FOURTH INTERNATIONAL CONFERENCE 
ON BIPOLAR DISORDER

Poster Abstracts (26-50)

26. Non-Manic Depressive Bipolar Disorders. The Need of Validating Cycloid Psychoses
M.G. Cetkovich-Bakmas; G.V. Vázquez, Center of Biological Psychiatry, Buenos Aires, Argentina

The concept of “Atypical Psychosis” arises from Wernicke´s observation of the impossibility of classifying some psychotic states within Krapelin´s dichotomic model.
  Kleist termed them “Marginal Psychosis” and Leonhard developed the concept of “Cycloid Psychosis” due to their resemblance to true cyclic psychosis, i.e: Manic Depressive Illness. Cycloid Psychosis has some clearly differentiated traits:
  a) Phasic Course with full recovery from each episode and, mostly, good prognosis in long-term outcome. b) Bipolar Structure, with polymorphous but characteristic syndromes; c) No residual syndromes, with some reduction in mental strength and ability of coping with stress; d) High trend to show psychotic features that push misdiagnosis with schizophrenic disorders.
  Cycloid psychosis could affect different aspects of psychic life 1) the one that displays paranoid anxiety alternating with mystic delusional ecstatic states, is called Anxiety-Happiness psychosis. 2) Confusional Psychosis affects cognitive processes alternating delusional and pressured incoherent speech states with perplexity ones. 3) Motility Psychosis: High excitability of expressive motions alternating with strong inhibition without catatonic features (rigidity, oposicionismus, negativismus and tremor).
  In the acute phase of cycloid psychoses, patients fulfill DSMIV criteria for Schizophreniphrom Disorder, Bipolar Disorder or Bipolar Mixed States.
  Epidemiological, genetic and biological data supports the idea of Cycloid Psychosis as a separate group of entities from Shchizophrenia and Manic Depressive Illness.
  The goal of this paper is to present a project on the validation of the concept of Cycloid Psychosis looking forward to improve phenotypes definition for clinical research and to stress the possibility that some of the called mixed states of Bipolar Disorders could be separated as Cycloid States.
Key Words:  cycloid psychosis, nosography, differential diagnosis.

27. Alteratated cAMP-dependent Protein Kinase Subunit Immunolabeling in Postmortem Brain from Patients with Bipolar Affective Disorder
A. Chang, P. P. Li, S. Kish and J. J. Warsh, Sections of Biochemical Psychiatry (AC. PL and JJW) and Human Brain Laboratory (SK), Centre for Addiction and Mental Health and Departments of Psychiatry, Pharmacology, and Institute of Medical Science, University of Toronto

Introduction: Increasing evidence implicates hyperfunctional cAMP-mediated signaling in the pathophysiology of bipolar disorder (BD). Previous findings of reduced [3H]-cAMP binding and elevated cAMP-dependent protein kinase (PKA) activity in discrete postmortem brain regions from BD patients suggest this downstream target of cAMP is also affected in this disorder. SAs sustained elevations in intracellular cAMP levels can induce adaptive cause cell-type specific changes in the relative abundance of PKA regulatory (R) and catalytic (C) subunits, and the abundance of PKA subunits that affects PKA activity. Thus, we sought , the main objective of this study was to determine if the levels of the C and specific R subunits of PKA subunits are also altered in BD brain.  Methods: Immunolabeling of PKA RI, RI, RII, RII, and C subunits was measured in cytosolic and particulate fractions of temporal (n=9) and parietal (n=7) cortices from BD patients and non-neurological nonpsychiatric controls matched on sex, age and postmortem interval. After separation by SDS-PAGE, PKA subunits were determined by Western blotting using subunit-specific antisera and enhanced chemiluminescent detection. Statistical analysis of differences in immunoreactivity levels between BD subjects and matched non-psychiatric healthy controls were performed by paired-sample two-tailed t-tests. Results: PKA RI, RII, RII and C subunit levels were significantly higher (86%, p = 0.015; 42%, p=0.009; 58%, p=0.014 and 30%, 0.034, respectively, [paired t-tests]) in the cytosolic fractions from BD temporal cortex compared with matched controls. A significant inverse correlation between PKA RI and RII levels and brain pH suggests the latter may be a confounding factors attributing contributing to the observed higher differences RI and RII in the levels of these particular subunits to changes in cAMP signaling. RII and C subunit levels did not correlate with pH, however, and comparison with a separate set of controls matched more closely on brain pH confirmed the higher immunoreactive levels of these subunits in BD temporal cortex. In contrast to temporal cortex, no significant differences were found in detected in the immunoreactive levels of any of the PKA subunit subunits assayed in either cytosolic or membrane fractions of parietal cortex in BD compared with controls. Higher C subunit levels in BD temporal cortex are consonant with the increased PKA activity previously reported in these same BD brains. The elevated cytosolic RII levels in BD temporal cortex may reflect a compensatory response to the putative increased cAMP signaling in this disorder, in an attempt to regulate the catalytic activity, as RII exhibits a higher sensitivity for cAMP induction compared with the other PKA R subunits, at least in some cell models.
  Supported by a NAMI Research Institute Stanley Foundation Award (JJW) and MRC Doctoral Research Award (AC)
Key Words:  signal transduction, cAMP signaling, postmortem brain.

28. Clinical Response to Quetiapine Add-on for Treatment Refractory Bipolar Disorder
K. Chisholm, T. Suppes, University of Texas Southwestern Medical Center, Dallas, Texas, U.S.A.

Introduction: A retrospective evaluation of the effectiveness and tolerability of quetiapine as an adjunctive treatment in 40 patients with bipolar disorder.  Methods: In this case series, response to quetiapine was assessed over a minimum of 2 weeks and up to a 6 month period using symptom ratings for depression (Inventory of Depressive Symptoms; Rush et al 1985); mania (Young Mania Rating Scale; Young et al 1978); and the Clinical Global Impressions Scale modified for Bipolar Illness (CGI-BP; Spearing et al 1997). Patients’ other psychotropic medications were required to remain stable during the study period, except for decreases due to side effects.  Results: The charts of 40 consecutive patients with bipolar disorder (1997-2000) who had quetiapine added to their medication regimens for clinical symptoms were reviewed. Patients received monthly prospective ratings for symptoms of bipolar disorder. These ratings and monthly evaluation of side effects were collated retrospectively. The majority of patients started on quetiapine had Bipolar I disorder. Quetiapine was initiated for a broad spectrum of symptoms including hypomania, mood lability, psychotic symptoms, and decreased sleep. The quetiapine dosage ranged from 25-600mg a day. Patients were required to receive a minimum of two weeks exposure to quetiapine to be included in the study. The most common side effect was sedation.  Conclusions: Results from this retrospective chart review on quetiapine add-on treatment will be presented. Controlled studies of quetiapine’s acute and long-term efficacy and side effects in bipolar disorder appear warranted.
Key Words:  bipolar disorder, quetiapine, treatment.

29. Psychoeducation and Prevention of Relapse in Bipolar Disorders: Preliminary Results
F.Colom, A. Martínez-Arán, M.Reinares, A.Benabarre, B.Corbella, E.Vieta, Bipolar Disorders Program, Department of Psychiatry, Hospital Clinic Barcelona, Barcelona, Spain

Introduction: The role of psychotherapy in bipolar disorder is still a controversial subject in constant progress and nowadays little is known about efficacy of psychotherapy in treating bipolar illness. Psychoanalysis, psychoeducation, family therapy, cognitive-behavioral therapy and interpersonal therapy have been used in the treatment of bipolar patients. To date, none have established efficacy in controlled clinical trials regarding aspects such as hospitalizations, recurrences or suicidal behavior, as medication alone does. Nonetheless, the psychoeducational approach combined with several cognitive-behavioral techniques, either in group or individually, seems to be the most promising, focusing on information, treatment compliance, and illness management skills. Methods: 58 consecutive euthymic (YMRS<6, HDRS<8) bipolar patients were randomly assigned either to control or experimental group. Control group patients received standard pharmacological treatment. Experimental group patients received standard pharmacological treatment plus psychoeducation. The psychoeducational program was composed by 21 sessions of 90 minutes duration of group therapy (between 8 and 12 patients per group) in which special emphasis was made on treatment compliance, illness awareness and early identification of new episodes. Results on outcome were compared after a one-year follow-up. Results: At the one-year follow-up, psychoeducated patients showed a significantly lower number of total episodes than control group patients (1.19 vs. 3.03, p<0.002), a lower number of depressive (0.28 vs. 1.27, p<0.001) and manic (0.14 vs. 0.65, p<0.003) episodes and less hospitalizations (0.09 vs. 0.72, p<0.004). Conclusions: Although these are just preliminary results from a study with a bigger sample, the benefits of psychoeducation on the course of bipolar disorders in terms of reduction of recurrences and hospitalizations are clear. Psychoeducation should be systematically adjuncted to the pharmacological treatment of bipolar disorders.
Key Words:  psychoeducation, prevention, psychotherapy.

30. Early Sexual Abuse and Clinical Features in Bipolar Disorder
R.G. Cooke, D.K. Whitney, E. Munco, S.V. Parikh, C. Miller, L. Lee, J.L. Kennedy, Center for Addiction and Mental Health, Toronto, Ontario, Canada

Early sexual abuse may be a risk factor for adult psychiatric disorders (Wexler, 1997), or may be linked to specific clinical features of the disorders (e.g. Levitan et al., 1998). Leverich et al. (2000) found that early sexual abuse was associated with a number of indicators of severity in patients with bipolar disorder (BD). Hypothesis: That our BD patients with a history of early sexual abuse would show clinical features similar to those reported by Leverich and colleagues, including more rapid-cycling and earlier age of onset, as well as more dysphoric mania, and more treatment refractoriness, than BD subjects with no abuse histories. Methods: Participants (N=250, 63% female, mean age 19.92 + SD = 7.6) were probands in a genetic study of BD underway at the University of Toronto. All subjects met DSM-IV criteria for bipolar I or II. The diagnosis was confirmed, and clinical features such as type I vs II BD, age of onset of BD, and the presence or absence of rapid-cycling were determined, by SCID. Specific details of early sexual abuse were systematically recorded and grouped according to exposure in childhood (< 12 years of age) or adolescence (12 to 17 years of age). Results: Fourteen subjects (5.6%) reported abuse in childhood; an additional 20 subjects reported abuse beginning in adolescence (total 13.6%). Compared to subjects with no abuse history, both those with childhood abuse (p=0.046), and those with childhood and/or adolescent abuse (p=0.036) were more likely to exhibit rapid cycling (chi-square analysis, one-way test). No relationship was observed between sexual abuse history, and current age, age of onset of BD, or type I vs II BD. Further analyses are underway. Conclusions: A history of sexual abuse in childhood and/or adolescence was associated with rapid-cycling in subjects with BD.
Key Words:  bipolar disorder, sexual abuse, trauma.

31. Carbamazepine but Not Lithium Salts Prevents the Development of Behavioural Supersensitivity to Dopamine Agonists Induced by Chronic Imipramine
P.S. D'Aquila, A.T. Peana, O. Tanda, M. Collu, P. Devoto, G.Serra, Università di Sassari, Sassari, Italy

Introduction: Antidepressant-induced supersensitivity of the mesolimbic dopamine system, as revealed by an increased sensitivity to the locomotor effect of dopamine agonists, has been suggested on the one hand to play a role in the mechanism of action of antidepressant drugs (1), and, on the other hand, to be one of the neural substrates underlying antidepressant-related mania, namely rapid cycling and antidepressant-induced mood switches (2). Lithium salts and carbamazepine are effective antimanic treatments, although their therapeutic spectra do not completely overlap. Indeed, carbamazepine appears to be effective in some patients resistant to lithium, and several clinical reports suggest that it might be effective in rapid cyclers, which are particularly resistant to lithium (3). In the present paper we report data from two studies on the effect of lithium chloride and carbamazepine on the development of the behavioural supersensitivity to the locomotor effect of the dopamine D2 like dopamine receptor agonist quinpirole. Methods: Experiments were performed on male Sprague-Dawley rats. The subjects of both lithium and carbamazepine studies have been allocated into four groups according to a design involving two between groups factors: imipramine (vehicle and imipramine) and diet (controls and either lithium chloride or carbamazepine diet). Chronic treatments have been performed for four (lithium study) or three weeks (carbamazepine study). Imipramine was administered intraperitoneally at the dose of 20 mg/kg daily. Lithium chloride and carbamazepine were administered in the diet (pellets with 60 mEq/kg LiCl or 5 g/kg carbamazepine). Twenty four hours after the end of chronic treatments the animals were individually placed into the motility apparatus. After one hour of habituation to the motility cages they were challenged with a subcutaneous injection of quinpirole (0.15 mg/kg) and their activity was further recorded for 45 minutes.  Results: The results show that a) imipramine, as expected, potentiated the locomotor response to quinpirole; b) lithium failed to influence quinpirole-induced motility both in vehicle- and in imipramine-treated rats; c) carbamazepine prevented the development of imipramine-induced supersensitivity.  Discussion: According to the view that antidepressant-induced potentiation of dopaminergic transmission might constitute one of the neurobiological mechanisms underlying antidepressant-related mania, the present results are consistent with the observation that carbamazepine, as opposite to lithium, has been shown to be effective in the treatment of rapid cyclers in uncontrolled studies and small controlled studies, although it must be pointed out that there are no controlled trials showing that carbamazepine is more effective than lithium in this category of patients (3). Moreover, it might be predicted that the addition of carbamazepine to antidepressants in the treatment of bipolar depression should prevent both antidepressant-induced switch from depression to mania and rapid cycling in bipolar patients treated with antidepressant medications. This prediction is consistent with clinical data showing that addition of antidepressant drugs to mood stabilisers in bipolar mania does not result in an increase of mood switches (4), although it should be stressed that no systematic comparison between lithium and carbamazepine has been performed in this study.
Key Words:  dopamine, lithium, carbamazepine.

References
1. Serra G, Collu M, D'Aquila PS, De Montis MG, Gessa GL (1990) Brain Res 527: 234-243.
2. D'Aquila PS, Collu M, Gessa GL, Serra G (2000) Eur J Pharmacol 340: 121-132.
3. Soares JC (2000) Int Clin Psychopharmacol 15: 183-196.
4. Boerlin HL, Gitlin MJ, Zoellner LA, Hammen CL (1998) J Clin Psychiatry 59: 374-379.

32. Episodic Memory Dysfunction in Bipolar Disorder
T.D. Deckersbach, Harvard Medical School, Boston, Massachusetts, U.S.A; N. Reilly-Harrington, Harvard Medical School, Boston, Massachusetts, U.S.A; G. Sachs, Harvard Medical School, Boston, Massachusetts, U.S.A.

Introduction: There is growing evidence showing that bipolar disorder patients do not only exhibit cognitive impairments during mood episodes but also when euthymic. Studies investigating cognitive functioning in euthymic bipolar patients have reported impairments in episodic long-term memory. The pattern of episodic memory impairment indicates that euthymic bipolar patients have difficulties learning new information. We hypothesized that this reflects difficulties organizing information appropriately during encoding (the process of establishing a new memory representation) which makes it difficult to retrieve episodic memories later on.  Method: Study participants were 20 euthymic DSM-IV bipolar I disorder patients and 30 normal control participants, matched for age, gender and education. Bipolar and control participants completed the California Verbal Learning Test (CVLT). The CVLT assesses memory for a list of 16 shopping items presented in five study trials, with free recall after each trial as well as short- and long-delayed free recall. The CVLT also provides a measure of semantic clustering demonstrating how subjects organize the word list during learning (encoding).  Results: Univariate ANOVAs indicated that bipolar patients recalled fewer words over the first five successive learning trials (F=10.54, df=1,48, p=.002), at short-delayed free recall (F=6.33, df=1,48, p=.015) and long-delayed free recall (F=6.65, df=1,48, p=.013). Patients also demonstrated less organization of the word list than controls during the five study trials (“semantic clustering”: F=8.88, df=1,48, p=.005). However, group differences in learning (trial 1-5: F=3.37, df=1,47, p=.07) and long-delayed free recall (F=1.76, df=1,47, p=.19) did not remain significant when group differences in semantic clustering were partialled out. This indicates that bipolar patients’ learning and long-delayed recall difficulties were mediated by impairments in verbal organization during learning. Our pattern of results is consistent with impairments reported for patients with frontal lobe dysfunction.
Key Words:  cognitive dysfunction, neuropsychological impairment, episodic memory.

33. Quetiapine as Adjunctive Treatment for Adolescent Mania
M.P. DelBello, H.L. Rosenberg, A.M. Hudepohl, S.M. Strakowski

Introduction: Quetiapine is an atypical antipsychotic with a unique receptor-binding profile. It is effective and well tolerated in adults and adolescents with psychotic disorders. Several case reports suggest that it may be effective as an adjunctive treatment for patients with bipolar disorder. No controlled studies have evaluated efficacy, safety and tolerability of quetiapine in these patients. This double-blind, placebo controlled study examines the use of quetiapine as an adjunct to divalproex for acute mania in adolescents with bipolar disorder. An analysis of our blinded, pooled safety data is presented. Method: The study design called for a total enrollment of 30 manic bipolar adolescents (ages 12-18); to date, 20 patients have completed the study. Adolescents were recruited from inpatient units and evaluated using the Washington University K-SADS. Patients were randomized to quetiapine or placebo for 6 weeks. All received an initial divalproex dose of 20mg/kg (mean level 93ug/ml), adjusted to achieve a therapeutic blood level (80-120ug/ml), and an initial dose of 25mg bid of quetiapine, titrated over one week, to a maximum of 150mg tid. Safety and tolerability were assessed weekly using the Simpson Angus, Barnes Akathisia and Abnormal Involuntary Movement Scales and laboratory measures. Weekly efficacy measures included the YMRS, CDRS, PANSS, and CGI/CGI-C. Results: Quetiapine dosages ranged from 250-450mg/day. According to the blinded, pooled data, the most common adverse events were headache (55%), nausea and vomiting (55%), and sedation (45%). Most adverse events were mild; no serious events occurred. No EPS, QTc prolongation,orthostatic hypotension, or change in thyroid function tests were noted. Mean change in prolactin was -1.9 + 8.7ng/ml. Discussion: Preliminary data suggest that quetiapine is safe and well tolerated as adjunctive therapy to divalproex in manic adolescents with bipolar disorder. Unblinded efficacy, safety and tolerability data for the entire sample of 30 will be presented.
Key Words:  bipolar disorder, adolescent, quetiapine.

34. A Case Control Study in the Use of Telemedicine for Treatment Adherence in Remote and Rural Bipolar Patients’
R. D’Souza, H. Hustig, The University of Sydney, Broken Hill, Australia

Aim: To study retrospectively the outcomes of remote and rural patients with a bipolar disorder admitted to a tertiary psychiatric centre. Patients who received continuing care and discharge plans through telemedicine were compared with the outcomes of a matched sample of patients from the unit ,who were discharged using conventional discharge summaries to their primary care physician. Method: 20 rural inpatients who had received Telemedicine for discharge plans, continuing care and follow up over a 12 month period (TMG) and 21 rural inpatients who had discharge summaries sent to their primary care physician with ongoing care from the primary care physician (CG) were selected after matching for age, sex and axis I diagnosis.  A questionnaire which studied 1. Satisfaction with the medication prescribed 2. Adverse affects of medication 3. Compliance with treatment. 4. Satisfaction with their case managers 5. satisfaction with psychiatric management. Case records were studied. Case managers were consulted and rehospitalisation records reviewed.  Results: Response rate higher with TM group. (TMG 90% CG 70%) Significantly higher scores with the TMG in areas of satisfaction with medication and case managers. TMG had higher compliance with treatment and significantly lower side effects with medication prescribed. TMG had significantly lower rehospitalisation than CG. Thus better outcomes and quality of life. Both groups were satisfied with their diagnosis and treatment at the tertiary centre.  Conclusion: Telemedicine for remote and rural patients with bipolar disorder can enhance outcomes of treatment. Telemedicine established and maintained therapeutic alliance with the rural case managers. Over all there were improved longitudinal outcomes, thus resulting in improved quality of life and a significant positive impact on the health economics for the service and the bipolar patient.
Key Words:  telemedicine, treatment adherence, rural.

35. An Open Label Study of Resperidone and Serteraline in the Treatment Of Rural Patients with Bipolar 1 Depression
R. D’Souza, C. Kingston, M. Scurrah, J. Williams, The University of Sydney, Broken Hill, Australia

Objective: To study the efficacy, effectiveness and safety of the combination use of a novel antipsychotic Respiradone and an SSRI Serteraline in patients with a Diagnosis of Bipolar disorder presenting with depression.  Method: 21 patients with a past diagnosis of Bipolar 1 illness, who had presented to a rural psychiatric service over a period of 16 weeks with symptoms of depression, were treated both as inpatients and out patients after assessment by the psychiatrist. These patients were reviewed and followed for a 24 weeks Patients seen at baseline and at treatment weeks 1,2,3,4, 6 and 8 then at 12, 16, 20 and 24.
Inclusion Criteria:
• 16 to 65 years
• Diagnosis confirmed with SCID
• Score of 20 + on MADRS
• Past history of Mania
Evaluation Assessment:
• Patient disposition including discontinuation analysis.
• Model doses.
• Efficacy evaluation using MADRS
• Abnormal Movements Simpson Argus
• Switching as evidenced on YMRS
• Consumer satisfaction.
• Weight monitoring
• Rehospitalisation
Results: Efficacy with reduction in MADRS score starting as early as at visit 2 Minimal Switching with I patient switching as evidenced by increase in YMRS above 15 Side effects of Akathisia was experienced by 5 patients who responded to lowered dose of antipsychotic + Diazepam. Minimal weight increase. Acceptable rehospitalisation- 2 patients readmitted over 24 weeks.  Conclusion: Treatment of Bipolar depression is divided. Some use a mood stabilizer while others add an antidepressant. The issue of switching has been discussed. Atypicals are currently not approved for treatment of Bipolar illness in Australia. In areas where monitoring of serum lithium levels and accessibility of medical resources are limited there is a use for atypicals. This study suggests that Resperidone can be used in combination with Serteraline in this group of patients with good results and safety.
Key Words:  bipolar depression, atypical, SSRI.

36. The Maudsley Bipolar Disorder Project: The Effect of Age of Onset on the Neuropsychological Profile of Biplolar I Disorder
S. Donaldson, L.H. Goldstein, V. Raymont, S. Frangou, Section of Neurobiology of Psychosis, Institute of Psychiatry, London, United Kingdom

Introduction: Impaired cognition has been reported in patients with Bipolar 1 affective disorder. Our aim was to explore the effect of age of onset on patients’ cognition in a representative treatment sample.  Methods: From the casenotes of 425 patients, 63 fulfilled DSM-IV criteria for bipolar I disorder (41.3% male), confirmed by personal interviews using the SCID. Forty-one patients (20 males) agreed to participate in a detailed neuropsychological assessment that included the WAIS-R, WMS-III, NART and the WCST. There was no significant difference in the clinical and demographic data between participating and non-participating patients. Results: Patients’ mean age was 43.1 years and mean age at onset was 26.53 years (sd 10.64). They were on: lithium (25%), antipsychotics (25.0%), sodium valproate (13.5%), carbamazapine (7.7%), benzodiazapines (7.69%), tricyclic antidepressants (7.7%) and SSRIs (11.5%). Patients whose age of onset fell one standard deviation below or above the median where respectively designated as early (< 21) and late onset (> 26) cases, resulting in 7 patients being excluded from the analysis. The early onset-group comprised 16 subjects (11 males), with an average 13.6 (sd 2.99) years of education and 22.83 years (sd 11.97) of illness duration. The late-onset group comprised 18 subjects (5 males) with an average of 12.8 (sd 2.54) years of education and 11.25 years (sd 7.83) of illness duration. WAIS-R and NART estimated IQ was 102.56 (sd 17.05) and 106.06 (sd 13.9) respectively. The late onset group scores were 98.19 (sd 16.29) and 105.25 (sd 12.1) respectively. In the early onset group, memory as assessed by the WMS-III was in the low-average range, but importantly, was significantly lower than that predicted by the intelligence scores. The late-onset group scored in the average range. Conversely, on the WCST the early-onset group completed significantly more categories and made fewer errors.
Key Words:  age of onset, neurocognition.

37. Prospective Longitudinal Follow-Up Study of the Offspring of Lithium Responsive and Lithium Non-Responsive Bipolar Parents
A. Duffy, M. Alda, S. Kutcher, C. Robertson, P. Cavazzoni, E. Grof, and P. Grof Department of Psychiatry, Dalhusie University, Halifax, Canada

Background: The descriptions of clinical course among bipolar youth vary significantly and differ markedly from the findings described in classical studies of manic-depressive adults. This difference may reflect heterogeneity of bipolar disorders. Response to lithium monotherapy identifies a homogeneous genetic subgroup of bipolar adults with a distinctive clinical course. In this poster, we present a prospective study of the offspring of lithium responsive bipolar parents and contrast them to a more heterogeneous high-risk offspring group.  Methods: High-risk offspring of two groups of bipolar parents were studied. One group of parents was identified by a response to prophylactic lithium treatment, the other by non-response. One parent met RDC and DSM-IV criteria for bipolar I disorder and the other parent had no lifetime psychiatric illness determined on the basis of SADS-L interviews. The response of the parent to long-term lithium treatment was determined according to strict research criteria. While blind to family affiliation and lithium response, interviewers assessed all eligible offspring aged 8 to 25 years and one or both parents according to KSADS-PL and reviewed them in diagnostic consensus meetings. Offspring were then re-interviewed over a six-year period.  Results: Offspring of lithium responders manifested typical mood disorders with good to excellent premorbid functioning and an episodic course. Comorbid conditions tended to resolve prior to the mood disorder. In contrast, offspring of lithium non-responders manifested a broader range of psychopathology, and those with mood disorders tended to experience a chronic course. In this group, comorbid illnesses occurred concurrently with the mood disorder. Clinical course among affected offspring was predicted by the disease course of the parent.  Conclusions: The pattern of clinical course, remitting or non-remitting, appears to be inherited. Among offspring of lithium responsive parents an early onset subgroup with a typical episodic clinical course can be identified. The relevance of the findings is discussed with emphasis on the treatment response of affected offspring.
Key Words:  high risk, early onset bipolar disorder, family study.

38. New Information Confirming the Importance of Dosing and Rash with Lamotrigine
N.L. Earl, J. Ascher and J. Messenheimer, GlaxoSmithKline Research and Development, Research Triangle Park, NC, USA

Objective: Due to the large increase in the number of patients enrolled in lamotrigine (LTG) clinical trials over the past three years, a re-examination of the LTG clinical trial database was warranted to better understand the relative risk of serious rash with this new agent for the treatment of bipolar disorder.  Methods: The incidence of serious rash (defined as any skin reaction associated with patient hospitalization and LTG discontinuation, or any case reported as possible Stevens-Johnson syndrome [SJS] or toxic epidermal necrolysis [TEN]) was examined in all controlled clinical trials conducted in adults. The effect of dosing regimen was also examined.  Results: From a datebase including 10,611 adult patients (2,645 of them bipolar patients), 28 (0.26% or 2.6 per 1000 patients) had rash classified as serious and 11 (0.10% or 1/1000) had rash classified as SJS (none as TEN). Of the 5,798 adult patients (2,469 of them bipolar patients) for whom current dosing guidelines were utilized, the corresponding incidence rates for serious rash and SJS were 0.12% (1.2/1000) and 0.05% (0.5/1000), respectively.  Conclusion: These controlled data demonstrate a clear effect of dosing on the incidence of serious rash (including SJS) with LTG. The risk of these serious dermatologic events can be significantly reduced by adherence to dosing instructions (including starting dose and rate of escalation) contained in the package insert.
Key Words:  lamotrigine, skin rash, Stevens-Johnson Syndrome.

39. Antidepressant Exposure in Children Diagnosed with Bipolar Disorder
R.S. El-Mallakh, M.D.,* D. Cicero, J. Holman, M.D., J. Robertson, M.D., Mood Disorders Research Program, Department of Psychiatry and Behavioral Sciences, University of Louisville School of Medicine, Louisville, Kentucky, U.S.A.

Bipolar disorder is being increasingly diagnosed in children and adolescents. The reasons for this are multiple and complex and include such things as increased recognition. However, the question of whether physician practices might be influencing the course of the illness needs to be addressed. Specifically, given that antidepressants may precipitate mania, and given the increased use of antidepressants in youths, it is reasonable to ponder if antidepressant administration might be inducing earlier manic episodes. All consecutive admissions with a diagnosis of bipolar disorder were reviewed. Information regarding previous exposure to antidepressants and stimulants was collected. The mean age of diagnosis of bipolar disorder in our cohort was 12 +-SD 3.47 years. Children who received prior antidepressants and/or stimulant treatments had an earlier diagnosis (10.7+-3.05 years) than children never exposed to these medications (12.7+-4.3 years; one tailed t=-1.33, df=22, p=0.099, power=0.93). Stimulants appeared to be tolerated for a longer duration than antidepressants before the diagnosis of manic-depression (55.5 +-20.42 vs. 6.7+-8.22, t=6.6, DF=12, p=0.0001). Concomitant mood stabilizers and antidepressant treatment was better tolerated than antidepressant treatment alone (81.8% vs. 16.7%, z=3.58, p<0.01). The retrospective nature of the design, the lack of use of structured diagnostic interview, and the small sample size from which sufficient data could be analyzed are the significant shortcomings in this study. Children exposed to antidepressants appear to be diagnosed with bipolar disorder earlier than children never exposed to this class of medications. While far from conclusive, these data are consistent with the hypothesis that antidepressant treatment may precipitate a manic episode earlier than might occur spontaneously.
Key Words:  manic-depression, children, antidepressants, antidepressant-induced mania.

40. Intracerebroventricular Ouabain Models Mania in Rats
R.S. El-Mallakh, M.O. Huff, X.P. Li, S. Decker, R.S. Levy, Mood Disorders Research Program, Department of Psychiatry and Behavioral Sciences, University of Louisville School of Medicine, Louisville, Kentucky, U.S.A.

Background: Human bipolar illness is characterized by mood state and diagnosis-associated abnormalities of cellular cation concentrations and ion transport. These include reduced sodium pump activity and expression and increased intracellular sodium. If these alterations are related to the pathophysiology of the disease then one would expect that modeling them in vivo would produce lithium-preventable behavioral abnormalities.  Methods: Ouabain, a potent inhibitor of the sodium pump, was administered intracerebroventricularly to male rats previously fed food containing lithium or plain rodent chow. Locomotion was quantified in an open field. Results: Ouabain increased locomotion 305 % over baseline. Lithium pretreatment totally prevented the ouabain-induced hyperlocomotor response (121.7 % vs. 115.7 % for controls, ns). Conclusion: Inhibition of central nervous system sodium pump with ouabain produces a plausible animal model of mania. This model may be useful for preclinical screening of potential mood stabilizers.
Key Words:  manic-depression, animal model, lithium.

41. Baseline Body Mass Index as a Correlate of Outcome in Patients with Bipolar I Disorder
A. Fagiolini, E. Frank, P.R. Houck, and D.J. Kupfer, University of Pittsburgh School of Medicine, Western Psychiatric Institute and Clinic, Pittsburgh, Pennsylvania, USA

Obesity is a major public health concern in the United States and its prevalence is increasing. This report from the Pittsburgh Study of Maintenance Therapies in Bipolar Disorder (MTBD) examines the relationship of body mass index at entry into the study and the likelihood of remitting from an acute affective episode and of having a recurrence once the acute episode has remitted. Baseline BMI (body mass index) was calculated on 175 MTBD subjects. The subjects were then classified as 12 (7%) underweight, 54 (31%) normal weight, 47 (27%) over weight and 62 (35%) obese. The underweight subjects were excluded from this analysis because their sample size was inadequate.
  Sixty-nine % of the normal weight, 77% of the over weight, and 71% of the obese subjects completed acute phase and started maintenance (preventative) treatment. There was no difference among the 3 major groups of normal weight, overweight and obese subjects in the rate of completing the acute phase of the study and starting the maintenance (preventative) treatment.
  Examining the preventative phase, there was a higher rate of recurrence among the obese group at 52%, compared to the rate of recurrence of the normal weight subjects of 38% and to the rate of recurrence of the over weight patients of 28%. A Kaplan-Meier survival analyses showed a significant higher risk of recurrence among the subjects who were obese at baseline (Log rank=6.45, df=2, p<.04; see figure). We conclude that obesity is positively related to the risk of developing a recurrence of a major affective episode in patients with bipolar 1 disorder.

Key Words:   bipolar, weight, recurrence.

42. Combination Lithium Carbonate and Divalproex Sodium Treatment of Pediatric Bipolar Disorder
R.L. Findling, B.L. Gracious, N.K. McNamara, J.R. Calabrese, Case Western Reserve University/University Hospitals of Cleveland, Cleveland, Ohio, U.S.A.

Introduction: Monotherapy with either lithium carbonate (Li+) or divalproex sodium (Depakote®, VPA) is effective in juvenile bipolarity. However, a significant proportion of patients may not respond to treatment with one medicine. However, if a youth becomes clinically stable on VPA and Li+ combination therapy, it is unclear whether both treatments are needed for maintenance treatment. The purpose of this ongoing study is twofold. The first is to examine the effectiveness of combination treatment with both Li+ and VPA. In addition, this study will compare the efficacy of Li+ versus VPA maintenance therapy. Methods: The trial is being conducted in 2 phases. In phase 1, youths ages 5-17 years with either BP-1 or BP-2 are treated with open-label VPA and Li+ for up to 20 weeks. If they tolerate pharmacotherapy and meet a priori criteria for stabilization for 4 consecutive weeks, they are then randomized to receive monotherapy with either VPA or Li+ for up to 76 weeks in a double-blind fashion. Results: Data are available for the first 63 youths with BP-1 and 3 youths with BP-2. Marked reductions in manic and depressive symptoms are evident by the end of 8 weeks of treatment (p< .0000). Ten patients are currently enrolled. Twenty-four have been randomized. The most common reasons for not being able to enter phase 2 include side effects and persistent psychiatric symptomatology. Conclusions: Symptoms of mania may be effectively and safely reduced acutely with combination Li+ and VPA pharmacotherapy. This trial will help determine whether or not Li+ or VPA monotherapy is an effective maintenance strategy.
Key Words:  medication, children, adolescents.

43. The Prevention of Bipolar Disorder in Genetically At-Risk Youths
R.L. Findling, B.L. Gracious, N.K. McNamara, J.R. Calabrese, Case Western Reserve University/University Hospitals of Cleveland, Cleveland, Ohio, U.S.A.

Introduction: Offspring of a parent with bipolar disorder (BP) are at high risk for developing this illness. In adult patients with BP, it has been shown that the severe symptoms of this illness do not have an abrupt onset. Rather, they initially become manifest in more modest expressions and gradually increase in severity over time. We have previously used the term “cyclotaxia” in order to describe the more modest mood symptoms that are present in these genetically high-risk youths who do not meet diagnostic symptom criteria for BP-1 or –2. The purpose of this study is to test the efficacy of divalproex sodium (Depakote ®, VPA) in preventing the full expression of bipolar disorder in these young patients with cyclotaxia.  Methods: This is a prospective, double blind, placebo-controlled study that is currently in progress. Youths between the ages 5-17 years of age who are the offspring of a parent with BP who have had significant symptoms of hypomania within the past 2 months who also meet diagnostic criteria for cyclothymia or BPNOS are eligible. Patients are randomized to receive VPA or placebo for up to 5 years.  Results: Thirty-two youths with a mean age of 11 years have been treated. At baseline, these subjects’ Children’s Global Assessment Scale was 54.10. The most common clinical reason for study discontinuation was continued mood symptomatology. Mania, hypomania, and continued cycling account for the mood states most often responsible for withdrawal from this study. Of note, no withdrawals have occurred due to adverse events. Conclusions: These preliminary results suggest that youths with cyclotaxia are functionally impaired and that therapeutic interventions for these children and teenagers are needed.
Key Words:  medication, children, adolescents.

44. Influence of Age and Executive Functioning on Verbal Memory of Patients with Unipolar and Bipolar Depression
P. Fossati, A.M. Ergis, J.F. Allilaire

Introduction: Despite many studies demonstrating memory and executive impairments in young and old depressed patients, the relationships between age, executive functioning and memory have not been evaluated in depression. The aim of this study was to investigate if older patients were more vulnerable than younger patients to the impact of depression on memory and if the differences between young and old depressed could be related to executive functioning.  Methods: Forty-nine inpatients, with unipolar (n=36) and bipolar (n=13) depression, ranging in age from 19 to 72 years were compared with 70 controls on a verbal memory task. Age cut-off of 45 years was used as a categorical variable to divide subjects into subgroups. A subset of patients (n=41) was also evaluated with the modified version of the Wisconsin Card Sorting Test and separated into a non-dysexecutive group and a group of patients with mild-executive impairment.  Results: Bipolar patients had experienced significantly more affective episodes than UP. The mean duration of illness and the mean number of hospitalizations were higher in BP than in UP. Despite these differences between UP and BP, BP patients performed at the same level than UP for all the cognitive scores. Depressed patients exhibited memory deficits with impaired free recall and normal cued recall and recognition. Both age and executive function influenced memory performance in depression, however neither group x age interaction nor age x executive status interaction were significant. The memory scores were related to age, psychomotor retardation and number of previous depressive episodes. Conclusion: Age and executive functioning influenced the memory performance of depressed patients. Older patients were not more vulnerable than younger patients to the impact of depression on memory. The memory deficits in depression may be associated with both trait and state factors and raise questions about the long-term cognitive functioning of patients with recurrent affective disorders.
Key Words:  memory, depression, aging.

45. Lifetime Influence of Panic Spectrum Symptoms on Treatment Outcome of Bipolar I Disorder
E. Frank, J. Cyranowski, P. Rucci, A. G. Mallinger, H. Swartz, M. E. Thase, D. J. Kupfer, Western Psychiatric Institute and Clinic, University of Pittsburgh Medical Center, Pittsburgh, PA, U.S.A.

Objectives: The authors tested the hypotheses that lifetime panic spectrum symptoms would be associated with higher levels of suicidal ideation and a poorer response to acute treatment among patients with bipolar 1 disorder.  Methods: A sample of 66 patients with bipolar I disorder completed a self-report measure of lifetime panic-agoraphobic spectrum symptoms (PAS-SR). Patients falling above and below a predefined clinical threshold for panic spectrum were compared with respect to clinical characteristics, the presence of suicidal ideation during acute treatment, and acute treatment response.  Results: Almost half of this outpatient sample reported panic spectrum features above the predefined threshold (i.e., PAS-SR score > 35). These lifetime symptoms were associated with higher levels of current depressive symptoms, higher levels of suicidal ideation, and a 20-week delay in remission of the acute episode. Patients with high PAS-SR scores were also more likely to be treated for depressed or mixed/cycling mood states, rather than pure mania. The negative clinical implications of lifetime panic spectrum symptoms, however, remained significant even after controlling for the predominant mood state treated.  Conclusions: Panic spectrum symptoms are common among patients with bipolar 1 disorder, and are associated with a more negative clinical picture and poorer treatment response. Panic spectrum features were particularly prevalent among patients treated for either depressed or mixed/cycling mood states. It was in these groups, moreover, that the presence of panic spectrum symptoms signaled the most negative clinical and prognostic outcomes. These findings support the utility of panic spectrum assessment among patients with bipolar 1 disorder, and highlight the need for alternate treatment strategies for bipolar depressed or mixed/cycling episodes in individuals with lifetime panic spectrum features.
Key Words:  panic symptomatology, treatment outcome, suicidal thinking.

46. Health Care Utilization in Bipolar Disorder & Comorbid Substance Abuse
M.A. Frye, L.L. Altshuler, A. Sanders, A. Cozzolino, S. Masseling, and J. Mintz

Of all Axis I diagnoses, the highest reported lifetime prevalence rate for substance abuse (60.7%) has been in bipolar I disorder (Regier 1989). This co-occurrence has been associated with greater illness morbidity and lithium non-response (Frye and Altshuler 1997). This retrospective study was conducted at the West LA VAMC to assess health care utilization in bipolar patients with and without substance abuse co-morbidity.
  Medical records from 100 outpatients (14f/86m) with a DSM-IV diagnosis of bipolar disorder were reviewed for the presence or absence of co-morbid substance abuse/dependence. Dependent variables (# psychiatric hospitalizations, # ER evaluations) from 1/1988-10/1994 were abstracted from the medical record. The data were analyzed using a 2 (alcohol) x 2 (drug) factorial regression design. Time at risk was entered as a co-variate, indexed by the date of the first visit or admission (whichever was earliest) in the record. As typically seen in count data such as these, the distributions were Poisson rather than normal, so the statistical analysis was a generalized linear model (SAS GENMED) specifying Poisson error.
  The 4 bipolar diagnostic groups were: No co-morbidity (n=49), Etoh (n=19),Drug (n=6), and Etoh&Drug (n=26). The mean number of hospitalizations (adjusting for duration of time in clinic within the study period) were: 3.15, 4.63, 4.02, and 8.01 respectively. Both alcohol (chi-square = 7.14, df=1, p=0.008) and drug co-morbidities (chi-square = 18.52, p<0.0001) were associated with an increased number of hospitalizations during follow-up. Similarly, the mean number of ER visits were: 7.09, 10.31, 8.59, and 18.82 respectively; both alcohol (chi-square = 48.86, df=1, p=0.0001) and drug abuse(chi-square = 14.54, p=0.0001) were associated with an increased number of ER visits.
  These data suggest a significant economic morbidity from the standpoint of health care utilization for bipolar patients who have illness complicated by substance abuse co-morbidity. Further study is encouraged to determine specific utilization patterns to better define substance abuse treatment programs for patients with bipolar disorder.
Key Words:  substance abuse, bipolar,  health care cost.

47. Bipolar Disorder: Assessing Treatment Response in a Naturalistic Setting
J. Garnham, A. Munro, A. Teehan, A. Duffy, M. MacDougall, M. Passmore, C. Slaney, M. Alda, Department of Psychiatry, Dalhousie University, Halifax, Canada

Purpose: Currently, several treatment modalities are used in the management of bipolar disorder, including anticonvulsant drugs (carbamazepine, valproate and lamotrigine), antipsychotic agents (risperidone and olanzapine), and lithium (a standard treatment used for the past 50 years). Varying degrees of research evidence support the efficacy of these treatments. The objective of this study was to evaluate these treatments in a naturalistic setting and compare the results with what is known from controlled clinical trials.  Methods: Patients with a diagnosis of Bipolar Disorder were randomly selected from the Mental Health Outpatient Clinic at the QE II Health Sciences Centre. Retrospective chart reviews were completed on each treatment group and treatment response was determined based on the semi-quantitative scale (0-10) that was developed by the International Group for the Study of Lithium (IGSLI) to assess treatment effect in a naturalistic setting. For each patient, their treatment response was evaluated only to those mood stabilizers received for a minimum of 6 months. Inter-reliability scores were calculated (concordance of ratings = 90%, based on previous data).  Results:  The results are summarized in the following table:

48. A Case Management Protocol for Patients with Bipolar Disorder
E. L. George, D.O. Taylor, and D.J. Miklowitz, University of Colorado at Boulder

The term case management is used quite often to describe a type of service that is provided for those with bipolar disorder. Given that case management is being used more frequently for individuals treated at mental health centers and through managed care companies, a thorough discussion of the structure and objections of case management for bipolar disorder is indicated.
  This poster describes a case management (CM) protocol for patients with bipolar disorder. We defined CM as a short-term supportive and educational treatment involving two or more sessions of family education and individual crisis intervention and psychoeducation on an as needed basis.
  Specific issues addressed in CM with bipolar individuals include medication noncompliance, self-medication with substances, suicidality, hospitalization, divorce, stigma of the disorder, circadian rhythms, seasonality, relapse prevention, and interpersonal/family conflicts. The content of the poster focuses on key issues in CM and how clinicians can address the issues of difficult or treatment-resistant patients.
Key Words:  case management, bipolar disorder, treatment development.

49. Seroquel Treatment of Rapid-Cycling Bipolar Disorder: An Open Prospective Study
S.N. Ghaemi, The Cambridge Hospital, Cambridge, MA, USA; J.F. Goldberg, New York Presbyterian Hospital, New York, NY, USA; J.Y. Ko, The Cambridge Hospital, Cambridge, MA, USA; L. Oakley, New York Presbyterian Hospital, New York, NY, USA.

Objective: To provide pilot data on effectiveness of quetiapine (Seroquel) in rapid-cycling bipolar disorder.  Method: This study is designed to recruit 40 bipolar I rapid-cycling subjects with up to one year open prospective follow-up of seroquel treatment, with or without concomitant mood stabilizer. So far, we have recruited 13 subjects (8 males, 5 females; 10 Caucasian, 3 non-Caucasian; mean age 43.8 ± 21.0 years), who have received regular Hamilton Depression Rating Scale (HDRS) and Young Mania Rating Scale (YMRS) assessments, along with Clinical Global Impression for Bipolar Disorder (CGI-BP) and daily mood charts ratings. Patients enter the study with any mood symptomatology severe enough to require added medication intervention. 8/11 received concomitant lithium and/or divalproex, and one person received concomitant lamotrigine. Results: The mean dose of seroquel was 140.9 ± 112.5 mg/d. Preliminary data analysis indicates that subjects experienced improvement on HDRS (15.3 ± 5.6 to 7.6 ± 4.3; paired t-test, t=2.73, p=0.03) and YMRS (15.6 ± 9.5 to 11.1 ± 8.4, paired t-test, t=1.10, p =0.35) ratings at week 8 of follow-up, with depressive symptoms improving statistically significantly. CGI-BP ratings indicated clinically significant improvement on both manic (mania severity decreased from 3.5 ± 1.3 to 2.1 ± 1.1; paired t-test, t=1.70, p=0.14) and depressive symptoms (depressive severity decreased from 3.7 ± 1.2 to 2.7 ± 1.8; paired t-test, t=1.23, p=0.27) at week 8. Based on CGI-BP improvement ratings, 4/7 (57.1%) patients treated for 8 weeks demonstrated moderate to marked improvement in overall bipolar illness. Drop-out and side effect data will be analyzed.  Conclusions: In this preliminary analysis of a partially recruited dataset, early indications are that seroquel improves rapid-cycling symptoms in 2 month follow-up of bipolar disorder. Surprisingly, the greatest magnitude of improvement seems to have occurred in depressive symptomatology. Data from this study with a larger sample and longer follow-up will also be presented.
Key Words:  bipolar disorder, rapid-cycling, seroquel.

50. Topiramate Treatment of Bipolar Spectrum Disorders: A Retrospective Chart Review
S.N. Ghaemi, The Cambridge Hospital, Cambridge, MA, U.S.A; S.G. Manwani, Boston University Medical Center, Boston, MA, U.S.A; J.J. Katzow, George Washington University, Washington, DC, U.S.A; J.Y. Ko, The Cambridge Hospital, Cambridge, MA, U.S.A; F.K. Goodwin, Center on Neuroscience, Medical Progress, and Society, George Washington University, Washington, DC, U.S.A.

Objective: To determine if topiramate is effective as treatment for bipolar spectrum disorders in a naturalistic setting.  Method: All charts of outpatients treated with topiramate (n =76) were reviewed and clinical response assessed retrospectively using the Clinical Global Impressions Scale for Improvement (CGI-I). This project was funded by a research grant from Janssen Pharmaceutica.  Results: Mild improvement was seen in 47% (n=36) and moderate to marked improvement in 13% (n=10). Responders received a higher mean dose (180mg/d) than non-responders (83.2 mg/d, p=0.002). Topiramate dose was also higher in those who lost weight (138.3 mg/d) than in those who did not (70mg/d, p=0.007). 50% experienced weight loss with a mean amount of 14.2 lbs. 82% (n=62) reported side-effects including cognitive effects, sedation, parasthesias, nausea, insomnia, head-ache and dizziness. 36% (n=27) of the total sample discontinued treatment because of adverse effects.  Conclusion: Topiramate led to significant weight loss in about half of this bipolar population, while also improving mood symptoms at least mildly in most patients. Topiramate response and weight loss were both dose related, with efficacy, in particular, associated with higher doses (mean 180 mg/d) than frequently used in current clinical practice.
Key Words:  bipolar disorder, drug therapy, topiramate.