behaviour might contribute to the assessment of suicide
risk, and translational results of neurobiological research
are beginning to be realised. Childhood adversity and a
familial history of suicide are associated with a
susceptibility to suicide, and a personal history of non-
fatal suicide attempts might show the existence of such a
diathesis to suicidal behaviour and thus need to be
assessed. This assessment can include measurement of
biological responses to ecologically relevant stimuli
related to mood, suicidal ideation, or emotional pain, and
characterisation of decision-making processes.
Treatment or prevention?
New treatments are emerging from early clinical trials for
the acute management of suicide risk. In the past, a
course of electroconvulsive therapy might have had a
short-term life-saving effect in suicidal patients with
severe or drug-resistant depression. More recently, a
subanaesthetic dose of ketamine was reported to have an
acute antisuicide effect that lasted for about 1 week.
Because of the association between suicidal behaviour and
serotonergic and noradrenergic disturbances, and because
most antidepressant drugs enhance serotonergic or
noradrenergic function or both, an antisuicidal effect
would be expected for these drugs. Direct evidence of such
an effect has not been shown, although one randomised
clinical trial of suicidal patients showed that a selective
serotonin-reuptake inhibitor had a better antidepressant
effect and greater reduction in suicidal ideation than did a
noradrenergic–dopaminergic drug, and that this thera­
peutic advantage is greatest in patients with the most
severe suicidal ideation.
A prosuicidal effect for anti­
depressant drugs in younger adults, adolescents, and
children, for which direct evidence was scarce, seems less
likely after findings from recent studies.
The probability
of serotonergic antidepressant drugs having beneficial
effects is supported by the discovery of plausible
underlying neurobiological mechanisms. Mouse studies
show that stress impairs neurogenesis in the one locus in
the human brain in which adult neurogenesis takes
place—the dentate gyrus of the hippocampus. In mice,
this effect can be blocked or corrected by giving them a
selective serotonin-reuptake inhibitor and is partly
mediated via hippocampal 5-HT
from post-mortem studies of the human brain confirm
that neurogenesis takes place in adulthood and that it
might have been enhanced in individuals with depression
who were treated with a selective serotonin-reuptake
inhibitor shortly before their death.
Investigators have also debated the effects of mood
stabilisers, lithium, and antipsychotic drugs on risk of
suicidal behaviour. The possibility that antiepileptic drugs
have adverse effects on risk of suicide might have been
A large amount of evidence now exists
that lithium reduces the risk of suicide and non-fatal
suicidal behaviour.
In addition to its serotonin-
enhancing effects, lithium has antiapoptotic effects, and
its use in long-term treatment might be associated with
increased grey-matter volume, effects that might decrease
suicide risk.
Clozapine is an antipsychotic drug that
blocks D2 and 5-HT
receptors and, in patients with
schizophrenia, it has a protective effect against suicidal
behaviour compared with olanzapine.
The antisuicidal
effects of lithium are independent of its effectiveness as a
mood stabiliser and the antisuicidal effect of clozapine is
independent of its effectiveness as an antipsychotic drug.
Therefore, both drug types could exert their antisuicidal
effect on some aspect of the diathesis.
The future
Since clinical predictors of suicide risk have poor
effectiveness, genomics and brain imaging are the most
promising new directions for detection of patients at
high risk for suicide. Because serotonergic abnormalities
can be detected in living patients, brain imaging might
help to identify people at risk of a more lethal suicide
attempt or suicide. Suicide prevention depends on
detection of such patients. Since a third of people who
die from suicide die from their first attempt, the aim for
prevention is to detect these patients before any attempt
is made.
Neuroimaging could delineate brain regions
and networks involved in suicide risk and could be a way
to track the effect of interventions, which target such
specific brain regions and neural networks. Given the
association between biochemical disturbances and
lethality of suicidal behaviour, further studies might
Figure 2:
Regions shown in neuroimaging and post-mortem studies to have structural or functional changes
associated with suicidal behaviour
Dotted lines indicate cortical-to-cortical connections; blue lines and arrows indicate the serotonergic pathways
from the dorsal raphe. Adapted fromTanaka and colleagues,
with permission fromMacmillan Publishers.
Medial prefrontal cortex and
anterior cingulate cortex
Ventral and dorsal
Globus pallidus,
substantia nigra
pars reticulate
Dorsal raphe
Orbitofrontal cortex
Dorsolateral prefrontal cortex
Ventral and dorsal
Inferior parietal cortex
1...,17,18,19,20,21,22,23,24,25,26 28,29,30,31,32,33,34,35,36,37,...56