Abnormal volumetric brain morphometry and cerebral blood flow in adolescents with depression

Abstract

BACKGROUND

Prior research has demonstrated that the brains of adolescents with depression exhibit distinct structural alterations. However, preliminary studies have documented the pathophysiological changes in certain brain regions, such as the cerebellum, highlighting a need for further research to support the current understanding of this disease.

AIM

To study brain changes in depressed adolescents.

METHODS

This study enrolled 34 adolescents with depression and 34 age-, sex-, and education-level-matched healthy control (HC) individuals. Structural and functional alterations were identified when comparing the brains of these two participant groups through voxel-based morphometry and cerebral blood flow (CBF) analysis, respectively. Associations between identified brain alterations and the severity of depressive symptoms were explored through Pearson correlation analyses.

RESULTS

The cerebellum, superior frontal gyrus, cingulate gyrus, pallidum, middle frontal gyrus, angular gyrus, thalamus, precentral gyrus, inferior temporal gyrus, superior temporal gyrus, inferior frontal gyrus, and supplementary motor areas of adolescents with depression showed an increase in brain volume compared to HC individuals. These patients with depression further presented with a pronounced drop in CBF in the left pallidum (group = 98, and peak t = - 4.4324), together with increased CBF in the right percental gyrus (PerCG) (group = 90, and peak t = 4.5382). In addition, 17-item Hamilton Depression Rating Scale scores were significantly correlated with the increased volume in the opercular portion of the left inferior frontal gyrus (r = - 0.5231, P < 0.01).

CONCLUSION

The right PerCG showed structural and CBF changes, indicating that research on this part of the brain could offer insight into the pathophysiological causes of impaired cognition.

Key Words: Voxel-based morphometry, Cerebral blood flow, Arterial spin labeling, Adolescent, Depression, The right percental gyrus

Core Tip: In this study, we first combined cerebral blood flow (CBF) and voxel-based morphometry (VBM) to study brain alterations in adolescent depression. We also found that the brain function (CBF) changes were mainly in the left pallidum and right precentral gyrus. Meanwhile, we detected alterations in the cerebellum. Our finding about a wide range of brain structure (VBM) changes in adolescent depression contributes to better treatment and prevention strategies for depression.

INTRODUCTION

Adolescents suffer from high rates of depression that contribute to adverse outcomes, including academic difficulties, substance abuse, behavioral issues, parental conflict, impaired peer interactions, and suicidality[1-3]. Depression rates are exceptionally high among children in middle school, with an incidence of up to 24.3%[4]. Clinical symptoms of depression include feelings of sadness, hopelessness, cognitive impairments, and reductions in motivation and pleasure. Overcoming depression can be particularly challenging among adolescents, with treatment failing to have any pronounced effect in 30%-50% of cases[5]. Recurrence rates are also very high[6], with most adolescents experiencing relapses multiple times in adulthood[7,8]. Adolescence is also a critical period for brain growth[9], and this disease can have a profound negative impact on such development[10]. Despite this fact, many studies have focused on the mental health of adults rather than exploring outcomes in vulnerable adolescent populations[11,12]. As a result, there is an urgent need for in-depth neurobiological research that investigates the pathophysiology of depression to develop effective treatments that can help free adolescents from the symptoms of this debilitating condition.

The results of imaging studies suggest that adolescents suffering from depression exhibit severe structural changes within their brains[13-15]. However, these reported findings have been inconsistent with some studies documenting the thinning of brain regions[16], whereas others note the thickening of certain areas[17]. These inconsistencies may result from a range of patient-specific confounding factors or pharmacological treatment effects. While cerebellar dysfunction has been documented in many disease states[18], changes in the cerebellum in depression remain overlooked mainly[19]. Recent research supports the cerebellum's function in the processing of emotion and its widely recognized role in the context of motor control[20]. Reductions in the gray matter volume in the cerebellum and anterior cerebellar vermis have been documented among adults with depression[21], and some evidence suggests the ability of cerebellar neurons to regulate depression-like behavioral development actively[22].

Given the poorly understood nature of the pathophysiological basis for depression among adolescents, efforts to detect and clarify morphological alterations associated with functional changes provide an opportunity to define these underlying pathogenic processes better. In a previous publication, we used cerebral blood flow (CBF) to assess Major Depressive Disorder(MDD) in adolescents and reported CBF alterations after treatment[22]. Voxel-based morphometry (VBM) is a technique that is frequently applied to detect structural changes within the brain[23], allowing for the voxel-by-voxel classification of brain volumes in different regions to enable the systematic, reproducible comparison of regions among multiple individuals more effectively than the traditional region of interest (ROI)-based strategies. VBM can assess anatomical differences throughout the brain with a short execution time[23,24] and is widely used in studies of alterations in depressed brains[25-27]. Still, there are also conflicting and heterogeneous results reported by VBM and magnetic resonance imaging (MRI) whole brain analysis[16,17]. These examples encourage the further development of VBM in MDD investigations. CBF can also be detected noninvasively and quantitatively through arterial spin labeling (ASL), in which arterial blood magnetic labeling is leveraged as an endogenous tracer[28]. Investigations have utilized VBM and voxel-based pathophysiology to analyze MDD[29]. Because structural alterations are usually accompanied by functional abnormalities[30,31], we decided to extend our study approach to incorporate VBM with CBF. The current study's findings are consistent with the high repetitive rate of change obtained so far regarding depression[32-34]. The cerebellum's role in depressive symptoms was confirmed in a recent study[35], and elevated pallidum CBF was also detected in patients with mild traumatic brain injury that may lead to cognitive decline[36]. These repeats suggest a potential association between the changes we detected using VBM combined with CBF and MDD dysfunction. Accordingly, in the present study, VBM and CBF strategies were used to evaluate changes in the brains of adolescents with depression.

Regular examinations of a broader range of brain regions are required to understand the pathophysiology of depression in adolescents properly, and unmedicated young people diagnosed with first-episode depression are the perfect model for studies of disease-related brain alterations. For the present analysis, these patients were thus the primary subjects of interest in hopes of detecting previously unrecognized yet clinically significant abnormalities present in depressed adolescents that have yet to undergo pharmacological treatment. These approaches offer a promising means of exploring the pathophysiology of adolescent depression while also validating and expanding upon previously published results[37].

Through the application of VBM and CBF techniques, a series of complex structural and functional alterations were documented within the brains of adolescent patients with depression compared to a healthy control (HC) population. The cerebellum was expected to undergo alterations because of its crucial involvement in developmental processes, and depression-related structural and functional changes were anticipated to coincide in several brain regions. These were the two primary outcomes that were expected from the present research.

MATERIALS AND METHODS

Participants

From August 2020 to July 2022, adolescents with depression were recruited from the Department of Psychiatry at Chongqing Medical University's First Affiliated Hospital. The Mini International Neuropsychiatric Interview for Children and Adolescents (MINI-KID) was used by senior psychiatrists to confirm the diagnosis of depression in all cases, and symptom severity was assessed with the 17-item Hamilton Depression Rating Scale (HAMD-17). Patients eligible for study participation were individuals of Han ethnicity who were right-handed, exhibited a HAMD score ≥ 17, had no history of antidepressant treatment, were experiencing first-episode depression, had no history of psychotropic drug use, and had no history of anesthetic or sedative, or analgesic drug use within 1 mo before the study initiation. Patients were excluded if they had a history of prior mental health conditions, including schizophrenia or bipolar disorder, a history of drug abuse or dependence, exhibited MRI contraindications, or had been diagnosed with organic brain diseases or severe physical ailments.

Participants in the HC trial were selected from age-, gender-, and education-level-matched subjects with no personal history of mental illness or psychosis among first-degree relatives. The same exclusion criteria were applied for both patients with depression and HCs. In total, 34 treatment-naive first-episode adolescent depression patients and 34 HCs from 12-17 years of age were recruited for study participation.

The First Affiliated Hospital of Chongqing Medical University approved this study (NO. 2017-157). The legal guardians of all participants provided written informed consent, and all participants were provided with a comprehensive overview of the details of the study.

General data collection

Sociodemographic data were collected using a questionnaire regarding participant age, sex, and educational status. For the MINI-KID, the parents of eligible study participants were contacted and informed about this study, after which questionnaire and interview responses were obtained electronically. One professional interviewer conducted the interview processing, which was completed in under two days[38].

HAMD-17 surveys were administered by two psychiatrists with similar levels of training to assess symptom severity based on appropriate guidelines. Two examiners blinded to participant details independently performed all subsequent scoring to ensure the absence of subjectivity.

MRI data collection

MRI scanning was performed with a Signa 3.0 Tesla MRI instrument (GE Medical Systems, WI, United States), yielding T1-weighted, T2-weighted, and T2-fluid-attenuated inversion recovery (T2-Flair) imaging sequences. T2-Flair data were used to detect any evidence of potential brain disorders, with patients exhibiting such conditions being eliminated from further study inclusion.

Structural images were acquired with no gap using the following settings: the number of axial slices = 156, time repetition (TR) = 8.4 ms, echo time (TE) = 3.3 ms, flip angle (FA) = 12°, slice thickness = 1.0 mm, the field of view (FOV) = 240 mm × 240 mm, matrix = 240 × 240, and voxel dimension = 1 mm × 1 mm× 1 mm. Parameters used for ASL image acquisition were as follows: NEX 3, TE 9.8 ms, TR 4639 ms, FOV 240 mm × 240 mm, slice thickness 4.0 mm, post label delay time 1525 ms, 3D spiral k-space filling, points 512, arms 8, acquisition scan slices 40. During imaging, participants were directed to remain awake with their eyes closed, and foam padding was employed to restrict head movement.

Image processing

FSL-VBM was used to analyze structural images[39]. After the initial recovery of these images, they were segmented into white matter (WM), gray matter (GM), and cerebrospinal fluid regions, followed by nonlinear registration-based alignment of these images to the Montreal Neurological Institute (MNI) standard space. Native GM images were then re-registered nonlinearly to the resultant averaged images to produce a template specific to the present study. Native GM images were then registered to this template and modified as appropriate based on local contraction or expansion associated with nonlinear spatial transformation. The altered GM images were then smoothed with an isotropic Gaussian kernel with a sigma of 3 mm (FWHM = 6.9 mm) as per FSL recommendations[39].

ASL images were analyzed following the earlier reported studies[22,28].

Statistical analysis

SPSS 22.0 (IL, United States) was used to analyze all data. Data are reported as means ± SD and were compared between groups with two-way ANOVA, while comparisons between males and females were made using chi-square tests.

Analyses of GM and WM volumes included group-based comparisons of larger and smaller volumes in HC individuals relative to adolescents with depression. These analyses employed grand mean scaling and absolute threshold masking. Voxel-level GM and WM density were compared between these groups with two-sample t-tests using SPM5. Random Gaussian field theory was used to estimate significant differences among groups using a familywise error-corrected alpha of 0.05. Clusters were displayed as statistical parametric maps in standard anatomical space. Participant age, education status, and total intracranial volume were utilized as covariates when conducting two-way ANOVA analyses of smoothed CBF images, with an initial P < 0.001 thresholds for cluster-level false discovery rate (FDR) multiple comparison correction (P < 0.05). Regions of the brain affected by sex or group interactions for CBF perfusion were then acquired with the xjview software (https://www.alivelearn.net/xjview). Brain areas with interactive effects were selected as ROIs, and the CBF values for these ROIs were extracted using DPABI[40].

Relationships between brain alterations and depressive symptom severity were explored through Pearson correlation analyses with P < 0.05 as the significance threshold.

RESULTS

Participant characteristics

Table 1 provides details regarding the characteristics of the study participants. No differences in age, sex, or education level were detected between depressed adolescents and HCs (P > 0.05), whereas HAMD-17 scores differed significantly between these groups (P < 0.001).

Table 1 Participant demographic and clinical characteristics.

Item	depressed adolescents (n = 34)	healthy controls (n = 34)	t/χ2	P value
Gender (F/M)	23/11	23/11	0a	1.00
Age (yr)	15.26 ± 0.99	15.68 ± 1.77	- 1.18b	0.24
Education (yr)	10.06 ± 1.98	10.44 ± 1.97	- 0.80b	0.43
HAMD-17	26.12 ± 4.43	1.03 ± 1.70	30.84b	< 0.001

^aχ² test.

^bTwo sample t-test.

The values are illustrated as mean ± SD. F: Female; M: Male; HAMD-17: 17-item Hamilton Depression Rating Scale.

Depression-related changes in VBM

FSL-VBM analyses exploring differences between groups highlighted increases in the volume of several regions of the brain following correction for multiple testing (P < 0.05), including the cerebellum, superior frontal gyrus, cingulate gyrus, pallidum, middle frontal gyrus, angular gyrus, thalamus, precentral gyrus, inferior temporal gyrus, superior temporal gyrus, inferior frontal gyrus, and supplementary motor area (Table 2, Figure 1). No regions of the brain exhibited depression-related decreases in brain volume.

Figure 1 Two-sample tests comparing healthy controls and adolescents with depression based on a whole-brain mask.

Table 2 Whole-brain group statistics for healthy controls and adolescents with depression.

Brain sub-region	Peak (MNI)			Number of voxels	P value (FWE corrected)
	X	Y	Z
HCs > depressed adolescents
Cerebellum_6_L	73	64	56	973	0
Frontal_sup_medial_L	81	16292	101106	925	0.01
Cingulum_mid_R	103	120	67	221	0.02
Pallidum_R	113	-6	49	211	0.02
Cingulum_mid_L	-4	168	92	146	0.04
Frontal_mid_L	67	70	118	113	0.03
Angular_R	119	116	92	88	0.03
Thalamus_L	85	136	102	74	0.03
Precentral_R	151	134	27	57	0.04
Temporal_inf_R	127	142	46	55	0.04
Temporal_Pole_Sup_L	61	130	98	42	0.04
Frontal- Inf_Oper_L	51	66	105	25	0.04
Occipital_Mid_L	63	126	148	24	0.04
Supp_Motor_Area_L	79	88	52	14	0.04
Cerebellum_3_R	101			13	0.04

HCs: Healthy controls; MNI: Montreal Neurological Institute; FWE: Family wise error.

Depression-related changes in CBF

Adolescents with depression exhibited pronounced decreases in CBF in the left pallidum as compared to HCs (group = 98, and peak t = -4.4324), together with an increase in the right percental gyrus (PerCG) (group = 90, and peak t = 4.5382) (Figure 2). This analysis was performed using a cluster cut-off P-value < 0.001.

Figure 2 Differences in cerebral blood flow when comparing healthy controls and adolescents with depression. Blue and red indicate reductions and increases in cerebral blood flow, respectively.

Correlations between brain alterations and depression severity

In patients with depression, HAMD-17 scores were significantly correlated with the increased volume of the opercular part of the left inferior frontal gyrus (IFG) (r = - 0.5231, P < 0.01). No correlations were detected between HAMD-17 scores and other changes in brain volume or CBF (P > 0.05).

DISCUSSION

The current findings showed that, in contrast to HCs, treatment-naive first-episode depression in adolescents is associated with CBF and brain volume modifications. Notably, increased volume of many brain regions was observed in adolescent depression patients, with increased CBF in the PerCG and decreased CBF in the left pallidum. HAMD-17 scores in these patients were also negatively correlated with the volume of the opercular portion of the left IFG. In particular, decreases in both volume and CBF were observed in the right PerCG, suggesting that this region may be an important area for future research to understand the pathophysiological bases of depression.

The PreCG regulates primary motor behaviors and has been reported to exhibit alterations in individuals with movement-related disorders, including amyotrophic lateral sclerosis[41] and akinetic-rigid Parkinson's disease[42]. Increased PreCG CBF is correlated with increased reaction time[43], pure apraxia of speech and other forms of impaired verbal fluency[44], and autism[45]. Several studies suggest a relationship between PreCGand emotion[46]. Depression can impair neural plasticity in the motor cortex, contributing to behavioral and cognitive alterations in affected patients[47]. Pathological alterations in the PreCG may thus be functionally linked to depression. An increase in PreCG GM volume has been reported[48,49], with the present results confirming these prior reports and highlighting a link between these changes and altered CBF.

The left IFG plays a crucial role in the functional and structural brain network that underlies language in neurocognitive models of language perception[50]. The opercular portion of the left IFG is a crucial node connected to multiple cerebellar, cortical, and subcortical structures within the language network[51]. In the present study, adolescents with depression exhibited a correlation between HAMD-17 scores and the volume of this opercular portion of the left IFG, consistent with its potential association with depression severity.

The cerebellum is thought to be closely related to movement, with additional experimental evidence highlighting its association with attention and motor systems[52], as well as with cognition[53]. This link could explain depression-related symptoms, such as cognitive impairment, disinterest, and poor attention. Adolescent depression may affect the cerebellum, contributing to frequent episodes in adulthood. However, it is essential to highlight that the reported increase in the cerebellum in the present study contradicts other published studies in which cerebellar volume was shown to be reduced[54,55]. These differences may be attributable to variations in disease severity, assessment methodology, diagnoses, or treatment approaches.

Other regions exhibiting abnormal findings included the cingulate gyrus, an essential component of the pathophysiological basis for depression related to the makeup of the default mode network[56]. In clinical research, cingulate gyrus abnormalities have been reported in various psychiatric disorders[57]. Altered WM connectivity in adolescent depression patients has also been described[58], contributing to reduced default mode network connectivity closely related to depression incidence[59-61].

There are various drawbacks to this study. Firstly, the number of participants was relatively small, partly due to parental reservations about fMRI examinations. Secondly, the effects of resting state physiological noise derived from the heart and respiratory rhythms cannot be eliminated from these analyses. Lastly, this cross-sectional analysis could not detect disease progression-related changes in brain morphological characteristics, underscoring a need for future longitudinal research with multiple follow-up visits per patient to monitor better such changes and how they respond to treatment.

CONCLUSION

In summary, the present results demonstrate that adolescent depression patients exhibited various structural changes in the brain and altered CBF in the left pallidum and right PerCG. Considering the finding that the right PerCG demonstrated structural and CBF changes associated with depression in these adolescent subjects, research on the right PerCG may provide fresh perspectives on the pathophysiology of this devastating psychological disorder. This identification of novel evidence regarding changes in the cerebellum and other brain areas provides robust imaging evidence supporting the hypothesis that these regions are involved in cognition and disease-related pathogenesis. The future application of these findings has the potential to guide the better prevention and treatment of depression among adolescents.

ARTICLE HIGHLIGHTS

Research background

The prevalence of depression in adolescents is high and research is scarce. Therefore, it is urgent to investigate depressed adolescents.

Research motivation

In-depth neurobiological studies investigating the pathophysiology of depression are urgently needed to develop effective treatments to help adolescents escape the symptoms of this debilitating illness. Several studies have documented the role of the cerebellum in psychiatric disorders, which is rarely mentioned in the imaging of depression, emphasizing the need for further research on the depressed brain.

Research objectives

We aimed to detect structural and functional changes in depressed adolescents. These changes may be relevant to better prevention and treatment of adolescent depression. We found that adolescents with depression exhibit various structural changes in the brain and alter cerebral blood flow in the left syphilitic spiral and right percent gyrus.

Research methods

This study recruited 34 adolescents with depression and 34 matched healthy control (HC) individuals. Voxel-based morphometry (VBM) discovers structural changes in the brain; Cerebral blood flow (CBF) explore functional changes in the brain; 17-item Hamilton Depression Rating Scale (HAMD-17) measures depression; t-test assess statistical differences.

Research results

We found that patients with adolescent depression exhibit various structural changes in the brain and altered CBF in the left pallidum and right percental gyrus. These findings may provide new insights into the pathophysiology of this disruptive psychological disorder and provide strong imaging evidence to support the hypothesis that these regions are involved in cognition and disease-related pathogenesis. Future applications of these findings have the potential to guide better prevention and treatment of depression in adolescents.

Research conclusions

New theory: Our study provides imaging evidence that supports the hypothesis that cerebellum is involved in cognition and disease-related pathogenesis. New method: We first combined VBM and CBF to examine the brains of depressed adolescents.

Research perspectives

Incorporating imaging data into the diagnosis of psychiatric disorders.