G Model

ENCEP-837; No. of Pages 8 ARTICLE IN PRESS

L’Encéphale xxx (2016) xxx–xxx

Disponible en ligne sur

ScienceDirect
www.sciencedirect.com

Review of literature

Lithium and suicide prevention in bipolar disorder

Lithium et prévention du suicide dans le trouble bipolaire
V. Benard a,b , G. Vaiva a,b , M. Masson c,d , P.A. Geoffroy e,f,g,h,∗,i
a
Centre hospitalier universitaire de Lille (CHRU), 59000 Lille, France
b
Université de Lille, 59000 Lille, France
c
Clinique médicale du Château-de-Garches, 92380 Garches, France
d
Service S.H.U., hôpital Sainte-Anne, 1, rue Cabanis, 75004 Paris, France
e
Inserm, U1144, 75006 Paris, France
f
Université Paris Descartes, UMR-S 1144, 75006 Paris, France
g
Université Paris Diderot, Sorbonne Paris Cité, UMR-S 1144, 75013 Paris, France
h
Pôle de psychiatrie et de médecine addictologique, groupe hospitalier Saint-Louis–Lariboisière–F.-Widal, AP–HP, 75475 Paris cedex 10, France
i
Fondation FondaMental, 94000 Créteil, France

a r t i c l e i n f o a b s t r a c t

Article history: Introduction. – Bipolar disorder (BD) is a severe and recurrent psychiatric disorder. The severity of prog-
Received 4 December 2014 nosis in BD is mainly linked to the high rate of suicide in this population. Indeed, patients with BD commit
Accepted 27 March 2015 suicide 20 to 30 times more frequently than the general population, and half of the BD population with
Available online xxx
an early age of onset have a history of suicide attempt. International therapeutic guidelines recommend
lithium (Li) as the first-line treatment in BD for its prophylactic action on depressive or manic episodes.
Keywords: In addition, Li is the only mood stabilizer that has demonstrated efficacy in suicide prevention. This effect
Bipolar disorder
of Li is unfortunately often unknown to psychiatrists. Thus, this review aims to highlight evidence about
Lithium carbonate
Lithium salts
the preventive action of Li on suicide in BD populations.
Suicidal ideation Methods. – We conducted a literature search between April 1968 and August 2014 in PubMed database
Suicide attempt using the following terms: “lithium” AND “suicide” OR “suicidality” OR “suicide attempt”.
Suicide Results. – As confirmed by a recent meta-analysis, many studies show that Li has a significant effect
Mood stabilizer treatment on the reduction of suicide attempts and deaths by suicide in comparison to antidepressants or other
mood-stabilisers in BD populations. Studies have demonstrated that long-term treatment with Li reduces
suicide attempts by about 10% and deaths by suicide by about 20%. The combination of Li and an antide-
pressant could reduce suicidal behaviours by reducing suicidal ideation prior to depressive symptoms. It
appears crucial for Li efficacy in suicide prevention to maintain the Li blood concentrations in the efficient
therapeutic zone and to instate long-term Li treatment. The “impulsive-aggressive” endophenotype is
associated with suicide in BD. The specific action of Li on the 5-HT serotoninergic system could explain the
specific anti-suicidal effects of Li via the modulation of impulsiveness and aggressiveness. Furthermore,
genetic variants of the glycogen synthase kinase 3␣/␤ (GSK3␣ and ␤; proteins inhibited by Li) seem to
be associated with more impulsiveness in BD populations.
Conclusion. – The anti-suicidal effect of Li has been very well demonstrated. By its specific action on the
serotoninergic system, treatment with Li significantly reduces “impulsive-aggressive” behaviour which
is a vulnerability factor common to suicide and BD. Long-term appropriately modulated treatment with
Li seems to have considerable impact on the reduction of suicidal behaviours, suicidal ideation and death
by suicide in the BD population.
© L’Encéphale, Paris, 2015.

r é s u m é

Mots clés : Introduction. – Le trouble bipolaire (TB) est une maladie psychiatrique sévère et fréquente. La sévérité du
Trouble bipolaire pronostic du TB est largement associée au taux élevé de suicides survenant dans cette population. En effet,
Lithium carbonate les suicides sont 20 à 30 fois plus fréquents chez les patients avec TB que dans la population générale, et

∗ Corresponding author.
E-mail address: pierre.a.geoffroy@gmail.com (P.A. Geoffroy).

http://dx.doi.org/10.1016/j.encep.2016.02.006
0013-7006/© L’Encéphale, Paris, 2015.

Please cite this article in press as: Benard V, et al. Lithium and suicide prevention in bipolar disorder. Encéphale (2016),
http://dx.doi.org/10.1016/j.encep.2016.02.006
G Model
ENCEP-837; No. of Pages 8 ARTICLE IN PRESS
2 V. Benard et al. / L’Encéphale xxx (2016) xxx–xxx

Sels de lithium un patient sur deux avec un âge de début précoce du TB présente un antécédent de tentative de sui-
Suicidalité cide. Dans les recommandations thérapeutiques internationales du TB, le lithium (Li) est le traitement
Tentative de suicide
de première intention pour son efficacité dans le traitement et la prévention des épisodes maniaques ou
Suicide
dépressifs. De plus, le Li est le seul traitement stabilisateur de l’humeur ayant déjà prouvé son efficacité
Traitement stabilisateur de l’humeur
dans la prévention du suicide. Néanmoins, cet effet du Li est souvent méconnu des psychiatres. Ainsi, cette
revue de la littérature à pour objectif la synthèse de l’efficacité du Li dans la prévention des conduites
suicidaires chez les patients souffrant de TB.
Méthode. – Nous avons réalisé une recherche bibliographique de la littérature scientifique publiée entre
avril 1968 et août 2014 à l’aide de la base de données PubMed, et en utilisant les mots clés suivants : «
lithium » AND « suicide » OR « suicidality » OR « suicide attempt ».
Résultats. – De nombreuses études, et une méta-analyse récente, confirment l’efficacité du Li sur la réduc-
tion des tentatives de suicide et la réduction des suicides chez des patients avec TB, en comparaison aux
autres stabilisateurs de l’humeur ou aux antidépresseurs. Il est démontré que l’utilisation au long cours du
Li permet de réduire de 10 % les tentatives de suicide et de 20 % le taux de suicide dans cette population.
L’adjonction du Li à un traitement antidépresseur en cours permettrait de réduire les comportements
suicidaires et ceci en diminuant les idées suicidaires avant que n’apparaisse la réduction des symptômes
dépressifs. Pour obtenir une efficacité dans la prévention du suicide, il apparaît indispensable de main-
tenir le traitement par Li au long cours et à des doses efficaces, permettant une lithiémie dans la zone
thérapeutique. De manière intéressante, il a été démontré que l’endophénotype « impulsif-agressif » est
largement associé au suicide dans les TB. L’action spécifique du Li sur le système sérotoninergique 5-HT
pourrait expliquer l’effet spécifique du Li dans la réduction des conduites suicidaires, ceci via la régulation
de l’impulsivité et de l’agressivité. Enfin, des variants génétiques de la glycogène synthase kinase 3␣/␤
(GSK3␣ et ␤, protéines inhibées par le Li) semblent être associés à une augmentation de l’impulsivité chez
les patients souffrant de TB.
Conclusion. – L’effet anti-suicidaire du Li est largement démontré par la littérature scientifique. Par son
action spécifique sur le système sérotoninergique, le traitement par Li permet de réduire significativement
les comportements « agressifs-impulsifs », qui sont des facteurs de vulnérabilité communs entre le suicide
et le TB. Le traitement par Li, au long cours et en zone thérapeutique efficace, semble avoir un impact
considérable dans la réduction des conduites suicidaires, des idées suicidaires et des suicides chez les
patients souffrant de TB.
© L’Encéphale, Paris, 2015.

1. Introduction
Bipolar disorder (BD) is a severe, cyclic, psychiatric illness
entailing recurrences of mood episodes referred to as “manic” or
depressive“. BD is a frequent illness, affecting 1% of the general
population if Types I and II BD are considered, with a prevalence
reaching 4.4% if wider spectrum BD is considered [1]. It is a severe
pathology, to the extent that it is the fourth-ranking contributor
to the general morbidity burden worldwide (whatever the age)
among mental, neurological and substance use disorders [2]. Unlike
the general population, BD patients have not seen an increase in life
expectancy over recent decades. It is indeed on average 20 years
below that of the general population [3]. The poor prognosis for this
illness is partly linked to high suicide levels, with half the subjects
with an early onset of the illness attempting suicide [4]. This high
prevalence of suicide attempt, which remains the strongest pre-
dictive factor for actual suicide, contributes to placing BD patients
among those with the highest suicide risk. Indeed, their suicide
rate is 20 to 30 times that of the general population [5]. Accord-
ing to WHO data, more than 800,000 individuals die from suicide
every year in the world, amounting to one every forty seconds [6].
Although 75% of suicides occur in countries with low to medium
income, in Europe the suicide rate was 11.4/100,000 in 2012, that
is to say slightly above the world average [6]. Suicide is the main
cause of death in the 15–29 age group, and the evaluation of suicide
risk thus appears as a major challenge for prevention, and all the
more so in case of BD [7].
Lithium (Li) has for decades been considered as the first-line
reference treatment for BD, and is included in all international rec-
ommendations [8]. It is the first mood-stabiliser treatment with
efficient therapeutic and preventive action in both manic and
depressive recurrent episodes [9], unlike the majority of treatments
that act on only one pole of bipolar disturbances [10]. Li also has an
effect on deaths from suicide among patients with a mood disorder
[11]. Thus Li remains the first-line treatment in BD, and appears to
be the only treatment to date that decreases suicide attempts and
suicide in these patients [8]. This emblematic anti-suicide effect
of Li is however not well known by psychiatric physicians. We
therefore set out to perform a synthesis of the scientific data in the
international literature on the prevention of suicide by Li among BD
patients. It can indeed be thought that enhanced awareness of the
anti-suicidal effect of this mood-stabiliser treatment could improve
care and also prognosis among our patients, reducing deaths from
suicide.
2. Methods
2.1. Definitions
Suicide can be described as intentionally causing one’s own
death, an act that is initiated and accomplished by a person who
has awareness of the fatal outcome and intends to succeed [12].
Self-harm acts with suicidal intent include intentional intake of
substances or medications, self-mutilation, scarification, indepen-
dently from motivation, from which the intention and desire to live
are either completely absent, or present only to varying degrees
[13,14].
2.2. Scientific literature search strategy
We conducted a bibliographic search between April 1968 and
August 2014 in the PubMed database for international scien-
tific publications on the medications for preventing suicide and
assessing the efficacy of Li in the reduction of suicidal behaviours.
We used the following MeSh terms: “lithium” AND “suicide” OR
“suicidality” OR “suicide attempt”. We chose to include only clini-
cal trials and the most recent literature reviews. Scientific articles
in English and French were included.

Please cite this article in press as: Benard V, et al. Lithium and suicide prevention in bipolar disorder. Encéphale (2016),
http://dx.doi.org/10.1016/j.encep.2016.02.006
G Model
ENCEP-837; No. of Pages 8 ARTICLE IN PRESS
V. Benard et al. / L’Encéphale xxx (2016) xxx–xxx
3. Results
3.1. Results of the literature search
Using the MeSh terms we found 597 results, including 195
reviews of the literature and 39 clinical trials in the recent literature.
We then selected scientific publications that explored the supe-
riority of Li in effectively reducing suicidal behaviours among BD
patients, and also publications where the main aim was to provide
an overview of the prevention of suicidal behaviours among BD
patients, pinpointing of the implication of Li (Table 1).
3.2. The anti-suicidal action of Li: clinical data
Suicide is at present one of the main causes of death among
patients with BD. In addition, the frequency of suicidal behaviours
that do not terminate in death, such as scarification and other forms
of self-mutilation without clear suicidal intent, but liable to lead to
a genuine suicide attempt, is high in this population [14]. This is
why, alongside suicide attempts, self-harm gestures are a useful
indicator that should not be neglected in the evaluation of suicide
risk [13].
Concerning BD and suicide, greater efficacy of Li in the reduc-
tion of suicide risk has been demonstrated [15]. It has indeed been
shown that patients with BD treated with Li have a five times lesser
risk of exhibiting suicidal behaviours (RR = 4.91, 3.82–6.31) [16]. In
addition, with long-term Li treatment, the risk of suicide attempt
and suicide were reduced by around 80%, and the lethal nature
of the means used for suicide attempt was reduced [16]. Several
studies also underlined the usefulness of long-term Li treatment,
estimating that it reduced the risk of suicide attempt ten times and
that of actual suicide 20 times. In addition, in cases of treatment dis-
continuation, suicidal behaviours [17,18] are reported. It appears
that Li could have an anti-suicide effect in long-term prescription,
but not in acute phase [15,19,20].
Oquendo et al. compared the anti-suicide effects of Li versus
Valproate among BD patients in depressive or mixed phase, with a
history of suicide attempt [21]. There was no death from suicide,
but there were more frequent suicidal behaviours in the Valproate
group than in the Li group. The authors nevertheless concluded that
the anti-suicide effect of Li was smaller than expected. They also
demonstrated that discontinuing treatment appeared to increase
suicide risk, which underlines the importance of therapeutic con-
tinuity in preventing suicide [21].
It has also been shown that the number of deaths from suicide
was larger in women with BD than in men (22% versus 15%), and
that 79% of suicides occurred among patients with type I BD. In addi-
tion, the suicide risk was greater under Divalproate treatment than
under Li (RR = 2.7), which shows the superiority of this treatment
for managing suicide risk [15].
Concerning suicidal behaviours, it has been shown that the pre-
vention of self-harm was more successful under Li than under
Carbamazepine (OR = 0.14) [22].
A meta-analysis of 43 randomised controlled trials confirmed
these results, focusing on the efficacy of long-term Li compared to
placebo in the reduction of suicide risk among patients with a mood
disorders according to DSM-IV criteria [22]. Cipriani et al., using
a sensitivity analysis, thus evidenced a protective effect of Li for
actual suicide and suicide attempt in comparison with a placebo:
OR 0.13 for the reduction in the number of suicides and OR 0.38 for
the reduction in deaths from whatever cause [23,24]. This means
a clear reduction in suicide risk with Li treatment compared to
placebo in this population.
Other studies considered suicide attempt rates before and after
Li treatment among patients with mood disorders [25,26]. They
were divided into three groups according to therapeutic response
Please cite this article in press as: Benard V, et al. Lithium and suicide prevention in bipolar disorder. Encéphale (2016),
http://dx.doi.org/10.1016/j.encep.2016.02.006
3
to Li after a mood episode (good, partial, and poor responders).
These studies showed a significant reduction in suicide attempts
after Li treatment in both responder and non-responder groups,
demonstrating an anti-suicide effect independently from its thy-
moregulatory action, whatever the level of severity of the BD
[26]. Subsequently, further studies confirmed that the anti-suicide
effect of Li is not correlated to its thymoregulatory proper-
ties [27,28]. It was indeed shown in these studies that certain
antidepressants and mood-stabilisers possessing similar speed
of action and efficacy to Li did not have the anti-suicide effect
of Li.
Li can also potentialise the effects of antidepressants in reduc-
ing suicidal ideation and behaviours. This was shown in a recent
randomised trial which hypothesised that the antidepressant
Citalopram, used among patients consulting for major depressive
episode, would reduce depressive symptoms and hence suicidal
ideation and behaviours, and that the co-prescription of Li would
be associated with a greater decrease in suicidal behaviours [29].
The study showed that the reduction in suicidal ideation and
behaviour preceded a reduction in depressive symptoms, leading
to the conclusion that reducing depression is not a primary, neces-
sary condition for altering suicidal behaviours. There was however
no significant difference in the total scores on the suicide eval-
uation scale S-STS (Sheehan-Suicidality Tracking Scale) between
Citalopram + Placebo and Citalopram + Li. However, if the thera-
peutic range of action of Li is considered, it has been shown that
there was a reduction of more than 50% on the S-STS score for the
Citalopram + Li group with lithium levels at or above 0.5 mEq/L.
Thus it appears necessary to define a therapeutic range on the
basis of plasma levels of lithium 12 hours from intake (≥ 0.5 mEq/L
and < 1.2 mEq/L) to obtain effective Li treatment in the reduction
of suicidal symptoms and suicide risk. It can be recalled that the
use of antidepressants is not advised among BD patients, and
requires extreme precaution (in particular on account of switch-
ing to manic episodes and the precipitation of thymic relapse)
[30]. In addition, the study results obtained in unipolar popula-
tions may not be transposable to BD, but they do at least underline
the trans-nosographic and fairly specific nature of Li efficacy in the
prevention of suicidal behaviours.
Work is at present underway to determine whether adding Li
treatment to standard therapy could be an effective strategy to
reduce suicide risk in the long term in patients with mood disorders
[31]. The results of this work could outline a strategy for increas-
ing Li over the long term in the prevention of suicidal behaviour
and suicide among patients with BD. Finally, Li appears efficacious
both in the reduction of deaths from suicide and in the reduction of
deaths from any cause among patients with a mood disorder [32].
3.3. Hypotheses on the anti-suicide effect of Lithium
A genetic vulnerability for suicidal behaviours is now acknowl-
edged, and the study of this vulnerability can entail the study of
suicide endophenotypes [33,34]. An endophenotype is a measur-
able component linking the illness considered and the genotype
of individuals with that illness. It thus gathers environmental,
genetic, hereditary and biological elements, enabling the study of
the neurobiological underpinnings of the pathology concerned [35].
Endophenotypes of the aggressive and impulsive type appear to
be strongly associated with suicide [16,32,36]. Thus, by acting on
these two types of behaviour, Li reduces both the number of sui-
cide attempts and the number of suicides among patients with BD
[37,38] (Fig. 1). This vulnerability to suicide combines environmen-
tal risk factors with the genetic factors, and it has thus been shown
that the main individual clinical risk factors are impulsiveness and
aggressiveness, alongside emotions linked to distress and despair
[39,40].
G Model
ENCEP-837; No. of Pages 8 ARTICLE IN PRESS
4 V. Benard et al. / L’Encéphale xxx (2016) xxx–xxx

Table 1
Lithium compared to other mood-stabilisers or antidepressants in the prevention of suicide among patients with mood disorders.

Study Year Number of subjects Follow-up Treatment Suicide attempt and suicide rates

Müller-Oerlinghausen 1992 68 in all 8 yrs Lithium 1 yr prophylaxis Number of suicides

et al. [17] 55 with Lithium Lithium n = 1/55
continuously Lithium discontinued n = 4/13
13 with Lithium Number suicidal behaviours
discontinued Lithium discontinued n = 11/13
Bocchetta et al. [18] 1998 100 10 yrs Lithium Number of suicides n = 10/100
Lithium discontinued n = 9/10
24 times > lithium taken continuously
Number suicidal behaviours: lithium
taken continuously 5–6 times < lithium
discontinued
Bauer et al. [24] 2000 29 20 weeks Lithium 0.5–1.0 mEq/L versus Number of suicides
placebo Lithium n = 0/14
Placebo n = 1/15
Decrease in suicides
Lithium OR = 0.14 (0.0–7.3)
Ahrens et 2001 167 BD Lithium Decrease in suicide risk
Müller-Oerlinghausen 3 groups according to Group 1 > Group 2 80% > Group 3 50%
[26] response to Lithium, Group Number suicide attempts/yr after and
1: 45 excellent responders before lithium
Group 2: 81 intermediate Group 1: RR = 0.10 vs 0.33
responders Group 2: RR = 0.06 vs 0.27
Group 3: 41 non-responders Group 3: RR = 0.02 vs 0.26
Müller-Oerlinghausen 2001 IGSL study n = 5616 10–15 yrs Lithium versus Suicide risk
[25] Meta-analysis 28 studies carbamazepine or RR = 8,6
Review n = 1700 amitriptiline Without lithium > Lithium
Post-hoc analysis n = 285 Number of suicides:
Lithium
n = 0/146
Carbamazepine:
n = 9/139
Bauer et al. [28] 2003 803 patients with MDE 1980–2002, Lithium versus placebo Decrease in suicide risk
Review 22 yrs In addition to antidepressant Enhancement of antidepressant effect
45% lithium
18% placebo
P < 0.001
Baldessarini et al. [16] 2006 Meta-analysis 1970–2006, Lithium 18 months versus no Decrease in suicide risk
Review 31 studies 36 yrs lithium Lithium RR = 4.91 (3.82–6.31)
85,229 patients
Lauterbach et al. [23] 2008 167 52 weeks Lithium 0.6–0.8 mEq/L versus Number of suicides
placebo Lithium n = 0/84
Placebo n = 3/83
Decrease in suicides
Lithium OR = 0.13 (0.01–1.27)
Khan et al. [29] 2011 80 MDE or BD 6 weeks Citalopram 20 mg/d + placebo Reduction score S-STS scale
40 in group 1: versus citalopram Group 1 S = 12 ± 3.7
citalopram + placebo 20 mg/d + lithium 300 mg/d Groupe2 S = 10.8 ± 2.3
40 group 2: Plasma levels: 0.4–0.6 mEq/L Previous suicide attempt
citalopram + lithium Group 1: 65%
Post-hoc analysis Group 2: 67.5%
Group 3: group 2 with Suicide attempt in previous month
lithium plasma level Group 1: 10%
monitoring Group 2: 2.5%
Suicide attempt in previous year
Group 1: 27.5%
Group 2: 15%
Oquendo et al. [21] 2011 98 BD 2.5 yrs Lithium: plasma levels No death from suicide
46 Lithium 0.4–1.0 mEq/L 495 days 18 suicide attempts
48 Valproate versus valproate: n = 14, women
depakinaemia 45–125 ␮g/mL Lithium n = 6
550 days Valproate n = 8
45 suicidal events
n = 35, men/women
Lithium n = 16
Valproate n = 19
No difference between groups
Cipriani et al. [31] 2013 210 12 months Reference Previous study: hypothesis rates for
In progress treatment + placebo versus suicide and attempted suicide 35%
reference treatment + lithium Present hypothesis: clinical advantage
150–300 mg/d for plasma with Lithium rate10%
levels 0,4–1,0 mEq/L
Foster et al. [15] 2013 1994–2001, Lithium versus no treatment Decrease in suicide risk
Review 17 yrs or divalproate Lithium vs no treatment RR = 4.91
(3.82–6.31)
Lithium vs divalproate RR = 2.7 (1.1–6.3)

BD: Bipolar disorder; MDE: Major Depressive Episode; NA: Not available; OR: Odds-Ratio; RR: Relative Risk; S-STS: Sheehan-Suicidality Tracking Scale.

Please cite this article in press as: Benard V, et al. Lithium and suicide prevention in bipolar disorder. Encéphale (2016),
http://dx.doi.org/10.1016/j.encep.2016.02.006
G Model
ENCEP-837; No. of Pages 8 ARTICLE IN PRESS
V. Benard et al. / L’Encéphale xxx (2016) xxx–xxx
Fig. 1. Neurobiological hypothesis for serotoninergic dysfunction in impulsive-aggressive behaviours, suicidal vulnerability and bipolar disorder: comorbid and co-
vulnerability links.
One recent hypothesis is the implication of Li in the nor-
malisation of impaired decisional processes, a major feature of
vulnerability towards suicide [41]. It is indeed interesting to
note that impaired decision-making processes could be linked
to emotional dysfunction, and thus constitute a neuropsycholo-
gical risk factor for suicidal behaviours [42,43]. In the cognitive
and emotional domains, there could be a hyperactivity of the
hypothalamus-pituitary-adrenal axis and cortical orbito-frontal
and dorsolateral anomalies [44,45]. In addition, recent on-going
work has shown better verbal learning performances among
patients with BD with a history of serious suicide attempt, and
alterations in these performances under Li [46]. Consequently, ver-
bal acquisition appears to be a cognitive function that is closely
associated with serious suicide attempt. A new hypothesis is the
implication of Li in certain alterations of verbal cognitive abili-
ties, which could enable a reduction in suicidal behaviours among
patients with BD [41].
In this review several studies evidenced the essential role of
the serotoninergic system and of impulsive-aggressive behaviours
in both BD and the risk of suicidal acting-out, which suggests
a common neurobiological and/or genetic component in these
pathologies and endophenotypes [47] (Fig. 1). It is indeed now well
known that Li facilitates the 5-HT serotoninergic neurotransmis-
sion involved in impulsive-aggressive behaviours, which therefore
points to anti-suicide action by Li. In pharmacological terms, the
serotoninergic system is implicated in affective regulation, more
particularly the 5-HT1B receptors. In the earlier literature, it has
already been shown that 5-HT1B serotoninergic receptors are a
specific molecular target for Li, and that this interaction is more
specifically linked to the therapeutic effects of Li than to its adverse
effects [48] (Fig. 1). It has also been evidenced in animals that Li has
an impact on secondary behavioural and neurochemical recovery
following stress implicating the serotoninergic system [49].
Knowledge about the molecular and pharmacogenetic actions
of Li is constantly improving [50,51]. There are indeed interactions
between Li and the metabolism of serotonin, magnesium and phos-
phoinositide, an inhibition of GSK3, action on circadian patterns,
and chronic neurotrophic and neuro-protector effects resulting
from the inhibition of the glutamatergic system [52–56].
Numerous studies have shown that mortality from suicide is
markedly higher in presence of a psychiatric disorder, and in
Please cite this article in press as: Benard V, et al. Lithium and suicide prevention in bipolar disorder. Encéphale (2016),
http://dx.doi.org/10.1016/j.encep.2016.02.006
5
particular in cases of BD or recurrent depression [57]. Impulsive-
ness is a major prognostic factor in BD, and a major factor in suicide
risk, and it has been evidenced that Li has a specific impact on
the reduction of this impulsiveness [58]. In genetic terms, it has
been shown that Li inhibits the ˛ and ˇ isoenzymes of glycogen
synthase kinase 3 (GSK3 ˛ and ˇ) and that variants of the GSK3 ˇ
gene could be associated with more impulsive behaviours among
BD patients [59]. In addition, variants of the GSK3 ˇ gene are also
thought to be associated with the risk of developing type I BD, with
the age of onset among women, and with therapeutic response
to Li [60]. Thus impulsiveness could be a shared comorbidity and
co-vulnerability factor between suicide and BD. The implication
of the 5-HTTLPR polymorphisms (the serotonin transporter gene)
in the association between impulsiveness and suicidal behaviours
has been established independently from socio-demographic and
clinical factors [61]. It has also been shown that the S-allele of the 5-
HTTLPR genotype was associated with suicidal behaviours but not
with BD, which opens up a new area of research [62] (Fig. 2). Indeed,
by highlighting molecular and genetic anomalies, in particular for
the independent variants of the 5-HTTLPR genotype, such as the
presence, the length and the transcription of the S-allele impli-
cated in suicidal behaviours, it is possible to suggest a different
approach to the therapeutic response to Li in suicidal behaviours
among patients with or without BD or recurrent depression.
Further to this, part of the anti-suicide effect of Li could also
be mediated by an anti-inflammatory action. There is indeed
increasing evidence for the role of inflammatory phenomena in the
physiopathology of both BD and suicidal behaviour [63,64]. It has
also been observed that there is an increase in inflammatory mark-
ers in behaviours associated with suicide such as aggressiveness,
impulsiveness and depression [64]. At molecular level, it is thus
possible that the anti-suicide effect operates via the well-known,
specific inhibition of GSK3 ␤ by Li [56], which leads to a reduction
in inflammation in BD patients [65].
Overall, Li treatment exhibits its superiority over other mood-
stabilisers in the prevention of suicide among BD patients. The
main biological hypothesis is a specific action on the serotonin-
ergic system, differing from the well-known effect of present-day
antidepressants, and enabling a reduction in impulsive-aggressive
behaviours, which appear to have a considerable impact in suicidal
behaviours, much greater than that of depressive symptoms.
G Model
ENCEP-837; No. of Pages 8 ARTICLE IN PRESS
6 V. Benard et al. / L’Encéphale xxx (2016) xxx–xxx
Fig. 2. Reduction in mortality from suicide among patients with BD: The impact of Lithium in the genetic vulnerability of GSK3 implicated in impulsive behaviours.
4. Discussion
With a view to improving the prevention of suicidal behaviours
and reducing suicide mortality among BD patients, Li, already a
first-line treatment in BD for its prophylactic properties in com-
bating thymic relapse, is the only thymo-regulator to have shown
efficacy in the prevention of suicide risk. Indeed, in comparison
to other moods stabilisers, Li has shown greater efficacy in the
prevention of relapse in either pole of BD, enhancing the reduc-
tion in death from suicide in this population [50]. Nevertheless,
adverse side effects have restricted medical prescription (renal
failure, hyperthyroidism, thyroid goitre, diabetes insipidus, etc.)
[66,67]. In addition, the arrival of new antipsychotic molecules in
the treatment of BD has contributed to a reduction in Li prescrip-
tions, despite high levels of proof of its action on thymic relapse
and its unique effect on suicidal behaviours.
Although the toxic does of Li is close to the therapeutic dose, it
is possible to avoid overdose, and hence adverse side effects, via a
dose-effect that can be monitored in plasma concentrations, thus
enabling the dose to be adapted. In addition, the biological surveil-
lance of Li treatment is fairly simple (THS, PTH, calcaemia, lithaemia
and renal function), enabling therapeutic efficacy to be optimised
and an improvement in clinical tolerance [68]. A recent study has
demonstrated the non-dangerous nature of Li prescription at low
doses and in the long term for renal function in patients over 65
Please cite this article in press as: Benard V, et al. Lithium and suicide prevention in bipolar disorder. Encéphale (2016),
http://dx.doi.org/10.1016/j.encep.2016.02.006
[69]. In the earlier literature, it was shown that low levels of lithium
are present in water in certain regions [70,71]. It is thus relevant to
hypothesise that a prescription at low or very low doses of Li could
provide well-tolerated, effective prevention of suicide risk in BD
patients. This is confirmed by the fact that the doses of Li used were
variable depending on the therapeutic effect sought, with a thera-
peutic target in the lower lithium plasma levels appearing sufficient
to prevent suicide [72].
The main comorbidity, linked to suicide rates and BD, concerns
aggressiveness and impulsiveness. The main hypotheses for the
action of Li in the reduction of suicidal behaviours are based on the
specific role of Li in serotoninergic activity, mainly via its action
on 5-HT1 receptors, which enables an increase in serotonin lev-
els in the body resulting in a decrease in aggressive and impulsive
symptoms, and hence in suicidal behaviours and thymic signs [61].
In addition, via its action on the serotoninergic system,
Li has a specific action on endophenotype dimensions of the
aggressiveness-impulsiveness type, which considerably reduces
suicidal behaviours and the number of suicide deaths among BD
patients [37]. Several studies have shown the importance of long-
term Li treatment with appropriate management of Li plasma levels
in the prevention of suicide risk in this population.
There are compelling arguments for the efficacy of Li on
impulsive and aggressive endophenotypes in BD patients. A
recent communication has evidenced the value of a dimensional
G Model
ENCEP-837; No. of Pages 8 ARTICLE IN PRESS
V. Benard et al. / L’Encéphale xxx (2016) xxx–xxx
prescription strategy for Li in order to obtain optimum preven-
tion of suicide and aggressive and impulsive behaviours [41]. Along
these lines, this team backed up the maintenance of Li treatment
among BD patients presenting suicide risk but who are not respon-
ding to its thymoregulatory effect. In this situation, maintaining Li
for its anti-suicide effect despite its lack of thymoregulatory effect
appears an interesting approach, alongside treatment with a mood-
stabiliser.
5. Conclusion
Because of its anti-suicide and thymoregulatory properties, Li
is the choice treatment to reduce suicidal behaviours, impulsive-
aggressive behaviours and thymic relapse in patients with BD.
Consequently, Li appears as an essential treatment enabling a
reduction in suicide mortality among these patients. Li acts on the
serotoninergic system, which explains its efficacy on the both the
thymic and the suicidal aspects. Indeed, by increasing serotonin-
ergic activity, Li provides a regression of impulsiveness, which is a
major co-vulnerability factor between BD and suicide.
Disclosure of interest
The authors declare that they have no competing interest.
References
[1] Merikangas KR, Akiskal HS, Angst J, et al. Lifetime and 12-month prevalence
of bipolar spectrum disorder in the National Comorbidity Survey replication.
Arch Gen Psychiatry 2007;64(5):543–52.
[2] Collins PY, Patel V, Joestl SS, et al. Grand challenges in global mental health.
Nature 2011;475(7354):27–30.
[3] Phillips ML, Kupfer DJ. Bipolar disorder diagnosis: challenges and future direc-
tions. Lancet 2013;381(9878):1663–71.
[4] Geoffroy PA, Etain B, Scott J, et al. Reconsideration of bipolar disorder as
a developmental disorder: importance of the time of onset. J Physiol Paris
2013;107(4):278–85.
[5] Pompili M, Gonda X, Serafini G, et al. Epidemiology of suicide in bipolar disor-
ders: a systematic review of the literature. Bipolar Disord 2013;15(5):457–90.
[6] WHO. Suicide prevention (SUPRE); 2014 http://www.who.int/mental health/
prevention/suicide/suicideprevent/en/.
[7] While D, Bickley H, Roscoe A, et al. Implementation of mental health
service recommendations in England and Wales and suicide rates,
1997–2006: a cross-sectional and before-and-after observational study.
Lancet 2012;379(9820):1005–12.
[8] Gershon S, Chengappa KNR, Malhi GS. Lithium specificity in bipolar illness: a
classic agent for the classic disorder. Bipolar Disord 2009;11(Suppl. 2):34–44.
[9] Severus E, Taylor MJ, Sauer C, et al. Lithium for prevention of mood episodes
in bipolar disorders: systematic review and meta-analysis. Int J Bipolar Disord
2014;2(1):15.
[10] Geddes JR, Miklowitz DJ. Treatment of bipolar disorder. Lancet
2013;381(9878):1672–82.
[11] Yerevanian BI, Choi YM. Impact of psychotropic drugs on suicide and suicidal
behaviors. Bipolar Disord 2013;15(5):594–621.
[12] Niedzwiedz C, Haw C, Hawton K, et al. The definition and epidemiology of
clusters of suicidal behavior: a systematic review. Suicide Life Threat Behav
2014;44(5):569–81.
[13] Caine DE. Self-harm behaviour: rethinking physical and mental health. Lancet
2012;380(9853):1536–8.
[14] Victor SE, Klonsky DE. Correlates of suicide attempts among self-injurers: a
meta-analysis. Clin Psychol Rev 2014;34(4):282–97.
[15] Foster TJ. Suicide prevention as a prerequisite for recovery from severe mental
illness. Int J Psychiatry Med 2013;46(1):15–25.
[16] Baldessarini RJ, Tondo L, Davis P, et al. Decreased risk of suicides and attempts
during long-term lithium treatment: a meta-analytic review. Bipolar Disord
2006;8(5 Pt 2):625–39.
[17] Müller-Oerlinghausen B, Müser-Causemann B, Volk J. Suicides and parasuicides
in a high-risk patient group on and off lithium long-term medication. J Affect
Disord 1992;25(4):261–9.
[18] Bocchetta A, Ardau R, Burrai C, et al. Suicidal behavior on and off lithium prophy-
laxis in a group of patients with prior suicide attempts. J Clin Psychopharmacol
1998;18(5):384–9.
[19] Depression in adults: the treatment and management of depression in adults.
Guidance and guidelines. NICE; 2009 http://www.nice.org.uk/guidance/CG90.
[20] Cipriani A, Barbui C, Butler R, et al. Depression in adults: drug and physical
treatments. Clin Evid 2011 [pii: 1003].
[21] Oquendo MA, Galfalvy HC, Currier D, et al. Treatment of suicide attempters
with bipolar disorder: a randomized clinical trial comparing lithium and
Please cite this article in press as: Benard V, et al. Lithium and suicide prevention in bipolar disorder. Encéphale (2016),
http://dx.doi.org/10.1016/j.encep.2016.02.006
7
valproate in the prevention of suicidal behavior. Am J Psychiatry 2011;168(10):
1050–6.
[22] Cipriani A, Hawton K, Stockton S, et al. Lithium in the prevention of sui-
cide in mood disorders: updated systematic review and meta-analysis. BMJ
2013;346:f3646.
[23] Lauterbach E, Felber W, Müller-Oerlinghausen B, et al. Adjunctive lithium
treatment in the prevention of suicidal behaviour in depressive disor-
ders: a randomised, placebo-controlled, 1-year trial. Acta Psychiatr Scand
2008;118(6):469–79.
[24] Bauer M1, Bschor T, Kunz D, et al. Double-blind, placebo-controlled trial of the
use of lithium to augment antidepressant medication in continuation treat-
ment of unipolar major depression. Am J Psychiatry 2000;157(9):1429–35.
[25] Müller-Oerlinghausen B. Arguments for the specificity of the antisuicidal effect
of lithium. Eur Arch Psychiatry Clin Neurosci 2001;251(Suppl. 2):II72–5.
[26] Ahrens B, Müller-Oerlinghausen B. Does lithium exert an independent antisui-
cidal effect? Pharmacopsychiatry 2001;34(4):132–6.
[27] Goodwin GM, Bowden CL, Calabrese JR, et al. A pooled analysis of 2 placebo-
controlled 18-month trials of lamotrigine and lithium maintenance in bipolar
I disorder. J Clin Psychiatry 2004;65(3):432–41.
[28] Bauer M, Forsthoff A, Baethge C, et al. Lithium augmentation therapy
in refractory depression-update 2002. Eur Arch Psychiatry Clin Neurosci
2003;253(3):132–9.
[29] Khan A, Khan SRF, Hobus J, et al. Differential pattern of response in mood symp-
toms and suicide risk measures in severely ill depressed patients assigned to
citalopram with placebo or citalopram combined with lithium: role of lithium
levels. J Psychiatr Res 2011;45(11):1489–96.
[30] Geoffroy PA, Bellivier F, Henry C. [Treatment of manic phases of bipo-
lar disorder: critical synthesis of international guidelines]. Encéphale
2014;40(4):330–7.
[31] Cipriani A, Girlanda F, Agrimi E, et al. Effectiveness of lithium in subjects with
treatment-resistant depression and suicide risk: a protocol for a randomised,
independent, pragmatic, multicentre, parallel-group, superiority clinical trial.
BMC Psychiatry 2013;13:212.
[32] Cipriani A, Pretty H, Hawton K, et al. Lithium in the prevention of suicidal
behavior and all-cause mortality in patients with mood disorders: a systematic
review of randomized trials. Am J Psychiatry 2005;162(10):1805–19.
[33] Labonte B, Turecki G. The epigenetics of suicide: explaining the bio-
logical effects of early life environmental adversity. Arch Suicide Res
2010;14(4):291–310.
[34] Mann JJ, Currier DM. Stress, genetics and epigenetic effects on the neurobiology
of suicidal behavior and depression. Eur Psychiatry 2010;25(5):268–71.
[35] Gottesman II, Gould TD. The endophenotype concept in psychiatry: etymology
and strategic intentions. Am J Psychiatry 2003;160(4):636–45.
[36] Pabis DJ, Stanislav SW. Pharmacotherapy of aggressive behavior. Ann Pharma-
cother 1996;30(3):278–87.
[37] Kovacsics CE, Gottesman II, Gould TD. Lithium’s antisuicidal efficacy: eluci-
dation of neurobiological targets using endophenotype strategies. Annu Rev
Pharmacol Toxicol 2009;49(1):175–98.
[38] Ernst CL, Goldberg JF. Antisuicide properties of psychotropic drugs: a critical
review. Harv Rev Psychiatry 2004;12(1):14–41.
[39] Courtet P, Gottesman II, Jollant F, et al. The neuroscience of suicidal behaviors:
what can we expect from endophenotype strategies? Transl Psychiatry 2011;1
[pii: e7].
[40] Pihlakoski L, Sourander A, Aromaa M, et al. The continuity of psychopathology
from early childhood to preadolescence: a prospective cohort study of 3–12-
year-old children. Eur Child Adolesc Psychiatry 2006;15(7):409–17.
[41] Courtet P, Guillaume S, Sénèque M, et al. Les indications actuelles du
lithium : la prévention du risque suicidaire. Ann Med Psychol Rev Psychiatr
2014;172(3):182–5.
[42] Jollant F, Bellivier F, Leboyer M, et al. Impaired decision making in suicide
attempters. Am J Psychiatry 2005;162(2):304–10.
[43] Richard-Devantoy S, Berlim MT, Jollant F. A meta-analysis of neuropsycholo-
gical markers of vulnerability to suicidal behavior in mood disorders. Psychol
Med 2014;44(08):1663–73.
[44] Tyano S, Zalsman G, Ofek H, et al. Plasma serotonin levels and suicidal behavior
in adolescents. Eur Neuropsychopharmacol 2006;16(1):49–57.
[45] Jollant F, Lawrence NL, Olié E, et al. The suicidal mind and brain: a review
of neuropsychological and neuroimaging studies. World J Biol Psychiatry
2011;12(5):319–39.
[46] Wingo AP, Wingo TS, Harvey PD, et al. Effects of lithium on cognitive perfor-
mance: a meta-analysis. J Clin Psychiatry 2009;70(11):1588–97.
[47] Swann AC, Dougherty DM, Pazzaglia PJ, et al. Increased impulsivity associated
with severity of suicide attempt history in patients with bipolar disorder. Am J
Psychiatry 2005;162(9):1680–7.
[48] Massot O, Rousselle JC, Fillion MP, et al. 5-HT1B receptors: a novel target
for lithium. Possible involvement in mood disorders. Neuropsychopharmacol
1999;21(4):530–41.
[49] Perveen T, Haider S, Mumtaz W, et al. Attenuation of stress-induced behavioral
deficits by lithium administration via serotonin metabolism. Pharmacol Rep PR
2013;65(2):336–42.
[50] Can A, Schulze TG, Gould TD. Molecular actions and clinical pharmacogenetics
of lithium therapy. Pharmacol Biochem Behav 2014;123:3–16.
[51] Geoffroy PA, Bellivier F, Leboyer M, et al. Can the response to mood stabilizers
be predicted in bipolar disorder? Front Biosci 2014;6:120–38.
[52] Mota de Freitas D, Castro MMCA, Geraldes CFGC. Is competition between
Li+ and Mg2+ the underlying theme in the proposed mechanisms for the
G Model
ENCEP-837; No. of Pages 8 ARTICLE IN PRESS
8 V. Benard et al. / L’Encéphale xxx (2016) xxx–xxx
pharmacological action of lithium salts in bipolar disorder? Acc Chem Res
2006;39(4):283–91.
[53] Berridge MJ. Inositol trisphosphate and calcium signalling mechanisms.
Biochim Biophys Acta 2009;1793(6):933–40.
[54] Urs NM, Snyder JC, Jacobsen JPR, et al. Deletion of GSK3 in D2R-expressing
neurons reveals distinct roles for arrestin signaling in antipsychotic and lithium
action. Proc Natl Acad Sci U S A 2012;109(50):20732–7.
[55] Chiu C-T, Wang Z, Hunsberger JG, et al. Therapeutic potential of mood sta-
bilizers lithium and valproic acid: beyond bipolar disorder. Pharmacol Rev
2013;65(1):105–42.
[56] Geoffroy PA, Etain B, Sportiche S, Bellivier F. Circadian biomarkers in patients
with bipolar disorder: promising putative predictors of lithium. Int J Bipolar
Disord 2014;2:5, http://dx.doi.org/10.1186/2194-7511-2-5.
[57] Stanley B, Jones J. Risk for suicidal behaviour in personality disorders. In:
Wasserman D, Wasserman C, editors. Oxford textbook of suicidology and sui-
cide prevention. Oxford University Press; 2009. http://www.oxfordmedicine.
com/view/10.1093/med/9780198570059.001.0001/med-9780198570059-
chapter-39.
[58] Kovacsics CE, Gottesman II, Gould TD. Lithium’s antisuicidal efficacy: eluci-
dation of neurobiological targets using endophenotype strategies. Annu Rev
Pharmacol Toxicol 2009;49:175–98.
[59] Jiménez E, Arias B, Mitjans M, et al. Association between GSK3b gene
and increased impulsivity in bipolar disorder. Eur Neuropsychopharmacol
2014;24(4):510–8.
[60] Lin Y-F, Huang M-C, Liu H-C. Glycogen synthase kinase 3b gene polymorphisms
may be associated with bipolar I disorder and the therapeutic response to
lithium. J Affect Disord 2013;147(1–3):401–6.
[61] Malloy-Diniz LF, Neves FS, de Moraes PHP, et al. The 5-HTTLPR polymorphism,
impulsivity and suicide behavior in euthymic bipolar patients. J Affect Disord
2011;133(1–2):221–6.
Please cite this article in press as: Benard V, et al. Lithium and suicide prevention in bipolar disorder. Encéphale (2016),
http://dx.doi.org/10.1016/j.encep.2016.02.006
[62] Neves FS, Silveira G, Romano-Silva MA, et al. Is the 5-HTTLPR polymor-
phism associated with bipolar disorder or with suicidal behavior of bipolar
disorder patients? Am J Med Genet B Neuropsychiatr Genet 2008;147B(1):
114–6.
[63] Stuart MJ, Baune BT. Chemokines and chemokine receptors in mood disorders,
schizophrenia, and cognitive impairment: a systematic review of biomarker
studies. Neurosci Biobehav Rev 2014;42:93–115.
[64] Vargas HO, Nunes SOV, Pizzo de Castro M, et al. Oxidative stress and lowered
total antioxidant status are associated with a history of suicide attempts. J Affect
Disord 2013;150(3):923–30.
[65] Beurel E, Jope RS. Inflammation and lithium: clues to mechanisms contributing
to suicide-linked traits. Transl Psychiatry 2014;4:e488.
[66] Llorca P-M, Courtet P, Martin P, et al. Dépistage et prise en charge du trouble
bipolaire : méthodologie. Encéphale 2010;36:S79–85.
[67] Oruch R, Elderbi MA, Khattab HA, et al. Lithium: a review of pharmacology, clin-
ical uses, and toxicity. Eur J Pharmacol 2014 [Internet] http://www.linkinghub.
elsevier.com/retrieve/pii/S0014299914004932.
[68] McKnight RF, Adida M, Budge K, et al. Lithium toxicity profile: a systematic
review and meta-analysis. Lancet 2012;379(9817):721–8.
[69] Aprahamian I, Santos FS, dos Santos B, et al. Long-term, low-dose lithium
treatment does not impair renal function in the elderly: a 2-year randomized,
placebo-controlled trial followed by single-blind extension. J Clin Psychiatry
2014;75(7):e672–8.
[70] Blüml V, Regier MD, Hlavin G, et al. Lithium in the public water supply and
suicide mortality in Texas. J Psychiatr Res 2013;47(3):407–11.
[71] Kapusta ND, Mossaheb N, Etzersdorfer E, et al. Lithium in drinking water and
suicide mortality. Br J Psychiatry 2011;198(5):346–50.
[72] Malhi GS, Tanious M, Das P, et al. Potential mechanisms of action of
lithium in bipolar disorder. Current understanding. CNS Drugs 2013;27(2):
135–53.