Major depressive disorder (MDD) is one of the most common psychiatric disorders, and lifetime prevalence reaches up to 16%, and 12-months prevalence is about 6.6%. MDD present various symptoms such as the dysregulation of mood, cognition, and behavior. In spite of high clinical prevalence of the disease, its pathogenesis has not been fully elucidated due to the diversity of clinical features and pathogenesis [1-3]. The relations between the low emotional regulation which is the major feature of major depression and the structural changes such as the medial prefrontal cortex, amygdala, hippocampus, the striatum, and thalamus of MDD’s brain have been proposed [4-8].

A number of magnetic resonance imaging (MRI) studies have attempted to investigate the structural brain changes of MDD. In particular, voxel-based morphometry (VBM) has been utilized to investigate morphological changes in the whole brain of MDD. VBM is a automated MRI analysis method to process morphological brain MRI data, and it minimizes operational bias, and objectively analyzes changes in the regional volume of brain tissues [8]. Many neuroimaging studies have been extracted various morphological features of human brain such as curvature [9, 10], fractal dimension [11, 12], thickness [13], and gyrification index [14]. Particularly, sulcal depth has been studied as an important neuroimaging biomarker for brain diseases and has been widely used to study the morphological characteristics of the cerebral folding [9, 15].

But, we could not find the sulcal depth related neuroimaging study in MDD. FreeSurfer (https://surfer.nmr.mgh.harvard.edu) is the most popular neuroimaging analysis software for MRI analysis; however, FreeSurfer did not provide sulcal depth analysis. CAT12 (https://neuro-jena.github.io/cat) is a useful neuroimaging software that provides structural MRI analysis, and has recently been added to the sulcal depth analysis pipeline. Therefore, this could be one of the reasons that sulcal depth studies have rarely been conducted in MDD.

Cortical gyrification is a complex process that occurs throughout life. It has been known that cortical regions are involved in the processes of brain development, aging, and brain disease-related degeneration. Sulcal depth is an important biomarker for brain morphology in neuroscience and neuropsychological diseases. The structural folding and deepening process of sulci are related to functional areas and occur during brain development, including early radial and later tangential growth. The deep sulcal areas are thought to be the first cortical folds in the early stages of brain development, and the formation of cortical folds could be related to genetic control and cytoarchitectonic areas [16]. The sulcal depth is sensitive to cerebral atrophy, thought to be related to reduced cortical thickness, gyral white matter volume, or tension of the cortico–cortical connections in the subcortical white matter could be secondary deformation due to disease progression. Several studies have analyzed brain morphological changes using sulcal depth, such as in schizophrenia or Alzheimer’s disease (AD) [17, 18]. These studies reported that generally reduced sulcal depths are related to cortical atrophy, and atrophy of the pial surface decreases the sulcal depth. Cortical structures, including the gyri and sulci, can be affected by genetic factors, neuropsychiatric diseases, and aging.

Previously, we performed various MRI study including volumetric MRI, VBM, surface-based vertex analysis, cortical thickness analysis based on old MDD cohort with Magnetom Trio 3T MRI [12-14]. In this study, new MRI data of MDD cohort collected using MAGNETOM Prisma 3T MRI which provides very high gradient performance with maximum gradient amplitude of 80 mT/m and a slew rate of 200. For the benefit of our new MRI scanner, high resolution T1 MRI (isotropic 1 mm) with high signal to noise ratio could be scanned in all subjects, and highly qualified T1 MRI is the most important factor to calculated sulcal depth in cerebral cortices.

In the present study, we hypothesized that there might be significant morphometric changes in the cerebral cortices between depressed patients and healthy controls (NMLs). To test this hypothesis, we investigated the morphometric differences between MDD patients and age and sex-matched NMLs using VBM and sulcal depth analyses.

Participants

We recruited 69 patients diagnosed with MDD from the outpatient Psychiatry Clinic at the Korea University Anam Hospital in Seoul, Republic of Korea between May 2018 and December 2021. The sample was confirmed as MDD by board certified psychiatrists (Ham B-JH, Han KM) using the Structured Clinical Interview from the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition Axis I disorders. Basic demographic information, including age, sex, and education level, as well as clinical information, were collected. The patients were aged between 19 and 65 years and had been diagnosed with MDD by board-certified psychiatrists using a structured clinical interview from the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Axis I disorders. The exclusion criteria were as follows: (i) any other major psychiatric comorbidity, including personality or substance use disorders; (ii) current psychotic features (e.g., delusions or hallucinations); (iii) history of serious and unstable medical illness; (iv) primary neurological illness, including head trauma with residual effects; and (v) any contraindication to magnetic resonance imaging (MRI) scan, such as metal implants or claustrophobia. Illness duration was also assessed in patients with MDD using the life-chart methodology, psychotropic medication history, and their current status. Illness duration was defined as the elapsed time since the patient had experienced their first mood episode, regardless of inter-episodic periods. We also recruited 23 age-matched normal controls (NML) via community advertisements. Controls were screened for major psychiatric histories and none had a psychiatric disorder. This study was approved by the Institutional Review Board of Korea University Anam Hospital (IRB No: 2018AN0118) and adhered to the principles of the Declaration of Helsinki.

MRI acquisition

All the subjects were scanned with a 3.0 Tesla Siemens MR scanner (Magnetom Prisma, Siemens Medical Solutions, Erlangen, Germany). coronal 3D T1-weighted magnetization prepared rapid gradient echo (MPRAGE) was obtained with the following parameters: ST=1.0 mm, no gap, number of slices=384 slices, TR/TE=2,300/2.26 milliseconds, number of signal averages=1, matrix=256×256, and FOV=256 mm×256 mm. Coronal MRI images were obtained perpendicular to the long axis of the anterior commissure to the posterior commissure in the midsagittal pilot.

Voxel based morphometry and sulcal depth processing

The MRIs were processed using the Computational Anatomy Toolbox (CAT12, http://www.neuro.uni-jena.de/cat12) for voxel-based morphometry (VBM) and sulcal depth analysis. For VBM, MRIs of all subjects were bias-corrected and spatially normalized to the Diffeomorphic Anatomical Registration Through Exponentiated Lie Algebra (DARTEL) template. Normalized MRI images were partitioned into gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF). The voxel values of the GM partitions were modulated to preserve the actual GM vales. Modulated and unmodulated GM were smoothed using an 8-mm full width at half maximum (FWHM) isotropic gaussian kernel (IGK). The values of individual total intracranial cavity volume (ICV) were automatically calculated from the GM, WM, and cerebrospinal fluid.

For sulcal depth, the CAT12 toolbox included the calculation of the central surface of the bilateral hemisphere based on projection-based thickness. The pipeline of surface analysis included topological correction, spherical mapping, spherical registration. The estimated GM and WM boundaries were constructed by classifying all WM voxels in the MRI. After computing the curvature of the surface, the surface representation was reconstructed to obtain the finest scale of the local surface curvature. The sulcal depth of individuals were calculated based on the Euclidean distance between the central surface and its convex hull. Mean values of sulcal depth were extracted Destrieux (aparc a2009s) atlas using the surface extract additional surface parameters function.

Statistical analyses

Statistical analyses were performed with SPSS 24 (SPSS Inc. USA). The differences in age, HDRS, years of education, and Intracranial volume between the normal controls and MDD patients were tested using the independent t-test. Group difference in sex was tested using chi-square. All tests were two-tailed, and the level of significance was p<0.05.

Analysis of covariance (ANCOVA), controlled for age, sex, and ICV, was applied to analyze regional volume changes of GM, and ANCOVA with the confounders of age and sex was applied to analyze regional concentration changes of GM. The correlation between the GM volume or concentration and disease duration, onset, education level, or HRSD scores was tested by partial correlation analysis with the confounders of age, sex, and ICV (for modulated VBM). The statistical threshold level was set to FDR-corrected p<0.05, and the extent threshold was set to kE >200 voxels (676 mm³; 1 voxel=3.375 mm³).

The group difference in sulcal depth was tested using ANCOVA, using age and sex as confounders with 5,000 permutations. The correlation between sulcal depth and disease duration, onset, education level, and HRSD scores was tested by partial correlation analysis with the confounders of age and sex. The significance level was set to a Holm-Bonferroni corrected p<0.05.

Characteristics of the participants

The Demographic characteristics of each participant, including age, sex, years of education, and Intracranial volume are summarized in Table 1. There were no significant differences in age or sex among the two groups. In the present dataset, for the MDD group, the mean age was 36.07, the mean percentage of females was 71.73%. For the NML group, the mean age was 36.78, and the mean percentage of females was 78.26%.

Voxel based morphometry and sulcal depth

There were no statistically significant differences in the ICV between the patients with MDD and the NML (NML vs. MDD; 1,467.9±143.68 cm³ vs. 1,407.0±120.14 cm³, p=0.067). GM (NML vs. MDD; 662.0±57.23 cm³ vs. 641.3±64.22 cm³, p=0.708). and WM (NML vs. MDD; 526.6±57.06 cm³ vs. 504.7±52.33 cm³, p=0.865) also not different between the two groups In sulcal depth analysis, depressed patients showed reduced sulcal depth in the areas of left posterior ramus of the lateral sulcus, superior frontal sulcus, supramarginal gyrus, central sulcus (Rolando’s fissure), and Heschl’s gyrus. The right posterior ramus of the lateral sulcus, temporal plane of the superior temporal gyrus, anterior transverse collateral sulcus, and central sulcus (Rolando fissure) were also reduced (Fig. 1, 2 and Table 2).

Sulcal depth was negatively correlated with HRSD scores by partial correlation analysis with age and sex. The results showed a negative correlation between sulcal depth and HRSD score in the right parahippocampal gyrus, parahippocampal part of the medial occipitotemporal gyrus, medial orbital sulcus (olfactory sulcus), medial occipitotemporal sulcus (collateral sulcus), and lingual sulcus (Fig. 3). However, there was no relationship between sulcal depth and disease duration, onset, or education level.

In VBM analyses, there were no significant differences between the two groups in the modulated (volume change) or unmodulated (concentration change) gray matter analyses and showed no relationship with disease duration, onset, or HRSD scores.

In this study, not only the central sulci, but also Heschl’s gyrus, posterior ramus of the lateral sulcus, superior frontal sulcus, and supramarginal gyrus of patients with MDD showed reduced sulcal depth compared to NML. As far as our knowledge, these findings are new in MDD. Recent, ENIGMA (Enhancing Neuro Imaging Genetics through Meta-Analysis) with MDD’s big MRI cohort (2148 MDD patients and 7957 healthy controls MRIs) reported reduced regional surface in the pericentral areas, and the study supports our findings [19]. During the developmental process gyrencephalic brain, the cerebral cortices show a predictable folding pattern. With primary sensory and primary motor cortices on the caudal and rostral banks, respectively, the central sulcus and surrounding region is perhaps the most thoroughly characterized portion of the cortex [20].

The pathological relation between depression and reduced sulcal depth in central sulci is not unclear. But, depressed patients usually suffer anhedonia which is the core feature of depression as well as other mental health disorders. Anhedonia is a severe condition that describes the absence of interest, enjoyment, and motivation. We recently reported higher cortical volume and fractal dimension in both the pre- and post-central gyri of professional basketball players on a college basketball team [11]. The higher volume or fractal dimension of elite basket players could suggest maturation of the primary motor and somatosensory cortices due to physical exercise. However, depressed individuals with anhedonia experience a loss of social or physical activities. Patients with MDD in this study showed moderate to severe HRSD scores and a mean disease duration of almost nine years. Therefore, the authors cautiously assume that the depressed patients in this study may have had less sensory or physical activity for a long time. So, the authors very carefully interpret the reduced sulcal depth in MDD could be the result of social and physical anhedonia related cerebral degenerations. To confirm this hypothesis, further controlled investigation would be needed.

In the correlation analysis between the sulcal depth and the HRSD score of MDD group, the negative correlation between HRSD score and the sulcal depth in areas including right parahippocampal gyrus, parahippocampal part of the medial occipito-temporal gyrus, medial orbital sulcus (olfactory sulcus), and medial occipito-temporal sulcus (collateral sulcus) and lingual sulcus (Fig. 2) was identified. Zamoscik et al. [21] reported that remitted depressed patients showed stronger functional connectivity than healthy controls between the parahippocampal gyri and the posterior cingulate cortex. Similarly, Couvy-Duchesne et al. [22] observed reduced surface area in the right lingual, fusiform and parahippocampal gyri in patients with anxiety-depression. Significantly shallower olfactory sulci in current and remitted MDD patients was also reported by Takahashi et al. These previous studies support of findings and the results indicate that the negatively correlated areas with HRSD in this study could be an important brain loci related to the clinical symptoms of MDD.

The limitation of this study is the difference of education levels between NML and MDD group. Patients with depression have been considered to have the tendency of lower education level [21, 22], and the MDD patients of this study also showed lower education level compared to NML. To figure it out the effect of education level on sulcal depth, we correlated these two variables by simple or partial correlation with confounder of age, sex, but we could not find any relations between the two variables. This result supports our previous study about hippocampal subfield volume of MDD witch did not correlated with education level [23], and other studies reported, the education years and the gray hippocampal volume of healthy elders was not related [24, 25], and the educational levels and gray matter volume of subcortical structures also did not demonstrate significant correlations [26, 27]. But, one study reported when only including healthy subjects older than 35 years old, the higher education group showed greater hippocampal volume compared to the lower education group. But, However, they could not find any significant difference when young subjects (younger than 35 years) did not show any significant difference [28, 29]. Further studies are necessary to make a consensus on this issue in future.

This work was supported by the National Research Foundation of Korea (NRF) funded by the Korean government (MSIP) (No. 2020R1I1A1A01067132), the National Research Foundation of Korea (NRF) funded by the Korean government (MSIP) (No. 2017M3C7A1079696).

Demographic and clinical data for patients with major depressive disorder and healthy controls

Reduced sulcal depth in patients with depression