www.elsevier.com/locate/jad

Journal of Affective Disorders 78 (2004) S29–S38

Symposium Abstracts

The International Society for Affective Disorders (ISAD)—2nd

Biennial Conference, Cancun, Mexico 5th–10th March 2004

Sunday, 7th of March 2004

Symposium 11

Cochrane reviews: Advancing the design of

clinical trials and guiding practice

S11.1 Fluoxetine and amitriptyline: First-line treat-

ments or reference compounds?

Corrado Barbui, Andrea Cipriani

Department of Medicine and Public Health,

Section of Psychiatry, University of Verona, Italy

In recent years fluoxetine and amitriptyline have

been extensively studied in meta-analyses of clinical

trials comparing new and old antidepressive (AD)

agents in subjects with depression. However, many

clinical and research questions remain unsolved. We

performed a re-analysis of amitriptyline clinical trials

to assess whether amitriptyline still has a place in the

pharmacological treatment of depression, and a re-

analysis of fluoxetine clinical trials to investigate

whether the design of fluoxetine trials has been biased

by the common belief that new drugs under evaluation

are better than reference ones (wish bias).

To address the first issue, we investigated the

contribution of study setting on outcome in clinical

trials comparing amitriptyline with any other AD. A

systematic review and meta-regression analysis of

amitriptyline randomised clinical trials was carried

out. The electronic search yielded 181 randomised

clinical trials, 47% enrolling inpatients and 53% out-

doi:10.1016/j.jad.2003.12.003

patients with depression. Both on a dichotomous and

continuous outcome amitriptyline resulted more ef-

fective than control agents in inpatients (Peto odds

ratio (OR) 1.22, 95%, Confidence Interval (CI) 1.04,

1.42; Standardised Mean Difference (SMD) 0.28,

95% CI 0.08, 0.46), but not in outpatients (Peto OR

1.01, 95% CI 0.88, 1.17; SMD 0.10, 95% CI � 0.02,

0.23). Among inpatients amitriptyline was significant-

ly more effective than tricyclics (TCAs) and non-

significantly more effective than the selective seroto-

nin re-uptake inhibitors (SSRIs). Among outpatients

no statistically significant differences emerged be-

tween amitriptyline and TCAs and between amitrip-

tyline and the SSRIs. Amitriptyline was less well

tolerated than control agents in outpatients (Peto OR

0.90, 95% CI 0.81, 0.99), but not in inpatients (Peto

OR 1.09, 95% CI 0.95, 1.25). These data suggest that

a reasonable approach could be the first-line prescrib-

ing of newer agents in the routine outpatient care of

depressive subjects, and the use of amitriptyline in

inpatients with depression.

To address the second issue, we compared fluox-

etine dose and outcome in trials where fluoxetine was

the experimental drug with trials where it was the

comparator. A systematic review of randomised con-

trolled trials comparing fluoxetine with any other

antidepressant in depressive patients was carried out.

Studies were allocated to one of the following two

groups: group 1, fluoxetine was the experimental

drug; group 2, fluoxetine was the control drug. The

systematic search yielded 103 randomised trials. Stud-

ies where fluoxetine was the experimental drug adop-

ted a higher dose regimen than group 2 studies. In the

efficacy analysis, the weighted rate of fluoxetine

responders was 70.1% (95% CI 67.4%, 72.8%) in

Symposium AbstractsS30

group 1 studies and 57.9% (95% CI 57.2%, 58.7%) in

group 2 studies. The weighted rate of fluoxetine

dropouts was significantly higher in group 1 studies.

A meta-regression analysis indicated that, after adjust-

ing for possible confounders (year and country of

publication, sample size, weeks of follow-up, setting,

diagnostic criteria), studies where fluoxetine was the

experimental agent were positively associated with

treatment effect, indicating a significant advantage for

fluoxetine. The evidence that the outcome of fluox-

etine trials varied according to whether this drug was

used as a new compound or a reference one suggests

the presence of bias. Systematic reviews and meta-

analyses represent an invaluable tool to highlight

pitfalls in the design of clinical trials and to guide

everyday practice.

Conflict of interest: none.

S11.2 Low dosage TCA for depression: What have

psychopharmacologists been doing for 50 years

Toshi Furukawa

Department of Psychiatry, Nagoya City University

Medical School, Nagoya, Japan

Background: Tricyclic antidepressants are still pre-

scribed as often as selective serotonin reuptake inhib-

itors and other newer antidepressants in the world.

Experts have often claimed that clinicians prescribe

tricyclic antidepressants at less than adequate dosages.

In order to determine how low dosage tricyclic anti-

depressants compare with placebo and with standard

dosage tricyclics in acute phase treatment of depres-

sion in terms of effects and side effects, we conducted

a systematic review (Furukawa et al., 2003) of rand-

omised trials comparing low dosage tricyclics (100

mg/day or less) with placebo or standard dosage

tricyclics in adults with depression.

Methods: We performed electronic search of the

Cochrane Collaboration Depression, Anxiety and

Neurosis Group Controlled Trials Register (which

incorporates group searches of MEDLINE,

EMBASE, CINAHL, PsycINFO, PSYNDEX and

LILACS, and group hand searches), conducted

SciSearch and reference search of included studies,

and contacted significant authors. Two independent

reviewers assessed eligibility and quality of the

studies and extracted data. Main outcome measures

included random effects model relative risk of

response in depression, according to the original

authors’ definition but usually defined as 50% or

greater reduction in depressive severity. Relative

risks of overall dropouts and dropouts due to side

effects were also examined.

Results: We identified 35 studies (2013 partici-

pants) comparing low dosage tricyclics against place-

bo and six studies (551 participants) comparing low

against standard dosage tricyclics. Low dosage tricy-

clics, mostly between 75 and 100 mg/day, were 1.65

(95% confidence interval: 1.36–2.0) and 1.47 (1.12–

1.94) times more likely than placebo to bring about

response at 4 weeks and 6–8 weeks. Several sensi-

tivity analyses confirmed these findings. On the other

hand, standard dosage tricyclics failed to bring about

more response but produced more dropouts due to

side effects than low dosage tricyclics.

Conclusions: The evidence suggests that adminis-

tration of low dose tricyclic is a defensible practice.

Tricyclic antidepressants at dosages below the here-

tofore recommended range are more effective than

placebo. They may or may not be as effective as but

certainly cause fewer dropouts due to side effects than

standard dosage tricyclics.

A very similar relationship between response and

tolerability has been shown for SSRIs as well (Bollini

et al., 1999): dosages of 21–40 mg/day of fluoxetine

or equivalent showed a greater response rate and an

even greater adverse event rate than dosages below

20 mg/day.

Professionals need remind themselves that, after 50

years of usage, they have not been able to establish the

minimum effective dosage and the effective dosage

ranges for antidepressants—a simple set of numbers

that every practising physician and patient would want

to know.

1) Furukawa, T., McGuire, H., Barbui, C., 2003.

Low dosage TCA for major depression (Cochrane

review). The Cochrane Library, Issue 3. Update Soft-

ware, Oxford.

2) Bollini, P., Pampallona, S., Tabaldi, G., Kupel-

nick, B., Munizza, C., 1999. Effectiveness of antide-

pressants: meta-analysis of dose–effect relationships

in randomized clinical trials. Br. J. Psychiatry 173,

297–303.

Symposium Abstracts S31

S11.3 Psychological debriefing for PTSD: What

have we really done?

Simon Wessely

Institute of Psychiatry

If asked most people will probably believe that

what is called ‘‘psychological debriefing’’ is helpful

in the immediate aftermath of disaster. Talking about

what has happening, explaining what we know about

psychological reactions, and allowing people to ven-

tilate their feelings, is a ‘‘good’’ thing. And indeed

single session psychological debriefing has become

exceptionally popular, a social movement.

But does it work? We carried out the first ever

systematic review and meta-analysis of RCTs in this

area. The results showed that there was no evidence

that psychological debriefing did what it is supposed

to do—reduce the rate of subsequent psychiatric

disorders, such as post-traumatic stress disorder.

And the two best and longest trials showed something

unexpected—an increase in the rate of psychiatric

disorder in those who had received the intervention.

I shall discuss the possible reasons for these unex-

pected findings, and also how the Cochrane review has

been used and abused by both the proponents and the

opponents of debriefing, counseling and psychological

treatments in general. Vast quantities of bathwaters

have been ejected, but quite a few babies as well.

S11.4 Psychological treatments for depression:

What should we be doing?

Rachel Churchill1, V. Hunot, M. McGuire, S.

Wessely

1Health Services Research Department,

Institute of Psychiatry, London, UK

Objectives: (1) To provide a qualitative summary

of all controlled trials of brief psychological therapy

for people with depression. (2) To provide a quanti-

tative synthesis of a proportion of trials, testing three

hypotheses: (i) that brief psychological treatments

would demonstrate greater efficacy than treatment as

usual; (ii) that efficacy of brief cognitive-behavioural

treatments would be greater than for other types of

brief therapy; (iii) that there would be clinical and

methodological heterogeneity between trials and ob-

served effects would be modified by pre-specified

clinical and methodological characteristics.

Methods: A systematic review and meta-analysis

was performed. Data sources were electronic biblio-

graphic databases, specialist journals and study bibli-

ographies. Authors were contacted for further

information. Included were RCTs and CCTs of brief

psychological treatments compared with treatment as

usual or another type of psychological treatment for

adults with depression. Trials were quality rated.

Results: Sixty-three of 76 potentially eligible studies

were included. Trials were generally of low overall

quality. Hypothesis (i) was supported by all the data.

Hypothesis (ii) was supported by the recovery data, but

the mean change and mean difference data demonstrat-

ed non-significant differences only. Considerable het-

erogeneity between studies was identified, providing

support for hypothesis (iii). Meta-regression analyses

suggested that much of the variability might be due to

type of comparison in trials used in the analyses for

hypothesis (i), and to baseline severity and number of

sessions in trials used in analyses for hypothesis (ii).

Conclusion: Brief psychological therapy appears to

be effective in the treatment of depression. CBT might

be more effective than other types of therapy, particu-

larly in less severely depressed patients and where

fewer sessions can be offered. However, much of the

primary research in this field is of poor quality, provid-

ing only limited guidance for clinical decision-making.

Symposium 12

Suicide and affective disorders

S12.1 Suicide, depression and the vulnerability for

suicidal behaviour

Kees van Heeringen

University Hospital Gent

Three levels of risk factors associated with suicidal

behaviour can be described. First, psychological autop-

sy studies have demonstrated that nearly all suicides

occur within the context of a psychiatric disorder. Axis-

Symposium AbstractsS32

I disorders with a particularly increased risk include

mood disorder, schizophrenia, and alcohol dependence.

The risk in association with the latter two disorders is

strongly increased in case of a co-morbid mood disor-

der. Secondly, a number of factors have been identified

as influencing the threshold for acting on suicidal

impulses, including exposure to suicide among peers

or in the media, availability and acceptability of mental

health care, and social support. Thirdly, it has become

clear that these (mainly state-dependent) characteristics

are not sufficient to explain why some but not all

depressed individuals commit suicide. There is now

increasing insight in an underlying and trait-dependent

predisposition or vulnerability to suicidal behaviour.

This presentation will focus on biological and cog-

nitive psychological aspects of this predisposition to

suicide. Biological research has mainly focused on two

aspects, i.e. a hyper-reactivity of the stress-system and

an impaired function of the serotonine neurotransmis-

sion system. These characteristics appear to be inter-

related, as the stress hormone cortisol has been shown

to have cytotoxic effects on the serotonergic system.

Studies in the cognitive psychological area have iden-

tified three core characteristics, which distinguish

depressed suicidal from depressed non-suicidal indi-

viduals. These psychological characteristics include

tendencies (1) to perceive oneself as a loser when

confronted with psychosocial adversity, (2) to perceive

no escape from this situation (leading to feelings of

entrapment and related to deficient problem solving),

and (3) to perceive no rescue (leading to the develop-

ment of feelings of hopelessness).

Recent research suggests that the results from bio-

logical and psychological approaches converge to a

considerable extent. Particularly for the first and third

characteristic, as mentioned above, a comprehensive

psychobiological can be described. With regard to the

first characteristic, the extent of activation of the stress-

system, as measured by 24-h cortisol production,

correlates with a personality characteristic that deter-

mines the sensitivity to social or interpersonal stimuli.

From neuro-anatomical, neuropsychological and neu-

robiological points of view this characteristic is related

to frontotemporal/hippocampal functioning, attention-

al biases, and the serotonine (5-HT1a) and noradrena-

line neurotransmission systems, as demonstrated by

functional neuro-imaging and neuropsychological

investigations. Regarding the third characteristic, func-

tional neuro-imaging studies have shown that the

development of feelings of hopelessness correlates

with serotonine (5-HT2a) receptor functioning in the

prefrontal cortex, which is involved in the generation

and choice of response alternatives when confronted

with particular problems, and with trait-dependent

regulation of affect. It thus appears that, at the least,

two inter-related clusters of psychobiological charac-

teristics are involved in a persisting vulnerability for

suicidal behaviour, i.e. a sensitivity for interpersonal

events (or ‘resilience’) associated with the activation of

the stress-system, and prefrontal serotonine (5-HT2a)

functioning, hopelessness and the regulation of affect.

These findings may have important implications

for prevention and treatment. The recognition of the

psychobiological characteristics, which differentiate

between depressed suicidal and depressed non-sui-

cidal patients, will substantially enhance our capacity

to predict suicidal behaviour among depressed indi-

viduals, and to delineate specific targets for treat-

ment. With regard to treatment the current state of

knowledge suggests that all three cognitive character-

istics have to be addressed in the psychological

treatment of suicidal individuals. In addition, the

relationship between psychological and biological

characteristics suggests that pharmacological treat-

ment may be useful, or even necessary, in preparing

the substrate that is involved in the neuropsycholog-

ical processes of memory and learning, which are the

essential components of effective psychotherapy.

S12.2 Functional neuro-imaging of suicidal

behaviour

Kurt Audenaert

University Hospital Gent

Abstract unavailable.

S12.3 Suicidal behaviour, depression and eating

disorders

Myriam Vervaet

University Hospital Gent

Abstract unavailable.

Symposium Abstracts S33

S12.4 Depression and suicidal behaviour in

adolescents

Alan Apter

University of Tel Aviv

Among teenagers, both attempted and completed

suicide is, in the great majority of cases, preceded by

depressive symptoms. Depressed young people who

attempt suicide often come from broken families and

have had one or more relatives who have committed

or attempted suicide. They have also, relatively often,

run away from home and thus been brought up

without favorable role models. Physical and mental

abuse, as well as sexual assault, is also more common

in this group. Young people who have attempted

suicide often have lasting problems at school and also

difficulties in achieving workable relationships with

their peers, compared with young people who are

depressed and have not attempted suicide. Abuse of

alcohol and drugs, impulsive behavior and asocial

behavior are additional risk factors for attempted

and completed suicide among depressed young peo-

ple. Owing to the high incidence of depression among

young people who have attempted suicide, it is

important to make a diagnosis and provide adequate

treatment at an early stage. Studies show that depres-

sive disturbances are more common among children

and young people than has previously been believed.

Unfortunately, many young people with depression

are not identified, partly because their depressive

symptoms are often atypical and partly because adults

do not readily recognize depressive symptoms in the

young, owing to their wish to see their children as

happy and healthy. Since the number of young people

with depression appears to have increased after the

Second World War, and the age of onset of depressive

disturbances has decreased, it is important to increase

the effort to detect depressions in order to be able to

prevent suicidal behavior. Major depression is most

easily diagnosed when it appears acutely in a previ-

ously healthy child and in these cases the symptoms

closely resemble those seen in adults. In many chil-

dren, however, the onset is insidious and the child

may show many other difficulties such as attention

deficit disorder or separation anxiety disorder before

becoming depressed. Mood disorders tend to be

chronic when they start at an early age and the

children come from families where there is a high

incidence of mood disorders and alcohol abuse. In

some cases the depressed adolescent may also be

psychotic and have hallucinations and delusions that

are usually mood congruent. When the psychotic

themes are related to suicide such as in command

hallucinations or delusions of guilt, the risk for suicide

is very high. Bipolar disorder was once thought to

occur only rarely in youth. However, approximately

20% of all bipolar patients have their first episode

during adolescence, with a peak age of onset between

15 and 19 years of age. Adolescents with bipolar

disorder are at increased risk for completed suicide.

Twenty percent of adolescents with bipolar disorder

made at least one medically significant suicide at-

tempt. In the adult literature, a large review of studies

examining depressive and manic depressive disorders

found the mean rate of completed suicides to be 19%.

Patients who are male, or who are in the depressed

phase of their illness, are at the highest risk.

S12.5 Can antidepressant medication prevent

suicide?

Goran Isacsson

Division of Psychiatry, Karolinska Institute NEURO-

TEC, Huddinge University Hospital, M59, S-141 86

Stockholm

E-mail: Goran.Isacsson@neurotec.ki.se

Method: Review of evidence, and an analysis of

Swedish toxicological data in 15,400 suicides, and

their relation to population prescription data.

Result: Several reports give evidence for that

lithium prevents suicide in bipolar patients (Tondo

et al., 2003). A long-term study of a cohort in Zurich

found that the amount of received antidepressants

was the best predictor for not having committed

suicide at follow-up (Angst et al., 2002). It has been

shown that only a minority of individuals committing

suicide have received adequate antidepressant treat-

ment (Isacsson et al., 1994). A recent report from

Australia demonstrated that suicide rates declined in

proportion to exposure for antidepressants among

different demographic groups (Hall et al., 2003). A

study from USA found that increased use of anti-

Symposium AbstractsS34

depressants was one of the most plausible factors

behind the decreased rate of youth suicide (Gould et

al., 2003). However, meta-analyses of clinical trials

have not found evidence for this, but these, as the

naturalistic studies above, do not allow for definitive

conclusions (Khan et al., 2000). A preliminary anal-

ysis of clinical trials of paroxetine in children has

raised concerns regarding a possible suicidality in-

ducing effect of the drug.

Based on epidemiological studies until 1991, it was

hypothesized by the author that a 5-fold increase in the

use of antidepressants would lower suicide rates by

25% (Isacsson, 1994). Ten years later, the use of

antidepressants in Sweden was five times higher than

in 1991 and the suicide rate was 31% lower (Isacsson,

2000). The proportion of suicides found positive for

antidepressants in forensic toxicology had increased

from 16% to 23%. Since alternative causal factors

could not be identified, this development supported

the hypothesis that antidepressants do prevent suicide.

When toxicological findings of specific antidepressant

drugs in 15,400 suicides were related to their use in the

population, it was found that paroxetine was the

substance least commonly found in suicides. In all

365 suicides under age 20, antidepressants were found

in 15 cases, seven of them SSRIs, none of these was

paroxetine.

Conclusion: Although definitive conclusions cannot

be drawn, the increased use of antidepressants is the

most probable cause of the decrease in suicide seen in

Sweden and in several other countries. No evidence

was found for a suicide provoking effect of paroxetine

regardless of age.

1. Tondo, L., Isacsson, G., Baldessarini, R., 2003.

Suicidal behaviour in bipolar disorder: risk and pre-

vention. CNS Drugs 17(7), 491–511.

2. Angst, F., Stassen, H.H., Clayton, P., Angst, J.,

2002. Mortality of patients with mood disorders:

follow-up over 34–38 years. J. Affect. Disord. 68,

167–181.

3. Isacsson, G., Holmgren, P., Wasserman, D.,

Bergman, U., 1994. Use of antidepressants among

people committing suicide in Sweden. Br. Med. J.

308, 506–509.

4. Hall, W.D., Mant, A., Mitchell, P.B., Rendle,

V.A., Hickie, I.B., McManus, P., 2003. Association

between antidepressant prescribing and suicide in

Australia, 1991–2000: trend analysis. Br. Med. J.

326(7397), 1008.

5. Gould, M.S., Greenberg, T., Velting, D.M.,

Shaffer, D., 2003. Youth suicide and preventive inter-

ventions: a review of the past 10 years. J. Am. Acad.

Child Adolesc. Psychiatry 42, 386–405.

6. Khan, A., Warner, H.A., Brown, W.A., 2000.

Symptom reduction and suicide risk in patients treated

with placebo in antidepressant clinical trials: an anal-

ysis of the Food and Drug Administration database.

Arch. Gen. Psychiatry 57(4), 311–317.

7. Isacsson, G., 1994. Depression, antidepressants

and suicide. A study of the role of antidepressants in the

prevention of suicide. [Thesis]. Karolinska Institute.

8. Isacsson, G., 2000. Suicide prevention—a med-

ical breakthrough? Acta Psychiatr. Scand. 102(2),

113–117.

Symposium 13

Genetics and pharmacogenetics of affective

disorders

S13.1 Genetics of BPD

Elliot Gershon

University of Chicago

As evidence from linkage and association studies

has accumulated, some broadly reproduced findings

have emerged. Through meta-analysis of published

whole-genome linkage studies, Gershon and Badner

(2002) demonstrated significant support for linkage,

independently, to Bipolar disorder and to Schizophre-

nia on two chromosomal regions, 13q and 22q. A third

region, on 8p, had significant support for linkage to

Schizophrenia. Positional cloning on 13q revealed

association of Bipolar disorder with a newly discovered

gene complex, G72/G30. This same complex had been

earlier associated with Schizophrenia, and is thus a

striking confirmation of the hypothesis that the over-

lapping linkage regions for each disorder would con-

tain the same susceptibility gene for both. The

neurobiological candidate gene approach has also

yielded a recent positive association, of brain-derived

neurotrophic factor, BDNF, with Bipolar disorder.

Symposium Abstracts S35

S13.2 Status of genetics of early onset recurrent

depression

Doug Levinson

University of Pennsylvania

Abstract unavailable.

S13.3 Pivotal issues in designing and interpreting

pharmacogenetic studies of antidepressant and

mood-stabilizing drugs.

Bernard Lerera, Ronnen H. Segmana,

Fabio Macciardib

aDepartment of Psychiatry, Hadassah, Hebrew

University Medical Center, Jerusalem, IsraelbDepartment of Medical Genetics, Medical Genetics,

University of Milan, School of Medicine, Italy

Little solid evidence exists upon which clinicians

can base a choice of the appropriate drug with which to

initiate treatment of patients with affective and other

psychiatric disorders. Consequently, there is consider-

able interest in the potential clinical application of

pharmacogenetic strategies to predict response to psy-

chotropic drugs and susceptibility to adverse effects

(Lerer, 2002). Such strategies are based on the hypoth-

esis that genetic variability in the enzymes which

control bioavailability of drugs and in the target

proteins upon which drugs act, result in variable

clinical effects that can be anticipated if the patient’s

genotypic characteristics are known when treatment is

initiated. A rapidly increasing number of studies on the

pharmacogenetics of antidepressant and mood stabi-

lizing drugs are being published or are under way

(Lerer and Macciardi, 2002). Some findings, such as

the role of genetic variability in the serotonin trans-

porter promoter in the response of depressed patients

to specific serotonin reuptake blockers, have received

fairly widespread support. However, efforts to move

the field forward rapidly do not always take into

account key design issues that can pivotally influence

the results of such studies. Population stratification and

admixture can lead to spurious results if ethnic back-

ground is not taken into consideration in planning

studies and analyzing data and possible cryptic strat-

ification is not examined in ostensibly matched sam-

ples. Another key issue is the differentiation of specific

from non-specific response. This is possible when the

study includes a placebo group but is considerably

more challenging when such a group is not part of the

design. Other key considerations include the simulta-

neous typing of several polymorphisms in a single

gene, which permits the analysis of haplotypes and

affords the study considerably more power and con-

sideration of epistatic (interactive) effects among

genes. The latter issue is of great potential importance

in that separate analysis of genes that act interactively

may not demonstrate effects on the phenotype that are

present when they are analysed jointly. The study of

gene–gene interactions requires samples considerably

larger than most being studied at present. A further

issue, not generally appreciated, is that the effect of

genes on a pharmacogenetic phenotype may be age

related and, for example, be demonstrable in older but

not in younger patients. Examples that illustrate these

issues will be presented from the work of the authors

and also from the published work of other investiga-

tors and their impact on the design and interpretation

of pharmacogenetic studies of antidepressant and

mood-stabilizing drugs will be discussed.

References

Lerer, B. (Ed.), 2002. Pharmacogenetics of Psycho-

tropic Drugs. Cambridge University Press, Cambridge.

Lerer, B., Macciardi, F., 2002. Pharmacogenetics

of antidepressant and mood stabilising drugs: a review

of candidate gene studies and future research direc-

tions. Int. J. Neuropsychopharmacol. 5, 255–275.

S13.4 Pharmacogenetics in geriatric major

depression.

Greer M. Murphy, Jr.

Department of Psychiatry and Behavioral Sciences,

Stanford University School of Medicine, Stanford, CA

94305-5485

E-mail: gmurphy@stanford.edu

Background: A major unanswered clinical question

is why some patients respond to antidepressant medi-

cations whereas others do not. Pharmacogenetics seeks

Symposium AbstractsS36

to individualize drug therapy based on DNA sequen-

ces. Patients carrying genetic variants resulting in poor

metabolism of antidepressants may experience more

side effects. Polymorphisms in genes for neurotrans-

mitter receptors, reuptake transporters, and signal

transduction molecules may enhance or interfere with

antidepressant efficacy.

Methods: We performed a multicenter pharmacoge-

netic study of 246 cognitively intact patients 65 years of

age and older with major depression who were treated

with either mirtazapine or paroxetine. Medications

were administered in a double-blind design over an

8 week acute phase with a 16 week extension phase.

Mood, cognition, and medication side effects were

quantified throughout the study. Genotypes were de-

termined for APOE ?4, serotonin 2A receptor poly-

morphisms, and the 5HTTLPR polymorphism at the

SERT locus. Oligonucleotide microarrays (‘‘Gene

Chips’’) were used to determine CYP2D6 genotypes.

Results: The apolipoprotein E ?4 allele was found

to be a predictor of antidepressant efficacy in the

elderly, whereas polymorphisms in the serotonin 2A

receptor gene were associated with antidepressant side

effects. The 5HTTLPR polymorphism was also found

to affect antidepressant efficacy and side effects.

CYP2D6 is important in the metabolism of many

antidepressants, but surprisingly CYP2D6 genetic

variation had little effect on outcome.

Conclusions: Genetic variation affecting antide-

pressant pharmacodynamics may be more important

than genetic variation affecting pharmacokinetics.

(Supported by Organon, NARSAD, VA Medical

Research, and the Pritzker Network. Disclosure: Re-

search support and consulting, Organon).

S13.5 Genetics and pharmacogenetics of lithium

response

Martin Alda

Dalhousie University

The long-term treatment of bipolar disorder (BD)

relies upon an ever-increasing number of medications.

Several factors likely play a role in determining which

patient responds to which treatment; these factors

include the subtype of the illness, pharmacological

effects of the medication, as well as patient’s genotype.

The biological mechanisms of psychotropic drugs have

been proposed as keys to understanding the nature of

disorders they treat, implying that there is a tight link

between the disease and the cure. Yet, the interplay

between these two appears complex. Several studies

suggest that genetic factors independent of those in-

volved in the aetiology of BD may be involved in the

treatment response. In this presentation, we will review

the present clinical and molecular data on response to

lithium in BD.

To this date, it is still not entirely clear to what extent

are various treatment specific to individual patients.

However, the preliminary data suggest that both acute

and long-term treatments for BD might be specific. For

example, patients who respond to lithium usually do

respond poorly to the anticonvulsants carbamazepine

or valproic acid. In a study comparing responders to

lithium and lamotrigine we found robust differences

with respect to clinical presentation, presence of co-

morbid conditions, course of the illness, and the family

history. The lithium responsive probands had episodic

clinical course prior to treatment while the lamotrigine

responders had chronic course and rapid cycling. The

rates of comorbidity were higher in lamotrigine res-

ponders, both for anxiety and substance abuse. In

families of lithium responders we found a higher

lifetime prevalence of BD, while relatives of lamotri-

gine responders had higher rates of schizoaffective

disorder, major depression, and anxiety/panic attacks.

Thus, patients responsive to two different mood stabil-

isers, may represent distinct subtypes of BD.

Several research groups have investigated the role of

genetic factors in treatment response to lithium. Con-

versely, the assumption that responders to certain treat-

ments could represent genetically distinct forms of BD

led us to investigate responders to lithium with the aim

to map genes for the illness. In a series of studies we

examined a number of candidate genes and performed a

genome scan. The most promising findings are those of

association with the phospholipase C and suggestive

linkage in several chromosomal regions. The genome

scan results suggest that several genes are involved,

each playing a different role in different families. In a

recent study, we investigated the effect of lithium on

gene expression in lymphoblasts from responders to

lithium. Using cDNA arrays we identified several

genes for confirmatory studies using Northern blots,

confirming all but one of the genes identified through

Symposium Abstracts S37

cDNA array. The expression of the alpha 1B receptor

gene was 2.5-fold higher in patients and was signifi-

cantly reduced by Li in patients, but not in controls.

Lithium also decreased an expression of several other

genes such as acetylcholine receptor alpha subunit and

phosphodiesterase 4D.

In conclusion, response to treatment in BD is a

complex process involving multiple factors related to

the patient and the treatment itself. Modern neurobio-

logical techniques are now available to help in under-

standing the various facets of this process, a necessary

step in better treatment of these conditions.

Symposium 14

Clinical research on ECT

S14.1 Efficacy and cognitive effects of ECT in

community settings

Joan Prudic, Mark Olfson, Steven C. Marcus, Rice

B. Fuller, Harold A. Sackeim

Columbia University

Background: Electroconvulsive therapy (ECT) is

the most effective short-term treatment for major

depression. However, there is considerable variability

among practitioners in treatment delivery, and efficacy

and cognitive side effects of the treatment in commu-

nity settings has not been examined.

Methods: In a prospective, naturalistic study con-

ducted at seven hospitals, 347 eligible patients partic-

ipated in at least one clinical evaluation after starting

ECT. Rates of response and remission and cognitive

side effects immediately following ECT, clinical out-

come over a 24-week follow-up, and longer term

cognitive outcome at 6 months post-ECT were exam-

ined in relation to patient characteristics and treatment

variables.

Results: The sites differed markedly in patient

features and ECT administration. There were no site

differences in clinical outcome. In contrast to the 70–

90% remission rates often reported with ECT, remis-

sion rates, depending on criteria, were 30.3–46.7%.

Longer duration of episode, comorbid personality

disorder, and schizoaffective disorder were associated

with poorer outcome. Among remitters, the relapse

rate was 64.3% during the follow-up. Relapse was

more frequent in patients with psychotic depression or

comorbid Axis I or Axis II disorders. Of the patients

who did not remit following ECT, only 23.4%

achieved remission without subsequent relapse during

the 24-week follow-up. Fundamental aspects of cog-

nitive function were differentially affected by right

unilateral versus bilateral electrode placement and

sine wave compared to brief pulse ECT.

Conclusions: The remission rate with ECT in com-

munity settings is substantially below that observed in

clinical trials. Providers frequently end the ECT course

with the view that patients have shown full benefit, yet

formal assessment shows significant residual symp-

toms. Patients who do not remit following ECT have a

poor prognosis, underscoring the need to achieve

maximal improvement with this modality. These data

demonstrate that bilateral ECT is associated with per-

sistent amnesia 6 months after the conclusion of an

index course.

S14.2 Personality disorder and ECT outcomes

Roger Haskett

University of Pittsburgh

Abstract unavailable.

S14.3 The Efficacy of ECT in psychosis and

suicidality

Charles H. Kellnera, Georgios Petridesa, b,

Mustafa Husainc, Teresa Rummansd, Max Finkb,

Rebecca Knappe, Martina Muellere,

Keith Rassmussend, Kevin O’Connord,

Hilary Bernsteine, Glenn Smithd, Melanie Biggsc,

John Rushc

aNew Jersey Medical School, University of Medicine

and Dentistry of New JerseybLong Island Jewish Health System, The Zucker

Hillside Hospital, NorthshorecSouthwestern Medical Center, University of TexasdMayo Clinic and FoundationeThe Medical University of South Carolina

*C.O.R.E., Consortium for Research in ECT

Symposium AbstractsS38

In this presentation data will be presented from

Phase I an ongoing multisite, NIMH-supported trial

comparing continuation ECT versus pharmacotherapy

(lithium and nortriptyline) [the ‘C.O.R.E.’ study].

Phase I represents the acute course of ECT given prior

to randomization to the two treatment arms in Phase II.

In the C.O.R.E. study, patients with unipolar major

depression, referred for ECT, receive a standardized

course of bilateral ECT 3� per week at 1.5� seizure

threshold. A HAMD24 score of z 21 is required for

study entry and remission criteria include two consec-

utive HAMD24 ratings of V 10, with z 60% reduction

from baseline. HAMD24 ratings are performed at

baseline and 24 h after each ECT.

We present data from the first 444 patients entered

into the trial. Patient demographics are as follows: age

(meanF S.D. = 55.6F 16.8, Gender (%female) = 68.2

(303/444), psychosis status (% psychotic) = 29.7 (132/

444), race (%white) = 91.7 (407/444).

Overall remission rate was 68.5% (304/444). Remis-

sion rate in those patientswith psychotic depressionwas

75% (99/132). Patients z 65 years of age had a remis-

sion rate of 71.3% (112/157). Patients responded very

rapidly to ECT. After six treatments (2 weeks) 34.9%

(155/444) of patients had reached remission criteria.

ECT resolved suicidality very rapidly. Eighty-one per-

cent of patients with high baseline suicide ratings were

no longer suicidal after six treatments.

These data, from one of the largest ECT datasets in

the modern era, confirm the high and rapid efficacy of

ECT in major depression, particularly in those with

psychotic depression. Suicidality was also rapidly

relieved in ECT.

S14.4 Redesigning the ECT stimulus: Towards a

more controlled and focal treatment

Harold Sackeim

New York State Psychiatric Institute

Abstract unavailable.