Abstract
Mood disorders, including bipolar disorder (BD) and major depressive disorder (MDD), are serious and often-recurring psychiatric conditions that commonly first manifest during childhood or adolescence (1, 2). Youth with mood disorders are four times more likely to attempt suicide (32% vs. 8% in the general population) (3), are at elevated risk for developing co-occurring psychiatric disorders (4), and frequently experience family maladjustment and exposure to significant early-life stress (5, 6).
Mood disorders, including bipolar disorder (BD) and major depressive disorder (MDD), are serious and often-recurring psychiatric conditions that commonly first manifest during childhood or adolescence (1, 2). Youth with mood disorders are four times more likely to attempt suicide (32% vs. 8% in the general population) (3), are at elevated risk for developing co-occurring psychiatric disorders (4), and frequently experience family maladjustment and exposure to significant early-life stress (5, 6). Pediatric mood disorders are also associated with academic impairment and reduced global functioning (7). Despite advances in accurately diagnosing and treating pediatric mood disorders, objective diagnostic and therapeutic biomarkers anchored in neurophysiological underpinnings of these psychiatric conditions are only beginning to emerge.
Neuroimaging studies in humans hold promise for a mechanistic understanding of aberrant structure and function in brain regions that contribute to the onset, persistence, and recurrence of mood disorders that start in childhood. For example, studies thus far have revealed altered interactions between prefrontal and subcortical brain regions that are central to mood disorders and putatively result in dysfunctional regulation of emotion and cognition over time (8) (Figure 4.1).
Figure 4.1 Candidate brain regions and circuits associated with aberrant responses to stress and reward in youth with or at risk for a major mood disorder.
Brain regions and circuits that subserve core symptoms in pediatric mood disorders may function pathologically or may function to compensate prevention of pathology when compared to healthy control (HC) youth. For example, aberrant emotion salience and regulation, reward processing, and cognition in mood disorders may result from anomalous reciprocal connections between the amygdala and dorsal and ventral prefrontal areas (8, 9), which may further result in anatomical differences compared to healthy youth, including amygdala volume reductions (10). These anomalies may further generate susceptibilities toward dysfunction in other brain regions related to emotion, including the anterior cingulate cortex (ACC), ventrolateral prefrontal cortex (VLPFC), striatum, thalamus, cerebellar vermis, and hippocampus (11). Alternatively, structural abnormalities, such as reductions in the amygdala and hippocampal volumes, may characterize mood disorders independently of functional abnormalities, including the subgenual anterior cingulate cortex (sgACC), dorsolateral prefrontal cortex (DLPFC), amygdala, and ventral striatum. Understanding variations in structure–function interactions in the brain over development may be critical to the pathophysiology of underlying mood disorders.
We subsequently review functional and structural magnetic resonance imaging (MRI) data that demonstrate aberrant responses to stress and reward and maladaptive developmental trajectories in pediatric BD and MDD. We conclude with implications of these neuroimaging markers for treatment of pediatric mood disorders.
4.1 Aberrant Responses to Stress and Reward
Acute stress causes a natural effect on attention, motivation, mood, perception, and other aspects of mental function. Although these brain responses to stress may be adaptive for most, in youth with BD and MDD, they may adversely affect the stress-response pathway and lead to maladaptive behaviors and outcomes. Similarly, mood disorders are centrally conceptualized by aberrant responses to reward- or goal-directed stimuli (9). Aberrant stress and reward responses can biologically interact, especially in the context of early exposures to trauma or adversity (6), and reveal themselves through key biomarkers. Selected biomarkers are outlined subsequently and summarized in Table 4.1.
Table 4.1 Selected brain biomarkers of aberrant responses to stress and reward likely leading to mood disorders in youth
| Region/network | Major depressive disorder (MDD)/bipolar disorder (BD) | Specificity/implication for mood disorder | Reference |
|---|---|---|---|
| Stress | |||
| Medial temporal lobe | BD | Reductions in amygdala and hippocampal volume* | (10, 45) |
| Connectivity between the right laterobasal amygdala and right hippocampus positively correlate with anxiety levels | (15) | ||
| MDD | Reductions in gray matter | (16) | |
| Anterior cingulate cortex (ACC), ventral medial prefrontal cortex (vmPFC), orbitofrontal cortex (OFC), amygdala | MDD | Across tasks assessing emotion processing, cognitive control, affective cognition, reward-processing, and resting state, elevated neural activity in ACC, vmPFC and OFC, and the amygdala | (26) |
| Reward | |||
| Striatum | BD | Increased putamen volume | (8) |
| Increased caudate volume | (8) | ||
| Bilateral caudate and left putamen volumes related inversely to age and pubertal status | |||
| Increased left nucleus accumbens (NAcc) | (48) | ||
| Smaller NAcc | (49) | ||
| MDD | Reductions in overall striatal volume | (22) | |
| Decreased striatal activity during reward feedback | (22) | ||
| Decreased activity in right caudate (body and head) and left caudate body during reward feedback | (22) | ||
| Decreased activity during reward anticipation bilaterally at caudate head and left putamen. Stronger blunting of activity in younger age studies | (22) | ||
| Decreased activity in caudate, putamen, and globus pallidus | (22) | ||
| Increased connectivity of ventral striatum predicts depression | (51) | ||
In pediatric BD, the most notable brain biomarker of aberrant response to stress is reduced medial temporal lobe volumes in the amygdala and hippocampus. Reduced amygdala volume in pediatric versus adult populations with BD is consistently supported by metanalytic data (10). Indeed, amygdala and hippocampal volumes are highly sensitized to stress exposure, as evident in studies of adults with BD with and without exposures to traumatic life events, so the prevalence of these biomarkers in pediatric BD, which commonly involves onsets of mania that may be associated with or triggered by a stressful life event, follows logically (12, 13).
It is reassuring that prefrontal cortical activation has been found to be relatively intact in children, compared to adults, with BD. In the context of exposure to stress, intact prefrontal function is critical for effective emotion regulation. In some studies, euthymic children with BD have exhibited prefrontal overactivation (e.g., in the ACC) during emotional paradigms, such as a 2-back visuospatial working memory task and an affective task involving the visualization of positively, neutrally, or negatively valenced pictures (1), suggesting that emotion dysregulation may interrupt typical cognitive function during times of stress.
Neuroimaging studies also show a statistical trend for decrease in hippocampal volume in youth with BD (14). Through inhibitory connections with other subcortical structures, the hippocampus is involved in the appraisal and regulation of stress and the generation of emotion and memory, indicating that youth with BD who have a smaller hippocampus may be more susceptible to stress. Reduced hippocampal volume is typically interpreted as resultant of enhanced cortisol release in response to hyperarousal, but might also be a preexistent risk factor (6). Of note, adversity may also contribute to reduced hippocampal volume, thus enhancing the likelihood of developing BD. In addition to volumetric comparisons, studies have also found that connectivity between the right laterobasal amygdala and right hippocampus is positively correlated with levels of anxiety (15). This further supports the role of the hippocampus in stress processing.
Reduced hippocampal volume is a recurring finding in youth with MDD. Reductions in hippocampal volume are a consequence of depressive symptoms detected as early as preschool (16). As with BD, this implicates an aberrant response to stress in adolescents with MDD and may represent a preexisting risk factor. Because of the significant role of the hippocampus in emotion regulation, the impact of decreased volume signifies a blunted response during emotion processing. Studies show that GM reductions in hippocampal volume are associated with an increased number of depressive episodes, greater symptom severity, and longer illness duration (17). Advancements in imaging resolution and processing have helped delineate amygdala–hippocampal boundaries, morphology, and subfields (18, 19), which provide additional granularity to explain progressive deficits in structure and function of the amygdala and hippocampus in relation to illness duration.
Similar to stress processing, reward processing activates limbic, striatal, and prefrontal systems in the brain. For healthy individuals, activity in the nucleus accumbens (NAcc) shell, ventral pallidum, parabrachial nucleus, orbitofrontal cortex (OFC), and insular cortex commonly instantiate reward-related pleasure. In youth with MDD and BD, however, a different neuroanatomical map is observed.
Reward processing in pediatric BD.
BD is defined by aberrant emotion and motivation that may lead to prompt risk-taking behaviors, and may be characterized by fluctuating experiences of hyperhedonia and anhedonia (20). Studies demonstrate that youth with BD compared with HCs report increased reward reactivity, greater arousal in reward conditions, and greater satisfaction with winning (21). These reports are substantiated in the pediatric brain through increased striatal volume and striatal hyperactivity, including enlargement of the putamen, caudate, and NAcc.
Youth with BD with higher levels of manic symptoms have decreased NAcc activation during reward anticipation compared to those with fewer manic symptoms (20), perhaps indicating that errors in reward prediction signaling may come from NAcc desensitization during reward activation. Together, these patterns of activation may clinically manifest in BD as grandiosity and dysregulated goal pursuit (20). In healthy youth offspring of parents with BD, increased OFC activation and increased pregenual anterior cingulate connectivity to the VLPFC while anticipating rewards suggest pre–illness-specific prefrontal regulatory mechanisms that may be deployed to mitigate reward activation and to keep mania symptoms at bay. Notably, in high-risk but not low-risk children, novelty seeking was associated with increased striatal and amygdalar activation in the anticipation of losses, and impulsivity was associated with increased striatal and insula activation in the receipt of rewards, suggesting potential targets for preventive intervention.
Reward processing in pediatric MDD.
Depression is also characterized by an inability to modulate behavior in response to intermittent rewards, possibly due to blunted phasic dopaminergic signaling critically implicated in reward learning (9). Recent findings indicate that ventral striatal (VS) blunting might constitute a risk factor for depression. Specifically, reduced reward-related VS activation predicted increased depressive symptoms over two years among adolescents (22).
Youth with MDD show decreased volumes in the striatum and decreased activation in the caudate, putamen, and globus pallidus during reward responses (22). This suggests decreased brain sensitivity to anticipating and consuming rewards, which is a defining characteristic of anhedonia and a core feature of MDD. Structural abnormalities in key prefrontal subregions, including the VLPFC, the DLPFC, the ACC and sgACC, and the OFC (23), may result in impairments in executive functions, emotion regulation, and reward processing in MDD.
In youth with MDD, abnormal limbic processing is attributed to decreased regulatory control of dorsal cortical regions (24) and altered functional activation and connectivity in extended medial prefrontal network regions, including the ACC, OFC, ventromedial PFC, and closely linked subcortical areas that include the amygdala (elevated amygdala activation) (25) and striatal regions. These prefrontal–limbic network findings have been observed in both medicated and unmedicated youth with MDD, and cover various domains of brain functioning, including tasks of cognitive control, emotional processing, affective cognition, stress, and reward-based decision-making (26). Similar to adults, depressed youth have been found to have alterations in prefrontal activation, including the medial and ventral subregions, resulting in impaired regulation of emotion during depression (26).
The neural network dysfunction observed in depression has interactive or consequent effects with multiple regulatory systems that may manifest through co-occurring conditions. For example, recent evidence suggests that a shared brain motivational network underpins the co-occurrence of depression and obesity among youth (27). Reward neural circuits (including regions such as the ACC and hippocampus) may underlie dysfunctional behavioral responses and increased sensitivity to rewarding aspects of ingesting high-calorie food that lead to disinhibitory behavior toward eating despite satiation. Youth with greater levels of insulin resistance (IR) show higher levels of anhedonia and food-seeking behaviors, reduced hippocampal and ACC volumes, and greater levels of ACC and hippocampal dysconnectivity to fronto-limbic reward networks at rest. Moreover, for youth with high levels of IR, thinner ACC and smaller hippocampal volumes were associated with more severe depressive symptoms. The opposite was found for youth with low levels of IR. The ACC–hippocampal motivational network that underlies depression and IR separately may signify an important neural collaboration that connects these two syndromes to each other. Reward neural systems are also perturbed by early-life stress such that in youth with depression and obesity, higher levels of abuse moderate the relation between reward network connectivity in the amygdala, insula, and NAcc and IR (28).
Collectively, these neuroimaging findings support mechanistic formulations of pediatric mood disorders as having aberrant responses to stress and reward mapped along abnormal structure and function of prefrontal–limbic and prefrontal–striatal networks. Stress-related reductions in medial temporal neuroanatomy and overactivity of striatal and limbic networks during tasks of emotion and reward processing are consistent across BD and MDD youth. Whereas reduced-reward system task activation and resting-state connectivity may represent a dimensional phenotype of depression severity common to bipolar and unipolar depression, categorical differences in reward system resting-state connectivity between unipolar and bipolar depression may represent a differential risk for mania. The biomarkers that emerge from these mechanisms hold promise for the development of novel and targeted interventions matched to individual profiles of stress and reward response.
4.2 Maladaptive Developmental Trajectories
Cortical GM maturation in childhood has been marked by early neuronal development and volume increases culminating in puberty, followed by discriminatory removal and myelination, and volume loss and thinning. This inverted-U-shaped course, in addition to cortical thickness, has been mapped to the evolution of emotion and cognitive regulation of emotion (29). Because many major psychiatric illnesses are viewed as neurodevelopmental, research on how mood disorders affect developmental trajectories is critical.
Disturbances in prefrontal–subcortical circuits in pediatric BD may manifest partly from structural volumetric deviations within regions in these circuits or come from distinct networks in BD that may result in less glial or neuronal support for specific brain structures (1). These interconnected functional and structural differences may contribute to the development or exacerbation of mood symptoms or represent markers of resilience in youth at familial risk for BD and MDD. For example, when compared with youth with a family history of BD or MDD who developed a mood disorder, those who were resilient had higher left inferior parietal lobe connectivity with visual cortical regions while processing happy faces and higher inferior frontal gyrus connectivity with frontal, temporal, and limbic regions while processing fearful faces (30). Through prospective evaluation of youth at risk for mood disorders, markers of conversion to a mood disorder versus resilience from developing a mood disorder may be delineated. Indeed, cortical GM loss that is typical during adolescence has been observed to occur at an accelerated rate in children with BD, suggesting disorder-related disruptions in typical neurodevelopment (31).
In a longitudinal study of youth with MDD, Luby and colleagues found that global reduction in GM volume, as indexed by cortical thinning, was directly associated with depression severity (29). Given that the age range covered in the study (preschool to adolescence) is characterized by a reduction in GM volume, putatively indicating synaptic pruning, this study suggests accelerated synaptic pruning in individuals who have experienced depressive symptoms (2). Moreover, their findings showing depression-related changes in volume and thickness of cortical GM present as early as middle childhood add to an expanding body of neuroimaging data in youth with MDD.
BD and MDD are often recurrent, and it is possible that an initial mood episode in youth may trigger key biological processes, in turn leading to subsequent mood episodes that may also be exacerbated by life stressors. Future investigation could use longitudinal modeling to examine precisely how the accumulation and timing of stressors along with the experience of mood symptoms influence brain development (2) or compare youth at familial risk for depression to those who have already developed depression (32). Table 4.2 summarizes studies that suggest maladaptive developmental trajectories, though additional longitudinal studies are needed to understand mood disorder onset and persistence over time, especially into adulthood.
Table 4.2 Selected brain biomarkers of maladaptive developmental trajectories likely leading to mood disorders in youth
| Region/network | Impact | Major depressive disorder (MDD)/bipolar disorder (BD) | Specificity/implication for mood disorder | Reference |
|---|---|---|---|---|
| Whole brain | Drives functional deficits during critical periods of neurodevelopment | BD | Volumetric reductions in GM and white matter in the cerebrum, but not in cerebrospinal fluid | (8, 45) |
| Parietal and temporal lobe volume reductions in regions pertaining to face recognition, attentional control, and memory | (45) | |||
| Significant reductions in superior temporal gyrus in the temporal lobe | (52) | |||
| MDD | Bilateral reductions in cortical GM thickness and in the volume of the right hemisphere, but needs further replication | (29) | ||
| Cortical thinning in anterior, subgenual, and posterior cingulate, and medial orbitofrontal cortex of the right hemisphere | (53) | |||
| Thinning in right pericalcarine gyrus, postcentral gyrus, superior parietal gyrus, and left supramarginal gyrus | (54) | |||
| Larger lateral ventricular volume, but smaller frontal lobe volume | (55) | |||
| General reductions in whole brain volumes | (56) | |||
| Prefrontal cortex (PFC) | Impairment in executive functions, processing of rewards and motivation, and regulation of emotion and attention (1) | BD | Reductions in dorsolateral prefrontal cortex (DLPFC) volumes | (48) |
| Reductions in ventrolateral PFC volumes | (57) | |||
| Reductions in ACC | (58) | |||
| Bilateral anterior and subgenual ACC reductions most pronounced after illness onset | (59) | |||
Subgenual anterior cingulate cortex (sgACC) volume reductions observed in the left hemisphere** | (60) | |||
| Amplified bilateral posterior sgACC volumes with past mood stabilizer exposure | (36) | |||
| MDD | Aberrations in rostral ACC (rACC) volume for boys** | (61) | ||
| Decreased GM rACC volumes in patients with comorbid MDD and borderline personality disorder*** | (62) | |||
| Reductions in GM volume in bilateral DLPFC | (63) | |||
| Amygdala | Impairment to perception and emotional valence, memory, and learning | BD | Decreased amygdala volume# | (8, 45) |
| MDD | Bilateral reductions of GM in the caudate nucleus | (64) | ||
| Thalamus and other regions | Impairments in learning, processing, and regulating social emotions (e.g., guilt) | MDD | Reduced GM in right superior and middle temporal gyri and in the thalamus | (64) |
| Decreased left and right anterior insula volumes in school-age children previously diagnosed with preschool-onset MDD | (65) | |||
| BD | None | (8, 45) |
* Studies mixed contingent upon specific region/hemisphere examined and sex of participants.
Related posts:
Chapter 1 – Brain Imaging Methods in Mood Disorders
Chapter 2 – Neuroanatomical Findings in Unipolar Depression and the Role of the Hippocampus
Chapter 8 – Magnetic Resonance Spectroscopy Investigations of Bioenergy and Mitochondrial Function in Mood Disorders
Chapter 9 – Imaging Glutamatergic and GABAergic Abnormalities in Mood Disorders
Chapter 15 – Magnetoencephalography Studies in Mood Disorders
Chapter 16 – An Overview of Machine Learning Applications in Mood Disorders
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
