International Journal of Mental Health & PsychiatryISSN: 2471-4372

All submissions of the EM system will be redirected to Online Manuscript Submission System. Authors are requested to submit articles directly to Online Manuscript Submission System of respective journal.

Research Article, Int J Ment Health Psychiatry Vol: 2 Issue: 3

Systems Neuroscience in Children and Adolescent Depression

Tomoya Hirota1, Gordana Milavić2, Fiona McNicholas3,4,5, Thomas Frodl6,7 and Norbert Skokauskas6,8*
1Department of Psychiatry Division of Child and Adolescent Psychiatry, University of California San Francisco, San Francisco, CA, United States
2National and Specialist Services, Michael Rutter Centre, Maudsley Hospital, London, United Kingdom
3Department of Psychiatry, University College Dublin, Ireland
4Lucena Clinic, Dublin, Ireland
5Department of Child Psychiatry, Our Lady's Children Hospital Crumlin, Dublin, Ireland
6Department of Psychiatry, Trinity College Dublin, Ireland
7Department of Psychiatry, Universität Regensburg, Regensburg, Germany
8Centre for Child and Adolescent Mental Health and Child Protection, Department of Neuroscience, Norges teknisk-naturvitenskapelige universitet (NTNU), Trondheim, Norway
Corresponding author : Norbert Skokauskas, NTNU
Faculty of Medicine, RKBU,Pb 8905, MTFS, N-7491 Trondheim, Norway
Tel: +47-405-36-900
E-mail: [email protected]
Received: April 11, 2016 Accepted: June 20, 2016 Published: June 24, 2016
Citation: Hirota T, Milavic G, McNicholas F, Frodl T, Skokauskas N (2016) Systems Neuroscience in Children and Adolescent Depression. Int J Ment Health Psychiatry 2:3. doi:10.4172/2471-4372.1000126


Depression is one of the most common psychological disorders in children and adolescents often presenting as a severe, chronic and recurring condition with a high risk of self-harm and suicide. Normal brain development during adolescence is necessary to put into context when investigating biology of adolescent depression. There is an uneven timeline for development going from ‘bottom up’ with the subcortical limbic areas developing first, followed by the prefrontal cortical (PFC) areas, which does not reach full functional maturity until the mid-20s. Amygdala and striatum have been reported to be reduced in depressed youth compared to controls. Functional brain changes have been identified in adults but less consistently in adolescent depression.

Keywords: Depression; Children; Adolescent; MRI, Genetics; Systems neuroscience


Depression; Children; Adolescent; MRI, Genetics; Systems neuroscience


Systems neuroscience uses a wide variety of approaches to study how networks of neurons form the bases of higher brain function. The field of systems and network neuroscience is evolving rapidly and this paper aims to highlight current advances in systems neuroscience from studies on child and adolescent depression.
For over half a century clinicians have been aware that children and adolescents can become depressed and require treatment [1-3]. There has, however, been a considerable delay before depressive disorders in children and adolescents became the focus of renewed academic study [4] with further delay in recognizing and treating depressive disorders in routine clinical practice [5]. It is now widely accepted that depression is one of the most common psychological disorders in children and adolescents often presenting as a severe, chronic and recurring condition with a high risk of self-harm and suicide. Depressive disorders run in families, are associated with adverse psycho-social circumstances and adverse life events, may lead to serious impairment in the child and adolescent’s current functioning and usher a poor prognosis and chronic course extending well into adulthood [4].
The clinical picture of depression in children and adolescent is, in the main, similar to that found in adults, but some differences stemming from biological, developmental and social attributes will influence the way depression presents in children, and particularly so in younger children. For instance, irritability, low frustration levels, temper tantrums, somatic complaints and social withdrawal are found more commonly in children [6,7]. Disinterest in general appearance, changes in peer relationships and social withdrawal, a drop in academic performance, self-harming and other risk taking behaviors may precede a more typical presentation of depression in adolescence. According to the World Health Organization (WHO) clinically significant depression is diagnosed when a child or adolescent presents with at least two weeks’ duration of depressed mood or irritability and either loss of interest or pleasure and/or tiredness and a number of other symptoms including poor or increased sleep, poor concentration or indecisiveness, low self-confidence, poor or increased appetite, suicidal thoughts or acts, agitation or slowing of movements, guilt or self-blame [8]. The degree of severity the depressive disorder is determined by the number of associated symptoms [8].
Prevalence rates of depressive disorders in children and adolescents vary across studies. The most quoted rates are Angold and Costello’s 12 month period prevalence rates [9] for major depression of 1% for prepubertal children and 3% for post pubertal children and adolescents. More recently Merikangas et al., looking at the life time prevalence of adolescent depression in the National Comorbidity Survey, Adolescent Supplement, a face to face study of 10,123 US adolescents, 13-18 used the modified version of WHO composite international diagnostic interview to establish rates of Major Depressive Disorder or Dysthymia [10]. The authors of the study found a total of 11.7% rate of depression with 8.7% having suffered severe impairment. Merikangas et al. noticed a female to male ratio of 15.9 to 7.7 in this age group as well as showed how the rates of depressive disorder increased as the young people grew older: 13-14 year olds had rates of major depressive disorders and dysthymia of 8.4%, 15-16 year olds 12.6% and 17-18 year olds 15.4%.
The transition to adulthood is a critical developmental period, marked by a series of tasks that include strengthening and expanding self-concept, forming stable intimate relationships, making career decisions, and achieving independence from parents. Homotypic and heterotypic continuities of depressive disorders from childhood to adulthood have been the subject of an increasing number of studies [11-18]. Heterotypical continuity means that an underlying (developmental) process or impairment stays the same, but the manifestations (i.e. depressive symptoms) do not. Homeotypic continuity is seen as the opposite of heterotypic continuity. Carballo and colleagues found that homotypic continuity of depressive disorder into adulthood was associated with an absence of psychiatric comorbidity prior to age 18 years. Also, adolescent onset depression by contrast to childhood onset depression showed greater homotypic continuity [19]. Luby established homotypic continuities from preschool to early school years18 but other studies point to the ‘complexity’ underlying heterotypic outcomes [20].
Depression in children or adolescent can be difficult to diagnose, and a clinician may need to see a patient over time to determine the appropriate diagnosis. A thorough mental health assessment including joint and separate interviews with the child and young person and their parents/carers is a pre-requisite to a comprehensive evaluation aimed at establishing the presence or otherwise of a depressive disorder. Psychoeducation, self-help and family support techniques are recommend at the outset of any intervention [21]. Watchful waiting, non-directive supportive therapy, group cognitive behavioural therapy and guided self-help are recommended for managing mild depression. Should the disorder become more severe or not respond to interventions at primary care level, evidence based psychological treatments including cognitive behavioural therapy, interpersonal psychotherapy and short-term family therapy should be introduced [22]. Should these interventions not be effective Fluoxetine (a selective serotonin reuptake inhibitors (SSRI) antidepressant) is recommended as the first line medication [21]. However itcan only be prescribed in conjunction with psychological treatments and only following an assessment and diagnosis by a child and adolescent psychiatrist [21]. SSRIs other than Fluoxetine, such as Sertraline or Citalopram, should only be used as a second line treatment. A meta-analysis of studies demonstrated the overall benefit of the use of Fluoxetine medication showing that there is consistent evidence for a statistically significant benefit but pointing to modest effect sizes. The response rates in the Fluoxetine trial groups varied between 41% and 61% and in the placebo groups between 20% and 35% [23].
The administration of antidepressants at a much earlier stage of treatment is indicated in severe depression as has been shown in more recent clinical trials [24,25]. The study of the Treatment of Resistant Depression in Adolescents (TORDIA) showed that in those young people where there was no response to one SSRIs, a switch to another antidepressant and the addition of CBT, showed better results than a switch to another antidepressant without CBT [26].
The safety and effectiveness debate with respect to the use of antidepressants continued after the publication of the NICE Guideline [27] in UK and Practice Parameters in US6. The TADS (The Treatment for Adolescents With Depression Study) extended findings established that all treatment arms were effective in reducing suicidal ideation but that the combined treatment of Fluoxetine and CBT reduced the risk of suicidal events in contrast to Fluoxetine treated patients who had more suicide related events [28]. The TADS predictors of suicidal events included higher levels of self-reported suicidal ideation and depression at baseline, only slight improvement in depression and acute interpersonal conflict (73% cases) [29].
Overall, “adverse events” analyses of clinical trials show that 1-3 individuals in 100 taking antidepressants will have onset or worsening of suicidality. It should be emphasized that there were no completed suicides in any of the quoted clinical trials. Electroconvulsive therapy (ECT) is rarely used in the treatment of depression in children and adolescents and apart from some anecdotal cases there are no randomized controlled trials using ECT in children and adolescents. Children and young people have a lower seizure threshold and the risk of prolonged seizures is increased. However procedures and practice have been progressively refined, including modern anaesthetic methods. Nevertheless fever than 1% of all patients treated with ECT are children and adolescents [30].

Normal Brain Structural and Functional Development

The minimal increase in brain size from childhood to adolescence clouds the significant re-modeling and restructuring of both white and grey matter. Significant increases in neuronal production and size of neurons (white matter) are matched with subsequent selective pruning which is genetically pre-programmed and subject to modification by experience; following ‘use it or lose it ‘principles. Around adolescence there is a peak in grey matter volume, only to reduce again in adulthood, following synaptic pruning. Myelination of neurons allow for faster and more efficient transmission with reduced energy expenditure involving fewer structural areas, alongside increased connectivity between the various neural systems. There is an uneven timeline for development going from ‘bottom up’ with the subcortical limbic areas developing first, followed by the prefrontal cortical (PFC) areas, which does not reach full functional maturity until the mid-20s. This uneven developmental timing exposes the adolescent to more influence by the earlier maturing subcortical system, which is reward driven, over the later developing ‘top down’ systems, which influence the executive functions of planning and impulse control [31]. This unbalance has been suggested for the higher risk of risk taking behaviours and substance misuse problems in adolescences compared to adulthood. Disturbance in the developmental progression can have wide ranging influence on subsequent behaviour, emotions and cognitive development. The depressant effect of substance misuses or secondary effects will not doubt confuse the biological picture found in adolescent depression, and make neurobiological comparisons with the adult depressed cohorts difficult. Neuroimaging studies have found a correlation between PFC maturation and cognitive mastery and control and behavioural inhibition [32] and therefore atypical neurodevelopmental trajectories may be harder to interpret in clinical groups.
In addition to structural findings, functional brain imagining studies have exposed differences in adolescent’s response when compared to either children or adults. When presented with emotionally laden images, adolescents show more pronounced amygdala activation, with a relative underuse of PFC areas [33]. Similarly adolescences recruit different and more numerous brain areas when faced with tasks requiring calculation and impulse control than adults, with a preference to use more caudal than frontal areas. This typical developmental differences need to be considered therefore when interpreting similarities or differences between adolescent or adult depression.
The neurotransmitter system also follows a distinctive development trajectory. The cholinergic and serotinergic systems are thought to be significant in facilitating neuronal maturation, plasticity and connectivity, particularly in the cortico-limbic pathways and develop earlier, declining during adolescence. Development of the dopamine rich meso-striatal and meso-cortico-limbic systems, responsible for reward directed behaviours, are more pronounced, and occur later with increased DOPA projections to the PFC as a result of subcortical dopamine pruning. The reduced serotinergic, GABAnergic and Glutaminergic inputs to the PFC in adolescence allow a relative dopamine dominated state which might explain the increased sensitivity to rewards and risk taking behaviour in adolescence [34] and may also explain the relative lack of efficacy of tricyclic’s antidepressant medication in youth versus adults. The age related variations in neurotransmitter system may also explain the lack of consistent findings when comparing blood or CSF levels between pediatric and adult cohorts. For example, lower levels of serotonin and its metabolite, 5HIAA in either blood or CSF have been found in adults, and have been linked with aggression and suicidal behaviour [35]. Post-mortem studies in adults have found abnormalities in brain serotonin system and in the few adolescent brains examined [36], but none were published specifically on depression. Studies of CSF 5-HIAA levels in children suggest similar inverse correlations between 5-HIAA levels and aggression [37] whereas platelet studies have been contradictory [38].

Neuroendocrine Studies

Many studies of adult patients have implicated the hypothalamicpituitary- axis in depression, with a blunted response following stimulation and higher levels in the CSF in depressed patients compared to controls, with some evidence of normalisation post treatment. This has been less consistently found in children or adolescents [39]. Non suppression of cortisol release by dexamethasone was almost considered diagnostic, occurring in 50-70% of depressed adults, especially associated with psychotic subtypes and past history of depression [38]. Although some studies have reported on similar findings in children and adolescence [40], this is by no means typical. The higher levels of cortisol found amongst adult depressed cohorts have not been identified in children [41]. Similarly confusing findings have been reported with regard to growth hormone levels. In depressed adults higher daytime and lower nocturnal secretion has been reported compared to non-depressed samples, yet in children [42] and adolescents [43], almost the reverse, with increased nocturnal GH release.

Structural Brain Changes

The most consistent evidence from neuroimaging studies suggest smaller PFC and Basal ganglia in depressed adults with corresponding decreased blood flow and metabolism [44]. Pediatric studies examining the amygdala and striatum have also reported reduced volumes in depressed youth compared to controls [45], but the findings are difficult to interpret given the lack of association with depressive severity, age of onset or chronicity, along with the variability of findings to adult studies. In contrast in a meta-analysis of MRI studies, volume reduction in PFC regions in depressed adult patients was associated with illness severity [46].
Studies, of both children and adults have found reduced Hippocampal volumes in depressed individuals [47] and in healthy adolescents with a family history of depression, with volumes correlating with subsequent risk of depression [48]. Given the role the hippocampus plays in mood regulation, sensitivity to stress and evidence of impaired neurogenesis and atrophy in chronically stressed animal models [17,49] these findings are of interest, and may be linked with both heritability to depression along with the exposure to stress if raised by a parent with depression.

Functional Brain Changes

Functional brain changes have been identified in adults but less consistently in adolescent depression. Following negative emotional stimuli, individuals with depression, or at risk of depression, have increased neuronal activity in the cortico-limbic and cortico-striatal circuits, and reduced activation following positive cues, at least in adults [50]. Findings in adolescences are less consistent. It is not clear whether these findings are state dependent, a risk state for developing depression, or as a result of the illness or its treatment [51].
It is difficult to disentangle the significant but normal neurodevelopment in adolescences from more pathological divergence of maturational trajectories, particularly given the need to consider individual variation, gene-environmental interaction and state trait phenomena. Substance misuse in common at the time of adolescence also creates its own deviations, with reduced white matter integrity and reduced density of gray matter. Taken together this makes the study of neurobiology of paediatric depression exciting and frustrating, yielding as much difference as similarity.


Little is unknown in pediatric populations as to how genes contribute to the development of depressive symptoms and how they interact with environment, despite high prevalence of pediatric depression. To investigate these enigmas, large amounts of genetic data, and sophisticated study design and analytical techniques are required. In this section we will first review genetic epidemiology, including family, twin, and adoption studies to shed light on familial aggregation, heritability, environmental influences, and correlation and interaction between gene and environment in Major depressive disorder (MDD). Then recent scientific achievements in molecular genetics, including linkage analysis and candidate gene studies will be highlighted.
Family studies
Several family studies have demonstrated higher levels of depression in the offspring of depressed parents in comparison to the offspring of parents with no psychopathology [52]. In a systematic literature review and meta-analysis of genetic epidemiology in adult populations with major depressive disorder, five family studies that met the inclusion criteria suggested there was strong evidence for an association between MDD in the probands and MDD in first-degree relatives [53]. According to the study a person with first-degree relatives with MDD has 2.84 increased likelihood of developing MDD. Following this, Rice et al. performed a systematic review and meta-analysis of family studies of MDD to investigate the familial risk for childhood. Four parent-to-offspring studies (i.e. studies of children of depressed parents) and six “bottom-up” studies (i.e. studies of the relatives of child probands with MDD) were included in a meta-analysis, revealing OR of 3.98 and 2.30, respectively, in comparison with normal control group53. This study suggested a strong familial aggregation in pediatric depression. However its familial risk was not substantially greater than that computed from adult study (OR 2.89). Retrospective studies indicate a stronger familiarity in populations with early-onset MDD than in ones with late-onset MDD, leading to the hypothesis that childhood-onset (prepubertal) MDD can harbor a stronger familial aggregation in comparison to adolescent- (postpubertal) and adult-onset MDD [54,55]. However, one study using samples from a longitudinal British cohort failed to prove this hypothesis, with relative risk of 1.0 and 0.8 in first and second degree relatives of probands with pre- versus postpubertal MDD, respectively and 0.8 and 1.0 in first and second degree relatives of probands with adult versus postpubertal MDD [56].
Twin studies
Twin and adoption studies are the two principal approaches to distinguish genetic and environmental contributions to MDD. The twin study requires affected probands from both pairs of monozygotic (MD) and dizygotic (DZ) twins, and presence (concordance) or absence (discordance) of the disorder in the co-twins of these affected probands is assessed. Concordance and discordance proportions are used to determine the relative contributions of genetic and environmental factors in the development of illness, which are conceptualized as genes, shared environment (that is, environmental factor that makes twins more alike – the socioeconomic status, for example), and unique environment (that is, non-shared environment that makes twins dissimilar).
In an adult study, heritability computed from data from five twin studies (three studies conducted in the community, one study done in the clinical setting, and one study using population based samples both in the community and the clinical setting) was 37% (95% CI =31 – 42%) [53]. On the contrary, heritability of depressive symptoms in youth varies across studies [52,57] depending on who rated symptoms and what measurements were selected. In the review by Rice et al [52], heritability estimates for parent-rated depressive symptoms varied from 30% to 80%, while lower heritability estimates were found for self-reported depressive symptoms. Despite these inconsistent results, only consistent finding from several twin studies is that the genetic influence is small in childhood and that it increases in adolescence [57,58], which might be due to uncovered gene-environment correlation, gene-environment interaction or additive genetic influence in adolescence [59]. Additionally, two studies from the national Swedish Twin Registry suggested higher heritability in women than in men (40% versus 30%: PMID11352363 and 42% versus 29% [60], and clear evidence was found for sex-specific genetic effects with genetic correlations estimated at + 0.55 and + 0.63.
Adoption studies
Despite the significant importance in genetic epidemiology of MDD, the fact that even MD twins can share much of the same environment poses a challenge to us in differentiating the relative contribution of genetic and environmental factors to the development of illness in twin studies. Adoption study is utilized to overcome this difficulty.
Three adult adoption studies on MDD were included in a systematic review by Sullivan et al, although results of them were not rigorous due to methodological issues (diagnosis of MDD was based on indirect sources rather than personal interviews) and limited statistical power [53]. In younger generation, three adoption studies investigating childhood depression (two studies examined internalizing problems, including depression and anxiety, and one study examined diagnosis of MDD), none of which provided rigorous evidence [61-63].
In summary early-onset depression may be more strongly genetically influenced than adult-onset MDD, suggesting strong familial aggregation (with less environmental contribution) and heritability.
Molecular genetics
Historically, the process of gene discovery required genetic linkage studies following the confirmation of familial aggregation described above. In the genomic era, however, we see different approaches to detect candidate genes using the rapid scientific advancement.
In a recent review article by Flint et al, 26 previously detected candidate genes were meta-analyzed, among which only the following seven genes yielded a significant (p<0.05) results: 5HTTP/SLC6A4, APOE, DRD4, GNB3, HTR1A, MTHFR, and SLC6A3 [64]. Despite these results, the authors reported that little conclusive evidence was yielded from candidate gene to support each gene’s involvement in depression based on the fact that odds ratio estimated over all candidate gene meta-analysis was 1.15.
In genetic linkage studies,two loci (the disease locus and the marker locus) are considered to be linked and inherited together more often than expected by chance within families if they are on the same chromosome and close enough together on it. In MDD, like other diseases, data from genetic linkage studies tend to generate series of publications because the authors acquire additional data. Among those studies, obvious heterogeneity and poor internal consistency are frequently noted, raising a concern for the existence of false positive or negative findings.
Genetic of Recurrent Early-Onset Depression (GenRED) study is one example that highlights these problems in genetic linkage studies in MDD. This study was conducted in six facilities in U.S., using a large sample of families with recurrent MDD [65], in which the authors used genome scan methods and tried to shed light on the idea that early onset and recurrence of depressive episodes might predict substantially increased risk of depression in first-degree relatives compared with the general population. Initial report from this study revealed 15q linkage, although this was not corroborated by the following paper including additional samples, which provided suggestive linkage results on chromosome17p1267 and chromosome 8p22-p21.3 [66,67]. To the best of our knowledge, no genetic linkage studies in pediatric MDD are identified via literature search.
Most recent candidate genes are studies using a case-control association study design. In this study design, genetic variants are usually selected based on a priori hypothesis, such as neurobiological plausibility (serotonin transporter for antidepressants, for example) or genomic location of a candidate gene (i.e. linkage peak). One small study consisting of 68 unrelated patients with DSM-IV depressive disorders and 68 unrelated healthy control subjects has reported significant association between short variant of 5-HTTLPR and pediatric-onset depressive disorders, including both major depressive disorder and dysthymia using a case-control design and a family-based association design [68].
With evidence of candidate genes, further attempts have been made to elucidate whether genetic vulnerability to MDD can be affected by environments. Using longitudinal data of representative samples from birth cohort in New Zealand, Caspi and colleagues concluded that individuals with genetic variation of the short allele of the serotonin transporter gene-linked polymorphic region (5-HTTLPR) were at elevated risk for the development of major depression in adult life in interaction with stressful life experiences in childhood [69], suggesting a potential gene-environment interaction in this illness. In the metaanalysis of published data conducted by Risch and colleagues, although significant association was found between the number of stressful events and depression (OR 1.41, 95% CI 1.25 – 1.57), the effect of 5- HTTLPR genotype alone failed to reach statistical significance on the development of depression. Furthermore, aggregate data from this meta-analysis did not suggest that 5-HTTLPR genotype interact with childhood stressful experiences on the risk of depression in later life (OR 1.01; CI 0.94-1.10) [70]. From a different perspective, thestudy by Kaufman et al revealed the risk of developing MDD in participants with short allele of 5-HTTLPR was removed in the presence of social support [71].
Genetic variants of the Val66Met polymorphism of brain-derived neutrophic factor (BDNF) gene were found to have potential interaction with environmental factors on the risk of major depression [72]. The three-way interaction among the Val66Met polymorphism of BDNF gene and the short variant of 5-HTTLPR has been reported to be associated with the higher risk of pediatric depression in children with maltreatment history in comparison to a normal healthy control group [73]. In this study, social support was also protective against the higher rate of depression following childhood maltreatment in children with the Met allele of BDNPgene. Similar moderating effect on childhood abuse and subsequent adult depression was found with regard to the corticotrophin releasing hormone receptor gene (CRHR1), but in males only [74].
Genetic study has been recently collaborated with neuroimaging. Polymorphism in 5HTTRL genes and variants in BDNF polymorphisms have been associated with both structural and functional changes in the cortico-limbic areas, including hippocampal size and amygdala activation in children with stress and exposed to emotionally laden cues [75,76].
Finally, as described above, Genome-wide association studies (GWAS) is currently available due to the rapid technological advances. This design enabled us to scan markers across the complete sets of genome to detect small variations which are called single nucleotide polymorphisms that occur more frequently in probands with a particular disease than in people without the disease. Flint et al summarized previously published nine studies in MDD revealing no significant findings have been reported in this illness [64]. The authors assumed the failure of GWAS studies to detect candidate loci was most likely attributable to underpowered sample size of each study. No such studies exclusive in pediatric populations with MDD have been conducted to the best of authors’ knowledge.

Final Remarks

Despite a growing body of research into the neurobiological correlates of depression, the evidence in children and adolescent is limited to the extent that no firm conclusions may yet be drawn. Studies that do exist are not always consistent, and are often hampered by ethical and methodological difficulties, particularly in the choice of controls, given then salience of age, gender, previous childhood experience and motivation along with definition and duration of depression. Equally it is recognised that adolescence, a developmental phase characterized by a significant increase in prevalence of depression is also a time of marked physical, cognitive and social change, occurring at different chronological and developmental ages. Thus standardizing for this in studies is difficult and might explain inconsistent findings in the few studies conducted. Lack of evidence does not mean similarities do not exist; only that more robustly conducted and replicated studies are needed. An understanding of normal brain development during adolescence is necessary to put into context possible variations found in studies of youth with or at risk for depression.


Track Your Manuscript