Structural imaging studies of suicidal behavior, using MRI, have focused on changes in white and gray matter.

Six studies demonstrated an association between suicidal behavior and white matter hyperintensities, i.e., deep white matter hyperintensities, periventricular hyperintensities, or subcortical matter hyperintensities (Ahearn et al. 2001; Ehrlich et al. 2004, 2005; Pompili et al. 2007, 2008; Serafini et al. 2011). Noteworthy is the study by Serafini et al. (2011), in which the relationship was studied between affective temperamental profiles, white matter hyperintensities, and suicidal behavior in patients with mood disorders. They found that patients with higher dysthymia and lower hyperthymia were more likely to have white matter hyperintensities and recent suicide attempts.

Monkul et al. (2007) (compared fronto-limbic brain structures between females diagnosed with a unipolar mood disorder and a history of one or more suicide attempt(s), unipolar females without such a history and female healthy controls. The presence of a history of suicide attempt(s) was associated with smaller bilateral orbitofrontal gray matter volumes and larger right amygdala volumes. There were no differences in gray matter volumes between unipolar patients without a history of suicide attempt(s) and healthy controls.

Three studies investigated gray and white matter in the brains of patients suffering from schizophrenia in association with suicidal behavior. Male patients with a history of suicide attempt(s), compared to those without such a history, showed a significant reduction in gray matter density in the left superior temporal gyrus and the left orbitofrontal cortex (Aguilar et al. 2008). Rusch et al. (2008) found significantly larger bilateral inferior frontal (and posterior orbital) white matter volumes in patients with a history of suicide attempts as compared to patients without such a history and to healthy controls. No other significant white or gray matter volume differences were observed. Spoletini et al. (2011) studied selected subcortical regions and found a significant increase in right amygdala volumes in association with a history of suicide attempts in individuals suffering from schizophrenia when compared to such individuals without a history of suicide attempts and healthy controls.

Numbers of suicidal behaviors and pituitary gland volume were correlated in young patients diagnosed with borderline personality disorder and with minimal exposure to treatment Jovev et al. (2008). Studying the association between a history of attempted suicide and anterior corpus callosum volumes, Matsuo et al. (2010) found no differences between bipolar patients with a history of suicide and those without such a history. Cyprien et al. (2011) however, studied the association between corpus callosum size and suicidal behavior in a large general population sample of elderly persons. While controlling for age, gender, childhood trauma, head trauma, and total brain volume, the area of the posterior third of the corpus callosum was significantly smaller in suicide attempters than in affective controls and healthy controls. Diminished interhemispheric connectivity may thus play a role in the pathophysiology of suicidal behavior.

Caplan et al. (2010) studied frontotemporal volumes in pediatric epilepsy patients according to the presence or absence of suicidal ideation. Suicidal ideation was associated with smaller right orbitofrontal gyrus white matter volumes and larger left temporal lobe gray matter volumes.

A history of suicide attempts in elderly male patients with late-onset depression was associated with decreased volumes of gray matter and white matter in the frontal, temporal, and parietal regions and the insula, lentiform nucleus, midbrain, and the cerebellum when compared to healthy controls. A particularly marked regional volume reduction was noticed in the dorsal medial prefrontal cortex (Hwang et al. 2010). Goodman et al. (2011) studied the volume of the anterior cingulate gyrus in adolescent patients with borderline personality disorder and comorbid depression and in healthy controls. They found smaller gray but not white matter volumes in the patients than in the controls, while, in the patient group, a greater number of suicide attempts was associated with smaller volumes of the anterior cingulate region.

Baldacara et al. (2011) found no effect of a history of suicide attempts on cerebellar volumes in euthymic bipolar type 1 patients.

Benedetti et al. (2011) studied gray matter volumes in currently depressed bipolar patients, some of whom had a history of suicide attempts, while others were treated with lithium. Suicide attempters showed reduced gray matter volumes in several brain areas, including the dorsolateral prefrontal cortex, orbitofrontal cortex, anterior cingulate, superior temporal cortex, parieto-occipital cortex, and basal ganglia. Noteworthy is the finding that long-term lithium treatment was associated with increased gray matter volumes in the same areas in which suicide attempters showed decreased gray matter volumes.

Wagner et al. (2011, 2012) used a different approach to study the vulnerability to suicidal behavior by comparing voxel-based morphometric properties in the brains of depressed patients at high risk of suicide (as indicated by a personal or familial history of suicidal behavior), depressed patients without such a risk and matched healthy controls. Patients with a high risk of suicide showed significantly decreased gray matter density in a fronto-striato-limbic network when compared to the healthy controls and in caudate and rostral anterior cingulate cortex when compared to nonhigh risk patients. The same research group investigated prefrontal cortical thickness in similar study groups (Wagner et al. 2011, 2012). Patients with high risk of suicide showed a significantly thinner cortex in the left dorsolateral, ventrolateral prefrontal cortex, and the anterior cingulate in contrast to nonhigh risk patients. Taken together, the findings provide evidence for structural brain alterations in depressed patients at high risk of suicide in the fronto-cingulo-striatal network that is strongly involved in reward processing and emotional control.

Within a group of persons with borderline personality disorders, those with a history of suicide attempts showed significantly decreased gray matter in the left insula when compared to nonattempters. High-lethality attempters had significant decreases in the right temporal gyrus, right orbitofrontal gyrus, right insular cortex, and right parahippocampal gyrus when compared to low-lethality attempters (Soloff et al. 2012).

Gray matter changes in brain regions in association with suicidal behavior in psychotic patients were studied by Giakoumatos et al. (2013)

Compared to nonattempters, attempters had significantly less gray matter volume in bilateral inferior temporal and superior temporal cortices, left superior parietal, thalamus and supramarginal regions, right insula, and superior frontal and rostral middle frontal regions. Among attempters, a history of high lethality attempts was associated with significantly smaller volumes in the left lingual gyrus and right cuneus. Compared to nonattempters, low-lethality attempters had significant decreases in the left supramarginal gyrus, thalamus, and the right insula.

Oquendo et al. (2013) published a pivotal functional imaging study of the involvement of the serotonergic neurotransmission system in suicidal behavior using ¹⁸F-FDG PET. Depressed high-lethality and low-lethality suicide attempters were scanned after a single-blind placebo and after the serotonin agonist fenfluramine hydrochloride administration on a second day. Secondary and proportional to the postsynaptic serotonin receptor stimulation, the anterior pituitary gland releases prolactin in the circulation following fenfluramine administration. Depressed high-lethality suicide attempters showed relative hypometabolism compared to low-lethality attempters in the ventral, medial, and lateral prefrontal cortex. This difference was more pronounced after fenfluramine administration. Lethality of the attempt appeared to be inversely correlated with metabolism in the ventromedial prefrontal cortex after challenge with fenfluramine. A lower rCMRglu correlated with higher lethality of suicidal behavior. They found a lower rCMRglu in high versus low-lethality suicide attempters. This hypometabolism in frontal cortex structures was related to the degree of suicide intent and impulsivity and not to depression.

Leyton et al. (2006) studied regional serotonin synthesis in the brain with PET and α-(¹¹C)-Methyl-L-Tryptophan trapping in high-lethality suicide attempters and in healthy controls. Suicide attempters showed reduced serotonin synthesis in the orbital and ventromedial prefrontal cortices, and α-(¹¹C)-Methyl-l-tryptophan trapping in these regions correlated negatively with suicide intent. Low serotonin synthesis in the prefrontal cortex may thus lower the threshold for suicidal behavior.

Four studies of suicidal behavior using SPECT have been published. Using ¹²³I-5-I-R91150, Audenaert et al. (2001) (studied 5-HT_2a receptor binding in the brains of recent suicide attempters and healthy controls. They found a significantly reduced binding index in the frontal cortex in the patient group. The binding index was significantly lower in the deliberate self-injury patients compared to the deliberate self-poisoning subjects. In a split-dose ^99mTc-ECD SPECT activation paradigm using a verbal fluency task, Audenaert et al. (2002) further studied recent suicide attempters. Attempters showed a blunted increase during verbal fluency tasks in perfusion in the left gyrus frontalis inferior, right gyrus parietalis inferior and bilateral gyrus cinguli anterior, the left and right gyrus temporalis medius, and the hypothalamic region.

Brain ^99mTc HMPAO SPECT scans of individuals who committed suicide between 10 days and 36 months after the SPECT scan were compared to the SPECT scans of depressed and healthy controls in two studies in a partly overlapping study population (Amen et al. 2009; Willeumier et al. 2011). Resting-state activity was lower in the suicide victims than in the controls in the premotor and primary motor cortex, corpus callosum, subgenual cingulate, and anterodorsal cortex. A significant area of low activity was the nucleus accumbens, extending into the ventromedial prefrontal cortex and the left and right putamen. During the Continuous Performance Test, the baseline perfusion deficits were attenuated in the depressed group and exacerbated in the suicide group during concentration.

Marchand et al. (2012) studied functional connectivity characteristics in association with attempted suicide. A network involving the bilateral striatum and anterior cortical midline structures was found to be associated with depressive symptom severity. Current suicidal ideation was associated with a similar but less extensive circuit in the left hemisphere. A distinct striatal motor/sensory network was associated with self-harm behaviors. Thus, a striatal-anterior cortical midline circuit likely plays a significant role in the expression of depressive symptoms and suicidal ideation. In contrast, a striatum-motor/sensory cortex network may be a trait marker of suicide-related behaviors.

Two studies have assessed brain correlates of mental pain in relation to suicidal behavior. Using ^99mTc-ECD SPECT, van Heeringen et al. (2010) (examined the effect of psychological pain severity on resting-state activity in patients with major depressive disorder (MDD) by comparing patients who scored high in psychological pain to those who scored low. Mental pain (measured with the Orbach and Mikulincer Mental Pain Scale), suicidal ideation (measured using the Hamilton Rating Scale for Depression), hopelessness (measured using Beck’s Hopelessness Scale), and regional cerebral blood flow as measured with single photon emission computed tomography were assessed in depressed individuals. Levels of mental pain were significantly and positively associated with suicidal ideation and levels of hopelessness. When compared with patients with low levels of mental pain, those with high levels of mental pain showed relatively increased perfusion in the right dorsolateral prefrontal cortex, occipital cortex and inferior frontal gyrus and in the left inferior temporal gyrus, and relatively decreased perfusion at the medulla. Reisch et al. (2010) tested the hypothesis that negative emotions experienced as psychological pain would exhibit decreased neural activity in the frontal cortex. Studying women who had attempted suicide in the two months prior to the study, the authors reported decreased prefrontal activity. Thus, Reisch et al. (2010) reported decreased PFC activity (left BA46, right BA10), whereas van Heeringen et al. (2010) (reported increased activity in the right PFC (BA9, BA44). Both studies used the Orbach and Mikulincer Mental Pain questionnaire to assess psychological pain, which has been shown to be reliable and have a high degree of validity, but it should be noted that different reference conditions were used.

Two other imaging studies investigated hopelessness, and its association with serotonergic disturbances. Using SPECT, van Heeringen et al. (2003) (studied the binding index of 5-HT_2a receptors in the frontal cortex of attempted suicide patients and normal controls using [I-123]5-I-R91150, a highly selective 5-HT_2a receptor ligand. Moreover, they measured personality characteristics (using Cloninger’s Temperament and Character Inventory) and levels of hopelessness (using Beck’s Hopelessness Scale), and studied the association between 5-HT_2a receptor binding index, hopelessness and these personality dimensions. When compared to normal controls, attempted suicide patients had a significantly lower binding potential of frontal 5-HT_2a receptors, a higher level of hopelessness, a higher score on the temperament dimension harm avoidance and lower scores on the character dimensions self-directedness and cooperativeness. A significant correlation was found between harm avoidance, hopelessness and binding index in the population of suicide attempters. Lower central serotonergic function, hopelessness and harm avoidance are interrelated phenomena, which may increase the probability of the occurrence of attempted suicide.

Meyer et al. (2004) used PET in order to measure the relationship between brain serotonin transporter binding potential with carbon 11-labeled DASB and negativistic dysfunctional attitudes during depression. Dysfunctional attitudes are negatively biased assumptions and beliefs regarding oneself, the world, and the future. Depressed subjects with highly negativistic dysfunctional attitudes had significantly higher 5-HTT binding potential compared with healthy subjects in brain regions mainly sampling serotonergic nerve terminals (prefrontal cortex, anterior cingulate, thalamus, bilateral caudate, and bilateral putamen). Increased 5-HTT binding potential was strongly associated with more negativistic dysfunctional attitudes in brain regions primarily sampling serotonergic nerve terminals (prefrontal cortex, anterior cingulate, thalamus, caudate, and putamen). The magnitude of regional 5-HTT binding potential thus may provide a vulnerability to low levels of extracellular serotonin and symptoms of extremely negativistic dysfunctional attitudes.

Yurgelun-Todd et al. (2011) studied the association between frontal white matter integrity and measures of impulsivity and suicidal ideation in male veterans with mild traumatic brain injury (n = 15) using DTI. Total and right cingulum fractional anisotropy values (reflecting fiber density, axonal diameter, and myelination in white matter) were found to correlate with current suicidal ideation. Mahon et al. (2012) used DTI to study white matter in the brains of bipolar patients with a history of suicide attempts, bipolar patients without such a history and healthy controls. Lower fractional anisotropy values were found in the left orbital frontal white matter in association with a history of suicide attempts, correlating with levels of motor impulsivity.