Table 46.1 Open in new tab Neuroimaging...

Table 46.1

Neuroimaging studies of suicide attempters

Author/study design	Study participants	Imaging modality/tracer, ligand	Major findings (as stated in paper)	Limitations
Structural MRI studies
Ahearn 2001 Case–control, retrospective	20 outpatients with MDD with SA history. 20 MDD without SA	T1- and T2-weighted images	Higher number of subcortical grey matter hyperintensities in those with history of SA.	No control for history of drug, alcohol use or method of SA. Comorbid diagnoses not specified.
Ehrlich 2005 Cross-sectional, retrospective	102 inpatients with MDD, 62/102 with SA history.	T2 weighted	Higher prevalence of PVH but not DWMH in patients with history of SA when compared to those without SA history. Severity of PVH a significant predictor of past SA.	No information on methods of SA. Three different scanners used. Group level differences in white matter hyperintensities (DWMH+PVH) between SA and non-attempters not significant.
Monkul 2007 Case–control, Brain volume	7 female MDD with history of SA 10 female MDD without history of SA 17 HC	T1 or T2? ROI: OFC, cingulate, amygdala, hippocampus	SA with decreased bilateral OFC grey matter volumes, and increased right amygdala volumes.	No information on methods or lethality of SA, or time from last attempt. 2 different image analysis programs were used.
Blood flow studies
Audenaert 2002 Activation Study: Verbal fluency test	20 MDD patients with recent SA (less than 7 days prior) 20 HC	SPECT/^{b, 99m}Tc-ECD ROI: none specified	During CFT, SA had blunted perfusion of left inferior PFC, R inferior parietal gyrus, L and R ACC. During LFT, SA had blunted perfusion in L and R med temporal gyrus, R ACC, and R hypothalamus.	No psychiatric control group. Patients on psychotropic drugs not excluded. Lower IQ in SA group. P values for regions of activation non-significant. Pixel by pixel analysis showed difference but not cluster level differences.
Fountoulakis 2004 Cross-sectional, retrospective	50 MDD patients Subgroups: 18 SA vs. 32 Non-SA; 17 No current thoughts of death 23 No specific thoughts of death 10 With suicidal thoughts	SPECT/^{a, 99m}Tc HMPAO ROI: cerebellum, thalamus, caudate nucleus, GP, FL, PL, TL, OL.	After Bonferroni correction, no difference in rCBF between SA versus non-SA. No difference between three subgroups based on thoughts of death.	Included patients with GAD and panic disorder, disorders which may involve the serotonin system. Did not study inferior frontal lobe.
Functional MRI studies
Jollant 2008	All euthymic male subjects 13 SA with MDD 14 MDD 16 HC	Event-related fMRI on 1.5 T magnet	Compared to MDD alone, SA with MDD showed increased activity in R lateral orbitofrontal cortex (BA 47) in response to angry faces Decreased activity in R superior frontal gyrus (BA 6) in response to angry faces Greater activity in R cerebellum in response to mild angry expressions. No differences in response to neutral or happy faces Greater activity in the R anterior cingulate gyrus (BA 32), extending to the medial frontal gyrus (BA 10) in response to mild happy versus neutral faces.	No description of comorbidities such as anxiety or cluster B personality disorders, common in SA and which may impact affect processing Time since most recent SA not given
Serotonin studies
Audenaert 2001 Case–control 5HT2a receptor study	9 SA (<8 days prior) Axis I diagnoses: 4 MDD 4 adjustment disorder 1 brief psychotic disorder 13 HC	SPECT, resting/ ^d[¹²³I]5-I-R91150 ROI: bilateral frontal cortex, OFC, dorsolateral PFC	After Bonferroni correction, frontal 5HT 2a BP lower in SA versus HC; 0.39 vs. 0.68.	No psychiatric control. No intent measure in SA. Multiple psychiatric diagnoses represented. Small groups, 3 patients in deliberate self-injury group. Use of SEM instead of SD in analysis. SA by overdose in 5/9 patients. Regions of PFC not specified. Multiple post-hoc comparisons with small sample sizes.
Van Heeringen 2003 Case–control Oquendo 2003 Cross-sectional. Activation study, fenfluramine challenge	See Audenaert 2001 25 MDD with SA (mean 4 years prior) 16 high-lethality SA 9 low-lethality SA	See Audenaert 2001 PET/^cFDG ROI: anterior cingulate and medial frontal gyri; anterior cingulate and right superior frontal gyrus.	Lower 5HT 2a receptor antagonist binding in PFC in SA (140.7) vs non- SA (168). Lower rCMRglu in ventral, medial, lateral PFC in high-lethality vs low-lethality SA. Lower VM PFC activity associated with lower impulsivity, intent, and lethality. Pre-fenfluramine: Lower rCMRglu in bilateral superior frontal, ACC, and inferior frontal gyri in high-lethality group. Post-fenfluramine: Lower rCMRglu in same areas above and superior frontal gyri.	See above. No psychiatric control group. No HC group. Comorbid diagnoses not discussed. Methods of suicide attempt not discussed.
Lindstrom 2004 Case–control 5HT transporter	12 SA with high intent Axis I diagnoses: 6 mood disorder 1 social phobia 3 adjustment disorder 12 HC	SPECT/cocaine analogue, ^f[¹²³I]-β CIT ROI: cerebellum and whole brain	No significant differences in whole brain BP of 5HTT. No significant differences in BP for violent SA vs. nonviolent SA vs controls.	Heterogeneous diagnoses. SPECT automatic scaling- increases error in ROI. No regional anatomic structures specified. β-CIT also binds the dopamine transporter.
Leyton 2006 Case–control Tryptophan uptake	10 high lethality SA (mean 14.7days) Methods: 8 by overdose, 1 by hanging, 1 by jumping Axis I diagnoses: 2 Mood disorder 6 Substance abuse 16 HC	PET/ α[¹¹C]Methyl-L-tryptophan ROI: medial OFG, left OFG, medial PFG	SA with decreased normalized tryptophan trapping in OFC and VM PFC. Increased tryptophan trapping seen in L thalamus, R paracentral lobule, L middle occipital cortex, hippocampal gyrus. Suicide intent negatively correlated with tryptophan trapping in OFG and R medial PFG.	No psychiatric controls. Unclear which toxins used in SA. Multiple comorbid diagnoses, including substance abuse. Utility of labelled tryptophan as a marker of serotonin synthesis has been questioned. In planned comparisons of trapping rate constants in VOI, main effect of group not significant.
Cannon 2006 Case–control 5HT transporter	18 BD, current MDE (8 SA history) 37 HC	PET/ ^g[¹¹C] DASB ROI: thalamus, striatum, insula, midbrain, sgACC, pgACC, DCC, PCC	In SA, increased pgACC binding and decreased midbrain binding compared to 10 without SA. Compared to controls, increased binding in thalamus, insula, DCC, and increased in midbrain.	Use of SEM distorts the small effect size. SPM reported with uncorrected p-values. No arterial sampling documented. Comorbid diagnoses include OCD, panic attacks. No account for cerebellum uptake (in bipolar disorder).
Oquendo 2007 Case–control 5HT transporter	18 BD, current MDE (9 SA history) 41 HC	PET/^h[¹¹C] McNeil 5652 ROI: midbrain, amygdala, hippocampus, thalamus, putamen, ACC	No difference in BP between SA and non-SA. Bipolar patients had lower 5HTT BP in midbrain, amygdala, hippocampus, thalamus, putamen and ACC. No correlation between depression severity and BP.	11% of patients with remission of symptoms during washout is concerning for change in synapses. Multiple comorbid diagnoses (OCD, PTSD, GAD, binge eating, simple phobia), some of which involve the serotonin system.

Author/study design	Study participants	Imaging modality/tracer, ligand	Major findings (as stated in paper)	Limitations
Structural MRI studies
Ahearn 2001 Case–control, retrospective	20 outpatients with MDD with SA history. 20 MDD without SA	T1- and T2-weighted images	Higher number of subcortical grey matter hyperintensities in those with history of SA.	No control for history of drug, alcohol use or method of SA. Comorbid diagnoses not specified.
Ehrlich 2005 Cross-sectional, retrospective	102 inpatients with MDD, 62/102 with SA history.	T2 weighted	Higher prevalence of PVH but not DWMH in patients with history of SA when compared to those without SA history. Severity of PVH a significant predictor of past SA.	No information on methods of SA. Three different scanners used. Group level differences in white matter hyperintensities (DWMH+PVH) between SA and non-attempters not significant.
Monkul 2007 Case–control, Brain volume	7 female MDD with history of SA 10 female MDD without history of SA 17 HC	T1 or T2? ROI: OFC, cingulate, amygdala, hippocampus	SA with decreased bilateral OFC grey matter volumes, and increased right amygdala volumes.	No information on methods or lethality of SA, or time from last attempt. 2 different image analysis programs were used.
Blood flow studies
Audenaert 2002 Activation Study: Verbal fluency test	20 MDD patients with recent SA (less than 7 days prior) 20 HC	SPECT/^{b, 99m}Tc-ECD ROI: none specified	During CFT, SA had blunted perfusion of left inferior PFC, R inferior parietal gyrus, L and R ACC. During LFT, SA had blunted perfusion in L and R med temporal gyrus, R ACC, and R hypothalamus.	No psychiatric control group. Patients on psychotropic drugs not excluded. Lower IQ in SA group. P values for regions of activation non-significant. Pixel by pixel analysis showed difference but not cluster level differences.
Fountoulakis 2004 Cross-sectional, retrospective	50 MDD patients Subgroups: 18 SA vs. 32 Non-SA; 17 No current thoughts of death 23 No specific thoughts of death 10 With suicidal thoughts	SPECT/^{a, 99m}Tc HMPAO ROI: cerebellum, thalamus, caudate nucleus, GP, FL, PL, TL, OL.	After Bonferroni correction, no difference in rCBF between SA versus non-SA. No difference between three subgroups based on thoughts of death.	Included patients with GAD and panic disorder, disorders which may involve the serotonin system. Did not study inferior frontal lobe.
Functional MRI studies
Jollant 2008	All euthymic male subjects 13 SA with MDD 14 MDD 16 HC	Event-related fMRI on 1.5 T magnet	Compared to MDD alone, SA with MDD showed increased activity in R lateral orbitofrontal cortex (BA 47) in response to angry faces Decreased activity in R superior frontal gyrus (BA 6) in response to angry faces Greater activity in R cerebellum in response to mild angry expressions. No differences in response to neutral or happy faces Greater activity in the R anterior cingulate gyrus (BA 32), extending to the medial frontal gyrus (BA 10) in response to mild happy versus neutral faces.	No description of comorbidities such as anxiety or cluster B personality disorders, common in SA and which may impact affect processing Time since most recent SA not given
Serotonin studies
Audenaert 2001 Case–control 5HT2a receptor study	9 SA (<8 days prior) Axis I diagnoses: 4 MDD 4 adjustment disorder 1 brief psychotic disorder 13 HC	SPECT, resting/ ^d[¹²³I]5-I-R91150 ROI: bilateral frontal cortex, OFC, dorsolateral PFC	After Bonferroni correction, frontal 5HT 2a BP lower in SA versus HC; 0.39 vs. 0.68.	No psychiatric control. No intent measure in SA. Multiple psychiatric diagnoses represented. Small groups, 3 patients in deliberate self-injury group. Use of SEM instead of SD in analysis. SA by overdose in 5/9 patients. Regions of PFC not specified. Multiple post-hoc comparisons with small sample sizes.
Van Heeringen 2003 Case–control Oquendo 2003 Cross-sectional. Activation study, fenfluramine challenge	See Audenaert 2001 25 MDD with SA (mean 4 years prior) 16 high-lethality SA 9 low-lethality SA	See Audenaert 2001 PET/^cFDG ROI: anterior cingulate and medial frontal gyri; anterior cingulate and right superior frontal gyrus.	Lower 5HT 2a receptor antagonist binding in PFC in SA (140.7) vs non- SA (168). Lower rCMRglu in ventral, medial, lateral PFC in high-lethality vs low-lethality SA. Lower VM PFC activity associated with lower impulsivity, intent, and lethality. Pre-fenfluramine: Lower rCMRglu in bilateral superior frontal, ACC, and inferior frontal gyri in high-lethality group. Post-fenfluramine: Lower rCMRglu in same areas above and superior frontal gyri.	See above. No psychiatric control group. No HC group. Comorbid diagnoses not discussed. Methods of suicide attempt not discussed.
Lindstrom 2004 Case–control 5HT transporter	12 SA with high intent Axis I diagnoses: 6 mood disorder 1 social phobia 3 adjustment disorder 12 HC	SPECT/cocaine analogue, ^f[¹²³I]-β CIT ROI: cerebellum and whole brain	No significant differences in whole brain BP of 5HTT. No significant differences in BP for violent SA vs. nonviolent SA vs controls.	Heterogeneous diagnoses. SPECT automatic scaling- increases error in ROI. No regional anatomic structures specified. β-CIT also binds the dopamine transporter.
Leyton 2006 Case–control Tryptophan uptake	10 high lethality SA (mean 14.7days) Methods: 8 by overdose, 1 by hanging, 1 by jumping Axis I diagnoses: 2 Mood disorder 6 Substance abuse 16 HC	PET/ α[¹¹C]Methyl-L-tryptophan ROI: medial OFG, left OFG, medial PFG	SA with decreased normalized tryptophan trapping in OFC and VM PFC. Increased tryptophan trapping seen in L thalamus, R paracentral lobule, L middle occipital cortex, hippocampal gyrus. Suicide intent negatively correlated with tryptophan trapping in OFG and R medial PFG.	No psychiatric controls. Unclear which toxins used in SA. Multiple comorbid diagnoses, including substance abuse. Utility of labelled tryptophan as a marker of serotonin synthesis has been questioned. In planned comparisons of trapping rate constants in VOI, main effect of group not significant.
Cannon 2006 Case–control 5HT transporter	18 BD, current MDE (8 SA history) 37 HC	PET/ ^g[¹¹C] DASB ROI: thalamus, striatum, insula, midbrain, sgACC, pgACC, DCC, PCC	In SA, increased pgACC binding and decreased midbrain binding compared to 10 without SA. Compared to controls, increased binding in thalamus, insula, DCC, and increased in midbrain.	Use of SEM distorts the small effect size. SPM reported with uncorrected p-values. No arterial sampling documented. Comorbid diagnoses include OCD, panic attacks. No account for cerebellum uptake (in bipolar disorder).
Oquendo 2007 Case–control 5HT transporter	18 BD, current MDE (9 SA history) 41 HC	PET/^h[¹¹C] McNeil 5652 ROI: midbrain, amygdala, hippocampus, thalamus, putamen, ACC	No difference in BP between SA and non-SA. Bipolar patients had lower 5HTT BP in midbrain, amygdala, hippocampus, thalamus, putamen and ACC. No correlation between depression severity and BP.	11% of patients with remission of symptoms during washout is concerning for change in synapses. Multiple comorbid diagnoses (OCD, PTSD, GAD, binge eating, simple phobia), some of which involve the serotonin system.

Author/study design

Study participants

Imaging modality/tracer, ligand

Major findings (as stated in paper)

Limitations

Structural MRI studies

Ahearn 2001

Case–control, retrospective

20 outpatients with MDD with SA history.

20 MDD without SA

T1- and T2-weighted images

Higher number of subcortical grey matter hyperintensities in those with history of SA.

No control for history of drug, alcohol use or method of SA.

Comorbid diagnoses not specified.

Ehrlich 2005

Cross-sectional, retrospective

102 inpatients with MDD, 62/102 with SA history.

T2 weighted

Higher prevalence of PVH but not DWMH in patients with history of SA when compared to those without SA history.

Severity of PVH a significant predictor of past SA.

No information on methods of SA.

Three different scanners used.

Group level differences in white matter hyperintensities (DWMH+PVH) between SA and non-attempters not significant.

Monkul 2007

Case–control, Brain volume

7 female MDD with history of SA

10 female MDD without history of SA 17 HC

T1 or T2?

ROI: OFC, cingulate, amygdala, hippocampus

SA with decreased bilateral OFC grey matter volumes, and increased right amygdala volumes.

No information on methods or lethality of SA, or time from last attempt.

2 different image analysis programs were used.

Blood flow studies

Audenaert 2002

Activation Study: Verbal fluency test

20 MDD patients with recent SA (less than 7 days prior)

20 HC

SPECT/^{b, 99m}Tc-ECD

ROI: none specified

During CFT, SA had blunted perfusion of left inferior PFC, R inferior parietal gyrus, L and R ACC.

During LFT, SA had blunted perfusion in L and R med temporal gyrus, R ACC, and R hypothalamus.

No psychiatric control group.

Patients on psychotropic drugs not excluded.

Lower IQ in SA group.

P values for regions of activation non-significant.

Pixel by pixel analysis showed difference but not cluster level differences.

Fountoulakis 2004

Cross-sectional, retrospective

50 MDD patients

Subgroups:

18 SA vs.

32 Non-SA;

17 No current thoughts of death

23 No specific thoughts of death

10 With suicidal thoughts

SPECT/^{a, 99m}Tc HMPAO

ROI: cerebellum, thalamus, caudate nucleus, GP, FL, PL, TL, OL.

After Bonferroni correction, no difference in rCBF between SA versus non-SA.

No difference between three subgroups based on thoughts of death.

Included patients with GAD and panic disorder, disorders which may involve the serotonin system.

Did not study inferior frontal lobe.

Functional MRI studies

Jollant 2008

All euthymic male subjects

13 SA with MDD

14 MDD

16 HC

Event-related fMRI on 1.5 T magnet

Compared to MDD alone, SA with MDD showed increased activity in R lateral orbitofrontal cortex (BA 47) in response to angry faces

Decreased activity in R superior frontal gyrus (BA 6) in response to angry faces

Greater activity in R cerebellum in response to mild angry expressions.

No differences in response to neutral or happy faces

Greater activity in the R anterior cingulate gyrus (BA 32), extending to the medial frontal gyrus (BA 10) in response to mild happy versus neutral faces.

No description of comorbidities such as anxiety or cluster B personality disorders, common in SA and which may impact affect processing

Time since most recent SA not given

Serotonin studies

Audenaert 2001

Case–control

5HT2a receptor study

9 SA

(<8 days prior)

Axis I diagnoses:

4 MDD

4 adjustment disorder

1 brief psychotic disorder

13 HC

SPECT, resting/

^d[¹²³I]5-I-R91150

ROI: bilateral frontal cortex, OFC, dorsolateral PFC

After Bonferroni correction, frontal 5HT 2a BP lower in SA versus HC; 0.39 vs. 0.68.

No psychiatric control.

No intent measure in SA.

Multiple psychiatric diagnoses represented.

Small groups, 3 patients in deliberate self-injury group.

Use of SEM instead of SD in analysis.

SA by overdose in 5/9 patients.

Regions of PFC not specified.

Multiple post-hoc comparisons with small sample sizes.

Van Heeringen 2003

Case–control

Oquendo 2003

Cross-sectional.

Activation study, fenfluramine challenge

See Audenaert 2001

25 MDD with SA (mean 4 years prior)

16 high-lethality SA

9 low-lethality SA

See Audenaert 2001

PET/^cFDG

ROI: anterior cingulate and medial frontal gyri; anterior cingulate and right superior frontal gyrus.

Lower 5HT 2a receptor antagonist binding in PFC in SA (140.7) vs non- SA (168).

Lower rCMRglu in ventral, medial, lateral PFC in high-lethality vs low-lethality SA.

Lower VM PFC activity associated with lower impulsivity, intent, and lethality.

Pre-fenfluramine: Lower rCMRglu in bilateral superior frontal, ACC, and inferior frontal gyri in high-lethality group.

Post-fenfluramine: Lower rCMRglu in same areas above and superior frontal gyri.

See above.

No psychiatric control group.

No HC group.

Comorbid diagnoses not discussed.

Methods of suicide attempt not discussed.

Lindstrom 2004

Case–control

5HT transporter

12 SA with high intent

Axis I diagnoses:

6 mood disorder

1 social phobia

3 adjustment disorder

12 HC

SPECT/cocaine analogue,

^f[¹²³I]-β CIT

ROI: cerebellum and whole brain

No significant differences in whole brain BP of 5HTT.

No significant differences in BP for violent SA vs. nonviolent SA vs controls.

Heterogeneous diagnoses.

SPECT automatic scaling- increases error in ROI.

No regional anatomic structures specified.

β-CIT also binds the dopamine transporter.

Leyton 2006

Case–control

Tryptophan uptake

10 high lethality SA

(mean 14.7days)

Methods: 8 by overdose, 1 by hanging, 1 by jumping

Axis I diagnoses:

2 Mood disorder

6 Substance abuse

16 HC

PET/

α[¹¹C]Methyl-L-tryptophan

ROI: medial OFG, left OFG, medial PFG

SA with decreased normalized tryptophan trapping in OFC and VM PFC.

Increased tryptophan trapping seen in L thalamus, R paracentral lobule, L middle occipital cortex, hippocampal gyrus.

Suicide intent negatively correlated with tryptophan trapping in OFG and R medial PFG.

No psychiatric controls.

Unclear which toxins used in SA.

Multiple comorbid diagnoses, including substance abuse.

Utility of labelled tryptophan as a marker of serotonin synthesis has been questioned.

In planned comparisons of trapping rate constants in VOI, main effect of group not significant.

Cannon 2006

Case–control

5HT transporter

18 BD, current MDE

(8 SA history)

37 HC

PET/

^g[¹¹C] DASB

ROI: thalamus, striatum, insula, midbrain, sgACC, pgACC, DCC, PCC

In SA, increased pgACC binding and decreased midbrain binding compared to 10 without SA.

Compared to controls, increased binding in thalamus, insula, DCC, and increased in midbrain.

Use of SEM distorts the small effect size.

SPM reported with uncorrected p-values.

No arterial sampling documented.

Comorbid diagnoses include OCD, panic attacks.

No account for cerebellum uptake (in bipolar disorder).

Oquendo 2007

Case–control

5HT transporter

18 BD, current MDE

(9 SA history)

41 HC

PET/^h[¹¹C] McNeil 5652

ROI: midbrain, amygdala, hippocampus, thalamus, putamen, ACC

No difference in BP between SA and non-SA.

Bipolar patients had lower 5HTT BP in midbrain, amygdala, hippocampus, thalamus, putamen and ACC.

No correlation between depression severity and BP.

11% of patients with remission of symptoms during washout is concerning for change in synapses.

Multiple comorbid diagnoses (OCD, PTSD, GAD, binge eating, simple phobia), some of which involve the serotonin system.

AD = antidepressant.

BD, bipolar disorder; BP, binding potential; CFT, category fluency test; CBF, cerebral blood flow; CV, cardiovascular risk factors; DR, dorsal raphe; DWMH, deep white matter hyperintensity; FL, frontal lobe; GAD, generalized anxiety disorder; GP, globus pallidus; HC, healthy controls; L, left; LFT, letter fluency test; MDD, major depressive disorder; MDE, major depressive episode; MPFC, medial prefrontal cortex; MRI,magnetic resonance imaging; OCD, obsessive–compulsive disorder; OFG, orbitofrontal gyrus; OL, occipital lobe; OPFC, orbital prefrontal cortex; PET, positron emission tomography; PFC, prefrontal cortex; PFG, prefrontal gyrus; pgACC, pregenual anterior cingulate cortex; PL, parietal lobe; PTSD, post-traumatic stress disorder; PVH, periventricular hyperintensity; R, right; rCBF, regional cerebral blood flow; rCMRglu, regional cerebral glucose utilization; ROI, regions of interest; SA, suicide attempters; SCH, subcortical grey matter hyperintensity; SEM, standard error of the mean; sgACC, subgenual anterior cingulate cortex; SPECT, single photon emission com-puted tomography; SPM, statistical parametric mapping; TL, temporal lobe; VM, ventromedial; VOI- voxel of interest; WM, white matter.

Tracers/ligands: ^a[^99mTc]-HMPAO, ^99mTc hexamethylpropyleneamine oxime;^b[^99mTc]-ECD, ^99mTc-Ethyl Cystine Dimer; ^cFDG, ¹⁸F- flourodeoxyglucose; ^d[¹²³I]5-I-R91150, 4-amino-N-[1-[3-(4-fluorophenoxy) propyl]-4-methyl-4-piperidinyl]-5-iodo-2-methoxybenzamide; ^f[¹²³I]-(β-CIT), ¹²³I-β-carbomethoxy-3-beta (4-iodophenyl)-tropane; ^g[¹¹C] DASB, [¹¹C]-3-amino-4-(2-dimethylaminomethylphenyl-sulfanyl)-benzonitrile; ^h[¹¹C] McNeil 5652, ¹¹C(+) trans 1;2;4;5;6; 10-β-hexahydro-6-[4-(methylthio)phenyl]-pyrrolo[2;1-a]isoquinoline.

This Feature Is Available To Subscribers Only