in unipolar depression, comparatively little is known about the role of such psychosocial factors in the course of bipolar mood disorders. However, the logic of cognitive diathesis-stress models and previous preliminary research suggest that cognitive vulnerability factors may predispose individuals with bipolar mood disorders to manic/hypomanic and depressive episodes when confronted with life events. The current study examined the interaction of cognitive style and life events in predicting the depressive and manic/hypomanic mood swings of undergraduates meeting RDC lifetime diagnoses (based on a modified-SADS-Lifetime interview) for Bipolar I (n=18), Bipolar II or Cyclothymia (n=43), Unipolar Depression (n=102), or no lifetime diagnosis (n=43). At two time points (averaging 1 month apart), subjects completed measures of depressive and manic symptoms (Beck Depression Inventory, MMPI Mania Scale), cognitive styles (Attributional Style Questionnaire, Dysfunctional Attitudes Scale), and major and minor, positive and negative life events (Life Experiences Survey, Hassles and Uplifts Scale). Hierarchical regression analyses indicated that subjects' attributional styles (as assessed at Time 1) interacted significantly with life events occurring between Times 1 and 2 to predict depressive symptoms at Time 2. Furthermore, support was found for the specific vulnerability hypotheses of Beck's Theory and Hopelessness Theory in which individuals are thought to be most susceptible to depression when experiencing events to which they attach strong personal meaning or significance. Hierarchical regression analyses indicated a significant interaction between autonomous cognitive style (as assessed at Time 1) and the occurrence of achievement-related stressors between Times 1 and 2 in predicting depressive symptoms at Time 2. While the interaction between cognitive style and life events in predicting manic/hypomanic symptomatology was nonsignificant in the current study, a significant main effect was found for cognitive style.

Comparisons of Cognitive Styles Across the Bipolar and Unipolar Spectrums

Authors: Noreen A. Reilly-Harrington, Lauren B. Alloy, and David M. Fresco

The role of cognitive processes in the phenomenology, cause, course, and treatment of unipolar depression has been the subject of scientifically fruitful investigation over the past two decades. However, relatively little is known about the characteristic cognitive patterns found in individuals with bipolar mood disorders. Previous research by the above investigators found that cyclothymics exhibit cognitive styles that are stable across depressive and hypomanic episodes and are as negative as those of dysthymic subjects. The present study sought to further explore cognitive patterns across the bipolar and unipolar spectrums. Subjects in the study were undergraduates who met RDC criteria for lifetime diagnoses (based on a modified-SADS-Lifetime interview) of Bipolar I (n=18), Bipolar II or cyclothymia (n=43), Unipolar Depression (n=102), or no lifetime diagnosis (n=43). Subjects were administered several measures of cognitive style including the Attributional Style Questionnaire (ASQ), Dysfunctional Attitudes Scale (DAS), Self-Consciousness Scale (SCS) and Cognitive Bias Questionnaire (CBQ). The general pattern of findings indicated that the Bipolar II and Cyclothymic group scored significantly differently and generally in a more dysfunctional direction than the other groups on measures of cognitive style. Specifically, the Bipolar II and Cyclothymic group scored significantly higher than unipolar or control subjects on the SCS private self-consciousness scale and significantly higher than control subjects on the SCS social anxiety scale. The Bipolar II and Cyclothymic group also scored significantly higher than controls on the CBQ depression scale. Interestingly, the Bipolar II and Cyclothymic group scored higher than the Bipolar I, Unipolar, or Control groups on the dimension of Autonomy as derived from the DAS. The implications of these findings will be discussed with respect to the role that cognitive factors may play in the bipolar mood disorders.

Gender Differences in Bipolar Disorder

Authors: Janine C. Robb, L. Trevor Young, Robert G. Cooke, Russell T. Joffe

The importance of gender in Bipolar Disorder (BD) has been widely acknowledged. The limited data available suggests that men and women do not differ in prevalency rates but may differ in the course of the illness. This study investigated gender differences in a large sample of patients with BD including measures on well-being and functioning.

Euthymic outpatients were systematically assessed. Measurements obtained included: SADA-LV, Hamilton Depression Rating scores, Young Mania Rating scores and Medical Outcome Survey SF-20. Across group differences were analyzed for statistical differences using Student t-tests, chi square and ANOVA.

Women and men with BD differ in both the clinical picture and reported well-being. Women have a later age of onset, are more likely to have a rapid cycling course, panic attacks and more severe suicidal behaviour. Women also report greater overall impairment in all MOS subscales and are significantly more impaired than men in the subscales of pain and physical health.

Evidence is presented to suggest specific clinical differences between men and women with BD. Further investigation and replication of these differences need to be addressed. As well, thought should be given to the gender sensitivity and bias of the current instruments available.

Suicide Among Psychiatric Hospital Inpatients: Evidence of Mood Cycling and Mixed States

Authors: Verinder Sharma, Emmanuel Persad, Karen Kueneman

This study examined the risk factors for suicide among inpatients in an Ontario provincial psychiatric hospital. Forty-four inpatients who had committed suicide during their hospital stay from 1969 to 1995 were compared with a sex, age and date of admission matched group of inpatient controls. The diagnosis for each patient was reviewed by the authors. The progress notes, including the nursing notes, of the index hospitalization were examined and information on the clinical course was extracted. DSM-IV criteria was applied to ascertain the presence of episodes of depression, mania, hypomania and mixed states. Suicide victims were more likely to have had a mood disorder, family history of psychiatric problems, mention of suicide risk in chart notes and a previous suicide attempt. The most common diagnosis among inpatients who committed suicide in this study was a mood disorder. Cycling between depression and hypomania was present in 13.6 percent of the suicide group and an additional 36.4 percent of the patients met the criteria for bipolar mixed states at the time of suicide. This was particularly the case among patients with delusional depression (n=8), 5 (62.5 %) of whom exhibited evidence of bipolarity. None of the patients in the control group experienced cycling or mixed states at any time during the hospitalization. The implications of these findings, including the possible role of antidepressants in the induction of cycling prior to suicide, will be discussed.

Gabapentin for Mood Instability Associated with Migrane

Authors: M. Shetty, D. Lynn, T. Kumaran

Introduction: Two anti-convulsants, valproate and carbamazepine, have been used successfully for mood stabilization. We looked at the effects of a new anti-convulsant, gabapentin, on mood stabilization on a specific group of patients.

Method: Our study included three patients who were chosen based on the following criteria: 1) Females aged 29 to 40, 2) Symptoms of irritability and dysphoria with premenstrual exacerbation, 3) Incomplete response to antidepressants, 4) Not meeting DSM-IV criteria for a diagnosis of Bipolar Disorder, 5) Migraine headaches with poor compliance to sumatriptan. Excluded from the study were patients: 1) Presently taking approved mood stabilizers, 2) Presently showing mania, hypomania or suicidal ideation, 3) Presently on gabapentin. These patients were on antidepressants and sumatriptan at the time of the study. Patients were given the Rickels and Howard Physician Questionnaire, which also included assessment of degree of psychopathology as observed by the clinician before starting gabapentin 900 mg/day in each case, and maintained this dose. In addition to weekly evaluation and psychotherapy by the senior author, a detailed assessment based on the rating scale was done at the end of three months. Patients have been followed for three to five months as of 12/1/96.

Results: 1) All patients reported a substantial decrease in their irritability and dysphoria and a decrease in premenstrual exacerbation, 2) Compliance to psychotherapy was much improved, 3) There were marked reductions in migraine attacks (sumatriptan was discontinued at the end of one month in each case, 5) Social functioning improved which was assessed based on the assessment of degree of psychopathology.

Discussion: This is a nonconventional use of gabapentin in a specific group of patients who do not have classical Bipolar Disorder. Gabapentin seemed to have mood stabilizing effects in this group and to reduce premenstual exacerbation of irritability. It also seemed to prevent migraine symptoms. This is a very preliminary study; its limitations include the small number of cases, open design, and lack of a control group. Our findings do suggest that a larger and more elaborate study would be appropriate.

Valproate as prophylaxis for steroid-induced mood disturbances: a case report

Authors: M. Shetty, D. Lynn

A 45-year-old Caucasian female with an established diagnosis of Multiple Sclerosis was admitted to an inpatient neurology service for management of an acute exacerbation. On admission, her medications were amitriptyline, amantadine and baclofen. Since 1980, she had experienced four exacerbations; these were variously treated with ACTH, methylprednisolone and prednisone. In each episode, steroid treatments were clinically effective, but she experienced severe irritability and mood instability. For management of the current episode, a course of intravenous methylprednisolone was planned. Because of poor irritability in response to steroids, we initiated an oral loading dose of valproate (20 mg/kg/day), prior to the initiation of steroids. We continued this oral dose for five days. She reported complete freedom from irritability or mood instability during this time; nursing and psychiatric clinical observations confirmed her euthymic status.

Steroid-induced psychiatric disturbances were originally described by Cushing in his classic reports. The most common manifestations are irritability, mania, depression and psychosis. One study showed a prophylactic effect of lithium in this setting. One case of a positive response to carbamazepine has been reported. Tricyclic antidepressants have been found to consistently exacerbate steroid-induced psychiatric symptoms.

We believe that irritability and other affective symptoms associated with steroid therapy represent a form of mixed mood disturbance rather than a simple depression. Thus, lithium and carbamazepine have been useful because of their mood stabilizing effects, whereas antidepressants have given negative results. We tried valproate because of its recently established effectiveness as a mood stabilizer. The oral loading dose strategy may provide advantages of rapidity and simplicity in this setting as compared with lithium. Additionally, valproate may be helpful for some patients who cannot tolerate lithium. This study has all the limitations of a single case design with an open trial and no control or comparison group, but the positive result does suggest that more elaborate studies on a larger scale with a double-blind design involving an appropriate comparison group would be appropriate at this time.

Effects of Divalproex Sodium on 5-HT1A Receptor Function in Male Healthy Humans

Authors: I-Shin Shiah, Lakshmi N. Yatham, Raymond W. Lam,

Athanasios P. Zis

Hypothermic and hormonal responses to a challenge with a selective 5-HT1A receptor agonist ipsapirone are considered to provide an index of 5-HT1A receptor function in humans. To explore the effects of divalproex sodium on 5-HT1A receptor function, we measured the hypothermic, ACTH and cortisol, and behavioral responses to ipsapirone in ten healthy male volunteers. After obtaining a blood sample for baseline hormone levels and measuring body temperature, a single dose of 0.3 mg/kg of ipsapirone was given orally to all the subjects and further bloods and temperature readings were obtained at regular intervals for 3 hours. The ipsapirone challenge test was repeated after the subjects had been treated with divalproex sodium (1000 mg, p.o., daily) for one week. The results showed that the hypothermia induced by ipsapirone was significantly attenuated by the divalproex sodium treatment, whereas the ACTH/cortisol release and the behavioral responses following ipsapirone challenges were not altered. Our findings suggest that divalproex sodium may enhance 5-HT neurotransmission in humans via a subsensitization of 5-HT1A autoreceptors but does not appear to have an effect on postsynaptic 5-HT1A receptors.

Effects of Lamotrigine on 5-HT1A Receptor Function in Male Healthy Humans

Authors: I-Shin Shiah, Lakshmi N. Yatham, Raymond W. Lam,

Athanasios P. Zis

Lamotrigine, a new anticonvulsant, has recently been reported to be effective in treating patients with bipolar mania, depression, and schizoaffective disorder, suggesting it is perhaps a mood stabilizer with antimanic and antidepressant properties. However, the mechanisms of action underlying its efficacy in mood disorders are still unknown. To explore the role of 5-HT1A receptor function in the mechanisms of action of lamotrigine, we compared the body temperature, plasma cortisol and behavioral responses to a challenge with a selective 5-HT1A receptor agonist ipsapirone in ten healthy male humans. After obtaining a blood sample for baseline hormone levels and measuring body temperature, a single dose of 0.3 mg/kg of ipsapirone was given orally to all the subjects and further bloods and temperature readings were obtained at regular intervals for 3 hours. The ipsapirone challenge test was repeated after the subjects had been treated with lamotrigine (100 mg, p.o., daily) for one week. The results showed that neither the hypothermic nor the plasma cortisol responses induced by ipsapirone were significantly altered by the lamotrigine treatment. Our findings provide no evidence to support that 5HT1A receptor function is involved in the mechanisms of action of lamotrigine.

Effects of Lithium & Amphetamine on Brain Inositol Metabolism as Measured by MRS

Authors: Peter Silverstone, Christopher Hanstock, Andrei Pukhovsky,
and Peter Allen

The mechanism of action of lithium has been the subject of much debate since it has effects on many multiple neurological components and systems. It has been suggested that the major action of lithium is to decrease the intracellular concentration of myo-inositol and increase that of its inositol monophosphate precursors. This is termed the "inositol-depletion" hypothesis. However, previous human magnetic resonance spectroscopy (MRS) studies, including our own, have not been supportive of this hypothesis. This may have been because the MRS scanners used were not powerful enough to detect small changes or because they were carried out in the unstimulated state. Previous animal studies have been very consistent in showing that much greater changes in myo-inositol and inositol monophosphate concentrations occur following stimulation of the phosphoinositol second messenger system. In the present double-blind placebo-controlled study we utilized 1H and 31P MRS to measure the concentration of both myo-inositol and phosphomonoesters (which contain the inositol monophosphate peak when using 31P MRS). We carried out three MRS examinations in these subjects: at baseline, after 7 days administration of lithium or placebo, and again on day 8 some 2 hours following oral administration of 20 mg dextroamphetamine. We used dextroamphetamine since it stimulates the phosphoinositol second messenger system. We found that in those subjects who received lithium (n=10) there was a significant increase in phosphomonoester (PME) concentrations following amphetamine administration as opposed to no change in those who received placebo (n=6). However, we found no changes in myo-inositol concentrations either following lithium alone or following dextroamphetamine. These results are in keeping with the "inositol-depletion" hypothesis for the actions of lithium, and are the first human brain findings to provide direct support for this hypothesis.

Variation in Suicide Risk among Bipolar Families

Authors: SG Simpson, MD, DF MacKinnon, MD, MG McInnis, MD, FJ McMahon, MD, JR De Paulo, MD

Bipolar disorders are highly heritable conditions which are associated with a high risk of suicide. We looked at suicide risk in 770 relatives in 79 bipolar families which were ascertained for a linkage study. Probands and all available relatives were directly examined by experienced psychiatrists using the Schedule for Affective Disorders and Schizophrenia-Lifetime version; best-estimate diagnoses were made by 2 non-interviewing psychiatrists using the RDC. Seventy-one families had bipolar (BP) I probands and 8 had bipolar (BP) II probands. There were 16 suicides in 15 families, all but one occurring in BP I families. Ten suicides occurred in the 27 families which had a paternal pattern of transmission of bipolar disorder (X = 7.2, p<0.01). Eighty-nine subjects attempted suicide; 53% had BP I and 28% had BP II. Fifty families had at least one attempter and half of these had more than one attempter. Forty-five percent of the 89 attempters made multiple attempts. The risk of attempting suicide was higher among relatives of the 30 probands who had attempted suicide, occurring in 23 % of their relatives versus 13% of relatives of probands with no reported attempts (p=0.056). Families with a completed suicide or with a proband who had attempted suicide appear to have an increased suicide risk. We are gathering more detailed clinical data on the suicides and attempts to try to define other family phenotypes which may be associated with increased suicide risk. We will test whether families with and without suicides and suicide attempts vary on other clinical variables (e.g., rates of comorbidity) or genetic variables (e.g., paternal versus maternal transmission of bipolar disorder). In the first 28 families to be genetically evaluated, we found linkage of bipolar disorder to markers on chromosome 18. Since our preliminary data on families with suicides suggest that paternal transmission of bipolar disorder may be associated with high risk of suicide, we are continuing to analyze the possible relationship between paternal parent of origin, linkage to chromosome 18, and suicide risk.

Fuzzy and Neurofuzzy Approaches for Predicting Serum Lithium Concentrations

Authors: B.A. Sproule, Pharm.D., M. Bazoon, M.Sc., I.B. Turksen, Ph.D.,
C.A. Naranjo, M.D.,

Our group is interested in developing fuzzy and neurofuzzy approaches for optimizing psychopharmacotherapy. Lithium is the mainstay of the treatment of bipolar disorder. We have previously used established fuzzy logic methodology to construct a model for predicting serum lithium concentrations (Clin Pharm Ther 1996, 59(2):157). The fuzzy sets for the input-output variables were different for each rule in the rulebase however, which reduced the overall potential for the model to be easily understood. Fuzzy logic modeling, emphasizes the qualitative nature of computation and is useful for complex systems or vague information. Fuzzy logic explains systems in understandable linguistic terms using if-then rules. For example, if the patient's renal function is "poor" then the dose of the drug should be "low". Artificial neural networks are parallel interconnected processing units which use an empirical, 'learning by example' approach to approximate functions. Both fuzzy logic and neural networks can be used to build models based on known input-output data, however, with neural networks the relationships between variables remain unknown. In our current study we propose a novel approach for building a fuzzy rulebase composed of uniform linguistic sets, for each input-output variables using concepts of fuzzy logic and artificial neural networks. Fuzzy clustering is used to create linguistic values of fuzzy rules. Each sorted crisp values of input and output variables are partitioned using Fuzzy C-Means (FCM) method. The number of fuzzy clusters for each input or output variable equals the number of linguistic values for that variable. A competitive learning mechanism called "leaking learning" is used to perform rule generation. The proposed network has two layers, an input layer and a Rule_layer. After training, the weights of each Rule_layer node encodes the existence of fuzzy rule and the product of the weights value connected to the Rule_layer node interpreted as the trust, ? l, that the rule ith been true. Finally a two layer neural network is used to implement fuzzy logic inference and adjust the t1 of the rules in the rulebase. The trained neural network infer new output values by processing new observed input data. The network "learn" to adapt to new circumstances by adjusting its weights that are the value of ? 1. The final rulebase had 26 rules consisting of 4 input variables (i.e. weight, serum creatinine, lithium dose and time since last dose) and 1 output (SLi, range 0.2-1.24 mmol/L). The performance of the model (rmse = 0.14 mmol/L, me = -0.02 mmol/L) was similar to the previous model (rmse = 0.13 mmol/L, me = 0.03 mmol/L). The new approach is more user friendly. Work is continuing to determine if neural networks may be used to further enhance fuzzy logic models by allowing them to be adaptable over time as new information becomes available and for testing model performance in real clinical situations. Work is underway to model the kinetic and dynamic properties of other CNS drugs.

A One-Year Randomized Trial Of Clozapine Vs. Usual Care In Patients With A History Of Mixed Mania

Authors: Trisha Suppes, A. John Rush, Andrew Webb, Thomas Carmody, Helena Kraemer

Patients with severe, persistent bipolar I or schizoaffective disorder, bipolar type (i.e. symptoms despite lithium [> 0.8 mEq/L] combined with an anticonvulsant at therapeutic levels and, if psychotic, a neuroleptic chlorpromazine > 500 mg or equivalent) entered a randomized trial of either "treatment-as-usual" (TAU) or clozapine for one year. TAU allowed any medication except clozapine; clozapine patients received clozapine plus any medication. Twenty-five women and 16 men were randomized. Two patients did not return for their first visit and are excluded from the analysis. The majority of patients were bipolar I (72%). Patients were age 22 at first treatment and 25 years old at first hospitalization using median values, and current median age was 38 years. Patients were evaluated monthly using clinical symptom scales: BPRS (18-item), HRS-D (24-item), the Bech-Rafaelsen Mania Scale (BRMS), the CGI, and a 40-item side effects check list. Regression analysis was used to develop slope estimates for each individual, and the means compared between treatment groups. The BPRS (p=0.009), BRMS (p=0.019), and CGI (p=0.015), all showed a significant difference over one year. The HRS-D (p=0.061 ) showed a trend towards significance.

These results support in a randomized, though open, clinical trial of naturalistic design that clozapine provided a sustained improvement in these severely ill bipolar patients. Importantly, significant improvement occurred within six months and was sustained over the next six months.

Supported by NARSAD (TS), Lattner Foundation (TS), NIMH K21 MH01221 (TS), Mental Health Connections at UTSWMC, a Sandoz donation of medication, MH41115 and MH53799 (TC).

Results presented at 1996 meetings of Society of Biological Psychiatry Annual Meeting and ACNP Annual Meeting.

Sleep Deprivation is Most Often Antidepressant for Bipolar Depression

Authors: Martin Szuba, Lewis Baxter, Jr., William Ball

One night of total sleep deprivation (TSD) or partial sleep deprivation (PSD) produces an acute antidepressant effect in 60% of depressed patients. We previously reported that bipolar depressed patients responded more frequently than unipolar patients. We sought to compare the effects of sleep deprivation on mood in depressed bipolars and depressed unipolars.

Twenty five bipolar and twenty unipolar depressed subjects underwent one night of either TSD (subject awake 38 consecutive hours from 0700 Day 1 until 2100 Day 2) or PSD (subject awake 19 consecutive hours from 0200 Day 2 until 2100 Day 2). Raters, blind to diagnosis and sleep deprivation condition, rated subjects twice daily with a validated subscale of the Hamilton Depression Scale (H6). Subjects completed the Profile of Mood States (POMS) every two hours while awake. Baseline, demographic, clinical and mood variables did not differ significantly between bipolar and unipolar subjects. Fifty percent of bipolar and 47% of unipolar subjects underwent TSD (p > .1).

According to previously published criteria, we classified 28/45 (62%) as "responders" to sleep deprivation. H6 scores of the responder group improved 50.7 + 21.9% while nonresponders' scores improved 1.4 + 26.4% (two-tailed t = 6.8, p = .00000003). Total POMS scores improved from 76.4 + 52.6 (Day 1 mean) to 48.4 + 42.2 (Day 2 mean) in responders (two-tailed t = 5.5, p = .00000006).

Among bipolar subjects, 20/26 (76.9%) and among unipolar subjects, 8/19 (42.2%) were responders (x2 = 5.7, p = .02). Neither the magnitude nor the timing of maximal mood response was different between bipolars and unipolars on H6 or POMS scales

These results confirm earlier reports that bipolars are more likely than unipolars to respond to one night of sleep deprivation with an antidepressant effect. However, the magnitude and timing of the response is similar, regardless of diagnosis.

Midnight Administration of Protirelin in Bipolar Depression

Authors: Martin Szuba, Antonio Fernando, Jay Amsterdam, Peter Whybrow,
Andrew Winokur

Some studies from the 1970's and 1980's suggest that TRH produces a quick, though transient antidepressant effect. Differences between diagnoses and route of administration may account for the disparities between these works. Remarkably little attention has been paid to the timing of administration. Most investigators gave TRH at 0800, when the pituitary is least responsive to TRH. The pituitary is much more responsive to TRH at 2300 than it is at 0800. We thus speculated that TRH infusion at midnight may produce more consistent antidepressant effects than previous work has shown.

In a controlled, double-blind protocol, fourteen adults with Bipolar Depression were studied for three days (baseline-Day 0; day after infusion-Day 1; and the second day after infusion-Day 2). Responses were rated with an abbreviated version of the Hamilton Depression Scale (H6). At midnight on the end of Day 0, 0.5 mg TRH or saline (placebo) was given intravenously.

Baseline demographic, clinical and mood variables were not different between the 7 TRH subjects (3 ? and 4 ? aged 41.9?13.2 y.o.) and the 7 placebo subjects (2 ? and 5 ? aged 42.6?5.6 y.o.). Over the two days post-infusion, six of the seven (86%) TRH subjects significantly improved, both clinically and statistically, while only one of the seven (14%) placebo subjects improved (Two-tailed, Fisher's Exact Test x2= 7.1, d.f. = 1, p = .03). H6 ratings of the TRH group dropped 70.2+35.9 %, while placebo subjects improved only 21.8 ? 21.5 % (Mann-Whitney U = 5.3, p = .02). No subjects became hypomanic.

TRH subjects responded significantly more frequently and significantly more robustly with an antidepressant effect than did placebo subjects. The results of this experiment suggest that administering TRH at midnight can rapidly improve bipolar depressive symptoms, without inducing hypomania or mania.

Alcohol Problems and Long-term Psychosocial Outcome of Chinese Bipolar Disorder

Authors: Shang-Ying Tsai, Chiao-Chicy Chen, Eng-Kung Yeh

One hundred and fifty-eight Chinese patients with bipolar disorder (DSM-III-R criteria) having been followed at various intervals for more than 15 years were included in this naturalistic study. Based on chart reviewing, 13 patients (8.2%) were found to have alcohol problems during their illness. Reliable histories about 101 patients of them were obtained from medical records, direct interviews, and family's confirmation. The lifetime prevalence of alcohol abuse among Chinese bipolar patients was 6.9%, and alcohol dependence 3.0%. There were 30% of our subjects impaired occupational functioning and nearly one-third, as the Global Assessrnent of Functioning Scale (APA, 1994) defined, clearly dysfunctional. In spite of markedly low prevalence of co-occurring alcohol use disorders, our data revealed that the Chinese bipolar patients were not superior to the Westerns with respect to psychosocial outcome (marriage, work, and social adjustment). In addition, there is significant difference between alcoholic and nonalcoholic groups in rate of rapid cycler(p<0.02), but not in variables regarding psychosocial functioning. We suggest that bipolar disorder itself rather than complicated alcohol use disorders has the impact on the long-term psychosocial outcome. Furthermore, the results support that bipolar disorder causes a subgroup with persistently psychosocial dysfunctioning.

The Antiepileptic Drug Lamotrigine May Limit Pathological Excitation by Modulating Calcium- and Potassium Currents

Authors: J. Walden, J. Wegerer, M. Berger, H. Grunze

Lamotrigine (LTG) is an antiepileptic drug gaining increasing interest as a potential mood stabilizer in psychiatry (cf. Walden et al., 1996). To further analyze the properties of LTG, we performed extracellular and whole cell patch clamp recordings from CA 1 pyramidal neurons in a hippocampal in vitro slice preparation.

Field potentials (FP) were elicited in a calcium-dependent model by omission of extracellular Mg2+. LTG (2? M) reduced the frequency of FP in a dose-dependent manner. In a subthreshold dose of 1 ? M, LTG combined with a subthreshold dose of the calcium antagonists verapamil (2 ? M) or carbamazepine (10 ? M) reduced reversibly the EFP frequency by 33.5 ? 11.4% for verapamil, and by 64.1? 33% for carbamazepine, respectively.

In the whole-cell patch clamp mode LTG (25 ? M) showed no effect on input resistance, AP threshold and maximal spiking frequency at 950 pA, but decreased AP overshoot. LTG (100 ? M) caused not only a decrease of the AP overshoot, but also a reduction of AP frequency for given current steps by reversibly increasing the interspike interval with unchanged AP duration.

Calcium antagonistic properties of LTG at voltage-operated calcium channels appear likely under epileptic conditions. Under more physiological stimulus conditions LTG may first increase the fast transient K+ current IA or the calcium dependent K+ current IAHP thus limiting excitation in cortical networks.

Reference: Walden, J., Grunze, H., Bingmann, D., Dh sing, R. (1992): Calcium antagonistic effects of carbamazepine as mechanism of action in neuropsychiatric disorders. EuropeanNeuropsychopharmacology 2, 455 Walden, J., Hesslinger,B.,van Calker,D.,Berger,M.: Addition of lamotrigine to valproate may enhance efficacy in the treatment of bipolar affective disorder. Pharmacopsychiatry 29, 193-195, 1996

Treatment of Patients With Bipolar Disorders with the New Antiepileptic Drugs Lamotrigine and Gabapentin

Authors: J. Walden, M. Berger, C. Norman, B. Hesslinger, H.Grunze,

The new antiepileptic drugs lamotrigine (LTG) and gabapentin are of special interest as potential mood stabilizers and antimanic drugs in psychiatry (cf. Calabrese et al., 1996; Walden et al.,1996). From its elementary mechanisms of action LTG is discussed to have effects on neuronal sodium and calcium channels and to decrease the glutamatergic neurotransmission (Leach et al.,1986). The exact mechanism of gabapentin is unclear, but an affinity to amino acid transporters is discussed (Hill et al., 1993).

We treated some patients with LTG in combination with valproate in acute mania and observed patients up to two years during such a prophylactic therapy. A considerable improvement of the patients’ condition was achieved during this drug regime and lasted over the total follow-up period of two years. The plasma concentrations were in the range of 57.1 to 95.2 mg/ and 1.9 to 6.2 mg/l for valproate and lamotrigine, respectively.

In first cases, we treated patients with acute mania with gabapentin. Oral dosage was 2000 to 3200 mg daily with plasma levels of about 4 mg/l. Although a co-administration with neuroleptics and benzodiazepines was necessary in the first weeks of medication some patients showed a significant psychopathological improvement.

The Response to an Apomorphine Challenge in Bipolar and Unipolar Depression

Authors: Walsh AES, McPherson HM, Silverstone T

Alterations in brain dopamine (DA) function have been implicated in the pathogenesis of depressive illness . Evidence from human studies suggest DA may be particularly important in certain types of depressive illness, including bipolar depression and depressive illness characterised by psychomotor retardation. The aim of the present study was to determine whether or not unipolar and bipolar depression can be distinguished on the basis of an apomorphine neuroendocrine challenge test.

Methods: unipolar, bipolar depressives and age, weight and sex matched controls underwent apomorphine challenge testing (0.008mg/kg) utilising a standard neuroendocrine procedure. Depressed patients were challenged: i) when depressed (baseline), ii) when euthymic or upon the development of mania. Controls were tested at baseline only.

Results: preliminary results to date on 12 subjects in each group show bipolar depressives were significantly older (Mean age=40.6yrs vs 30.3yrs; p=0.02) and heavier (Mean weight=81.3kg vs 66.1kg; p=0.04) compared to unipolar depressives, while baseline depression ratings were not significantly different between the two groups (M&A=30.8 (bipolar) vs 33.3 (unipolar); p=0.29; HamD=20.1 (bipolar) vs 22.8 (unipolar); p=0.15). Preliminary neuroendocrine results on 7 depressed subjects in each subgroup vs 7 controls suggest a trend towards an enhanced GH response to apomorphine, with unipolar GH responses > bipolar GH responses > control GH responses. These findings, suggesting enhanced sensitivity of DA receptors in depression are in contrast to the literature and will be discussed.

Seasonality of Manic Depressive Illness over Fifty Years

Authors: Diane K Whitney MD, Verinder Sharma MB BS, Karen Kueneman BA

Background: A seasonal pattern for affective disorders has long been of interest to clinicians. Previous studies have generally shown a peak incidence for mania in summer and for major depressive episodes in spring and autumn. However many studies are limited by the small number of patients and the brief time period of the study.

Objective: To investigate whether a seasonal pattern exists for admissions of manic depressive illness at a provincial psychiatric hospital.

Method: A review of the Case Conference Books was conducted to collect data for the decades 1920 to 1960. The admissions were divided according to season for each mood state (mania, depression, mixed). Computer data for more recent decades is in the process of being compiled.

Results: Over the fifty year period, there were 1482 admissions for mania, 1493 for depression and 552 for mixed states. Using chi square for analysis, no seasonal pattern of admissions for mania was evident. There was a preponderance of admissions for depression in spring and summer but this did not reach statistical significance. For mixed states there was a peak incidence of admissions in the summer (x2=15.45, p<.01).

Conclusions: The findings in this study do not support the results of previous studies that have shown a seasonal pattern for major depression and mania. The authors believe that the peak incidence for summer admissions of mixed states is a unique finding. Once the analysis of more recent decades is completed, further conclusions may be drawn.

+The present single case reports suggest that lamotrigine and gabapentin may be useful new agents for the treatment of patients with bipolar disorder who fail to respond to lithium or other antiepileptic drugs.

A Double-Blind, Placebo-Controlled Trial Comparing the Effect of Paroxetine and Imipramine in the Treatment of Bipolar Depression

Authors: Muriel L. Young, Cornelius D. Pitts, Rosemary Oakes, Ivan P. Gergel

The treatment of bipolar depression is a complex issue. Lithium therapy has been the mainstay of managing shifts into depression as well as the manic component of the disease. However, many of those treated with lithium and other antidepressant agents remain unresponsive. This multicentered, placebo-controlled trial compared the efficacy and safety of paroxetine and imipramine in treating bipolar depression in patients stabilized on lithium therapy. In this trial, 117 patients diagnosed with bipolar depression (DSM-III-R criteria) were randomized in double-blind fashion to receive paroxetine (n= 35), imipramine (n= 39) or placebo (n= 43) for a 10-week treatment period. The dosage range for paroxetine was 20-50mg daily and for imipramine, 50-300mg daily. Patients were stratified according to high (> 0.8mEq/L) and low (< 0.8mEq/L) serum lithium levels. The primary comparison of interest was paroxetine versus placebo. Baseline efficacy scores (HAMD and CGI, severity of illness) were similar across all treatment groups for both high and low lithium serum levels, as well as the total patient population. These scores indicate that patients entering the study were moderately ill. In the endpoint analysis of primary efficacy parameters (change from baseline in the HAMD and severity of illness CGI), both paroxetine and imipramine were statistically superior to placebo (p < 0.05) in the low lithium level stratification. Although there was no statistical separation between treatment groups in other stratifications, both active groups exhibited a numerical advantage over placebo. Based on a response criteria of HAMD < 7 or CGI, global improvement < 2, the secondary efficacy analysis showed no statistical separation for the secondary efficacy parameters between treatment groups. An assessment for the development of manic symptoms, indicated that no patients in the paroxetine group developed mania in any of the lithium stratifications or the total population. Conversely, the incidence for imipramine was 8.3%, 5.9% and 10.5% for the total population, high and low lithium stratifications, respectively. The safety profiles of paroxetine and imipramine showed that the incidence of anticholinergic events was higher in the imipramine group. Such events included: dry mouth, constipation, sweating and confusion. Gender related adverse events (impotence and abnormal ejaculation) were also higher in the imipramine group. Serum lithium levels remained in the medically accepted therapeutic range for all treatment groups. This study shows that paroxetine and imipramine are comparable in efficacy in the treatment of bipolar depression. This effectiveness is especially evident in patients maintained on low serum levels of lithium. When used in such patients, paroxetine may be associated with reduced incidence of mania, anticholinergic and gender related side effects. Clinically therefore, paroxetine presents an advantage for patients unable to tolerate high lithium serum levels or who may be refractory to the antidepressant effects of lithium.

Acute Treatment of Bipolar Depression with Gabapentin

Authors: L. Trevor Young, Janine C. Robb, Irene Patelis-Siotis,
Cathy MacDonald and Russell T. Joffe

Treatment approaches for depression in bipolar disorder are not well established. Anticonvulsants have shown to have mood stabilizing effects, and little is known as to the therapeutic effects of the newer anticonvulsants in bipolar disorder. Gabapentin has been suggested to have mood stabilizing effects and possibly antidepressant properties . This open label trial investigated the therapeutic effects of gabapentin in bipolar disorder types I and II, normocyclic and rapid cycling course.

Outpatients in a mood disorders program who met criteria for bipolar disorder, depressed phase and who had failed other conventional treatments with mood stabilizers were enrolled. At weekly visits, dose adjustments of gabapentin were based on clinical response and treatment-emergent side effects. Pre and post treatment Hamilton scores were compared, using a two tailed paired t-test.

Fifteen patients with bipolar disorder were treated with gabapentin over six weeks at varying doses. In 53% of those treated there was a significant reduction between baseline and six week Hamilton Depression rating scale scores. There was no induction of mania or destabilization of mood cyclicity overall.

This is a preliminary report that provides evidence for the use of gabapentin in the treatment of acute depression in bipolar disorder. These preliminary findings require confirmation and extension in larger controlled trials to determine continued efficacy and whether it will lead to longer term mood stabilization.

Valproate in the Treatment of Bipolar Disorder in Adolescents

Author: Linda Zamvil, M.D.

Bipolar Disorder is commonly considered an adult condition. However, as many as 35% of adults with Bipolar Disorder experience their first episode of this disorder during adolescence. Adolescent onset of Bipolar Disorder has been linked to greater incidence of psychoticism, comorbidity and atypical expression of symptoms. In addition, earlier age of onset of Bipolar Disorder has been demonstrated to reduce responsiveness to lithium carbonate. Research has demonstrated that approximately 2/3 of lithium refractory bipolar patients respond to Valproate, an anticonvulsant that has revealed mood stabilizing properties when administered to adult bipolar patients. In addition, the side effect profile of this medication has been quite favorable. This presentation will summarize the results of a prospective, flexible dose open trial study of Valproate with hospitalized adolescents who met the DSM-III-R and SADS (Schedule for Affective Disorders and Schizophrenia for school-aged children) for Bipolar Disorder. Symptom severity was measured by the Beck Depression Inventory, the Becks Hopelessness Scale, the Teenagers' Self-report Questionnaire, the Clinical Global Assessment Scale and the Clinical Global Impressions Scale. Evaluations were administered upon admission, the first day of medication and then at two week intervals by a trained member of the research staff.

-ten adolescent subjects

-five males

-five females

-age range 13-l8 years

-mean age 15.8 years

Attributable and Avoidable Mortality in Recurrent Affective Disorders

Authors: Claudia Schumann, Bernd Ahrens

In the last few years several studies have shown that long-term lithium treatment in recurrent mood disorders can significantly reduce excess mortality.

Since it is not possible to study the anti-suicidal property of lithium using a double-blind placebo design, the efficacy of lithium treatment in reducing excess mortality in affective disorders has to be measured in some other way, which was done in this study .

Taking into account prevalence rates in affective disorders and suicide rates in the general population, and making the assumption that 60% of those suicides can be considered as the result of affective disorders, in our study the following procedure was employed:

60% of the suicide rate per 100.000 inhabitants was imposed upon the equivalent number of patients diagnosed as suffering from affective disorders according to the prevalence rate for a specific age and sex group found in the ECA study.

Thus, expected suicides were calculated for a population of 827 patients with recurrent affective disorders being treated with lithium in four different lithium clinics. In the group, comprising 5,600 patient years under lithium treatment, a total of seven suicides were observed whilst 1.3 were expected in comparison to the general population. Taking, more appropriately, the adjusted numbers for affective disorders as a basis for the analyses, 37 expected suicides were calculated.

Therefore 30 suicides were avoided in a population of 827 long-term lithium treated patients. This finding supports the hypothesis that lithium is not only effective in the prevention of relapses of affective episodes, but also in the prevention of suicide.

Recurrence in Affective Disorder -A Case Register Study

Authors: Lars Vedel Kessing, Per K. Andersen, Tom G. Bolwig,
Preben Bo Mortensen

Background: In recent years, studies of the risk of recurrence in affective disorder in relation to the number of prior episodes have given contradictory results.

Method: Survival analysis was used to calculate the rate of recurrence after successive episodes in a case register study including all hospital admissions with primary affective disorder in Denmark during 1971-1993. Totally, 20,350 first-admission patients were discharged with a diagnosis of affective disorder, depressive or manic/circular type.

Results: The rate of recurrence increased with the number of previous episodes in both unipolar and bipolar disorder. Initially, the two types of disorders followed markedly different courses, but later in the course of the illness the rate of recurrence was the same for the two disorders. A subgroup of patients who developed many episodes seemed to have a constitutionally high risk of recurrence already from the first episode.

Conclusion: The course of severe unipolar and bipolar disorder seems to be progressive in nature despite the effect of treatment.

Keywords: affective disorder- unipolar- bipolar- case register- longitudinal data-recurrence

A Psychoeducational Group Program for Partners/Family Members of Patients with Bipolar Disorder

Authors: Bartha, C., Thomson, C., Parker, C.

Current literature in the area of psychoeducational groups for family members reports good outcomes for short-term groups, however many of these studies focus on the general psychiatric or schizophrenic population. This study evaluates the impact of a standardized, outpatient, psychoeducational group intervention designed specifically for partners and family members whose relatives have been diagnosed with bipolar disorder.

Family members, most often partners and parents of adult patients, are offered an eight week series of two hour groups facilitated by two social workers. The ill relative does not attend. The first two groups focus on providing information about the nature of the disorder, its etiology and treatment options. The remaining six groups are interactive and supportive, and focus on skill-building in the areas of stress management, communication, problem and crisis management. An important aspect of the group is the use of group process to deal with issues which frequently arise with bipolar disorder. For example, group members' concerns and experiences with mania, violence and suicidal behaviour are dealt with in discussions about stress and crisis management. Role-plays of relatives' experiences are used to illustrate different communication and crisis management strategies. The emotional impact of these experiences is processed and integrated with practical strategies for coping with future problems.

The group program is an integral part of our clinical service and has received excellent evaluations from group participants. More recently, it was designed as a study to formally evaluate outcome. The study employs a repeated measures design to evaluate the impact of the group intervention on family members' sense of burden, depression, anxiety and stress, as well as their sense of control over their lives. Measures are collected at the time the participant is placed on the waiting list, prior to group one, after group eight and 12 weeks later at follow-up. Preliminary data on these measures will be available. The investigators are also currently developing a manual which describes the program and discusses the clinical challenges of facilitating these groups.

Gabapentin as Mood Stabilizer in Adolescent Bipolar Disorder: A Case Report

Authors: Thau K. M.D., B. Schmed-Siegel M.D., A. Holzinger M.D.

Gabapentin is a novel anticonvulsant that is structurally related to gamma-aminobutyric acid (GABA). It has a unique spectrum of activity in animal seizure models and has demonstrable efficacy in patients with refractory epilepsy.

Although designed as a GABA-mimetic, gabapentin does not interact with any of the known pharmacological sites on either the GABAa or GABAb receptor, nor does it block GABA uptake or inhibit the GABA-metabolising enzyme GABA-transanimase. Some electrophysiological studies suggest that gabapentin may act as a partial agonist at the glycine modulatory site of the NMDA receptor.

In several controlled and uncontrolled clinical studies, the efficacy of carbamazepine and valproate in the prophylaxis of bipolar affective disorders and the treatment of the acute manic syndrome has been shown.

Naturalistic studies show that in clinical practice the most severe cases of bipolar disorder will be set on a combination of two to three different mood stabilizers. We would like to present a case of a 14 year old caucasian female, who experienced her first affective episode (depressive episode with suicide attempt) four years ago. Under sertraline she became manic and was put on lithium and valproid acid. Although the intensity of the mood swings decreased slightly, the side effects due to lithium were unbearable and lithium had to be discontinued. Instead carbamazepine was started. As the clinical course did not improve, after her next psychotic episode (with hallucinations), that required additional neuroleptic treatment (risperidon), we added a third mood stabilizer, gabapentin. The mood swings improved significantly over the next six months. Due to difficulties with the patient’s compliance, we discontinued carbamezepine and valproid acid. As mood stabilizer she remains on 2400mg gabapentin per day. As she still was suffering from depressive symptoms and rare hallucinations, paroxetine and olanzepine were added.

In summary, the course of the illness showed under gabapentin a marked improvement. Because of the strong side effects of lithium and the nonresponse to lithium as well as to carbamazepine and valproate, gabapentin can be in some patients an important alternative mood stabilizer. The course will be presented in calendar form.

Bipolar Education Treatment Trial (BETT): Preliminary Results

Authors: Sagar V. Parikh, M.D.,C.M., FRCP, Alice Kusznir, M.Ed., BScOT,
Robert G. Cooke, M.D.,M.Sc., FRCPC, Carol M. Parker, B.S.W., M.S.W.,
Elizabeth (Billie) Pryer, B.Sc.,B.Sc.N.,R.N.

Epidemiologic surveys reveal that many individuals affected with bipolar disorder do not adequately utilize health services. Treatment studies also show that medication noncompliance affects at least 50% of bipolar patients. Such noncompliance often results in illness relapses, hospitalizations, and even suicide. Health education can offer a helpful paradigm for treatment. The Bipolar Education Treatment Trial (BETT) seeks to examine the utility of psychoeducation and its affect on medication compliance, relapse and readmission rate as well as quality of life. The BETT consists of five intensive, interactive educational sessions conducted by a psychiatrist, nurse,occupational therapist, and social worker. The session with the psychiatrist features a discussion of the treatment options as well as key aspects and course of the illness. A discussion of the management of symptoms, strategies for improving functioning, and the role of the family in the management of the illness together with education about community resources, is offered in the remaining sessions. The goal of these sessions is to teach self-recognition of symptoms, appropriate use of community resources and coping strategies. Starting in January 1997, 60 newly diagnosed or currently untreated individuals with bipolar disorder are being enrolled in the BETT as part of a randomized, controlled trial, with 30 receiving the course and 30 receiving 'usual care'. Both groups will continue to receive medical management of their bipolar disorder from their referring physician. The two groups will be compared for clinical features, symptoms, medication compliance, and blood levels of mood stabilizers at 3 and 6 months post intervention. Additional outcome measures include a knowledge questionnaire, a self-report of disability, health service and community resources utilization, a quality of life measure as well as an impact and satisfaction survey. This presentation will offer details of the study, and pilot and preliminary data.

Linkage Studies of Lithium Responsivene Bipolar Disorder and Candidate Genes

Authors: Turecki, G; Alda, M.; Grof, P.; Martin, R.; Cavazzoni, P.; Duffy, A.; Grof, E.; Rouleau, G.

Conflicting results have been a major problem in linkage studies of Bipolar Disorder (BD) and heterogeneity has been largely identified as one of the main reasons for the lack of replication. We have been studying excellent long-term lithium responsive bipolar families as a method to reduce heterogeneity. In this study, 25 families with 79 genotyped affected individuals were investigated. Diagnoses were made using the SADS-L with RDC criteria, and individuals with BD, schizoaffective disorder and recurrent major depression were considered affected. To be considered lithium responsive, all patients must have had a high risk of recurrence and must have been maintained on lithium exclusively, with no further episodes for a minimum of three years. Linkage studies were carried out using parametric (assuming a recessive mode of inheritance with sex specific penetrance consistent with parameters obtained from segregation analysis in this population) and nonparametic methods. Several candidate genes and regions were tested, based on hypothesized mechanisms implied in the fisiopathology of bipolar disorder, as well as on published linkage reports. No indication of linkage was found between any of the tested markers and lithium responsive bipolar disorder.

Lamotrigine in Treatment of Refractory Bipolar Depression

Authors: Lakshmi N. Yatham, Vivek Kusumakar

Lamotrigine is a new anticonvulsant which has recently been reported to have antidepressant effect and antimanic properties. We examined the efficacy of lamotrigine in treatment of a group of patients with refractory bipolar depression. Twenty-two bipolar depressed patients who were refractory to treatment with a combination of divalproex sodium (DVP) and another mood stabilizer or DVP and antidepressant for 6 weeks were treated with an addition of lamotrigine to DVP. All patients were on DVP monotherapy for at least 2 weeks at baseline. Patients were seen at weekly intervals (weeks "0" to "6") and HAM-D 21 item was completed at each visit. Lamotrigine 25 mg bid was added to DVP at week "0" and the dose was increased to 50 mg bid at week 2 if HAM-D 21 item score was > 15. Sixteen out of 22 (72%) responded by the end of week 4 when response was defined as > 50% reduction in HAM-D score compared to baseline. All patients tolerated medications well and none developed rash. None switched to hypomania or mania. The results of this preliminary study suggest that lamotrigine may be useful in patients with bipolar depression.

Functional Health in Bipolar Disorder: Results fom the Ontario Health Survey - Mental Health Supplement

Authors: Cooke RG, Lin E, Parikh SV, Kusznir A, Scott E

Bipolar disorder (BD) can be associated with a broad range of disruptive effects on social, occupational and recreational functioning and subjective well-being. Most information on the impact of BD on these aspects of quality of life has been gleaned from clinical populations. For example, bipolar disordered patients in our clinic have been shown to experience difficulties in multiple domains of functioning and well-being, as severe, where comparative data is available, as difficulties reported by patients with other serious psychiatric or medical illnesses in our centre or elsewhere. However, there are also reports from epidemiologic and community-based studies (e.g. Romans and MacPherson, 1991) demonstrating the deleterious effect of BD on selected areas of functional health.

In the current study, we used data collected in a large scale epidemiological investigation to quantify the impact of BD on a number of dimensions of functional health and well-being in individuals living in the community.

The Mental Health Supplement to the Ontario Health Survey (Offord et al, 1994) was a community epidemiologic survey of 9953 provinical residents age 15 years and over, intended to assess the prevalence, and associated disability and service utilization, of selected major mental disorders. In the current study, we compared interviewees aged 18-65 years, meeting DSM-IV criteria for BD (N=88), with subjects meeting criteria for major depressive disorder (N=590), anxiety disorders (panic disorder, agoraphobia, generalized anxiety disorder) (N=1427) and no psychiatric disorder (N=5287), on survey items relating to social, occupational and recreational functioning, subjective well-being and life satisfaction. Group differences were assessed using ANOVA and non-parametric tests (chi-square, Kruskal-Wallis) as appropriate, correcting for age and sex differences where necessary.

Compared to the other groups, subjects with BD reported highly-statistically significantly increased dissatisfaction and impaired performance and/or enjoyment in numerous areas of social, occupational and recreational functioning. These results highlight the serious impact of BD on quality of life in subjects living in the community.

The Combination of Lithium and Divalproex Sodium in Lithium-Resistant Manic Adolescents

Authors: Vivian Kafantaris, M.D., Brian Klee, M.D., Robert Dicker, M.D., Daniel J. Coletti, M.S., Gina Padula, M.D., Neil Smoke, D.O., Cheryl Halpern Colvin, Ph.D., and Susanne Choe, B.S.

Background: Monotherapy with lithium or divalproex sodium (VPA) is often ineffective in the treatment of adolescents with mania. In other studies of VPA in adolescent mania, all or nearly all subjects received concurrent antipsychotic treatment (Papatheodorou et al., 1995; West et al., 1994). To minimize the use of antipsychotics, we are assessing the efficacy of combining VPA and lithium in lithium nonresponders.

Method: Eligible subjects in this ongoing trial are 12-18 years old, in a manic or mixed episode, with inadequate response to lithium at therapeutic levels for four weeks or more. VPA is added, with target serum levels of 50-120 mcg/ml. Subjects are rated weekly for four weeks on standardized rating instruments.

Results: Eleven subjects (4 males, 7 females, mean age=16.35 years) have been enrolled to date. All but one had mixed episodes. Six initially presented with psychosis; only two required haloperidol throughout the investigation. At study entry, the mean serum lithium level was 0.96Eq/L (SD=0.16). At week four, the mean serum level of VPA was 72.77 mcg/ml (SD=17.02). Gastrointenstinal complaints (n=5), sedation (n=4), and chest pain (n=1) slowed the rate of dose increments.

Mean MRS scores delined from 23.09 (SD=9.16) at baseline to 13.45 (SD=6.62) at week four (t=3.23, df=10, p=.009). Seven subjects had a decline of greater than 25% on the MRS. Scores on the 17-item Hamilton Depression Rating Scale also decreased, from 12.82 (SD=7.74) at baseline to 6.63 (SD=2.83) at week four (Z=2.49, p=.013), with nine declining by greater than 25%.

Conclusion: Significant reductions were observed in both manic and depressive symptoms when VPA was added to lithium. Although the sample is small and lacks a comparison group, results suggest that lithium and VPA in combination may be effective in this population, with less frequent need for antipsychotic medication.

Alcohol Abuse and Dependence: Immunocytochemical Basis of Comorbidity with Mood Disorders

Authors: M. Chrysanthou-Piterou, M.R. Issidorides

In studies of the relation between alcoholism and mood disorders, the percent of comorbidity varies between 40%-67% (Kaplan et al., 1994). Alcoholism is more frequent in bipolar patients, it responds to lithium therapy, and may be a secondary complication of bipolar disorder (Winokur et al., 1995). In this study we sought to define a possible neurobiological common denominator of the two conditions. Previous findings in postmortem tissue of patients with mood disorders revealed immunoreactivity of ubiquitin - a stress protein - in dopamine neurons of substantia nigra. A pilot study of three alcoholics revealed also ubiquitin immunoreactivity in the dopamine neurons of the substantia nigra. The above findings prompted the investigation of a large sample of chronic alcoholics (33-80y), in order to study the distribution of ubiquitin immunoreactivity in the midbrain. The application of a polyclonal ubiquitin antibody to paraffin sections of the midbrain of 27 patients and 10 matched controls showed immunoreactivity distributed in two patterns: in the younger age group (33-50y), with cirrhotic livers, staining was localized in neuronal dendrites, occasionally in the cytoplasm and most often along the cell membranes. In the older age group (50-80y) the neurons were negative and most immunoreactivity was confined to dystrophic neurites. These findings suggest that the dopamine system, which subserves reward processes and is known to be implicated in memory networks, is more vulnerable to alcohol in the younger alcoholics than in the older ones. Lesions in alcoholism and dysfunction in mood disorders with emergence of stress proteins in the dopamine system, appear to represent a biological common denominator of the two conditions. Judging from the lack of localization of ubiquitin immunoreactivity in the older age group, we propose that this particular age group of abusers can develop protective compensatory adaptation mechanisms to the effects of alcohol.

Utrastuctural and Immunocytochemical Evidence of Membrane Fluidity and Noradrenaline Loss in Depression

Authors: M. R. Issidorides, M. Chrysanthou-Piterou, S. Havaki, G. D. Pappas

Extensive studies have shown that a large fall in noradrenaline (NA) concentration in the locus coeruleus (LC) is critically involved in mediating behavioral depression. With the potassium permanganate (KMnO4) method for electron microscopy (EM), designed to demonstrate catecholamines as dense cores inside vesicles (Hokfelt and Jonsson, 1968), we have shown that in the normal human LC perikarya numerous large globules contain dense cores, presumably indicating the presence of NA (Issidorides et al., 1996). This is a new phenotype, compared to animals where few and small dense core globules in man could represent a back-up storage compartment of neurotransmitter necessary for meeting the increased demands to cope with unpredictable stresses, inherent in man’s ecological (social) environment. Our objective was to confirm the involvement of the globules in the fall of NA in depression by studying ten suicides with a diagnosis of major depression. We applied the KMnO4/EM method, as well as a rabbit monoclonal anti-serum to NA, visualized by colloidal gold EM immunostaining. We found that the globules in the LC neurons of the suicide cases were as numerous as those in normal controls, but their electron density was greatly reduced, indicating depletion of NA. This was confirmed by the decreased immunoreactivity of NA. Abnormalities in the ultrastructure of the double membranes surrounding the dense bodies, such as dissociations, splits and blebs, indicating increased membrane fluidity, support the hypothesis that defective membrane chemistry and structure may be the cause of this neurotransmitter leakage. Membrane fluidization, a result of low cholesterol, could be an important biological substratum in major depression for the wide spectrum of dysfunctions which are expressed in many body systems aside from the brain.

Medications Used for Mood Disorders From 1907 to 1952 at Eginition Hospital: Relevant Cellular Effects

Authors: J. Mantonakis, A. Deffner, and M.R. Issidorides

Eginition Hospital was founded in 1904 as the first Neuropsychiatry Clinic of the University of Athens. The hospital records filed in the archives are dated from 1906. Treatments were entered in the records from 1907. The purpose of the present study was to examine to what extent the old medications, persistently used to treat mood disorders before 1952, held clues for the understanding of the core dysfunction in depression and mania. A random sample of patients’ records from 1907 to 1952 was consulted and cases with the symptomatology of mood disorder (totalling 795) were studied. All compounds appearing in the records as treatments were tabulated according to frequency of use. In a decreasing order the following were used in major depressive disorder: sodium cacodylate, laudanum, barbiturates, calcibromine, bromides, scopolamine, opium tincture, etc. In the manic episodes the order was: bromides, scopolamine, chloral hydrate, sodium cacodylate, barbiturates, morphine, hypertonic NaCl solution, etc. Furthermore, we tabulated and listed according to dates of administration of the drugs and the diagnosis of the patients, not only the medications, but also all other therapeutic modalities which appeared in the records over the years, such as hydrotherapy, fever therapy, cardiazol shock and, since 1947, E.C.T. Similarities were found between the action of some treatments and that of modern therapeutic approaches to mood disorders. The cellular effects of some of these drugs and medications on patients’ blood cells studied in vitro point to their beneficial actions on the impaired immune function underlying mood disorders.

Ultrastructural Correlates of the Biochemical Effects of Lithium on Gene Expression

Authors: S. Sidiropoulou-Skokou, S. Havaki and M.R. Issidorides

Bosch et al. (1992) showed that lithium treatment of hepatoma cells resulted in a decrease in the level of a specific mRNA, i.e. in a decrease of gene expression. In addition, acute lithium treatment of rats also decreased the expression of the same hepatic gene. These are the first observations of a direct inhibitory effect of lithium on gene transcription, the mechanism of which remains unclear. Our previous in vitro experiments have shown that rat tissue slices, exposed to lithium carbonate, display nuclear chromatin condensation which is known to block gene expression. Given the fact that lithium administration to manic depressive patients may cause hepatic damage, we sought to verify whether similar condensations in liver chromatin could represent the necessary conformational change, affecting gene expression, and hence, be responsible for the observed liver damage in bipolar patients. Following incubation of alternate slices of rat liver in saline containing 1,1 mEq/L lithium carbonate, and in plain saline, for 2 h, we processed the tissues for electron microscopy with a staining method designed to distinguish condensed from dispersed chromatin, by staining with phosphotungstic acid hematoxylin applied in toto.

Observations of the thin sections showed that the majority of nuclei in the lithium-treated liver tissue displayed unusual condensations of the chromatin: large masses attached to the nuclear membrane, segregated in the center of the nucleus and often capping the nucleoli. This distribution is typical of chromatin condensation induced by lithium in immature neuronal cells (D’Mello et al., 1994). Therefore, the condensing effect of the drug on the chromatin, producing steric hindrance to RNA polymerases, is a legitimate cause for the reduced gene expression leading to organ pathology or to the therapeutic modulation of gene expression in the brain.

This work was supported by the Theodor Theohar Cozzika Foundation

Histone Biosynthetic Profile of Lymphocytes: Peripheral Marker of the Phases of Bipolar Disorder

Authors: T.G. Sourlingas, M.R. Issidorides, A. Karpouza, G. Trikkas,
K.E. Sekeri -Pataryas

According to previous studies, affective disorders have been associated with abnormalities of the immune system, which may be studied in leukocytes of peripheral blood. The aim of the present investigation was the biochemical analysis of lymphocytes from patients with bipolar disorder, so as to define biochemical markers which could characterize the physiological state of lymphocytes from these patients. It is well known that lymphocytes of normal individuals are in the Go resting phase of the cell cycle. Histone synthesis is characteristically different during the Go, G1/G2 and the S phase of the cell cycle. As such, it can be used as a biochemical parameter with which to distinguish between resting and actively cycling cells. In addition, specific ratios of the various histones are required for the maintenance of genome integrity, which is affected by abnormalities of the immune system. In order to investigate the physiological state of lymphocytes from patients with bipolar disorder, total histone synthesis, as well as histone variant synthesis were analyzed in peripheral blood lymphocytes from 12 patients and 7 normal control subjects, after incubation of cells with radioactive protein precursors, extraction of histones from cells, two dimensional electrophoresis, and determination of incorporated radioactivity. It was found that the lymphocytes of patients in the normothymic phase had values similar to those of normal control subjects. In contrast, the lymphocytes of patients in either the depressed or the manic phase of the illness were found to have values intermediate to those of cells in the active and the resting state of the cell cycle. The results of this study show that the synthesis of histones may be used as a novel biochemical parameter to distinguish and differentiate among the three phases of the illness.

Lithium Ratio and Bipolar Disorder

Authors: Giovanni Poletti, Guido Mazzotti, Luciano Poletti, Carla Gallo

The diagnostic criteria for bipolar disorder (BD), are based on clinical observations and depend on patient evolution. The difficulty to obtain an early and exact diagnosis, render necessary to find laboratory assays capable to support the clinical diagnosis of BD. We propose the lithium ratio (LR) as a biological marker of BD. Contrary to other investigations, we have succeeded to discrimate patients with BD (n = 28, LR = 0.216 ? 0.071) from those with confounding symptomatology (NBP) (n = 32, LR = 0.151 ? 0.035) and from a control nonpsychiatric population (n = 18, LR = 0.138 ? 0.013). In addition, taking as cutoff point LR = 0.175, it is possible to separate the BP and NBP populations with a sensitivity of 75% and a specificity of 81%.

Our method differs from the originally described in two aspects: 1) it introduces a modification to the method of determination of LR and 2) the average of four LR taken weekly over a month is used as the diagnostic value, lowering in this way the effect of interacting factors.

This study is being continued, in order to increase the sample, and in order to be able to evaluate the predictive value of LR for the response to lithium therapy.

Olanzapine Versus Haloperidol in the Treatment of Schizoaffective Bipolar Patients

Authors: Mauricio Tohen, M.D., Dr.P.H., Todd M. Sanger, Ph.D.,
Gary D. Tollefson, M.D., Ph.D., Susan L. McElroy, M.D.

Olanzapine is a new atypical antipsychotic agent that has affinity to both 5-HT2A and D2 receptors but binds more potently to the 5HT2A receptor by a factor of 3:1. In a sub-sample of a large multisite blind parallel study comparing olanzapine against haloperidol, 73 patients with schizoaffective disorder including currently bipolar (N = 28) currently mixed (N = 43) currently depressed (N = 52) and euthymic (N = 48) were assessed with the Brief Psychiatric Rating Scale (BPRS) and the Montgomery-Asberg Depression Rating Scale (MADRS rating scale at baseline and at week 6. Five items of the BPRS scale (euphoria, hyperactivity, agitation, irritability, psychosis) were utilized to assess "manic" symptoms. Patients with schizoaffective bipolar disorder currently manic randomized to olanzapine had a mean change of 6.06 compared to 3.36 for the haloperidol group (p = .251). Patients with schizoaffective bipolar disorder mixed had a mean change of 1.47 for the olanzapine group and 2.82 for the haloperidol group (p = .439). With the MADRS rating scale, patients with schizoaffective bipolar disorder currently depressed had a mean change of 8.57 in the olanzapine group and a worsening change of 6.63 in the haloperidol group (p = 0.002). Olanzapine appears to have mood stabilizing properties in patients with schizoaffective disorder. Further studies need to be completed to replicate these findings.

Double-Blind, Multi-Centre Trial of Carbamazepine Versus Lithium as Prophylaxis for Treatment-Naive Bipolar Patients

Authors: EGTM Hartong, P Moleman, WA Nolen, CAL Hoogduin

Ninety-eight patients with bipolar affective and schizoaffective disorder (DSM-III-R) without prior prophylactic treatment, were treated for two years with carbamazepine (CBZ) or lithiumcarbonate (Li) in a double-blind, multi-centre trial. Dosage was adjusted to obtain serum-levels of 6-10 U/L (1 U/L = 0.1 mmol/L Li or 1 mg/L CBZ). Thus the double-blind was preserved. No additional treatment other than restricted doses of benzodiazepines were allowed.

Patients were included after diagnostic interview by the central investigator (CINV:EH) in the presence of the local investigator (LINV) after obtaining written informed consent. Patients were recruited from 1990. Last patient evaluation was mid 1996, after which investigators were unblinded. Continued double-blind medication was provided for patients who had finished the protocol.

Primary outcome criterion was an episode according to DSM-II-R. If relapse occurred, antidepressants or neuroleptics could be added to the blind medication. After recovery within four weeks, additional medication was stopped within three months. These patients were classified as partial responders. If recovery did not occur within four weeks or a second relapse occurred, patients were classified as nonresponders and evaluation ended. LINV rated global mood monthly in a clinical interview. CINV and LINV rated patients on Bech-Rafaelsen Mania and Melancholia Scales at baseline, every six months, in case of relapse and at the end of the study, complemented with the Comprehensive Psychopathological Rating Scale by CINV and a four-point side effects list by LINV. Laboratory testing was performed at regular intervals.

Of 154 eligible patients 98 (45 males, 53 females) were included. Ninety-five had bipolar disorder, and 3, schizoaffective disorder. Forty-one patients completed two years, 6 were partial responders, 18 nonresponders. First efficacy results will be presented.

Refractory Depression (A New Model for Definition)

Authors: Gupta RK, Burrows G, Thase M

In 1974, Keilholz reviewed the prevalence and treatment of resistant depression and concluded that about 85% of all depressed patients responded to serial trials of antidepressant therapy and electroconvulsive therapy (ECT). In 1995, Thase and Rush reviewed a much larger literature and reached a similar conclusion. Thus, over the past 20 years, the point prevalance of refractory depression apparently has not changed.

Remick has drawn attention to several conflicting definitions and concepts between absolute and relative treatment resistant depression, as well as the combination of dysthymic disorders and a major depression. Guscott and Grof have pointed out that "treatment resistant depression is not a unitary diagnosis but a phenomenon of labeling by treating clinicians." Similarly, Dyck criticised the current approach to treatment resistant depression as a syndrome description that is "confounded by clinicans’ treatment outcome expectations" and "...the lack of any systematic methodologically sound, well controlled studies of a homogeneous population of patients."

The classification of refractory depression in 1996 may be viewed as analogous to the diagnosis of the depressive disorders in the 1960’s and 1970’s. Although we suspect that the Thase and Rush staging classification will prove to be rated reliably, we see it as incomplete if it is not complemented by assessment of medical status, compliance and treatment adequate, as well as subsequent developments clarifying illness pathophysiology.

We prefer using the term "resistant depression" in a treatment-specific context. For example, "patient X has been resistant to adequate trials of paroxetine, moclobemide and imipramine". For clarity’s sake, we suggest that the term refractory depression be reserved for major depressive syndromes that are not responsive to adequate trials of SSRIs, TCAs, at least two other MAOIs, and ECT. If such treatments are administered in a careful, sequential manner, a response rate of as high as 95% can be expected.

A NEW MODEL FOR DEFINITION OF REFRACTORY DEPRESSION. In the interest of conceptual clarity, failure of the illness symptoms to respond to a standard course of ECT should constitute an essential component of this condition. Alternatively one could borrow the concept from general medicine and a depression, if it fails to remit following a course of ECT, may be labeled as "refractory depression in the first degree" and the one that fails to remit intensive inpatient trial of treatment with antidepressant medicines be referred to as "refractory depression in the second degree". The other components of this model are:

Evidence of persistent clinical symptoms at the syndrome level particularly those suggesting physiological disturbance, for example, appetite disturbance and weight changes.

Evidence of a definite deterioration (suboptimal functioning) in mental, physical and social well being as compared to the patient’s baseline status.

A well adjusted premordbid personality

This suggestion has the merit that it fits in with the DSM-IV multi-axial classification system.

Platelet Membrane Phosphoinositide Measured in Bipolar Disorder Patients

Authors: Jair C. Soares, Christine S. Dippold, Alan G. Mallinger

Intracellular signal transduction abnormalities may be present in bipolar disorder patients. The intracellular phosphatidylinositol (PI) pathway has been investigated. Lithium may be effective by inhibiting critical steps in this pathway. We have used a method to determine the relative membrane volume measures of phosphoinositides in platelets (Mallinger et al, 1993), that consists of thin-layer high performance chromatography, followed by scanning laser densitometry. The relative membrane volumes of phosphatidylinositol (PI), phosphahatidylinositol-4-phosphate (PIP), phosphatidylinositol-4,5-bisphosphate (PIP2), besides 5 other individual classes of membrane phospholipids can be determined. There are previous findings from our lab (Brown et al, 1993) suggesting that unmedicated bipolar disorder patients in the manic state have increased PIP2 levels compared to normal controls. To follow up on these findings, and further investigate whether this abnormaility is a state or a trait finding, we have recruited unmedicated bipolar disorder patients, in the manic, depressed, and euthymic states, and a group of euthymic patients on lithium monotherapy. Seven lithium-treated euthymic patients (3 males, 4 females - mean age? S.D. = 37.7? 9.5) were compared to 8 normal controls (4 males, 4 females - mean age? S.D. = 30.8? 8.5). The relative PIP2, PI, and PIP values in the platelet membranes of patients and normal controls were 0.26? 0.08 and 0.34? 0.07, 4.5? 1.2 and 5.4? 1.5, and 0.7? .3 and 0.7? 0.2, respectively. The differences between the euthymic lithium-treated and normal control groups were not statistically significant for any of the phosphoinositides measured, but there was a trend to decreased PIP2 values in patients (t value = 1.96, p = 0.07). These findings are consistent with a normal-functioning PI pathway in the lithium-treated euthymic group, that could result from a normalization of mood, and/or the effects of lithium-treatment. We are currently expanding this sample of euthymic lithium-treated patients, and recruiting a subsample of unmedicated bipolar patients that are being studied before and after lithium treatment. The preliminary findings for the pre- and post-lithium comparison, and the drug-free patient groups will also be presented.

Olanzapine in the Treatment of Schizoaffective Disorder

Authors: Tollefson G, Sanger T, Graffeo K, Kuntz A

Study HGAJ was an international, multicenter, double-blind trial which compared the efficacy and safety of a single dose-range of olanzapine (OLZ) (5-20 mg) to a single dose-range of haloperidol (HAL) (5-20 mg) in the treatment of 1,996 patients with a DSM III-R diagnosis of schizophrenia, schizophreniform disorder and schizoaffective disorder. During the acute phase (6 weeks), a total of 300 patients with schizoaffective disorder were enrolled (177 schizoaffective, bipolar type (OLZ:120, HAL:57) and 123 schizoaffective, depressive type (OLZ:76, HAL:47)). The patient and illness characteristics, and the efficacy data of the patient population with schizoaffective disorder will be presented in this poster.

The OLZ-treated patients with schizoaffective disorder and the subgroup of patients with schizoaffective disorder, bipolar type experienced statistically significantly greater improvement from baseline to endpoint of the acute phase in the primary and secondary efficacy scales: BPRS total, PANSS total, PANSS negative, CGI severity and MADRS total scores when compared with the HAL treatment group (p<.01 for all schizoaffective patients; p<.05 for the schizoaffective disorder, bipolar type patients). In the subgroup of patients with the diagnosis of schizoaffective, depressed type, the mean improvement in the OLZ-treated patients was only marginally statistically significantly greater than that in the HAL-treated patients in every score mentioned above. However, the observed differences between the two treatment groups were consistent with those observed in the total schizoaffective patient population.

A statistically significantly greater percent of OLZ-treated patients continued into the double-blind extension phase thant that of HAL-treated patients (OLZ:43%, HAL:24%;p = .001). The acute therapy gains on the OLZ-treated patients with schizoaffective disorder were sustained through 52 weeks as assessed by the same efficacy scores.

We conclude that OLZ is statistically significantly superior to HAL in the treatment of patients suffering from schizoaffective disorder.

Stressful Life Events and Social Rhythm Disruption in the Onset of Manic and Depressive Bipolar Episodes: A Preliminary Investigation

Authors: Susan Malkoff-Schwartz, Ellen Frank, Barbara Anderson, Joel T. Sherrill, Lori Siegel, Declan Patterson, David J. Kupfer

Background: Although an association between stressful life events and onset of bipolar episodes has often been reported, methodologic limitations in many prior investigations render the nature of this relationship unclear. For example, whether severely stressful life events relate differentially to the onset of manic versus depressive bipolar episodes remains to be elucidated. Additionally, the association between onset of bipolar episodes and types of stressful life events that disrupt social routines and, potentially, sleep, has not yet been investigated.

Methods: Thirty-nine bipolar patients whose index episodes were characterized as primarily manic (N=20) or primarily depressed (N=19) were interviewed with the Bedford College Life Events and Difficulties Schedule to determine the presence of severe events and major difficulties during an 8-week pre-onset period and during an earlier, 8-week episode-free control period. Life events during these time periods were also rated for degree of social rhythm disruption. This rating, developed in our research group, reflects the degree to which a given event potentially and acutely affects the sleep-wake cycle.

Results: Rates of patients experiencing at least one event characterized by social rhythm disruption during the 8-week pre-onset period were significantly greater than either corresponding control period rates in manic subjects or pre-onset rates in depressed subjects. Rates of patients with at least one severely stressful life event or major difficulty during the pre-onset period did not differ significantly from corresponding rates during the control period for either manic or depressed subjects.

Conclusions: Consistent with the hypotheses that social rhythm disruption may precipitate onset of affective episodes, and that sleep reduction may be the final common pathway for onset of mania, we found evidence that life events characterized by disruptions in social routines are associated with the onset of manic episodes in patients with bipolar disorder.

Bipolar Depression: An Underestimated Treatment Challenge

Authors: Stefanie A. Hlastala, Ellen Frank, Alan G. Mallinger, Michael E. Thase, Angela Ritenour, and David J. Kupfer

Objective: While the manic phase of bipolar illness is often more acutely disabling, bipolar depression may have a more persistent, albeit insidious impact. Yet only a few studies have compared the outcomes of bipolar patients experiencing depressive and manic episodes. We sought to determine whether depressive and mixed/cycling episodes were as responsive to standardized pharmacotherapeutic interventions as were manic episodes in bipolar 1 patients.

Method: As part of the Maintenance Therapies in Bipolar Disorder (MH29618-E. Frank, PI) study, forty-two acutely ill bipolar 1 patients who had been randomly assigned to one of two preliminary phase non-pharmacologic treatment stategies (interpersonal and social rhythm therapy (IPSRT) or a standard medication clinic approach) were treated according to a standardized pharmacotherapeutic protocol. Symptom severity was measured weekly with the Hamilton Depression Rating Scale and the Bech-Rafaelsen Mania Scale in order to assess symptomatic remission. Survival analysis with the proportional hazards model was performed on time to remission.

Results: Manic patients were significantly more likely to achieve clinical remission than the depressed patients (100% vs. 59%) and did so significantly more rapidly (median weeks to remission = 17 vs. 40; p=0.006). The difference in proportion remitting and time to remission between the depressed and mixed/cycling groups was not statistically significant. No significant effect for non-pharmacologic treatment assignment was found.

Conclusions: These results point to the need to develop more effective treatments for bipolar depression. They also suggest that psychotherapy may have a limited impact in the acute phase treatment of bipolar episodes.

Group Cognitive Therapy for Bipolar Disorder

Author: A.G. Palmer

The limitations notwithstanding, studies of group therapy for bipolar disorder indicate that the pessimism in the psychoanalytic literature is unwarranted. Effective mood stabilising medications can be enhanced by various psychosocial interventions to reduce adverse psychosocial consequences and be cost effective in terms of hospital admissions. Our early exploratory study (1995) provided information about Group Cognitive Therapy as an adjunct to mood stabilising medication. Cognitive techniques showed some promise in addressing symptoms which interfere with work and social functioning. It was, however, limited by a small sample size and lack of a comparison group.

We have now extended this study to give a larger sample size; that is, five consecutive therapy groups. In addition, patients receiving group cognitive therapy as an addition to treatment they would usually receive were compared with patients receiving ‘treatment as usual’ only. Preliminary findings will be presented. The problems involved in evaluating a treatment package with a severe chronically fluctuating condition such as bipolar disorder will be discussed. We will also present our clinical impressions, addressing issues such as the therapeutic alliance, the use of highly structured approach and which techniques worked within a group format.

Cognitive Therapy for Individuals Suffering form Bipolar Disorder

Author: Professor Jan Scott

The aims of therapy in bipolar disorder (BD) are to alleviate acute symptoms, restore psychosocial functioning and prevent relapse and recurrence (Scott, 1995). Whilst the central plank of treatment for BD sufferers remains pharmacotherapy, the increased emphasis on stress-diathesis models has facilitated a greater acceptance of psychosocial interventions, such as cognitive behaviour therapy (CBT), for individuals with severe mental disorders. This poster describes an MRC funded pilot study on the use of CBT with 30 individuals suffering from bipolar disorder.

The model of CBT we have developed comprises 20 sessions of treatment of gradually reducing frequency. The general goals of our approach are to increase or enhance non-pharmacological coping skills, to help individuals recognise and manage psychosocial stressors, to teach CBT strategies to deal with cognitive and behavioural problems and to develop self-management and relapse prevention techniques (Scott, 1996).

The poster describes the outcome of this treatment programme including measures of attitudes to treatment, self-esteem, and problem-solving skills, as well as more global measures of overall functioning. The approach appears to be acceptable and helpful with about 60% of the individuals referred to the study and a larger scale randomised controlled trial is now planned.

References

Scott J (1995) Psychotherapy for bipolar disorder: An unmet need?.

British Journal of Psychiatry, 167, 581-89.

Scott J (1996) The role of cognitive behaviour therapy in bipolar disorder. Behavioural & Cognitive Psychotherapy, 24, 195-208.

Buproprion Plus SSRI Combination Therapy in Treatment-Resistant Depression

Authors: Ketter, TA, Winsberg ME, DeGolia SG, Dunai M, Tate DL, Strong CM

BACKGROUND: Bipolar depression is often resistant to mood stabilizers, and may fail to respond adequately to the addition of an antidepressant.

METHOD: We studied nine (4 women, 5 men, mean age 47.6) bipolar patients (4 BP1, 3 BP2, 2 BPNOS), including two rapid cyclers. Mean illness and current depressive episode durations were 28.2 years and 40.9 weeks, respectively. Six patients were on lithium (mean dose 854 mg/d, level 0.65), five on valproate (mean dose 750 mg/day, level 63.0) (three of whom were also on lithium), and one on carbamazepine (800 mg/d, level 8.4). Patients continued to have depressive symptoms despite a mean of 51.8 weeks therapy with SSRIs or bupropion. In the seven patients on SSRIs (5 on paroxetine, mean dose 30 mg/day; 2 on sertraline, mean dose 50 mg/day), bupropion (mean dose 193 mg/day) was openly added (mean duration 9.3 weeks). In the two patients on bupropion (mean dose 300 mg/day), paroxetine, 10 mg/day, or venlafaxine, 150 mg/day, were added (mean duration 17.0 weeks).

RESULTS: Combination therapy was generally well-tolerated. Four patients had marked improvement, five little change, and two (both of the rapid cyclers), minimal deterioration. Mean Hamilton Depression Ratings tended to decrease from 16.4 to 11.8 (p = .10) in the entire group, decreased from 16.3 to 4.5 (p < .0001) in responders, and went from 16.6 to 17.6 (p = NS) in nonresponders.

CONCLUSION: Combination of bupropion with SSRIs and venlafaxine may be well tolerated and help some patients with treatment-resistant bipolar depression. This uncontrolled open pilot study must be viewed with caution, and randomized double-blind placebo controlled studies appear warranted to further explore the safety and efficacy of this treatment strategy.

Supported by the Stanley Foundation Research Awards Program.

Follow-up of Lithium-Refractory Patients Treated wit hLithium-Carbamazepine Combination

Authors: A. Bocchetta, C. Chillotti, R. Ardau, G. Severino, M. Del Zompo

Introduction: We have recently reported the effects of carbamazepine augmentation in a group of 22 patients affected by bipolar disorder who had failed on lithium alone (Bocchetta et al., in press). Patients had been followed up prospectively for at least two years of combined treatment. Fifteen patients had no further recurrences during the observation period. One additional patient featured reemergence of affective episodes after several years of satisfactory benefit. Such a relatively low rate of delayed tolerance compared with the 50% rate found by Post et al. (1990) in a group of lithium-refractory patients treated with carbamazepine (alone or in combination with other psychotropic drugs) prompted us to extend the follow-up period of the same group of patients.

Objectives: The aim of this study is to evaluate whether the patients who had responded to carbamazepine augmentation during the initial years showed or not a pattern of continued improvement.

Methods: An illness index of morbidity (duration x severity) was calculated for each patient for the additional follow-up period in accordance with the method of Post et al (1990).

Results: Out of 15 patients who had no recurrence during the previous observation period, 3 manifested a new (moderate to severe) manic episode during additional one-year follow-up. They must be added to the one who had already manifested delayed tolerance during the initial study. Overall, 4/16 = 25% showed loss of prophylaxis after initial improvement for at least two years.

Conclusions: The increasing rate of failures with extension of the observation period might eventually lead to results similar to Post et al. (1990), who first reported escape from carbamazepine prophylaxis in a relevant proportion of initial responders.

References: R.M. Post, G.S. Leverich, A.S. Rosoff, L.L. Altshuler, J. Clin. Psychopharmacol. 10, 318-27, 1990. A. Bocchetta, C. Chillotti, G. Severino, R. Ardau, M. Del Zompo, J. Clin. Psychopharmacol, in press.

Introduction | Poster Sessions | Audio Recordings
Bipolar Conferences Home

© The Department of Psychiatry, University of Pittsburgh School of Medicine