Insights into structural deviations in attention deficit hyperactivity disorder (ADHD) and comorbidities using big data-derived brain charts: a cross-sectional study

Introduction

Attention-deficit hyperactivity disorder (ADHD) is a widespread neurodevelopmental condition typically manifesting during early childhood. Clinically, it is characterized by a triad of core symptoms: inattention, hyperactivity, and impulsivity. These symptoms typically manifest before the age of 12 and can persist into adulthood. ADHD, depression, Tourette syndrome (TS), and autism spectrum disorder (ASD) comprise a major cause of health-related disabilities in children and adolescents (1). The presence of ADHD in conjunction with any of these conditions can significantly worsen the overall situation. Research findings indicate that the prevalence of comorbid depression among ADHD patients is notably high (2), which is considered a consequence of ADHD-related impairments and adverse environmental factors (3). Children with comorbid ADHD and depression necessitate more intensive interventions compared to those with ADHD alone, as they endure higher levels of stress and encounter more psychosocial and familial challenges. TS is a frequent comorbidity in children with ADHD, irrespective of medication status (4). The co-occurrence of ADHD and TS is frequently associated with greater levels of social and psychopathological impairment (5). It is well-acknowledged that the detrimental impact of comorbid ADHD and ASD on daily functioning is profound, particularly in the realms of social interaction and communication, as well as on the broader spectrum of psychopathology (6). Overall, the intricate comorbidity of ADHD with depression, TS, and ASD presents substantial challenges in the screening, diagnosis, and management of these conditions.

Despite the ongoing challenge of pinpointing the exact sources of comorbidity and its neural substrates, it is clear that brain differences play a pivotal role in etiological models. In the case of ADHD and depression, there are distinctive neural markers within the frontoparietal network (7), with the dysregulation of emotion acting as a bridge between these two disorders (8). Research also indicates that TS and ADHD share comparable alterations in local neuronal activity, which result in varied neuronal and behavioral outcomes (9). The development of both ADHD and TS involves a complex interplay of networks, spanning cortical and striatal regions as well as the basal ganglia (5). The co-occurrence of TS with ADHD is further associated with an exacerbated form of sensory processing dysfunction (10). Neurocognitive perspectives suggest that the commonalities in comorbidity with ADHD and ASD are deeply ingrained in the functional connectivity networks that govern executive control (11). Moreover, abnormalities in white matter microstructure, specifically within the splenium of the corpus callosum, may be a shared feature among individuals with ADHD and ASD (12). Grasping the nature of these brain-related differences is an essential step toward enhancing our understanding of the underlying etiology.

However, most data regarding brain differences are obtained through case-control designs. Although these designs provide information on the degree of inter-group differences at the population level, they offer less insight into individual variability. The heterogeneity and overlap among neurodevelopmental disorders, along with differences in research methods and sample sizes, are significant contributors to the lack of consistent findings in studies of ADHD’s neurodevelopmental trajectory. Normative modeling offers a framework with a common baseline for studying different neurodevelopmental conditions, allowing for a better quantification of individual differences and addressing challenges such as heterogeneity, gender differences, and the integration of multi-site datasets. Recently, advancements in neuroimaging and statistical techniques, as well as the availability of large datasets, have led to the creation of brain charts that define reference standards for brain structural measurements (13). Similar to the use of pediatric growth charts, brain charts can identify individual-specific variations or differences associated with neurodevelopmental conditions, even before related traits are clinically expressed (14). Data obtained for the development and validation of these brain charts support their utility in dissecting biological heterogeneity in clinical diseases (e.g., ADHD, depression, ASD, and neurodegenerative diseases) and summarizing the metrics of brain development (representing deviation patterns in individual brain regions) (13-15). Big data-derived brain charts provide an objective reference standard for brain development and can be used to quantify deviations in individual brain development trajectories, offering a new perspective for studying ADHD and its comorbidities.

As a next step to showcase the clinical potential of brain charts, the current study aimed to leverage normative brain chart modeling to delve into regional brain development deviations in ADHD and comorbid groups, while also assessing the correlations between structural deviations and clinical symptoms. We hypothesized that children with ADHD and its comorbidities—depression, TS, and ASD—would display unique regional brain development deviations compared to those with ADHD alone, as evidenced by normative brain chart analysis. These specific brain structural deviations were presumed to correlate with the severity and expression of ADHD symptoms. Big data-generated brain charts were expected to uncover measurable disparities in brain development paths, offering a groundbreaking method for comprehending the origins and diversity of ADHD and associated conditions. We present this article in accordance with the STROBE reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2024-2707/rc).

Methods

Participants

This is a large, population-based cross-sectional study with an observational design. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Board of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology (TJ-IRB20221150). Informed consent was documented in writing from both the guardians and the minors involved.

The participants were children who were diagnosed with ADHD and sought treatment at the Pediatric Health Care outpatient service or the Neurology Department of Tongji Hospital between January 2022 and August 2023. Their diagnosis and subsequent inclusion in the study were confirmed by two senior physicians, adhering to the ADHD criteria outlined in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-V) (16). The criteria for participation included an age range of 6 to 16 years, Han Chinese descent, right-handedness, and no record of past head trauma, somatic illness, or epilepsy. A cohort of age-appropriate, healthy control children was assembled from those undergoing routine health check-ups at the hospital. Neuroimaging data were excluded if any of the following conditions were present: braces, incidental findings that significantly altered brain morphology, or poor image quality due to poor patient cooperation.

Behavioral assessment

The Swanson, Nolan, and Pelham scale version IV (SNAP-IV) (17) was widely used to assess symptoms of ADHD. The severity of ADHD features was assessed using the Chinese version of a three-factor structure of the SNAP-IV scale. Intelligence quotient (IQ) was measured with an age-appropriate Chinese-Wechsler Intelligence Scale for Children-Fourth Edition (WISC-IV) (18). For participants with comorbid TS, the severity of TS was assessed using the Yale Global Tic Severity Scale (YGTSS) (19). In cases where ASD was also present, the Childhood Autism Rating Scale (CARS) and the Autism Behavior Checklist (ABC) (20) were employed to gauge the core symptoms of ASD in these children. If children exhibited comorbid depressive symptoms, they were administered the Children’s Depression Inventory (CDI) (21) as part of their involvement in the study.

Magnetic resonance imaging (MRI) acquisition and processing

Neuroimaging data were acquired on a 3.0 Tesla GE Discovery MR750 MRI system (GE Healthcare, Milwaukee, WI, USA). Sagittal three-dimensional (3D) T1-weighted images were acquired by a brain volume sequence [sequence parameters: repetition time/echo time (TR/TE) =8.16/3.18 ms; inversion time =450 ms; flip angle =12°; field of view (FOV) =256×256 mm; matrix =256×256; slice thickness =1 mm; no gap; 165 sagittal slices]. FreeSurfer image analysis suite (version 6.0, RRID:SCR_001847) was used for cortical reconstruction and volumetric segmentation for this study, and we exclusively employed FreeSurfer’s default registration procedures. FreeSurfer output directories included summary files that represent left and right hemisphere cortical thickness values of the Destrieux parcellation and subcortical volumetric values (aseg.stats, lh.aparc.a2009s.stats, and rh.aparc.a2009s.stats).

All pre-trained models and associated code for the normative modeling of brain charts were made available online through GitHub (https://github.com/predictive-clinical-neuroscience/braincharts). The normative modeling was conducted using Python (RRID:SCR_013141) and the PCNtoolkit package (version 0.20). Bayesian linear regression (BLR), enhanced with likelihood warping, was utilized to predict cortical thickness and subcortical volume, taking into account covariates such as age, sex, and site. The model encompasses both healthy controls and individuals with ADHD, aiming to forecast normative brain development across clinical and non-clinical groups. Deviation scores (Z-scores) were calculated for each participant and brain area based on the formula:

Znd=ynd−y^ndσd2+(σ*2)d[1]

Here, ynd represents the actual response (the observed measurement for a specific brain area), y^nd is the predicted mean, whereas σd2 and (σ*2)d denote the variance components of the model, which account for measurement noise and model uncertainty, respectively. These Z-scores indicate how much an individual’s brain measurement deviates from the predicted value, standardized by the combined effects of noise and uncertainty. This enables us to pinpoint atypical brain measurements in populations. We evaluated the model’s fit for each brain region by assessing the explained variance (a measure of central tendency), the mean squared log-loss (evaluating both central tendency and variance), and the skewness and kurtosis of the deviation scores. These additional metrics help us to understand not only the model’s fit to the data but also whether the shape of the regression function corresponds well with the true underlying patterns of brain development. For a comprehensive mathematical treatment and discussion of this methodology, refer to the relevant literature (14).

To ensure accurate estimation, we used an in-house MRI dataset consisting of T1-weighted images from healthy controls, which we adapted to meet our site’s standards. This adaptation set serves as the local calibration set for the model (15), accounting for our specific imaging protocol and scanner characteristics. Employing this dataset mitigates the potential impact of scanner or acquisition protocol variations on model predictions when applied to external data (ADHD dataset).

The brain charts used in our approach automatically adjust for age and sex effects by incorporating them as covariates in the model. These charts utilize a B-spline basis expansion to capture non-linear age effects and apply a warping procedure to account for sex-specific differences in brain structure. By directly modeling these factors, the brain charts can predict brain development tailored to each individual’s sex and age. The likelihood warping technique further corrects for non-Gaussian effects in the data, enabling the model to account for complex developmental patterns that may not conform to simple Gaussian distributions. This ensures that the model provides unbiased normative predictions, free from age or sex biases, leading to more accurate individual deviation assessments regardless of demographic background.

Statistical analyses

Principal component analysis (PCA) serves to categorize interrelated variables, thereby reducing data dimensionality. This approach involves projecting each data point onto a subset of the initial principal components, thereby achieving a reduced-dimensional representation while retaining a significant portion of the dataset’s variability (22). By diminishing the number of variables within the dataset, PCA addresses subsequent statistical challenges and enhances the comprehension of the data. We assessed the deviations from typical development in cortical and subcortical regions using the normative modeling of brain charts. PCA was performed on the deviation scores of all cortical and subcortical brain regions. The principal component 1 indicates the direction that maximizes the variance of the projected data. Further analysis of principal component 1 was conducted to examine whether there were any intergroup differences between ADHD and its comorbid groups, as well as to assess its correlation with behavioral assessments.

Furthermore, we delved into whether there were intergroup differences in the deviation scores for each brain region. Prior to conducting analysis of variance (ANOVA), we verified the assumption of homogeneity of variances using Levene’s test for each brain region across the four clinical groups. The results indicated that the variance was homogeneous across groups, as evidenced by P values greater than 0.05 for all brain regions. Subsequently, we utilized ANOVA to compare the differences in brain development trajectories between the pure ADHD group and other comorbid groups (ADHD with TS, ADHD with depression, ADHD with ASD). The false discovery rate (FDR) was managed using the Benjamini–Hochberg procedure. Utilizing Pearson correlation, we further investigated whether deviations from typical development in cortical and subcortical regions, as evaluated by brain charts, were associated with the clinical severity of the condition.

To encapsulate the individual variations within each clinical group, deviation scores were summarized by first categorizing them into positive and negative deviations. We then counted the number of cases with extreme deviations in a given region of interest (ROI). Extreme deviations were defined as Z>2 for positive deviations and Z<−2 for negative deviations. This count was divided by the group size to calculate the percentage of individuals with extreme deviations in that brain area. The Z threshold of 2 was chosen to be consistent with established literature on brain charts (14), ensuring comparability with existing research. Based on the observed data distribution, we noted that no brain region exhibited a percentage of individuals with extreme deviations exceeding 50%. Therefore, to maintain objectivity and simplicity, we adopted 10% decrements and selected brain regions where the proportion of individuals with extreme deviations exceeded 40% as core areas with significantly deviated development trajectories. To determine if there were significant differences in the distribution of extreme deviation values among brain regions between the ADHD group and the other comorbid groups, we also conducted classical Chi-squared tests on the percentages of individuals with extreme deviations and applied a threshold to the results at a Benjamini-Hochberg adjusted P value (P_FDR) <0.05.

Results

Demographic and behavioral characteristics

A total of 581 children with ADHD who had completed evaluations and had acceptable MRI images were recruited into the study. Patients with ADHD combined with additional neurodevelopmental disorders, including anxiety, sensory sensitivities, and developmental coordination disorder, among others, as well as those with three or more comorbid conditions, were excluded from the study (n=122). Based on behavioral assessments, 459 children were further categorized into a pure ADHD group and other comorbid groups (ADHD + TS, ADHD + Depression, ADHD + ASD). There were no overlapping cases among the ADHD groups with comorbid conditions. Our in-house dataset comprised N=251 healthy controls, with an age range spanning from 6 to 14 years, having a mean age and standard deviation of 9.53±1.76 years, and a male proportion of approximately 62%. Demographic and behavioral characteristics of the cases in the pure ADHD group and comorbid groups are detailed in Table 1. Brain charts, which span the entire human lifespan, offer the capability to quantify individual variations against centiles of variation in a reference population. Each participant’s brain development deviation is predicted against norms that match their own sex and age, thus eliminating age and sex as confounding factors in statistical analyses. Comparative analysis between groups revealed significant differences in full scale-IQ scores (F=15.03, P<0.01), whereas no significant intergroup differences were observed in scores of inattentiveness (F=0.52, P=0.67), hyperactivity/impulsivity (F=1.47, P=0.22), or oppositional defiance (F=2.45, P=0.06).

Deviations from typical development

In the PCA, we extracted principal component 1, which accounted for 19.07% of the explained variance. The loadings for individual brain regions corresponding to principal component 1 are detailed in Figures S1,S2. The amplitude of these loadings indicates the intensity of the correlation between each brain region and principal component 1. A higher absolute value of the loading indicates a more significant contribution of the brain region to this principal component. Although there were no significant differences in principal component 1 across ADHD and its comorbid groups, the principal component 1 demonstrated significant correlations with the SNAP-IV subscales for inattention (r=−0.17, P<0.001) and oppositional defiant disorder (r=−0.10, P=0.04).

In the analysis of each brain region, significant intergroup differences in the deviations of cortical thickness were observed in the left cuneus gyrus (F=6.50, P_FDR =0.03) and the lingual part of the medial occipito-temporal gyrus (lh_G_oc-temp_med-Lingual) (F=5.75, P_FDR =0.04). Further multiple comparisons of the cortical thickness in the left cuneus gyrus revealed no difference in deviation scores between the ADHD + Depression group and the pure ADHD group, a trend towards lower deviation scores in the ADHD + TS group compared to the pure ADHD group, and significantly higher deviation scores in the ADHD + ASD group compared to the other three groups (least significant difference-corrected P<0.01) (Figure 1A). Additional multiple comparisons of the cortical thickness in the left oc-temp-med-Lingual area showed no difference in deviation scores between the ADHD + Depression group and the pure ADHD group, lower deviation scores in the ADHD + TS group compared to the pure ADHD group (least significant difference corrected P<0.05), and significantly higher deviation scores in the ADHD + ASD group compared to the other three groups (least significant difference corrected P<0.01) (Figure 1B). There were no significant intergroup differences in the deviations of subcortical volume. Additionally, we found that the deviation scores in multiple cortical brain regions were significantly correlated with the SNAP-IV subscales for inattention (specific details provided in Table 2). There were no significant correlations between the typical developmental deviations in brain regions and other clinical scores.

Figure 1 Cortical regions with significant intergroup differences in the score of deviation from typical development. Significant intergroup differences in the deviations of cortical thickness were observed in the left cuneus gyrus (A) and the lingual part of the medial occipito-temporal gyrus (lh_G_oc-temp_med-Lingual) (B). *, P<0.05 values corrected for multiple comparisons using the least significant difference method; **, P<0.01 values corrected for multiple comparisons using the least significant difference method. ADHD, attention deficit hyperactivity disorder; ASD, autism spectrum disorder; G, gyrus; lh, left hemisphere; med, medial; oc-temp, occipito-temporal; TS, Tourette syndrome.

Core areas with significant structural deviations

In this study, deviation scores were summarized for each clinical group by separating them into positive and negative deviations. Extreme core brain regions with a proportion of extreme negative deviations exceeding 40% were found only in the analysis of extreme negative deviations. Figure 2 illustrates the core brain regions with significant negative deviations in the development trajectory of cortical thickness. Compared to the other groups, the ADHD + TS group had the largest number of extreme core brain regions, including the left middle frontal sulcus (lh_S_front_mid), left inferior temporal gyrus (lh_G_temporal_inf), bilateral superior frontal gyrus (lh_G_front_sup and rh_G_front_sup), left superior frontal sulcus (lh_S_front_sup), and left middle temporal gyrus (lh_G_temporal_mid). The left middle temporal gyrus was an extreme core brain region shared by all four groups. Chi-squared tests were further conducted to examine the distribution differences of extreme negative deviations among groups. Significant differences were only found in the left middle frontal sulcus (P_FDR =0.04), with the ADHD + TS group showing a significantly higher proportion than the ADHD + Depression group (P_FDR <0.01).

Figure 2 The core brain regions with significant negative deviations in the development trajectory of cortical thickness. The x-axis represents the proportion of extreme negative deviations. * indicates a significant difference in the distribution of extreme negative deviations. ADHD, attention deficit hyperactivity disorder; ASD, autism spectrum disorder; front, frontal; G, gyrus; inf, inferior; lh, left hemisphere; mid, middle; rh, right hemisphere; S, sulcus; sup, superior; TS, Tourette syndrome.

Discussion

The current study employed brain chart normative modeling to deeply analyze the developmental differences in specific brain regions between ADHD and its comorbid groups, and explored the relationship between structural deviations and clinical symptoms. We identified pronounced intergroup disparities in cortical thickness within the left cuneus gyrus and the medial occipito-temporal gyrus. The left middle temporal gyrus emerged as a brain region with extreme deviations shared across all groups. The ADHD + TS group exhibited a greater number of brain regions exhibiting extreme deviations when contrasted with the other groups. Furthermore, the study revealed that the ADHD + TS group had a substantially higher proportion of negative deviations in the left middle frontal sulcus than the ADHD + Depression group. Both the principal component 1 of structural deviations and the deviation scores across multiple cortical brain regions exhibited significant correlations with the SNAP-IV subscales for inattention. These findings contribute novel insights into the neurobiological underpinnings of ADHD and its comorbidities, and they provide a foundation for the advancement of personalized diagnostic and therapeutic strategies.

Specifically, our observations revealed substantial intergroup disparities in cortical thickness within the left cuneus gyrus and the lingual portion of the medial occipito-temporal gyrus. This suggests that there may be distinct neuroanatomical substrates underlying the comorbidity of ADHD with TS, depression, and ASD. In comparison to the ADHD + ASD group, the cortical thickness in the left cuneus gyrus and the lingual part of the medial occipito-temporal gyrus was significantly reduced in the groups with only ADHD, ADHD combined with depression, and ADHD comorbid with TS. The cuneus gyrus and the lingual portion of the medial occipito-temporal gyrus play pivotal roles in visual processing and attentional control. Research has confirmed correlations between attentional deficits and the visual processing areas (cuneus and temporo-occipital gyrus) within the cerebral cortex of individuals with ADHD (23,24). Consistent with expectations, the modifications in cortical thickness observed in these brain regions could potentially underlie the central symptoms of ADHD, such as inattention and impulsivity, as well as the visual processing impairments frequently noted in children affected by ADHD. Furthermore, the cuneus, temporal, and lingual gyrus are key areas for sensory perceptions in mental and psychological disorders (25-27). The reduced development in the left cuneus gyrus and the lingual part of the medial occipito-temporal gyrus in patients with ADHD comorbid with depression may be related to deficits in emotional regulation. Previous reports have highlighted the cuneus as a significant functional hub in the neural model of depression (28). Volume of the cuneus has been found to be associated with improved inhibitory control in patients with depression (29). The cuneus is also part of the salience network, which has been reported to be altered in depression (30). Thinning of the occipito-temporal cortex has also been reported to be associated with depressive behaviors (27). In comparison, patients with ADHD + TS exhibit the most pronounced reduction in developmental deviation in the left cuneus gyrus and the lingual part of the medial occipito-temporal gyrus. This finding is consistent with the literature reporting that TS patients not only have brain differences in cognitive and motor domains, but these differences also extend into the perceptual domain (31). The reduced cortical thickness in these two brain regions may lead to functional impairments in visual processing and spatial cognition, which may be related to deficiencies in perception and motor control as well as impulsivity control in patients with ADHD comorbid with TS.

Further analysis revealed a high proportion of individuals across all groups exhibiting extreme negative deviations in the left middle temporal gyrus. The middle temporal gyrus plays a role in the perception of moving objects in the visual field and is also involved in tasks such as semantic and language comprehension (32,33). The structure and function of the middle temporal gyrus undergo changes throughout the lifespan, which have been identified as biomarkers of brain development (34). ADHD patients, who exhibit cognitive impairments and impulsive behaviors, have been reported to have an altered topological organization of brain structural covariance networks, including the middle temporal gyrus (35). As a key region in the “social brain” network, the middle temporal gyrus has been extensively associated with ASD (36,37). TS is characterized by tics that are typically preceded by uncomfortable urges that build until the tic is performed, and the severity of premonitory urges has been linked to the middle temporal gyrus region (38). As a core brain region for cognitive processing of emotions and sensations, the middle temporal gyrus has been reported to have gray matter morphology that is a significant neurostructural correlate of depression (39,40). Our results may reflect an overlapping neurobiological feature of these disorders, namely the developmental abnormality of the middle temporal gyrus, providing important insights into the mechanisms of comorbidity.

Our findings revealed that the ADHD + TS group exhibited a greater number of brain regions with extreme deviations when compared to the other groups, particularly a significantly higher proportion of negative deviations in the left middle frontal sulcus than the ADHD + Depression group. The left middle frontal sulcus is associated with executive functions, including working memory and cognitive flexibility. Abnormalities in the frontal lobes of ADHD and TS patients have been extensively investigated. These studies have confirmed that ADHD patients exhibit structural differences in the prefrontal cortex and other regions associated with attention, impulse control, and executive functions (41,42). TS patients, on the other hand, may demonstrate more widespread neural network abnormalities, including areas related to motor control and emotional regulation (43-45). Consequently, the greater number of extreme deviations observed in the ADHD + TS group may be related to the shared and unique neurobiological features of these disorders. The specific attention paid to the proportion of negative deviations in the brain regions of comorbid patients in this study represents a different focus to that commonly found in broader research contexts. Overall, our results suggest that the dispersed involvement of multiple cortical regions involved in different functions can explain the heterogeneity of ADHD and its comorbid symptoms. This aligns with existing literature, which posits that ADHD is a complex neurodevelopmental disorder characterized by the heterogeneity of symptoms and neurobiological features (46). This heterogeneity is manifested not only in the diversity of clinical symptoms but also in the widespread nature of brain structural abnormalities (47).

The principal component 1 of structure deviations, which captures the overall pattern of structural differences across the brain, was found to be significantly correlated with inattention symptoms. This indicates that a global neural pattern may underlie the inattention phenotype in ADHD. Moreover, the deviation scores across multiple cortical brain regions, including the frontal cortex, occipital lobe, parietal lobe, and cingulate gyrus, were also significantly correlated with inattention symptoms. These findings suggest that specific brain regions are integral to the manifestation of inattention symptoms in ADHD. The identification of these brain regions associated with inattention symptoms could provide a roadmap for the development of targeted interventions. For instance, cognitive training exercises designed to enhance frontoparietal function might help to improve executive control and attentional focus (48). Similarly, neurofeedback techniques could be employed to modulate activity in the cingulate gyrus, which is implicated in maintaining task-related attention (49). Interventions that target the occipital lobe might address the visual processing deficits that can contribute to inattention (50). By tailoring treatments to the specific neural deficits observed in each individual with ADHD, we can aim for more personalized and effective therapeutic outcomes. It is anticipated that such interventions could not only alleviate the immediate symptoms of inattention but also promote long-term cognitive development and functional improvement in individuals with ADHD.

This study is subject to several limitations. Firstly, the sample size employed was relatively modest, which may constrain the broader applicability of our results. It is imperative to replicate the study with larger and more diverse cohorts to ascertain the validity and robustness of our findings. Secondly, in the present study, PCA was applied to the deviation scores of cortical and subcortical brain regions as a dimensionality reduction technique to identify the principal modes of variation. However, considering the complex and potentially nonlinear nature of brain structure deviations, coupled with the inherent linearity assumption of PCA, future research would benefit from exploring nonlinear dimensionality reduction techniques. This approach may reveal additional meaningful patterns, offering a more nuanced understanding of structural variations. Finally, our investigation was confined to assessing brain structure, omitting an examination of functional connectivity and other neuroimaging metrics. Future research should explore functional connectivity and additional neuroimaging parameters in conjunction with brain structural assessments to achieve a more nuanced understanding of the neurobiological substrates of ADHD and associated comorbid conditions.

Conclusions

Our findings provide valuable insights into the neurodevelopmental trajectories of ADHD and its comorbid disorders. Brain charts offer a powerful tool for characterizing individual differences in brain development and identifying potential neurobiological substrates of neurodevelopmental disorders. Further research utilizing brain charts and other neuroimaging techniques is needed to improve our understanding of the complex etiology of ADHD and its comorbidities. Further exploration of the potential of brain charts for predicting the course and prognosis of ADHD and its comorbidities may ultimately lead to more effective diagnostic and treatment strategies.

Acknowledgments

The authors would like to express sincere gratitude to Dr. Ning Zheng at Clinical and Technical Support, Philips Healthcare, Wuhan, China, for engaging in useful discussions.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2024-2707/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2024-2707/dss

Funding: This study was supported by the National Natural Science Foundation of China (No. 82471965 to T.T.) and the Hubei Provincial Natural Science Foundation of China (No. 2023AFB862 to D.L.). The funders of the study had no role in study design, planning, data analysis, data interpretation, or writing of the report.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2024-2707/coif). T.T. received support from the National Natural Science Foundation of China (No. 82471965). D.L. received supports from the Hubei Provincial Natural Science Foundation of China (No. 2023AFB862). The funders of the study had no role in study design, planning, data analysis, data interpretation, or writing of the report. The research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Board of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology (TJ-IRB20221150). Informed consent was documented in writing from both the guardians and the minors involved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Yang J, He X, Qian L, Zhao B, Fan Y, Gao F, Yan B, Zhu F, Ma X. Association between plasma proteome and childhood neurodevelopmental disorders: A two-sample Mendelian randomization analysis. EBioMedicine 2022;78:103948. [Crossref] [PubMed]
2. Thapar A, Livingston LA, Eyre O, Riglin L. Practitioner Review: Attention-deficit hyperactivity disorder and autism spectrum disorder - the importance of depression. J Child Psychol Psychiatry 2023;64:4-15. [Crossref] [PubMed]
3. Blackman GL, Ostrander R, Herman KC. Children with ADHD and depression: a multisource, multimethod assessment of clinical, social, and academic functioning. J Atten Disord 2005;8:195-207. [Crossref] [PubMed]
4. Ogundele MO, Ayyash HF. Review of the evidence for the management of co-morbid Tic disorders in children and adolescents with attention deficit hyperactivity disorder. World J Clin Pediatr 2018;7:36-42. [Crossref] [PubMed]
5. El Malhany N, Gulisano M, Rizzo R, Curatolo P. Tourette syndrome and comorbid ADHD: causes and consequences. Eur J Pediatr 2015;174:279-88. [Crossref] [PubMed]
6. Lebeña A, Faresjö Å, Faresjö T, Ludvigsson J. Clinical implications of ADHD, ASD, and their co-occurrence in early adulthood-the prospective ABIS-study. BMC Psychiatry 2023;23:851. [Crossref] [PubMed]
7. Vilgis V, Yee D, Silk TJ, Vance A. Distinct Neural Profiles of Frontoparietal Networks in Boys with ADHD and Boys with Persistent Depressive Disorder. Cogn Affect Behav Neurosci 2022;22:1183-98. [Crossref] [PubMed]
8. Seymour KE, Miller L. ADHD and Depression: the Role of Poor Frustration Tolerance. Curr Dev Disord Rep 2017;4:14-8. [Crossref] [PubMed]
9. Israelashvili M, Yael D, Vinner E, Belelovsky K, Bar-Gad I. Common neuronal mechanisms underlying tics and hyperactivity. Cortex 2020;127:231-47. [Crossref] [PubMed]
10. Isaacs D, Riordan H. Sensory hypersensitivity in Tourette syndrome: A review. Brain Dev 2020;42:627-38. [Crossref] [PubMed]
11. Vaidya CJ, Klein C. Comorbidity of Attention-Deficit Hyperactivity Disorder and Autism Spectrum Disorders: Current Status and Promising Directions. Curr Top Behav Neurosci 2022;57:159-77. [Crossref] [PubMed]
12. Zhao Y, Yang L, Gong G, Cao Q, Liu J. Identify aberrant white matter microstructure in ASD, ADHD and other neurodevelopmental disorders: A meta-analysis of diffusion tensor imaging studies. Prog Neuropsychopharmacol Biol Psychiatry 2022;113:110477. [Crossref] [PubMed]
13. Bethlehem RAI, Seidlitz J, White SR, Vogel JW, Anderson KM, Adamson C, et al. Brain charts for the human lifespan. Nature 2022;604:525-33. Erratum in: Nature 2022;610:E6. [Crossref] [PubMed]
14. Rutherford S, Fraza C, Dinga R, Kia SM, Wolfers T, Zabihi M, et al. Charting brain growth and aging at high spatial precision. Elife 2022;11:e72904. [Crossref] [PubMed]
15. Nárai Á, Hermann P, Rádosi A, Vakli P, Weiss B, Réthelyi JM, Bunford N, Vidnyánszky Z. Amygdala Volume is Associated with ADHD Risk and Severity Beyond Comorbidities in Adolescents: Clinical Testing of Brain Chart Reference Standards. Res Child Adolesc Psychopathol 2024;52:1063-74. [Crossref] [PubMed]
16. Regier DA, Kuhl EA, Kupfer DJ. The DSM-5: Classification and criteria changes. World Psychiatry 2013;12:92-8. [Crossref] [PubMed]
17. Hall CL, Guo B, Valentine AZ, Groom MJ, Daley D, Sayal K, Hollis C. The Validity of the SNAP-IV in Children Displaying ADHD Symptoms. Assessment 2020;27:1258-71. [Crossref] [PubMed]
18. Zhang Z, Wang J, He L, Huang Z, Sun L, Zhang Y, Zhang X. Assessing regional homogeneity and cognitive function alterations in pediatric brain tumor patients: a resting-state functional magnetic resonance imaging study. Quant Imaging Med Surg 2024;14:8686-95. [Crossref] [PubMed]
19. Zheng Y, Zhang ZJ, Han XM, Ding Y, Chen YY, Wang XF, Wei XW, Wang MJ, Cheng Y, Nie ZH, Zhao M, Zheng XX. A proprietary herbal medicine (5-Ling Granule) for Tourette syndrome: a randomized controlled trial. J Child Psychol Psychiatry 2016;57:74-83. [Crossref] [PubMed]
20. Chu L, Shen L, Ma C, Chen J, Tian Y, Zhang C, Gong Z, Li M, Wang C, Pan L, Zhu P, Wu D, Wang Y, Yu G. Effects of a Nonwearable Digital Therapeutic Intervention on Preschoolers With Autism Spectrum Disorder in China: Open-Label Randomized Controlled Trial. J Med Internet Res 2023;25:e45836. [Crossref] [PubMed]
21. Ji L, Lili S, Jing W, Yanyan H, Min W, Juan X, Hongmei S. Appearance concern and depression in adolescent girls with systemic lupus erythematous. Clin Rheumatol 2012;31:1671-5. [Crossref] [PubMed]
22. Higgins V, Hooshmand S, Adeli K. Principal component and correlation analysis of biochemical and endocrine markers in a healthy pediatric population (CALIPER). Clin Biochem 2019;66:29-36. [Crossref] [PubMed]
23. Lei D, Ma J, Du X, Shen G, Jin X, Gong Q. Microstructural abnormalities in the combined and inattentive subtypes of attention deficit hyperactivity disorder: a diffusion tensor imaging study. Sci Rep 2014;4:6875. [Crossref] [PubMed]
24. Carmona S, Hoekzema E, Castellanos FX, García-García D, Lage-Castellanos A, Van Dijk KR, Navas-Sánchez FJ, Martínez K, Desco M, Sepulcre J. Sensation-to-cognition cortical streams in attention-deficit/hyperactivity disorder. Hum Brain Mapp 2015;36:2544-57. [Crossref] [PubMed]
25. Heitzeg MM, Cope LM, Martz ME, Hardee JE, Zucker RA. Brain activation to negative stimuli mediates a relationship between adolescent marijuana use and later emotional functioning. Dev Cogn Neurosci 2015;16:71-83.
26. Nicol K, Pope M, Romaniuk L, Hall J. Childhood trauma, midbrain activation and psychotic symptoms in borderline personality disorder. Transl Psychiatry 2015;5:e559. [Crossref] [PubMed]
27. Neufeld NH, Kaczkurkin AN, Sotiras A, Mulsant BH, Dickie EW, Flint AJ, Meyers BS, Alexopoulos GS, Rothschild AJ, Whyte EM, Mah L, Nierenberg J, Hoptman MJ, Davatzikos C, Satterthwaite TD, Voineskos AN. Structural brain networks in remitted psychotic depression. Neuropsychopharmacology 2020;45:1223-31. [Crossref] [PubMed]
28. Tomasi D, Volkow ND. Functional connectivity hubs in the human brain. Neuroimage 2011;57:908-17. [Crossref] [PubMed]
29. Haldane M, Cunningham G, Androutsos C, Frangou S. Structural brain correlates of response inhibition in Bipolar Disorder I. J Psychopharmacol 2008;22:138-43. [Crossref] [PubMed]
30. Anand A, Jones SE, Lowe M, Karne H, Koirala P. Resting State Functional Connectivity of Dorsal Raphe Nucleus and Ventral Tegmental Area in Medication-Free Young Adults With Major Depression. Front Psychiatry 2019;9:765. [Crossref] [PubMed]
31. Polyanska L, Critchley HD, Rae CL. Centrality of prefrontal and motor preparation cortices to Tourette Syndrome revealed by meta-analysis of task-based neuroimaging studies. Neuroimage Clin 2017;16:257-67. [Crossref] [PubMed]
32. Petrides M. On the evolution of polysensory superior temporal sulcus and middle temporal gyrus: A key component of the semantic system in the human brain. J Comp Neurol 2023;531:1987-95. [Crossref] [PubMed]
33. Briggs RG, Tanglay O, Dadario NB, Young IM, Fonseka RD, Hormovas J, Dhanaraj V, Lin YH, Kim SJ, Bouvette A, Chakraborty AR, Milligan TM, Abraham CJ, Anderson CD, O'Donoghue DL, Sughrue ME. The Unique Fiber Anatomy of Middle Temporal Gyrus Default Mode Connectivity. Oper Neurosurg 2021;21:E8-E14. [Crossref] [PubMed]
34. Xu J, Zhang J, Li J, Wang H, Chen J, Lyu H, Hu Q. Structural and Functional Trajectories of Middle Temporal Gyrus Sub-Regions During Life Span: A Potential Biomarker of Brain Development and Aging. Front Aging Neurosci 2022;14:799260. [Crossref] [PubMed]
35. Chen Y, Lei D, Cao H, Niu R, Chen F, Chen L, Zhou J, Hu X, Huang X, Guo L, Sweeney JA, Gong Q. Altered single-subject gray matter structural networks in drug-naïve attention deficit hyperactivity disorder children. Hum Brain Mapp 2022;43:1256-64. [Crossref] [PubMed]
36. Xu J, Wang C, Xu Z, Li T, Chen F, Chen K, Gao J, Wang J, Hu Q. Specific Functional Connectivity Patterns of Middle Temporal Gyrus Subregions in Children and Adults with Autism Spectrum Disorder. Autism Res 2020;13:410-22. [Crossref] [PubMed]
37. Bai C, Wang Y, Zhang Y, Wang X, Chen Z, Yu W, Zhang H, Li X, Zhu K, Wang Y, Zhang T. Abnormal gray matter volume and functional connectivity patterns in social cognition-related brain regions of young children with autism spectrum disorder. Autism Res 2023;16:1124-37. [Crossref] [PubMed]
38. Bhikram T, Crawley A, Arnold P, Abi-Jaoude E, Sandor P. Neuroimaging the emotional modulation of urge inhibition in Tourette Syndrome. Cortex 2021;135:341-51. [Crossref] [PubMed]
39. Wang S, Zhao Y, Wang X, Yang X, Cheng B, Pan N, Suo X, Gong Q. Emotional intelligence mediates the association between middle temporal gyrus gray matter volume and social anxiety in late adolescence. Eur Child Adolesc Psychiatry 2021;30:1857-69. [Crossref] [PubMed]
40. Wang Y, Gong N, Fu C. Major depression disorder diagnosis and analysis based on structural magnetic resonance imaging and deep learning. J Integr Neurosci 2021;20:977-84. [Crossref] [PubMed]
41. Bayard F, Nymberg Thunell C, Abé C, Almeida R, Banaschewski T, Barker G, et al. Distinct brain structure and behavior related to ADHD and conduct disorder traits. Mol Psychiatry 2020;25:3020-33. [Crossref] [PubMed]
42. Krain AL, Castellanos FX. Brain development and ADHD. Clin Psychol Rev 2006;26:433-44. [Crossref] [PubMed]
43. Greene DJ, Williams Iii AC, Koller JM, Schlaggar BL, Black KJThe Tourette Association of America Neuroimaging Consortium. Brain structure in pediatric Tourette syndrome. Mol Psychiatry 2017;22:972-80. Erratum in: Mol Psychiatry 2020;25:3112. [Crossref] [PubMed]
44. Jackson SR, Parkinson A, Jung J, Ryan SE, Morgan PS, Hollis C, Jackson GM. Compensatory neural reorganization in Tourette syndrome. Curr Biol 2011;21:580-5. [Crossref] [PubMed]
45. Fahim C, Yoon U, Das S, Lyttelton O, Chen J, Arnaoutelis R, Rouleau G, Sandor P, Frey K, Brandner C, Evans AC. Somatosensory-motor bodily representation cortical thinning in Tourette: effects of tic severity, age and gender. Cortex 2010;46:750-60. [Crossref] [PubMed]
46. Luo Y, Weibman D, Halperin JM, Li X. A Review of Heterogeneity in Attention Deficit/Hyperactivity Disorder (ADHD). Front Hum Neurosci 2019;13:42. [Crossref] [PubMed]
47. Hoogman M, Muetzel R, Guimaraes JP, Shumskaya E, Mennes M, Zwiers MP, et al. Brain Imaging of the Cortex in ADHD: A Coordinated Analysis of Large-Scale Clinical and Population-Based Samples. Am J Psychiatry 2019;176:531-42. [Crossref] [PubMed]
48. Karbach J, Unger K. Executive control training from middle childhood to adolescence. Front Psychol 2014;5:390. [Crossref] [PubMed]
49. Zilverstand A, Sorger B, Slaats-Willemse D, Kan CC, Goebel R, Buitelaar JK. fMRI Neurofeedback Training for Increasing Anterior Cingulate Cortex Activation in Adult Attention Deficit Hyperactivity Disorder. An Exploratory Randomized, Single-Blinded Study. PLoS One 2017;12:e0170795. [Crossref] [PubMed]
50. Lu K, Yueh K, Hu H, Guo M, Liu Y. A Novel Neurofeedback Attentional Enhancement Approach Based on Virtual Reality. Annu Int Conf IEEE Eng Med Biol Soc 2022;2022:5140-3. [Crossref] [PubMed]