BNN 2000 UpdateBi

pola

r N

etw

ork

New

s..

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

....

..

A Quar ter ly Pub l i ca t ion o f the B ipo lar Network

March 2000

Vol. 6, Issue 1

Several important changes have occurred in thefunding and production of the Bipolar NetworkNews newsletter in the last few months. Because ofthese changes, there were only two issues printedfor 1999. We apologize for any misunderstandingsconcerning our publication schedule. Both this is-sue and the last issue are longer than usual to fill inany information gaps we may have missed. We willresume publishing four issues per year; this issue isthe first issue of 2000.

Funding Support

We are most appreciative for the contribution of anunrestricted grant by Janssen Pharmaceuticals, a sub-sidiary of the Johnson & Johnson company, to con-tinue to publish the BNN. This unrestricted grantfor the BNN is directed at better informing patients,their families, physicians, researchers, and all othersinterested in bipolar illness of the latest developmentsin the field. We value this generous support and itsunrestricted nature.

BNN Content

As in previous editions of the BNN, we will try toprovide detailed updates on the most current find-ings in bipolar illness, with a particular focus onactivities in the Stanley Foundation Bipolar Net-work and the Intramural Program of the NationalInstitute of Mental Health from which most of theactive work of the editors emanates. In addition, wewill try to note new developments in the illness asrevealed by the Systematic Treatment EnhancementProgram (STEP) and other NIMH, Stanley, and Na-tional Alliance for Research on Schizophrenia andDepression (NARSAD) initiatives that are focusedon acquiring new data on the neurobiologicalmechanisms involved in bipolar illness, and the de-velopment of more effective treatments.

Also as in previous editions of the BNN, we willinclude case reports using the NIMH-Life ChartMethod (NIMH-LCM™) of a patient’s course ofillness and response to treatment illustrating an im-portant characteristic or treatment principle aboutthe illness. We will also continue to review selectedhighlights from national and international meetingsto preview some of the new developments andemerging trends in the field around the world.

Disclaimer

It is important to emphasize that in many instances,the findings noted in the BNN are highly prelimi-nary and/or based on prepublication presentationsand abstracts. Although we have and will continueto try to accurately report preliminary findings, wecannot be responsible for errors of fact or interpreta-tion in this preliminary communication format.Thus, patients and physicians should not act clini-cally on the basis of information provided in the BNNwithout further materials from formal, peer-reviewedpublications in the scientific literature. In particu-lar, patients should always consult with their treat-ing physicians about any potential alterations in theirtherapeutic approach raised by information in theBNN or other preliminary material.

The views expressed in the BNN are solely thoseof the editors and individual contributors, and donot represent the views of the NIMH, the StanleyFoundation, any scientific or academic society, orany pharmaceutical firm. We are proud of our closealliance with treatment advocacy groups includingthe National Depressive and Manic-Depressive As-sociation (NDMDA) and the National Alliance forthe Mentally Ill (NAMI), but the scientific report-ing and commentary in the BNN also do not reflectthe policies of these organizations.

We will continue to print a short disclaimer warn-ing in each subsequent issue, but patients should beaware that any potential ideas (either good ones or

(Continued on page 2)

In This Issue:

New Research:Gabapentin in Bipolar Disorder..........2

Meeting Highlight:Intl. Conf. on Bipolar Disorder............3

Life Chart Highlight:Lamotrigine+Gabapentin.....................6

Meeting Highlight:ACNP Annual Meeting.........................8

Investigator Spotlight:Drs. Bolwig and Kessing....................12

Page 2

BNN V.6, Iss. 1

Bipolar Network News

Editors-in-Chief:Robert M. Post, MDGabriele S. Leverich, MSW

The BNN is published four times ayear by an international group ofinvestigators working with patientswith bipolar disorder to betterunderstand the long-term courseand treatment of the illness. Thegoal of the Network is to helpdevelop new and more effectivetreatments for bipolar disorder.

We welcome any comments orsuggestions you may have. Forcommunication or to be placed onthe mailing list, please contact us at:

Bipolar Network NewsNIMH10 Center Drive MSC 1272Bldg. 10, Room 3N212Bethesda, MD 20892-1272

Telephone: (800) 518-SFBN (7326)Fax: (301) 402-0052Website: www.bipolarnetwork.orgE-mail: info@bipolarnetwork.org

BNN 2000 Update(Continued from page 1)

treatment misconceptions) generatedfrom reading the BNN need to be care-fully reviewed with their treating physi-cian; only he or she can assume medicalresponsibility for treatment decisions somade.

The articles and information printedin the BNN are not copyrighted and maybe reproduced without permission. Wewelcome comments about the content ofthe BNN and the expression of alterna-tive points of view on any of the clinicalor scientific material noted or referredto. ■

Gabapentin (Neurontin®) in the acute treatment ofrefractory bipolar disorder

Lori Altshuler, Paul Keck Jr., Susan McElroy, Trisha Suppes, E Sherwood Brown,Kirk Denicoff, Mark Frye, Michael Gitlin, Sun Hwang, Robyn Goodman, GabrieleLeverich, Willem Nolen, Ralph Kupka, and Robert Post

Full Article Appeared in the Journal: Bipolar Disorders 1: 61–65, 1999

BACKGROUND: Gabapentin, a new anti-epileptic agent, has been anecdotallyreported to be effective as an adjunctive treatment in bipolar illness. We systemati-cally assessed the response rate in bipolar patients being treated with open adjunc-tive gabapentin for manic and depressive symptoms or rapid cycling not responsiveto standard treatments.

METHOD: Twenty-eight bipolar patients experiencing manic (n = 18), depressive(n = 5), or rapid-cycling (n = 5) symptoms inadequately responsive to at least onemood stabilizer were treated in an open fashion with adjunctive gabapentin at oneof four sites. Illness response was assessed every two weeks using the Clinical GlobalImpressions Scale modified for Bipolar Disorder (CGI-BP). A ‘positive response’was operationalized as a CGI response of much or very much improved.

RESULTS: Of the 28 patients treated with open adjunctive gabapentin, an overallpositive response was seen in 20 patients (72%). Fourteen (78%) of the 18 patientstreated for hypomania (n = 6) or mania (n = 12) had a positive response to a dosagerange of 600–3600 mg/day. Patients with hypomania responded fastest, with a posi-tive response achieved in 12.7 ± 7.2 days. Patients with classic mania had a meantime to positive response of 25 ± 12 days, and in patients with mixed mania it was31.8 ± 20.9 days. The five patients treated for depression had a positive responsewithin 21 ± 13.9 days. Only one of five patients with rapid cycling had a positiveresponse, however. Gabapentin was generally well tolerated, with the most commonside effect being sedation (n = 5; 18%).

CONCLUSIONS: Gabapentin appears to have positive effects when used as anopen adjunctive treatment for residual manic and depressive symptoms, but notrapid cycling. Prospective double-blind studies are needed to further delineate itsacute and prophylactic efficacy in conjunction with other mood stabilizers.

Note: Two double-blind trials of gabapentin were negative—response rates did notexceed those of placebo. These studies included: 1) the NIMH study of Frye et al.(2000; J Clin Psychopharmacol, in press) where lamotrigine (Lamictal®) was superiorto both gabapentin and placebo in a six-week monotherapy trial for refractory pa-tients; and 2) the failure of gabapentin to exceed placebo in an outpatient study ofacute mania (Pande et al., 1999; see p. 4). In the study of Frye et al., better gabapentinresponse was associated with younger age, shorter duration of illness, and lowerinitial body weight (Obrocea et al., unpublished data). Clearer definitions of theutility of gabapentin in bipolar disorder requires much further systematic investiga-tion. ■

New Research

Page 3

March 2000

Lithium Update

DRS. R. BALDESSARINI, L. TONDO, AND J.HENNEN of Harvard Medical School re-viewed the literature on patients who con-tinued long-term treatment with lithiumversus those who discontinued. Suicidalacts rose 22-fold, and fatalities increased14-fold, within the first year after discon-tinuing lithium. They concluded fromtheir data that there are “…powerful,long-sustained protective effects oflithium against suicidal behavior.…” Aslow taper of lithium during discontinu-ation is somewhat helpful in decreasingthe rate of suicide, but the risk is still 11-fold higher than for those who continueon lithium.

Note: These data continue to indicate thegreat importance of lithium in treatment regi-mens for bipolar patients, even thoughlithium, in monotherapy, is not completelyeffective for mood regulation in some patients.Despite the growing availability of other pu-tative mood stabilizers, a continued role forlithium as adjunctive therapy, if notmonotherapy, remains high on the list of treat-ment approaches.

Third International Conference on Bipolar DisorderPittsburgh, Pennsylvania, June 17th�19th, 1999

DRS. BALDESSARINI AND TONDO alsoreported data contrary to numerous pre-vious studies in the literature, that nei-ther a long latency until institution of pro-phylactic lithium therapy (i.e., after manyepisodes have been experienced) nor rapidcycling conferred a negative prognosis forultimate lithium response.

DR. A. VIGUERA et al. of HarvardMedical School reported similar rates ofrelapse in 42 pregnant (52%) and 59 non-pregnant (58%) women who discontin-ued their lithium for 40 weeks. However,in the post-partum period (i.e., weeks 41–64 after lithium discontinuation) relapseswere 3.3 times more frequent in the preg-nant versus non-pregnant women (80%vs. 24%, respectively).

DR. M. BAUER and colleagues at theUniversity of Berlin reported an increasedprevalence of goiter in 100 lithium-treatedpatients that was related to duration anddose of lithium treatment and was pre-vented by thyroxine (T4, Synthroid®)treatment.

Note: These data, together with those ofFrye et al. (1999; Am J Psychiatry 156:1909–1914) reporting that lithium-treated patientswith low free T4 had more depression andcycling, continue to suggest several positives(in efficacy and side effects) for consideringthyroid augmentation strategies.

DR. S. GHAEMI et al. of MassachusettsGeneral Hospital reported data from a 1.7year open-label comparative trial ofdivalproex (Depakote®) vs. lithium in thetreatment of bipolar disorder. They foundboth to be equally effective in 43 patients;lithium nonresponders responded well todivalproex, and vice versa. Divalproexmonotherapy was notably effective intreating depressive symptoms.

Antidepressants

DR. C. GUILLE et al. of Massachusetts Gen-eral Hospital reported on a continuingstudy of bupropion (Wellbutrin®) vs. de-sipramine (Norpramin®) in bipolar de-pression (Sachs et al., 1994; J Clin Psy-chiatry 55:391–393), and found that overthe one-year follow-up, the rate of switch-ing into hypomania and mania was stillsignificantly greater in the desipraminegroup (37%) compared with thebupropion group (13%, p<0.05).

DR. R. JOFFE et al. of McMaster Uni-versity, Canada, reported that for 25 pa-tients in a randomized, controlled six-week trial, adjunctive treatment with theantidepressant paroxetine (Paxil®) wasequally effective for depression breakingthrough ongoing mood stabilizer treat-ment as the addition of a second moodstabilizer, typically valproate to lithiumor vice versa.

Note: Although rapid cyclers were not in-cluded, this is a highly important study andshould help refocus attention on the relativesafety and efficacy of the newer antidepres-sants for bipolar depression.

DR. L. ALTSHULER and colleagues atthe UCLA Neuropsychiatric Institutefound that the risk of depressive relapsefor 27 bipolar patients after antidepres-sant discontinuation was 67% versus 39%in the 18 patients who remained on anti-depressants.

Note: Although this was a retrospectivechart review and not a randomized study, itsuggests that the previous recommendation—discontinuing antidepressants as soon as pos-sible in patients with episodes of depressionbreaking through mood stabilizers—shouldcontinue to be reevaluated as new data be-come available. (Continued on page 4)

Meeting Highlights

The Third InternationalConference on BipolarDisorder, like the previous two

conferences in 1994 and 1997, pro-duced a wide variety of interesting andexciting new findings in the treatmentof bipolar disorder. Among the manyexcellent presentations were the follow-ing highlights:

Page 4

BNN V.6, Iss. 1

Meeting Highlights: 3rd Intl. Bipolar Conf.(Continued from page 3)

New Treatments

DR. A. PANDE of Parke-Davis Pharmaceu-tical Research presented data indicatingthat gabapentin (Neurontin®) was not aneffective treatment for acute mania. In a10-week, double-blind, placebo-con-trolled trial of adjunctive gabapentin, flex-ibly dosed between 900 and 3600 mg/day, gabapentin was not superior to pla-cebo as an adjunctive treatment for bipo-lar I symptoms of hypomania, mania, ormixed states. These data are convergentwith the data from the NIMH double-blind study (Frye et al., 2000; J ClinPsychopharmacol, in press) indicating thatlamotrigine (Lamictal®) was more effec-tive in a group of patients with a varietyof refractory affective disorders than ei-ther gabapentin or placebo (neither ofwhich differed statistically from eachother).

In contrast to Dr. Pande, DR. M.HARDOY et al. of the University of Flo-rence, Italy, reported a 60% moderate tomarked response rate for open-label, ad-junctive gabapentin in the treatment ofbipolar and schizoaffective disorders, aresponse rate similar to the results fromopen-label, adjunctive studies in the lit-erature. Gabapentin has also been shownto be effective in some anxiety disordersincluding social phobia, and is widelyused for adjunctive treatment in pain syn-dromes. Thus, its spectrum of therapeu-tic efficacy in adjunctive treatment foranxiety and depression associated withbipolar disorder remains to be furtherdelineated.

DR. S. HOOPES, (private practice,Boise, Idaho) in a study of lamotrigine inthe treatment of bipolar depression andother affective disorders in 218 patients,found that lamotrigine combined wellwith antidepressants, neuroleptics,lithium, and other anticonvulsants, andproduced high rates of response in com-

bination (69%) and in monotherapy(42%). These data parallel the open anddouble-blind data of Calabrese et al.(1999; J Clin Psychiatry 60:79–88).

DR. A. LAURENZA and colleagues atGlaxoWellcome Research performed adouble-blind, placebo-controlled study in437 outpatients that supported the effi-cacy of lamotrigine in unipolar depres-sion, finding it equal to desipramine andsuperior to placebo.

DR. T. SANGER et al. of Eli Lilly andCompany reported data from a study us-ing olanzapine (Zyprexa®) in the treat-ment of rapid cycling bipolar I patients.Twenty-five rapid cycling patients wererandomized to placebo and 19 toolanzapine; olanzapine (which has nowbeen FDA-approved for the treatment ofmania) was statistically significantly su-perior to placebo in mean reductions onthe Young Mania Rating Scale.

DR. R. CHENGAPPA et al. from West-ern Psychiatric Institute in Pittsburgh re-ported data on topiramate (Topamax®) asan adjunctive treatment for bipolar dis-order, finding a 56% much or very muchimproved response rate in 16 patients af-ter six weeks; four of the 16 patients hadtransient parasthesias (i.e., an unpleasantsense of touch) and two had word-find-ing difficulties. All patients lost weight(average of 10 lbs). These findings are verysimilar to those in the Stanley Founda-tion Bipolar Network as reported by Dr.S. McElroy et al. (2000; Biol Psychiatry,in press) in 56 patients.

DR. V. KUSUMAKAR and colleagues atDalhousie University, Canada, also foundpositive results of open-label topiramateaugmentation in female patients with re-fractory rapid cycling bipolar disorder.Eight of 15 patients who completed the16-week study achieved euthymia, andalmost half the patients lost 1–5% weight.

DRS. A. SCHAFFER AND A. LEVITT

(University of Toronto), and DR. R. JOFFE

(McMaster University) reported onmexiletine (Mexitil®) in treatment-resis-tant bipolar disorder. This agent, whichhas anticonvulsant, antiarrhythmic, andanalgesic properties, was examined in aninitial open-label study of 13 rapid cy-cling patients at doses ranging between200–1200 mg/day. A full response wasobserved in 46% of subjects.

Note: This promising study of mexiletinein rapid cycling patients maintained on othermood stabilizers deserves further follow-up.

DR. L. SHAFFER et al. of Sutter Com-munity Hospital reported data from theuse of primidone (Mysoline®) in the treat-ment of refractory bipolar disorder. Theyreported a modicum of success, i.e., eight(31%) of 26 patients treated with adjunc-tive primidone experienced a persistentpositive therapeutic effect.

Note: These data parallel the previousdata of Hayes (1993; Ann Clin Psychiatry5:35–44) showing a 33% response toprimidone.

DR. U. YAROSLAVSKY et al. of Ben-Gurion University, Israel, conducted adouble-blind controlled trial of phenytoin(Dilantin®) vs. placebo in mania for fiveweeks and found that there was signifi-cant therapeutic benefit for phenytoin,beginning on week three. They concludedthat “the results suggest that antimanicproperties may be true for allanticonvulsants.”

Psychotherapy

DR. D. HIRSHFELD-BECKER et al. at Mas-sachusetts General Hospital, in a studyof short-term adjunctive cognitive-behav-ioral therapy for bipolar disorder, foundsignificant decreases in the rate of relapsein those so treated. These data parallelthose of DR. E. FRANK at the University of

(Continued on page 5)

Page 5

March 2000

Meeting Highlights: 3rd Intl. Bipolar Conf.(Continued from page 4)

Pittsburgh, presented in the opening ses-sion of the conference, showing decreaseddepressive symptomatology in response tointerpersonal and social rhythm therapy.

DR. H. LAM DOMINIC et al. from theInstitute of Psychiatry, London, also pre-sented data on cognitive therapy for bi-polar illness. In a pilot study of relapseprevention with six-month post-therapyfollow-up data, they found there werefewer bipolar episodes, higher social func-tioning, and better coping strategies forbipolar prodromes in the cognitivetherapy group compared with the con-trol group that received treatment asusual. There was no evidence that im-provement in the therapy group was dueto more medication being prescribed.

DR. R. MORRISS of Royal Preston Hos-pital, UK, conducted a randomized con-trolled trial of teaching bipolar disorderpatients to identify early symptoms of re-lapse and obtain early treatment. Thismethod was highly effective in reducingthe number of days in relapse with maniaat 6 months (93%, p=0.007) and at 18months (58%, p=0.02).

Note: The above four studies clearly sup-port the beneficial effects of focused psycho-therapies as an important adjunctive treatmentapproach for bipolar illness.

Other New Research

DR. L. ALLOY et al. of Temple Universityreported data on lifetime comorbidity in17–24 year olds with bipolar spectrumdisorders. They reported higher lifetimerates of anxiety, substance abuse, Atten-tion Deficit-Hyperactivity Disorder(ADHD), and eating disorders in bipo-lar than in unipolar depressed partici-pants. Women with bipolar disorder hadhigher rates of anxiety (85% vs. 41%),ADHD (18% vs. 5%), and eating disor-ders (15% vs. 0%) than men with bipo-lar disorder, but not substance abuse dis-orders (41% vs. 32%). These data con-

tinue to point out the high rate ofcomorbidity of anxiety and substanceabuse disorders in individuals of all ageswith bipolar illness.

Note: This high rate of comorbidityshould lead to a comprehensive, prospectivetreatment approach to these conditions as wellas considering bipolar illness as a specific riskfactor for substance abuse disorders in ado-lescents.

DR. D. AXELSON et al. of the Univer-sity of Pittsburgh gave a clinical presen-tation of pediatric bipolar disorder, basedon Kiddie Schedule for Affective Disor-ders (KSADS) interviews from 142 pa-tients with a lifetime diagnosis of bipolardisorder from 1986 to 1995. They re-ported that the median episode durationof manic symptoms was one to two daysin 117 patients who met criteria for ma-nia or hypomania during the currentmood episode. They concluded from thedata that pediatric bipolar disorder intheir outpatient, predominantly adoles-cent sample “…presented primarily withbrief episodes and prominent depressivesymptoms. Suicidality, psychosis, andcomorbid disruptive behavior disorderswere common features...The presence ofmildly elated mood and increased energymay be important for distinguishing bi-polar patients from other mood and anxi-ety disorder patients.”

DR. L. GYULAI of the University ofPennsylvania reported that bone mineraldensity did not acutely decrease in pre-and postmenopausal women with refrac-tory affective disorder treated with high-dose levothyroxine (T4). ■

Complete abstracts of the ThirdInternational Conference on BipolarDisorder were published in thejournal Bipolar Disorders, aninternational journal of psychiatryand neurosciences (Vol. 1, Suppl. 1,1999; www.munksgaard.dk).

The following internet websites mightbe helpful to all those interested inthe research and treatment of bipolarillness:

www.nami.org

With more than 208,000 members,the National Alliance for the Men-tally Ill (NAMI) is the nation�s lead-ing grassroots advocacy organizationsolely dedicated to improving the livesof persons with severe mental ill-nesses including schizophrenia, bipo-lar disorder (manic-depressive illness),major depression, obsessive-compul-sive disorder, and severe anxiety dis-orders.

www.ndmda.org

The National Depressive and Manic-Depressive Association (NDMDA) isthe nation�s largest patient-run, ill-ness-specific organization. The mis-sion of the National Depressive andManic-Depressive Association is toeducate patients, families, profession-als, and the public concerning the na-ture of depressive and manic-depres-sive illnesses as treatable medical dis-eases; to foster self-help for patientsand families; to eliminate discrimina-tion and stigma; to improve access tocare; and to advocate for research to-ward the elimination of these ill-nesses.

www.narsad.org

The National Alliance for Researchon Schizophrenia and Depression(NARSAD) raises and distributesfunds for scientific research into thecauses, cures, treatments, and pre-vention of brain disorders, primarilythe schizophrenias, depressions, andbipolar disorders. NARSAD hasawarded $82.13 million to fund 2,130grants to 1,048 scientists in 146 uni-versities and medical research insti-tutions.

//www: bipolar

Page 6

BNN V.6, Iss. 1

Response to lamotrigine and gabapentincombination therapy?

As illustrated in the life chart (p. 7),this 44-year old male patient hadrecurrent and mild bouts of (win-

ter) depression from 1960 until 1971,when they became disabling. The courseof his illness changed in 1975 to morecontinuous hypomania, followed by se-vere depressions of many months’ dura-tion beginning in 1981. Treatment withantidepressants in 1983 (amitriptyline[Elavil®], and maprotiline [Ludiomil®)was associated with the induction of amore severe mania than had previouslybeen observed, but no further manic epi-sodes occurred despite a variety of subse-quent antidepressant clinical trials.

The antidepressants amitriptyline andnortriptyline (Pamelor®) were inadequatefor preventing depressions in 1987 and1989, the latter of which did not remituntil the combination of desipramine(Norpramin®) and fluoxetine (Prozac®)was instituted. With the onset of anotherdepression in 1991, the monoamine oxi-dase inhibitor (MAOI) phenelzine(Nardil®) in combination with lithiumwas instituted without success, and onlya partial response to electroconvulsivetherapy (ECT) was observed in 1994. Theserious depression of 1996 did not re-spond to treatment with two mood sta-bilizers (lithium and valproic acid[Depakote®]), the addition of the antide-pressant amoxapine (Asendin®), augmen-tation with the psychomotor stimulantmethylphenidate (Ritalin®), and thyroidhormone (T4, Synthroid®).

NIMH Double-Blind Trial

The patient was admitted to the NationalInstitute of Mental Health (NIMH) (p.7, bottom) and gave written informedconsent for participation in a double-blind randomized clinical trial compar-ing placebo, lamotrigine (Lamictal®), andgabapentin (Neurontin®) (see BNN Vol.4, Iss. 4). As illustrated, the patientshowed no response in Phase I duringplacebo administration, but in Phase IIhad a substantial, but partial, response tolamotrigine (i.e., a ‘B’ for “much im-proved” on the Clinical Global Impres-sions [CGI] Scale). Then, in Phase III,he showed some loss of response duringgabapentin administration (resulting in aCGI rating of ‘C’ or “minimally im-proved”). In Phase IV for response con-firmation, the Phase II drug wasreinitiated on a double-blind basis (laterrevealed to be lamotrigine), although inthis instance the response was not as ro-bust as previously observed (receiving a‘C’ on the CGI). At this juncture,gabapentin was added to lamotrigine,which resulted in rapid, substantial im-provement to a virtual symptom-freebaseline (i.e., an ‘A’ or rating of “verymuch improved” on the CGI).

Although we highlight this case as apotential clinical response to lamotrigineand gabapentin in the presence of onlypartial responses to monotherapy, the dataconveyed and inferences that can bedrawn from this life chart are not unam-biguous. On first view, it appears that apartial response to lamotrigine was con-verted to a more complete one with theaddition of gabapentin. However, it is alsopossible that this improvement resulted

from a natural “cycling out” of a depres-sive episode, as might be expected on thebasis of prior durations of depression (p.7, middle line). It is noteworthy that thetime frame on the time line or date line ischanged on the bottom line row, so thatmonths rather than years are illustrated(i.e., 5/97 – 12/97).

The patient continued in remissionfrom November 1997 to July 1998, but amoderate depressive episode ensued de-spite ongoing continuation prophylaxiswith both lamotrigine and gabapentin atthe same doses previously utilized (notshown). Although the depression was lesssevere than observed in previous episodes,the failure of lamotrigine and gabapentinto prevent the onset of the next episodefurther supports the possibility that thepresumptive response to the addition ofgabapentin to lamotrigine was related tothe natural course of illness.

Discussion

Despite this ambiguity, we present thiscase for discussion for a number of rea-sons. Several other partial responders tolamotrigine appeared to benefit from theaddition of gabapentin at the NIMH. Thetwo drugs have very different mechanismsof action which could result in additiveeffects. That is, gabapentin increases thelevels of the major inhibitory neurotrans-mitter in brain, gamma-aminobutyricacid (GABA), whereas lamotrigine isthought to act in part by decreasing therelease of the excitatory amino acidglutamate in the brain. In addition, thetwo drugs have very different mechanismsof anticonvulsant effects on amygdala-

(Continued on page 9)

Life Chart Highlight

Page 7

March 2000

Figure 1: Multiple antidepressants, mood stabilizers, and ECT failed to adequately block severe depressive recurrencesfrom 1983 to 1996 in this patient with bipolar II illness. At the NIMH (bottom row), daily prospective LCM ratings duringthe double-blind randomized protocol of Frye et al. (2000; J Clin Psychopharmacol, in press) revealed: no response toplacebo (CGI - �minimally worse� = �E� ); moderate response to lamotrigine (CGI - �much improved� = �B� ); slightdeterioration while on gabapentin, but still slightly improved from earlier �worst phase� (i.e., CGI - �minimally improved� =�C� ); beginning of a re-response in phase IV response confirmation mode when the phase II medication (lamotrigine) wasblindly added again; and then a complete response to the lamotrigine+gabapentin combination (CGI - �very muchimproved� = �A� ).

Mania

Mania

Mania

Depre

ssio

nD

epre

ssio

n

Depre

ssio

n

Mild

Mild

Mild

Mild

Mild

Mild

Moderate

Moderate

Low Moderate

High Moderate

High Moderate

Low Moderate

Moderate

Moderate

Severe

Severe

Severe

Severe

Severe

Severe

Response to Lamotrigine and Gabapentin Combination after Partial Response to Monotherapy

1955

1977

1957

1979

1959

1981

1961

1983

1963

1985

1965

1987

1967

1989

1969

1991

1971

1993

1973

1995

1975

VVVVVV

T4XXXX X

PROSPECTIVE LCM DATA AT NIMH

RETROSPECTIVE LCM DATA

05/97 06/97 07/97 08/97 09/97 10/97 11/97 12/97

Placebo Gabapentin Gabapentin

Lamotrigine Lamotrigine

GGGGGGGGGGGGGGGGGGGGG

LLLLLLLLLLLLLLLLLLLL LLLLLL LLLLLLLLLLLLLLLLLLLL LLLLLLLLLLLLLLLLLL

GGGGGGGGGGGGGGGGGG

NortriptylineDesipramine

Amitriptyline

MaprotilineFluoxetine Fluoxetine

Lithium Lithium

Carbamazepine

PhenelzineECT

Valproic Acid

Methylphenidate

Amoxapine

CGI Improvement

E B C A

Page 8

BNN V.6, Iss. 1

The 38th Annual Meeting of the ACNPtook place from December 12–16, 1999,in Acapulco, Mexico. A variety ofneuropsychopharmacological findingspertinent to bipolar disorder and the restof the affective disorders were presented:

Drug Mechanisms

DR. H. MANJI et al. of Wayne State Uni-versity updated their data concerning thepossible neurotrophic effects of lithiumand valproate (Depakote®). They previ-ously reported findings that lithium in-creased brain levels of n-acetylaspartate(NAA), a putative marker of neuronal vi-ability and function. At the meeting, theyreported that valproate increased NAAas well, and robustly increased neuriteoutgrowth. This group also reported thatlithium increased the number of new cellsin the adult dentate gyrus.

American College of Neuropsychopharmacology (ACNP)Acapulco, Mexico, December 12th�16th, 1999

Note: Lithium and valproate also increasebrain-derived neurotrophic factor (BDNF)and Bcl2 (both cell survival factors) and de-crease Bax and p53 (cell death factors).

DR. R. DUMAN of Yale University re-ported that chronic antidepressant treat-ment increases cyclic-AMP response ele-ment binding protein phosphorylation(CREB-P) and the expression of BDNFin hippocampus and cerebral cortex.Chronic administration of differentclasses of antidepressants, including nor-epinephrine and serotonin selectivereuptake inhibitors, monoamine oxidaseinhibitors, and electroconvulsive seizures,was found to increase the number of newcells in the subgranular zone of the hip-pocampus. The increased cell number wasnot observed after acute treatment, con-sistent with the therapeutic time courseof antidepressants. In addition, this effectappeared to be pharmacologically specificto antidepressant treatment in thatchronic administration of an antipsy-chotic (haloperidol) or an opiate (mor-phine) did not increase the number of newcells. These findings are consistent withthe possibility that the action of antide-pressant treatment is mediated, in part,by increased survival of neurons.

Note: These data suggest important rolesfor lithium, valproate, and antidepressants ina cascade of biochemical effects culminatingin alterations in gene expression; some of thesealterations may involve changes in the micro-structure of the brain by increasing cell sur-vival or even the development of new neu-rons in adulthood (neurogenesis).

DR. L. MARANGELL and colleagues atBaylor University reported on inhibitoryeffects of omega-3 fatty acids on proteinkinase C (PKC) activity in vitro. Prelimi-nary data suggest that omega-3 fatty ac-ids may be effective mood stabilizers forpatients with bipolar disorder (Stoll et al.,1999; Arch Gen Psychiatry 56:407–412).Both lithium and valproic acid are knownto inhibit PKC activity after subchronicadministration in cell culture and in vivo.The current study was undertaken to de-termine the effects of omega-3 fatty ac-ids�eicosapentaenoic acid (EPA) anddocosahexaenoic acid (DHA)�on PKCactivity, in vitro. Dr. Marangell and col-leagues found that both DHA and EPA,as well as the combination of DHA andEPA, inhibited PKC activity at concen-trations as low as 4 mmoles/L. In con-trast, lithium, valproic acid and arachi-donic acid had no effect on the measureof PKC activity. They concluded thatDHA and EPA significantly inhibit theactivity of the catalytic domain of PKCbeta in a manner distinct from lithiumand valproic acid.

Stress and Trauma

DR. E. GOULD at Princeton Universitypresented further evidence ofneurogenesis in postnatal weeks of lifein both the rodent and the marmoset.She found that in the rodent, the odorof a predator (fox odor, a potent rodentstressor) increased glucocorticoid recep-tors and decreased neurogenesis; socialstress in the marmoset and the tree shrewalso decreased neurogenesis. Dr. Gouldstated that these findings, taken withthose of Dr. Duman (above) that antide-

(Continued on page 9)

Meeting Highlights

The American College ofNeuropsychopharmacology(ACNP), founded in 1961, is

a professional organization of some 600leading scientists. Members are selectedprimarily on the basis of their originalresearch contributions to the field ofneuropsychopharmacology, whichinvolves the evaluation of the effects ofnatural and synthetic compounds uponthe brain, mind, and human behavior.The official journal of the ACNP isNeuropsychopharmacology, publishedmonthly by Elsevier Science, Inc. Tolearn more about the ACNP, visit theirwebsite at: www.acnp.org.

Page 9

March 2000

pressants increase neurogenesis, suggestthat not only do stress and antidepres-sants have opposite effects on a varietyof neurotrophic factors such as BDNF,but they also have opposing effects onengendering new neurons.

A number of investigators spoke of thelong-term effects of early trauma on be-havior, neuroendocrinology, and brainfunction of the adult rodent. DRS. M.

PLOTSKY AND C. NEMEROFF of EmoryUniversity gave a presentation on mater-nal separation as a vulnerability factor fordisorders of emotional regulation. Dis-ruption of maternal behavior and patternsof care-giving appear to exert profoundand long-lasting effects on the offspring.Studies of Long-Evans rats and rhesusmonkeys exposed to neonatal maternal

separation display dysfunction of the hy-pothalamic-pituitary-adrenal axis, anxi-ety-like behavior, mild cognitive impair-ments, anhedonia, and proclivity to theacquisition of cocaine and alcohol self-administration.

Many humans experiencing repeatedtraumatic events early in life appear tosuppress their rate of cortisol production,yet they too show a higher incidence ofdepression, anxiety, and substance abuse,as well as suicide attempts. The new datapresented by Dr. Nemeroff mirror thoseof Dr. R. Yehuda and others, indicatingthat patients with a history of early severetrauma appear to have a blunted cortisolresponse to a variety of challenges com-pared with controls.

DR. J. BREMNER and colleagues at YaleUniversity reported that abused womenwith post-traumatic stress disorder(PTSD) had lower cortisol levels follow-ing both corticotropin-releasing factor(CRF) and adrenocorticotropic hormone(ACTH) challenge, and that abusedwomen with and without PTSD had ablunted ACTH response to CRF chal-lenge. In a second study, they found thatwomen with PTSD had a decrease in righthippocampal (as well as medial prefron-tal) function during traumatic remindersof childhood abuse. The findings wereconsistent with hippocampal atrophy anddysfunction in PTSD, and suggest long-term dysregulation involving increasedcentral CRF activity but decreased corti-sol in the periphery. (Continued on page 10)

Meeting Highlights: ACNP(Continued from page 8)

kindled seizures, wherein gabapentinincreases the threshold for seizures andlamotrigine is able to inhibit the spreadof seizures without changing the thresh-old.

Two double-blind clinical trials ofgabapentin have failed to reveal superi-ority to placebo. The NIMH clinicaltrial in refractory affective disorders(Frye et al., 2000; J Clin Psycho-pharmacol, in press) indicated thatlamotrigine was superior to bothgabapentin and placebo. Used asmonotherapy for six weeks, gabapentinwas not substantially different from pla-cebo in this highly refractory subgroupof patients. In addition, Dr. A. Pande ofParke-Davis Pharmaceutical Research,reported that gabapentin was no moreeffective than placebo as an adjunctivetreatment for acute mania (see “Meet-ing Highlights,” p. 4).

Nevertheless, a considerable litera-ture of open-label gabapentin augmen-tation of other mood stabilizers contin-

ues to suggest its potential utility for somesubgroups of bipolar patients. This util-ity is convergent with the other docu-mented effects of gabapentin, such as ef-ficacy in anxiety and social phobia syn-dromes, as well as in paroxysmal pain syn-dromes, each of which has a moderatelyhigh incidence of comorbidity with bi-polar illness. Gabapentin also does nothave many pharmacokinetic interactionswith other drugs, because it is largely se-creted unchanged in the kidney. In addi-tion, its side-effects profile is generallyrelatively benign when used in combina-tion with other agents.

Brodie and Yuen (1997; Epilepsy Res26:423–432) and others in the neurologi-cal literature suggest the excellent efficacyand tolerability of the lamotrigine-val-proate combination. However, because ofincreased risk of serious rash whenlamotrigine and valproate are used incombination, this may not be a high pri-ority treatment option for some patients.To the extent that gabapentin shows some

biochemical effects similar to valproate(i.e., it increases brain GABA levels), itis possible that the potentiative effectsobserved for lamotrigine and valproatecould also extend to lamotrigine andgabapentin.

This proposition remains to be moresystematically explored in more formalclinical trials. However, at present, itwould appear that gabapentinmonotherapy has little place as a primarymood stabilizer in bipolar illness, but itsuse as an adjunct may be associated withsubstantial clinical improvement and,occasionally, a complete remission, asoccurred in this patient, at least tran-siently. Clearly, the optimal choice forcombination therapy in patients with re-fractory affective disorders will requiremuch further research, but the currentcase at least raises the possibility of com-bining two drugs with very differentmechanisms of action in order to achievea more complete response than with ei-ther alone. ■■■■■

Life Chart Highlight(Continued from page 6)

Page 10

BNN V.6, Iss. 1

DR. M. MEANEY of McGill University,Canada, presented a dramatic example ofthe potential impact of early environmen-tal experience on lifelong changes in be-havior and gene expression, changes thatwere previously thought to be mediatedby hereditary genetics. Dr. Meaney foundthat in contrast to rat pups undergoing3-hour separation stress, rat pups thatwere separated for 15 minutes a dayevoked increased licking responses fromtheir mothers upon their return and wereprotected against age-related cognitive de-cline and loss of hippocampal cell volume.Because the licking component appearedcrucial, he examined spontaneous lickingbehavior and found that some mothersshowed this trait in great quantity, whereasothers were minimal lickers. The offspringof rat pups that were minimally lickedwere hypercortisolemic and showed in-creased anxiety in a variety of open fieldstudies. When minimally-licked rat pupsreproduced, their offspring also werehypercortisolemic and anxious. However,when these rat pups were cross-fosteredand brought up by a high-licking mother,they were protected againsthypercortisolemia and age-related cogni-tive decline and, in turn, were able to passthis more positive trait on to their off-spring transgenerationally.

Note: If analogous changes are possiblein humans, these data would appear to havetremendous social, cultural, and political im-plications because effective interventions inone single generation could have very long-lasting effects on the behavior and stress vul-nerability of many subsequent generations.

DR. M. DAVIS of Emory University hasbegun to study the potential for genetherapy of neuropsychiatric syndromes.He was able to increase CREB levels spe-cifically in the basolateral amygdala viaherpes simplex viral vector-mediated genetransfer, resulting in near normal long-term memory following massed trials of

conditioned fear training which usuallyresult in poor learning. These results showthat functional CREB activity in theamygdala may serve as a molecular switchfor the formation of long-term emotionalmemory as reflected in fear conditioning.

Brain Imaging

DR. M. PHELPS from UCLA gave a pre-sentation on “Molecular imaging: frommetabolism to gene expression withpositron emission tomography (PET),”raising the possibility of ultimately us-ing functional brain imaging to studygene expression in humans, and provid-ing initial evidence from their data thatthis may soon be possible.

DR. T. KETTER of Stanford Universitypresented data from his work in the Bio-logical Psychiatry Branch of the NationalInstitute of Mental Health (NIMH) aswell as his data from the new AffectiveDisorders Clinic at Stanford. At theNIMH, he found using PET that thosepatients with baseline hyperactivity infronto-temporal areas of brain, particu-larly the left insula, responded better tocarbamazepine (Tegretol®), whereas thosewho were hypometabolic in this area re-sponded better to the L-type calciumchannel blocker nimodipine(Nimotop®). Relative hypoactivity atbaseline was also associated with responseto lamotrigine (Lamictal®), gabapentin(Neurontin®), and 20 Hz (as opposed to1 Hz) repeated transcranial magneticstimulation (rTMS) of the brain. Lesschronically ill and less treatment-refrac-tory depressed bipolar disorder outpa-tients at Stanford compared to controlshad anterior paralimbic and cerebello-posterior cortical hypermetabolism bothat rest and in a (30 minute) sustained self-induced sad state. In antidepressant re-sponders to valproate, decreases in me-tabolism with the sadness task comparedto controls were blunted but were en-hanced in the nonresponders todivalproex.

The DR. H. SACKEIM group at Colum-bia University reported data that showedthat those depressed patients who have thegreatest decrease in blood flow in frontalcortex, cingulate gyrus, and temporal lobefollowing electroconvulsive therapy(ECT) have the best antidepressant re-sponse. Thus, it is uncertain which areasof brain showing alterations are relatedto the primary pathophysiological processof depression, and which are compensa-tory and adaptive and in need of enhance-ment.

New Treatments

At a panel session addressing endogenousglutamate receptor ligands as novel leadsfor schizophrenia research and therapy,it was noted that a number of investiga-tors have thought that increasingglutamate agonism indirectly could behelpful in the treatment of schizophre-nia because the N-methyl-D-aspartate(NMDA)-type glutamate antagonistphencyclidine (PCP, or“angel dust”) ishighly psychotomimetic. A presentationby DR. D. GOFF of Harvard Universityreviewed the clinical implications of en-dogenous and exogenous glutamate re-ceptor ligands. Attempts to use d-serineor cycloserine have been partially success-ful and, at this meeting for the first time,it was reported that d-alanine, whichmore readily crosses the blood-brain bar-rier, and is converted to d-serine in thebrain, is also effective in initial studies.A six-week trial of d-alanine showed im-provement in positive and negative symp-toms as well as in cognition in patientswith schizophrenia.

Note: These data provide another new tar-get of therapeutics in schizophrenia in addi-tion to attempts to decrease the enzymenaladase, as suggested at the meeting by Coyleet al. of Harvard University, or to decreaselevels of quinolinic acid, as suggested at themeeting by Schwarcz and associates at theUniversity of Maryland.

(Continued on page 11)

Meeting Highlights: ACNP(Continued from page 9)

Page 11

March 2000

Meeting Highlights: ACNP(Continued from page 10)

DR. J FRAZIER and colleagues atMcLean Hospital presented data on astudy of olanzapine (Zyprexa®) in thetreatment of bipolar disorder in juveniles.Despite widespread use of antipsychoticsin juveniles with bipolar disorder, fewstudies using these medications have beenconducted in this patient population. Theprimary objective of this study was to as-sess safety and efficacy of the novel anti-psychotic olanzapine after up to 8 weeksof open-label treatment. Twenty-three bi-polar disorder patients (currently manicor mixed), ages 5 to 14, receivedolanzapine (dose range: 2.5–20 mg/day).The response rate (using a >=30% im-provement on the Young Mania RatingScale) was 60.9%. Extrapyramidal symp-tom measures were not (statistically) sig-nificantly different from baseline levels.However, significant increases were ob-served in weight (4.98 ± 2.32 kg,p<0.001).

DR. M. SAJATOVIC of Case Western Re-serve University reported on quetiapine(Seroquel®) in neuroleptic-dependentmood disorders. In this prospective, open-label, 12-week trial of quetiapine therapyin patients with neuroleptic-dependentmood disorders, twenty individuals (10with bipolar disorder and 10 withschizoaffective disorder) receivedquetiapine therapy at a mean dosage of202 mg/day. Overall, patients did well onquetiapine therapy with significant im-provement compared to previous antipsy-chotic medication therapy. Most indi-viduals in the study had substantial de-

pressive symptoms, which improved onquetiapine therapy. Simpson-Angus rat-ings of neurological side effects also de-creased significantly from a baseline of 5.5to an endpoint score of 1.9 (p = 0.02).

DR. E. VIETA of the University ofBarcelona, Spain, presented data onrisperidone (Risperdal®) as add-ontherapy in bipolar disorder. Risperidonehas shown efficacy in treating affectivesymptoms of schizophrenia and in pa-tients with schizoaffective disorder andbipolar disorder. They conducted an openstudy in 598 patients with eitherschizoaffective disorder, bipolar type, orbipolar disorder, type I or II. Inclusioncriteria were age between 18 and 65 yearsand DSM-IV symptoms of an episode ofacute mania, hypomania, or mixed symp-toms, and a Young Mania Rating Scalescore >7. Each patient was receivingmood-stabilizing medication. Of the pa-tients initially recruited, 541 were eligiblefor evaluation using various measures oftreatment efficacy. During the 6-monthfollow-up, 111 patients dropped out forseveral reasons, including lost to follow-up (4%), side effects (3%), hospitaliza-tion (3%), lack of response (3%), poorcompliance (1%), patient’s decision (1%),and other (6%).

At 6 months, the mean dose ofrisperidone was 3.9 mg/day. According toa last-observation-carried-forward analy-sis, significant improvements were seenin scores on the Clinical Global Impres-sions scale (p<0.0001), total and positive,negative, and general psychopathology

subscales of the Positive and Negative Syn-drome Scale (p<0.0001), Young ManiaRating Scale (p<0.0001), and HamiltonRating Scale for Depression (p<0.0001).Significant improvements were also seenin most neurological side effects, includ-ing scores on the Udvalg for KliniskeUndersogelser subscale for neurologicalside effects (p<0.0001). By the end of thestudy period, over 40% of patients wereasymptomatic. These results suggest thatrisperidone may have mood stabilizingproperties and prove useful in the treat-ment of patients with bipolar disorder andschizoaffective disorder, in conjunctionwith mood-stabilizing medication.

In a special plenary symposium cel-ebrating the first 50 years of experiencewith lithium, DR. MULLER-

OERLINGHAUSEN of the Freie University,Berlin, indicated that although there is a30 to 50-fold increased risk of suicide inthe unipolar and bipolar affective disor-ders compared with the general popula-tion, lithium greatly lowers this risk backtoward that of the general population. Inthis regard, lithium appears superior tocarbamazepine because, in a prospectiverandomized study, no suicidal acts oc-curred in the lithium-treated group (146patients), but four suicides and five at-tempts were reported in 139 patients oncarbamazepine. These data are consistentwith a recent meta-analysis of Dr.Baldessarini and colleagues (see “MeetingHighlights,” p. 3). ■

DISCLAIMER:Although the editors of the BNN have made every effort to report accurate information, much of the work detailed here is in summary orprepublication form, and therefore cannot be taken as verified data. The BNN can thus assume no liability for errors of fact, omission, orlack of balance. Patients should consult with their physicians, and physicians with the published literature, before making any treatmentdecisions based on information given in this issue or in any issue of the BNN.

Page 12

BNN V.6, Iss. 1

DR. TOM G. BOLWIG obtained his medi-cal degree in 1964 and his advanced de-gree in 1977 from the University ofCopenhagen; he was Senior Lecturer inPsychiatry from 1977 to 1982, and hasbeen a Professor of Clinical Psychiatry from1982 to the present. Dr. Bolwig’s career isdistinguished by both its breadth anddepth. He has made major contributionsto clinical psychiatry, psychopharmacol-

ogy, and experimental neurobiology. He has published 215 papersand edited three books. He is also the Associate Editor of the Jour-nal of ECT and Acta Psychiatrica Scandinavica. Because ECT iswidely used in the treatment of refractory depression, Dr. Bolwigand his colleagues have studied its potential mechanisms of action,revealing alterations in neuropeptides, receptors, and the activationof glial cells following this treatment in experimental animals. Hisfinding evidence of glial activation with seizures is particularly in-teresting in relation to recent findings of deficient numbers and

Dr. Tom G. Bolwig, Dr. Lars V. KessingRigshospitalet, Copenhagen, Denmark

decreased activation of glial cells in depression reported by manygroups, including in autopsy specimens of depressed patients in theStanley Foundation Brain Bank collection.

Dr. Bolwig has also demonstrated that repeated episodes of al-cohol withdrawal would result in increased seizure susceptibility, aspredicted from the kindling model. In examining the Danish caseregistry in collaboration with Dr. Lars Kessing, he has found thestrongest evidence to date of sensitization phenomena in the affec-tive disorders (see p. 13). These data add to a growing list of evi-dence that suggests the importance of early and effective interven-tion and long-term prophylaxis in the recurrent affective disordersbased on the finding that the number of prior episodes is a strongpredictor of subsequent relapse and rehospitalization. We look for-ward to further important contributions of Dr. Bolwig and col-leagues to the better understanding and treatment of the refractoryaffective disorders. ■

Investigator Spotlight

DR. LARS V. KESSING received his M.D.from Copenhagen University in 1986,becoming a specialist in psychiatry in1999. He has studied the psychiatricsymptomatology of the human immu-nodeficiency virus (HIV), the conse-quences of epilepsy surgery, and cogni-tive impairments in primary affectivedisorder in relation to illness character-istics.

His series of studies (along with Dr. Bolwig) based on thedata in more than 20,000 patients in the Danish case registryhas added a crucial perspective to factors that contribute to therecurrence of affective disorders. He found that the best predic-tor of the incidence and rate of rehospitalization in both unipo-lar and bipolar disorder was the number of prior episdodes (seep. 13). Again, these data speak to the importance of long-termprophylaxis in attempting to alter this course of recurrent affec-tive disorders.

Most recently, Dr. Kessing has studied factors related to thedevelopment of dementia in affective disorder patients as wellas individual heterogeneity of illness characteristics that are as-sociated with recurrence. These data will enable one to betteranticipate high degrees of illness vulnerability and recommendmore effective long-term prophylactic strategies. Dr. Kessingwill defend his thesis in Medical Science at the University ofCopenhagen this summer, and we look forward to his contin-ued major contributions to the field. ■

Dr. Bolwig and Dr. Kessing are the lead investigators at the newStanley Foundation Bipolar Network European Center inDenmark. An example of their groundbreaking collaborativework is summarized on the following page.

Page 13

March 2000

Recurrence in affective disorder I. Case register study

Lars Vedel Kessing, Per Kragh Andersen, Preben Bo Mortensen, Tom Gert Bolwig

Full Article Appeared in the Journal: Br J Psychiatry 172: 23–28, 1998

BACKGROUND: In recent years, studies of the risk of recurrence in affective disorder in relation to the number of priorepisodes have given contradictory results. Some investigators have found that the duration of time to recurrence decreases as afunction of the number of prior episodes; other investigators have failed to demonstrate this relation in bipolar disorder. Toinvestigate changes in the risk of recurrence with successive episodes it is necessary to study a large sample of patients followedthrough many years. Psychiatric case registers make this possible and data can now be analysed with survival analysis. The aimof this study was to describe the rate of recurrence, expressed as the rate of readmission after successive discharges in unipolarand bipolar disorders, by applying life-table methods in a larger sample from a psychiatric case register.

METHOD: In Denmark, all psychiatric admissions have been registered in a nationwide register. Since there are no privatepsychiatric hospitals or clinics in Denmark, all admissions for the 5.1 million inhabitants are included in the register. Thestudy sample was defined as all inpatients with a manic-depressive main diagnosis, depressive or manic/cyclic episode, at firstdischarge. The sample was divided into two, according to the type of manic-depressive disorder at a given time (unipolar orbipolar). The type of disorder was thus time-dependent and some patients changed type during the study. The risk ofreadmission originates from a sum of two other risk factors: the risk of relapse, i.e., the return of an episode during remission(before recovery) and the risk of recurrence, i.e., the appearance of a new episode during recovery. The first eight weeks afterdischarge was defined as the period of remission, and readmission after these first eight weeks as a new episode (i.e.,recurrence).

RESULTS: A total of 20,350 patients constituted the study sample meeting the diagnoses requirements. Time to recurrencedecreased with the number of previous episodes for unipolar as well as for bipolar patients. Initially, the two types of disordersfollowed markedly different courses, but later in the course of the illness the rate of recurrence was the same for the twodisorders. The progressive course of the illness was partly due to a selection phenomenon, since patients with many episodesdemonstrated a high risk of recurrence from the first episode. However, this selection bias did not explain the wholedeteriorating pattern, since the high initial risk of recurrence for patients with many episodes also increased further with everynew episode. Further analysis revealed that other selection biases were caused by age and gender, since younger patients andfemale patients constituted an increasing proportion of patients and in addition experienced greater risk of recurrence as thenumber of episodes increased. However, if the sample was divided on gender and on age categories, all subpopulationsdemonstrated the same deteriorating pattern for each new episode. (Kessing et al., 1998; Br J Psychiatry 172:29–34)

CONCLUSIONS: The course of unipolar and bipolar disorder as reflected in hospitalizations seems to be progressive innature despite naturalistic treatment in the community, and stresses the great need for further, earlier, and more effectiveprophylactic interventions. ■

Lars Vedel Kessing, MD, Tom Gert Bolwig, MD, Department of Psychiatry, University of Copenhagen, Rigshospitalet; PerKragh Andersen, PhD, Department of Biostatistics, University of Copenhagen; Preben Bo Mortensen, Department ofPsychiatric Demography, University of Aarhus, Psychiatric Hospital, Risskov, Denmark

Bipolar Network NewsNIMH10 Center Drive MSC 1272Bldg. 10, Room 3N212Bethesda, Maryland 20892-1272www.bipolarnetwork.org

CHANGE SERVICE REQUESTED

In This Issue:

Third International Conference on Bipolar DisorderAmerican College of Neuropsychopharmacology Annual Meeting

Lamotrigine-Gabapentin Combination TherapyInvestigator Spotlight: Drs. Bolwig and Kessing

BNN V.6, Iss. 1