Journal CMEs

Print Friendly 

Structural and Functional
Neuroimaging of Pediatric Depression

Vilma Gabbay, MD, Raul R. Silva, MD, F. Xavier Castellanos, MD,
Beth Rabinovitz, BA, and Oded Gonen, PhD
Needs Assessment:
Neuroimaging technology has been increasingly used to investigate the underlying neurobiology of psychiatric disorders such as pediatric depression. As this field is rapidly progressing, it is difficult to stay abreast with new developments. Our goal is to provide clinicians with current knowledge regarding the structural and functional neuroimaging findings in pediatric depression. This information will allow medical professionals to view neuroimaging data critically and to understand methodological concerns in neuroimaging research involving pediatric depression.  

 Learning Objectives:
 •  Identify suitable neuroimaging methods for the study of major depressive disorder in the pediatric population.

 •  Critically evaluate studies of neuroimaging in pediatric depression.

 •  List major brain structures implicated in pediatric depression.

 •  Identify functional correlates of pediatric depression.

 Target Audience:
Primary care physicians and psychiatrists.

 Accreditation Statement:
Mount Sinai School of Medicine is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians.

 Mount Sinai School of Medicine designates this educational activity for a maximum of 3.0 Category 1 credit(s) toward the AMA Physician’s Recognition Award. Each physician should claim only those credits that he/she actually spent in the educational activity. Credits will be calculated by the MSSM OCME and provided for the journal upon completion of agenda.

 It is the policy of Mount Sinai School of Medicine to ensure fair balance, independence, objectivity, and scientific rigor in all its sponsored activities. All faculty participating in sponsored activities are expected to disclose to the audience any real or apparent conflict-of-interest related to the content of their presentation, and any discussion of unlabeled or investigational use of any commercial product or device not yet approved in the United States.

 To receive credit for this activity:
Read this article and the two CME-designated accompanying articles, reflect on the information presented, and then complete the CME quiz. To obtain credits, you should score 70% or better. Termination date: September 30, 2007. The estimated time to complete all three articles and the quiz is 3 hours.


Pediatric major depressive disorder (MDD) is a common disease associated with significant morbidity and mortality. Newly available noninvasive neuroimaging techniques provide unique opportunities to illuminate the underlying neurobiological factors of MDD. This article reviews structural and functional neuroimaging data in pediatric MDD. In general, neuroimaging studies in pediatric MDD tend to confirm findings in adult depression implicating the prefrontal cortex, amygdala, and hippocampus. These brain regions are linked and believed to be critical in modulating emotional responses. However, neuroimaging research in pediatric MDD is still in its infancy, and inconsistencies are rife. These inconsistencies are largely due to the small samples and lack of agreement regarding methodology in ascertainment as well as in imaging. Greater focus on careful delineation of clinically and neurobiologically defined subgroups will likely lead to improved understanding of the pathophysiology of MDD.


Major depressive disorder (MDD), a serious public health concern, affects both pediatric and adult populations. This common psychiatric ailment has an estimated lifetime prevalence of approximately 15% for adolescents 15–18 years of age.1 Pediatric MDD has serious consequences, including social and academic impairment. Most critically, attempted and completed suicides are the third leading cause of death among youths 15–19 years of age.2-7 Pediatric MDD is also a strong predictor of MDD in adulthood, which carries its own burden of disadvantage.8,9 The importance of specific neurobiological research in pediatric MDD has been increasingly recognized over the past decade. Neuroimaging technology has provided unique tools for direct structural and functional imaging of the working human brain. In adult MDD, neuroimaging research has implicated specific brain regions, including the anterior cingulate cortex, orbital cortex, basal ganglia, amygdala, and hippocampus, as well as disturbances in pathways linking cortical, subcortical, and limbic sites.10

Safety concerns regarding radiation exposure have limited the use of neuroimaging techniques such as computerized tomography (CT), positron emission tomography (PET), or singlephoton emission computed tomography (SPECT) in pediatric populations. Fortunately, noninvasive imaging techniques such as structural magnetic resonance imaging (MRI), functional MRI (fMRI), proton magnetic resonance spectroscopy (1H-MRS), and diffusion tensor imaging (DTI) do not involve radiation exposure, nor do they require injections, thus alleviating safety concerns.

Neuroimaging research in pediatric MDD is still in its initial stages, and inconsistencies among different research groups are not uncommon. However, the field is rapidly progressing and has the potential to contribute to our understanding of the underlying pathophysiological processes of pediatric MDD. This type of research could potentially foster preventive therapeutic options and contribute to the identification of at risk individuals. This article reviews structural and functional neuroimaging data in pediatric MDD.

Critical Assessment of Neuroimaging Findings

Despite commendable advances in pediatric neuroimaging research, particularly in the past decade, several limitations require mention. First, small sample sizes are still the rule, and these yield insufficient statistical power. Second, comparison of results across studies is constrained by variability in subject selection. This is especially problematic in a heterogeneous clinical syndrome such as pediatric MDD that most likely reflects a common final clinical pathway of multiple etiologies.11 For example, limiting adult MDD samples to patients with familial MDD yields relatively more informative neuroimaging findings in specific brain regions.10,12 In addition, because pediatric MDD is associated with significant morbidity and mortality, recruiting medication-naïve patients in North America has become more challenging. Studies often include children and adolescents who are currently taking antidepressants, who have been exposed previously to medications, and who have never been medicated. This may further complicate comparisons across studies. Finally, the field has not yet adopted standard quantitative analytical methods that allow direct comparisons across studies. Current methods include handtracing of individual regions of interest, fully automated methods, and semiautomated methods. Fully automated methods maximize test-retest reliability but are best applied to large well-defined brain regions. Semi-automated methods combine the strengths and weaknesses of the other two alternatives.

Neuroanatomical Correlates of Pediatric Major Depressive Disorder

Structural MRI allows the assessment of neuromorphology and neuromorphometry in pediatric MDD. Most structural MRI studies in pediatric MDD have focused on the frontal cortex, the hippocampus, and the amygdala. The corpus callosum, the pituitary gland, and white matter abnormalities have also been examined. These findings are discussed in turn.

Frontal Cortex

Anatomic hypotheses of the substrates of MDD have generally focused on the role of the frontal brain, particularly on the prefrontal cortex (PFC) (Figure 1A). Evidence based on animal and adult studies suggests that the PFC acts  as the executive branch of emotions.13,14 Neuroimaging studies have confirmed the role of the frontal lobe/PFC in adult MDD.10

There are several structural studies of frontal cortex in pediatric MDD (Table 1).15-18 In a retrospective chart review study, structural MRI images of 65 hospitalized children and adolescents with depressive disorders (56 with MDD, 9 with dysthymia) were compared to 18 non-depressed psychiatric controls.15

Psychiatric diagnoses among the controls included conduct/oppositional defiant disorder (n=11), attention-deficit/ hyperactivity disorder (n=2), posttrau matic stress disorder (PTSD; n=3), and adjustment disorder (n=2). As the original digital MRI data were not available for analysis, results are based on redigitization of brain images from films. The researchers discovered decreased frontal lobe/cerebral volume ratios and increased lateral ventricle/cerebral volume ratios in the hospitalized pediatric MDD group compared to psychiatric controls. These results should be viewed with caution in light of several design limitations, including the lack of structured diagnostic measures, the lack of a healthy control group, and an inherently less sensitive method of image analysis. In addition, medication history, which may affect imaging findings, was not documented.

A later study by the same group supported the involvement of the frontal lobe in pediatric MDD16 by examining frontal lobe morphometry in adolescents with MDD (n=19) versus healthy comparisons (n=38). Three MDD subjects were treated in the past with antidepressants, but were antidepressant free for at least 3 months at the time of the scan. The authors found significantly smaller whole brain volumes in the pediatric MDD group, as well as significantly smaller frontal white matter, compared to the healthy comparison group. It is noteworthy that 73% of MDD subjects and 16% of the controls had a first-degree relative with a mood disorder. However, an effect for familial loading was not found.

Nolan and colleagues17 focused on the PFC in 22 psychotropic-naïve children and adolescents 9–17 years of age with MDD, and 22 healthy comparisons. Twelve MDD subjects had familial MDD (with at least one first-degree relative with MDD), but none had a familial history of bipolar disorder. While the intracranial PFC volumes did not significantly differ between the MDD and control groups, the nonfamilial MDD subjects were found to have larger leftsided total PFC volumes and larger prefrontal white matter, compared to familial MDD subjects and healthy comparisons. Familial MDD subjects were found to have smaller left-sided gray matter volumes compared to subjects with nonfamilial MDD. These findings are intriguing in light of similar findings in adult MDD in which striking reductions in mean gray matter volumes of the subgenual PFC (sgPFC; located ventral to the genu of the corpus callosum, Figure 1A) were demonstrated only in familial MDD12 and in familial bipolar disorder.19 These gray-matter reductions were attributed to reduced glial cell density in the sgPFC in postmortem study.20

Botteron and colleagues18 examined sgPFC volumes in thirty young women (17–23 years of age) with adolescent onset MDD compared to eight matched controls. Females with adolescent onset MDD compared to normal controls had reduced left sgPFC volumes.

These studies support the view that regions of the frontal cortex play a role in the pathophysiological mechanisms underlying adult and pediatric MDD.

Amygdala and Hippocampus

The temporo-limbic structures comprising the amygdala and hippocampus (Figures 1B and 1C) are also critical in emotional regulation processes.13,21 Excitatory projections enable the amygdala to directly activate the PFC. Current evidence suggests that the amygdala is critical in the reception and production of emotional responses and in the establishment of conditioned fear.14

The amygdala and hippocampus have been the focus of many structural and functional imaging studies in adult MDD. Several structural studies have examined these brain regions in pediatric MDD (Table 2).2224 MacMillan and colleagues22 examined the amygdala and hippocampus volumes in antidepressant-naïve MDD children and adolescents (n=23; 8–17 years of age) and case-matched healthy comparisons (n=23). No significant group differences in amygdala and hippocampus volumes were found. However, significantly larger left (14%; P=.004) and right (11%; P=.026) amygdala:hippocampal volume ratios were found in the MDD group compared to control.22

In a later study focusing only on hippocampal volumes in adolescents with MDD (n=17, ages=13–18) and matched comparisons (n=17), relatively smaller hippocampus volumes were found in the MDD group.23 This difference was more significant in the left hippocampus (17%; P=.001) compared to the right hippocampus (P=.047). Two MDD subjects had comorbid substance abuse disorder and three were on antidepressants. Small left hippocampal volumes were confirmed when analyses were repeated using only the medication-naïve subjects (n=14). In addition, duration of MDD episode was found to be correlated with left hippocampal volume.23

In a recent study, Rosso and colleagues24 examined amygdala and hippocampal volumes in children and adolescents (n=20) with MDD and in healthy controls (n=24). They reported decreased amygdala volumes in the MDD group compared to controls. No significant difference in hippocampal volumes was found between the two cohorts. In a study of children and adolescents with PTSD, of whom half had MDD, no hippocampal volumes reductions were found either.25 In adult MDD, smaller hippocampal volumes have been reported in some but not all studies. In two adult studies, hippocampal atrophy was found to be correlated with the duration of MDD, suggesting that these changes are secondary to recurrence or chronicity of the illness.26,27 Conflicting results may also be related to difficulties in determining the precise boundaries of these structures, spatial resolution limits, and sample differences (ie, age, comorbid disorder, medication history).  

Corpus Callosum

Lyoo and colleagues28 examined corpus callosum (Figure 1A) structure in females with early onset minor depression (n=40; 18–25 years of age) and in healthy comparisons (n=42). The authors found that the genu of the corpus callosum in the depression group was significantly smaller compared to the healthy controls. The corpus callosum was also found to be smaller in offspring of mothers with a history of MDD.29 If confirmed, this finding may represent a developmental marker of premorbid risk for MDD.

Pituitary Gland

The role of the pituitary gland in MDD has been inferred from evidence of abnormalities of the hypothalamic-pituitary-adrenal (HPA) axis in individuals with MDD. MacMaster and Kusamaker30 measured pituitary volumes in adolescents with MDD (n=17, 14–17 years of age), and age- and sex-matched controls (n=17). Larger pituitary volumes were found in the MDD group (P=.02) even when MDD subjects treated with medications were excluded.

White-Matter Hyperintensities

White matter signal hyperintensities (WMH), detected by T2-weighted clinical MRI, reflect brain regions with increased water density. The clinical importance of these white-matter lesions has not been determined. Neuroimaging studies have examined the prevalence of WMH in adults with several psychiatric disorders. There are also two studies that were conducted in children and adolescents with MDD. Lyoo and colleagues examined WMH in children and adolescents admitted to inpatient psychiatric units compared with nonpsychiatric patients. The pediatric MDD group (n=94, 7–17 years of age) was found to have significantly greater numbers of WMH compared to subjects without a psychiatric diagnosis (n=83).30 WMH were mainly located in frontal lobes. As noted, small whitematter volumes were also found in adolescents with MDD.16 The same group conducted a retrospective study in which prevalence and severity of WMH were compared to history of suicidal attempts among  inpatients with a variety of psychiatric disorders for whom an MRI scan had been performed as part of the medical work-up.32 WMH were found to be associated with a history of suicide attempts only in the pediatric MDD group (n=48, 12–17 years of age). These findings are interesting in light of decreased oligodendrocyte density found in the amygdala of patients with MDD. Oligodendrocytes play the key role in myelination. Although there is no clear evidence of a myelin disorder in mood disorders, myelin basic protein is decreased in PFC in adults with MDD and schizophrenia.33

Functional Neuroimaging Studies in Pediatric Major Depressive Disorder

Functional neuroimaging techniques such as PET and SPECT, which provide information on brain function in specific brain regions, are widely used in adult populations. These techniques allow the measurement of neurotrans mitters, specific brain receptors, as well as blood flow and perfusion. The use of these technologies has been limited in pediatric MDD due to safety and ethical concerns (ie, even minimal radiation exposure and the requirement to inject radioactive isotopes). We know of only two studies that have used SPECT in pediatric MDD. Tutus and colleagues34 examined medication-free adolescents with MDD (n=13; 11–15 years of age) and comparisons (n=11; 12–15 years of age) using SPECT. The authors found reduced perfusion in the left anterolateral and left temporal cortical areas in the MDD group compared to controls. These abnormalities were not found in follow-up scans after depressive symptoms had subsided, suggesting that abnormalities may be a state-related marker for this disorder.34 Another small study conducted by Kowatch and colleagues,35 using SPECT (seven subjects per group), found regional hyperperfusion in the right mesial temporal lobe, and hypoperfusion in the parietal lobe, thalamus, and caudate. Above all, the authors emphasized the need for replication with larger samples prior to reaching firm conclusions. However, such larger samples are unlikely, particularly in the United States. Instead, the noninvasive neuroimaging techniques fMRI and 1H-MRS have been used in pediatric MDD.

Functional Magnetic Resonance in Pediatric Major Depressive Disorder

fMRI allows for the identification of specific brain regions which are activated during the performance of various cognitive tasks by the subtle change in their hemodynamics. Investigating pediatric MDD, Thomas and colleagues36 examined amygdala activity in five girls with MDD (8–16 years of age), five girls with anxiety disorders, and five healthy comparison girls matched for age. Two of the MDD girls had a comorbid disorder of generalized anxiety disorder. Girls with MDD were found to have reduced amygdala activation when processing fearful or neutral faces compared with the anxious and healthy comparison girls, suggesting amygdala involvement in pediatric MDD. This small but pioneering study is consistent with structural data implicating the amygdala in pediatric MDD, but clearly calls for more and larger studies.

Proton Magnetic Resonance Spectroscopy in Pediatric Major Depressive Disorder

Regional reductions in numbers of glia and neurons, increased WMH, and abnormalities of cerebral blood flow and metabolism have been well documented in MDD. These findings suggest that impaired cellular resilience may underlie MDD.10,12,37 1HMRS allows the in vivo non-invasive assessment of a variety of neurochemicals which reflect neuronal and glia integrity (Figure 2). These qualities make 1H-MRS ideally suited for use in pediatric MDD since it allows the early detection of neurochemical alterations and may contribute to the identification of at risk individuals. Among the metabolites that 1H-MRS quantifies are choline (Cho, reflects membrane lipid breakdown), N-acetylaspartate (NAA, associated with neuronal integrity and viability), creatine (Cr, may indicate abnormal energy metabolism and decreased overall cell density), γ-aminobutyric acid (GABA), and glutamine. There are discrepancies among 1 H-MRS studies of MDD as to whether absolute metabolite levels and/or ratios are higher or lower in pediatric and adult MDD. Conflicting results have been attributed to methodological differences in imaging technique (single voxel versus multivoxel; voxel size), medication history at time of scan, and the heterogeneous nature of MDD samples.

The few 1H-MRS studies on pediatric MDD in the current literature are in general agreement with structural neuroimaging studies, suggesting the role of PFC38-42 and the amygdala42 in pediatric MDD (Table 3).38,39,41-45

The Frontal Lobe in Proton Magnetic Resonance Spectroscopy Studies of Pediatric Major Depressive Disorder

Steingard and colleagues38 examined the orbitofrontal cortex in adolescents with MDD (n=17) and healthy comparisons (n=28). Four MDD subjects were treated with medication at the time of the scan. Significantly higher ratios of Cho/Cr (P=.032) and Cho/ NAA (P=.04) in the left orbitofrontal cortex were found in adolescents with MDD compared to healthy comparisons. Farchione and colleagues39 examined the dorsolateral PFC (DLPFC) in children and adolescents with MDD (n=11, 10–16 years of age) and matched healthy comparisons (n=11). All MDD subjects were psychotropic naïve. Higher absolute Cho levels were found in the left DLPFC in the MDD group compared to healthy comparisons. These findings contrast a recently published study in which Cho was lower in the left DLPFC in children and adolescents with MDD (n=14, 9–17 years of age, including six treated with psychotropic medications at the time of the scan), when compared to healthy controls (n=22, 8–17 years of age).42 1H-MRS can quantify GABA and glutamine, amino acids which have been increasingly implicated in MDD. 1H-MRS provides only limited ability to assign unequivocal resonance peaks to these amino acids. The fitted combination of glutamine, glutamate, GABA, and homocarnosine are often referred to as Glx. Two 1 H-MRS studies from the same group examined Glx levels in the anterior cingulate cortex in pediatric MDD and controls.41,43 Both studies found decreased Glx levels in the anterior cingulate in the pediatric MDD group compared to healthy comparisons.

The Amygdala in Proton Magnetic Resonance Spectroscopy Studies of Pediatric MDD

Kusumakar and colleagues44 examined the neurochemistry of the amygdala in adolescents with MDD (n=11, 14–18 years of age), and in age and sex matched comparisons (n=11). Significantly decreased Cho/Cr ratios were found in the MDD group in the left amygdala region.

The Thalamus in Proton Magnetic Resonance Spectroscopy Studies of Pediatric Major Depressive Disorder

Smith and colleagues45 compared thalamic Cho levels in psychotropicnaïve children and adolescents with MDD (n=18, 9–17 years of age) and 18 matched comparisons. No differences were reported between the two groups.

Diffusion Tensor Imaging

DTI is a recently developed MRI method that measures the directionality of self-diffusion of water molecules. DTI allows the quantification of correlates of myelination (ie, white matter), one of the key components of neuronal maturation which continues through adolescence and which may be impaired and/or delayed in pediatric MDD. This possibility is especially relevant in light of structural neuroimaging studies which found white matter abnormalities in pediatric MDD.


Consistent with neuroimaging research in adult MDD, the current structural and functional neuroimaging literature implicates several key brain structures involved in pediatric MDD, including the prefrontal cortex, amygdala, and the hippocampus. There has been almost a complete lack of neuroimaging research that examines the basal ganglia in pediatric MDD, a brain region which is suggested to play a possible role in adult MDD. Methodological inconsistencies and low statistical power limit current neuroimaging findings. Extrapolating from the adult literature, future studies should strive to focus on specific clinical subgroups. Inclusion criteria such as familial MDD, psychotropic-naïve status, and specific age of onset (adolescent onset versus childhood onset) may improve the detection of neurobiological findings by decreasing phenotypic heterogeneity.10,46  Such refinements in study design should improve the yield of the powerful noninvasive functional neuroimaging technologies such as fMRI, MRS, and DTI, that can now be applied to pediatric MDD. PP


1. Kessler RC, Walters EE. Epidemiology of DSMIII-R major depression and minor depression among adolescents and young adults in the National Comorbidity Survey. Depress Anxiety. 1998;7(1):3-14. 

2. Kutcher SP, Marton P. Parameters of adolescent depression. A review. Psychiatr Clin North Am. 1989;12(4):895-918.

3. Fleming JE, Offord DR. Epidemiology of childhood depressive disorders: a critical review. J Am Acad Child Adolesc Psychiatry. 1990;29(4):571-580.

4. Weissman MM, Wolk S, Goldstein RB, et al. Depressed adolescents grown up. JAMA. 1999;281(18):1707-1713.

5. Lewinsohn PM, Hops H, Roberts RE, Seeley JR, Andrews JA. Adolescent psychopathology: I. Prevalence and incidence of depression and other DSM-III-R disorders in high school students. J Abnorm Psychol. 1993;102(1):133-144.

6. Rao U, Weissman MM, Martin JA, Hammond RW. Childhood depression and risk of suicide: a preliminary report of a longitudinal study. J Am Acad Child Adolesc Psychiatry. 1993;32(1):21-27.

7. Ryan ND, Puig-Antich J, Ambrosini P, et al. The clinical picture of major depression in children and adolescents. Arch Gen Psychiatry. 1987;44(10):854-861.

8. Pine DS, Cohen P, Gurley D, Brook J, Ma Y. The risk for early-adulthood anxiety and depressive disorders in adolescents with anxiety and depressive disorders. Arch Gen Psychiatry. 1998; 55(1):56-64.

9. Harrington R, Fudge H, Rutter M, Pickles A, Hill J. Adult outcomes of childhood and adolescent depression. I. Psychiatric status. Arch Gen Psychiatry. 1990;47(5):465-473.

10. Drevets WC. Neuroimaging studies of mood disorders. Biol Psychiatry. 2000;48(8):813-829.

11. Hasler G, Drevets WC, Manji HK, Charney DS. Discovering endophenotypes for major depression. Neuropsychopharmacology. 2004;29(10):1765-1781.

12. Drevets WC, Price JL, Simpson JR Jr., et al. Subgenual prefrontal cortex abnormalities in mood disorders. Nature. 1997;386(6627):824-827.

13. Davidson RJ, Irwin W. The functional neuroanatomy of emotion and affective style. Trends Cogn Sci. 1999;3(1):11-21.

14. Davidson RJ. Anxiety and affective style: role of prefrontal cortex and amygdala. Biol Psychiatry. 2002; 51(1):68-80.

15. Steingard RJ, Renshaw PF, Yurgelun-Todd D, et al. Structural abnormalities in brain magnetic resonance images of depressed children. J Am Acad Child Adolesc Psychiatry. 1996;35(3):307-311.

16. Steingard RJ, Renshaw PF, Hennen J, et al. Smaller frontal lobe white matter volumes in depressed adolescents. Biol Psychiatry. 2002;52(5):413-417.

17. Nolan CL, Moore GJ, Madden R, et al. Prefrontal cortical volume in childhood-onset major depression: preliminary findings. Arch Gen Psychiatry. 2002;59(2):173-179.

18. Botteron KN, Raichle ME, Drevets WC, Heath AC, Todd RD. Volumetric reduction in left subgenual prefrontal cortex in early onset depression. Biol Psychiatry. 2002;51(4):342-344.

19. Hirayasu Y, Shenton ME, Salisbury DF, et al. Subgenual cingulate cortex volume in first-episode psychosis. Am J Psychiatry. 1999;156(7):1091-1093.

20. Ongur D, Drevets WC, Price JL. Glial reduction in the subgenual prefrontal cortex in mood disorders. Proc Natl Acad Sci U S A. 1998;95(22):13290-13295.

21. Lane RD, Reiman EM, Bradley MM, et al. Neuroanatomical correlates of pleasant and unpleasant emotion. Neuropsychologia. 1997;35(11):1437-1444.

22. MacMillan S, Szeszko PR, Moore GJ, et al. Increased amygdala: hippocampal volume ratios associated with severity of anxiety in pediatric major depression. J Child Adolesc Psychopharmacol. 2003;13(1):65-73.

23. MacMaster FP, Kusumakar V. Hippocampal volume in early onset depression. BMC Med. 2004;2:2.

24. Rosso IM, Cintron CM, Steingard RJ, Renshaw PF, Young AD, Yurgelun-Todd DA. Amygdala and hippocampus volumes in pediatric major depression. Biol Psychiatry. 2005;57(1):21-26.

25. De Bellis MD, Keshavan MS, Clark DB, et al. A.E. Bennett Research Award. Developmental traumatology. Part II: Brain development. Biol Psychiatry. 1999;45(10):1271-1284.

26. Bremner JD, Narayan M, Anderson ER, Staib LH, Miller HL, Charney DS. Hippocampal volume reduction in major depression. Am J Psychiatry. 2000;157(1):115-118.

27. Sheline YI, Wang PW, Gado MH, Csernansky JG, Vannier MW. Hippocampal atrophy in recurrent major depression. Proc Natl Acad Sci U S A. 1996;93(9):3908-3913.

28. Lyoo IK, Kwon JS, Lee SJ et al. Decrease in genu of the corpus callosum in medication-naive, early-onset dysthymia and depressive personality disorder. Biol Psychiatry. 2002;52(12):1134-1143.

29. Martinez P, Ronsaville D, Gold PW, Hauser P, Drevets WC. Morphometric abnormalities in adolescent offspring of depressed mothers. Abstr Soc Neurosci. 2002;498:4.

30. MacMaster FP, Kusamaker V. MRI study of the pituitary gland in adolescent depression. J Psychiatr Res. 2004;38(3):231-236.

31. Lyoo IK, Lee HK, Jung JH, Noam GG, Renshaw PF. White matter hyperintensities on magnetic resonance imaging of the brain in children with psychiatric disorders. Compr Psychiatry. 2002;43(5):361-368.

32. Ehrlich S, Noam GG, Lyoo IK et al. White matter hyperintensities and their associations with suicidality in psychiatrically hospitalized children and adolescents. J Am Acad Child Adolesc Psychiatry. 2004;43(6):770-776.

33. Honer WG, Falkai P, Chen C, Arango V, Mann JJ, Dwork AJ. Synaptic and plasticity-associated proteins in anterior frontal cortex in severe mental illness. Neuroscience. 1999;91(4):1247-1255.

34. Tutus A, Kibar M, Sofuoglu S, Basturk M, Gonul AS. A technetium-99m hexamethylpropylene amine oxime brain single-photon emission tomography study in adolescent patients with major depressive disorder. Eur J Nucl Med. 1998;25(6):601-606.

35. Kowatch RA, Devous MD, Sr., Harvey DC et al. A SPECT HMPAO study of regional cerebral blood flow in depressed adolescents and normal controls. Prog Neuropsychopharmacol Biol Psychiatry. 1999;23(4):643-656.

36. Thomas KM, Drevets WC, Dahl RE, et al. Amygdala response to fearful faces in anxious and depressed children. Arch Gen Psychiatry. 2001;58(11):1057-1063.

37. Manji HK, Drevets WC, Charney DS. The cellular neurobiology of depression. Nat Med. 2001;7(5):541-547.

38. Steingard RJ, Yurgelun-Todd DA, Hennen J, et al. Increased orbitofrontal cortex levels of choline in depressed adolescents as detected by in vivo proton magnetic resonance spectroscopy. Biol Psychiatry. 2000;48(11):1053-1061.

39. Farchione TR, Moore GJ, Rosenberg DR. Proton magnetic resonance spectroscopic imaging in pediatric major depression. Biol Psychiatry. 2002;52(2):86-92.

40. Mirza Y, Tang J, Russell A, et al. Reduced anterior cingulate cortex glutamatergic concentrations in childhood major depression. J Am Acad Child Adolesc Psychiatry. 2004;43(3):341-348.

41. Rosenberg DR, Mirza Y, Russell A, et al. Reduced anterior cingulate glutamatergic concentrations in childhood OCD and major depression versus healthy controls. J Am Acad Child Adolesc Psychiatry. 2004;43(9):1146-1153.

42. Caetano SC, Fonseca M, Olvera RL, et al. Proton spectroscopy study of the left dorsolateral prefrontal cortex in pediatric depressed patients. Neurosci Lett. 2005;384(3):321-326.

43. Mirza Y, Tang J, Russell A, et al. Reduced anterior cingulate cortex glutamatergic concentrations in childhood major depression. J Am Acad Child Adolesc Psychiatry. 2004;43(3):341-348.

44. Kusumakar V, MacMaster FP, Gates L, Sparkes SJ, Khan SC. Left medial temporal cytosolic choline in early onset depression. Can J Psychiatry. 2001;46(10):959-964.

45. Smith EA, Russell A, Lorch E, et al. Increased medial thalamic choline found in pediatric patients with obsessive-compulsive disorder versus major depression or healthy control subjects: a magnetic resonance spectroscopy study. Biol Psychiatry. 2003;54(12):1399-1405.

46. Drevets WC, Ongur D, Price JL. Neuroimaging abnormalities in the subgenual prefrontal cortex: implications for the pathophysiology of familial mood disorders. Mol Psychiatry. 1998;3(3):190-191;220-226.

Dr. Gabbay is assistant professor in the Department of Psychiatry at New York University School of Medicine in New York City.

Dr. Silva is associate professor of psychiatry and the deputy director of the Division of Child and Adolescent Psychiatry at New York University School of Medicine/Bellevue Hospital Center.

Dr. Castellanos is the Brooke and Daniel Neidich Professor of Child and Adolescent Psychiatry, director of research, and director of the Institute for Pediatric Neuroscience in the Department of Psychiatry at New York University School of Medicine.

Ms. Rabinovitz is a research assistant in the Department of Psychiatry at New York University School of Medicine.

Dr. Gonen is professor of radiology, and physiology and neuroscience in the Department of Radiology at New York University School of Medicine.

Disclosure: Dr. Gabbay has received funding from the American Foundation for Suicide Prevention and the Tourette Syndrome Association. Dr. Silva is a consultant to Novartis; is on the speakers bureau for AstraZeneca, Novartis, and Ortho-McNeil; and receives grant support from the New York City Department of Mental Health, the New York State Office of Mental Health, the Red Cross, the Research Foundation for Mental Hygiene, and the Substance Abuse and Mental Health Services Administration. Dr. Castellanos receives grant support from the Stavros S. Niarchos Foundation, the National Institute of Mental Health, the National Institute on Drug Abuse, the National Alliance for Research in Schizophrenia and Affective Disorders, the National Institute for Neurological Disorders and Stroke, and McNeil Consumer & Specialty Pharmaceuticals Division of McNeil-PPC. Ms. Rabinovitz reports no affiliations with or financial interest in any organization that may pose a conflict of interest. Dr. Gonen serves on several National Institute of Health (NIH) review panels and currently holds two NIH R01 grants.

Please direct all correspondence to: Vilma Gabbay, MD, 577 First Ave, New York, NY 10016; Tel: 212-263-2731; Fax: 212-263-8662; Email: