Major Depressive Disorder (MDD) is a mental illness linked to severe consequences, including impaired social and occupational functioning, substance dependence, and suicide. Its core symptoms are persistent depressive mood and loss of interest (1). Globally, MDD impacts around 16% of individuals and stands as a primary contributor to disease burden (2, 3). The neurobiological mechanisms of MDD are not fully elucidated, and exploring these mechanisms and finding effective treatments are crucial.

It is well-known that the onset of MDD has many potential factors, with childhood trauma (CT) being the known risk factor (4). Increasing evidence suggests that people with a past of CT are more susceptible to psychiatric illnesses, including MDD (5), bipolar disorder, and post-traumatic stress disorder (6). Studies indicate that 46% of individuals diagnosed with MDD have experienced CT (7). Moreover, CT encompasses various forms such as emotional abuse, physical abuse, emotional neglect, and physical neglect; various types of CT may further elevate the risk of adverse outcomes later in life, like how physical abuse might affect an adult’s responsiveness to drug treatment (8). The impact of CT on MDD has been validated, and clinical research has shown that MDD patients with a background of CT exhibit higher suicide risks, increased severity of depressive symptoms, and functional deficits (9). Plentiful research indicates that CT is connected with immune, endocrine, and epigenetic activities, along with alterations in brain anatomy and functionality (10, 11). Despite the unique disease trajectory of MDD patients with CT, its neurobiological mechanisms remain unclear.

With the advancement of neuroimaging technologies, MRI studies suggest that healthy individuals with a history of CT show widespread cortical and subcortical morphological changes. Studies involving gray matter regions have found an association between CT experience and lower volumes of the hippocampus and amygdala (12–14). Beyond gray matter regions, several studies have reported specific white matter(WM) changes in individuals with an experience of CT, including the inferior longitudinal fasciculus, uncinate fasciculus, thalamic radiation, corona radiata, longitudinal fasciculi, cingulum, and corpus callosum (15–17). Notably, there is an anatomical overlap between the specific WM change areas in MDD patients and those mentioned above. This overlap might validate the close connection between CT and MDD, though it might also suggest a need to discern whether these changes are attributable to CT or MDD. Therefore, studying the impact of CT on MDD patients is necessary.

Many studies have proven that MDD patients with an experience of CT exhibit specific functional changes in the brain (18–21). However, only a few studies have explored the morphological changes in the brains of MDD patients with CT. These structural MRI study results are inconsistent (22, 23), with some finding changes and others not, possibly due to heterogeneity in study subjects (such as age, medication, duration of illness) and methodological differences in previous research. The effect of antidepressants on WM is unclear, as most past studies included medicated patients, making it difficult to determine whether these results are confounded by medication effects. Moreover, chronic or extended-duration illnesses may result in more pronounced neuroimaging alterations, and drug-naive patients with a shorter illness duration might help us better understand the early onset of the disease. Additionally, many studies merely demonstrate the impact of CT on the WM structure of the brain in MDD patients through correlation analysis (24, 25). Our article will refine the grouping based on CT and, on this basis, further investigate the effects of CT on participants and its interaction with the MDD diagnosis.

Diffusion Tensor Imaging (DTI) is among the most commonly employed techniques to assess the integrity of MDD brain WM, helping us better explore the disease’s underlying biological mechanisms and improve consensus on diagnosis and treatment (26). Fractional anisotropy (FA) is particularly responsive to alterations in microstructure and is often used to examine WM structural connectivity in MDD (27). Impaired WM integrity might be one of the most important pathogenic mechanisms in depression with CT, and it’s crucial to explore which fiber tracts are affected and whether WM structural changes mediate the relationship between CT and emotional/cognitive functional impairment. Among the primary analysis methods in DTI, region of interest (ROI) analysis and voxel-based morphometry present clear drawbacks: ROI analysis might restrict results to ‘regions of interest’, potentially overlooking other significant brain regions; voxel-based morphometry faces issues like registration, smoothing, and other MRI data processing problems (28). Tract-Based Spatial Statistics (TBSS) can comprehensively explore all brain regions associated with changes related to CT and is more sensitive, objective, and interpretable in DTI data analysis.

In this study, the TBSS method will be used to explore how CT and MDD affect WM structural changes and whether there is an interaction between them in medication-free MDD individuals versus comparable HCs. Additionally, we examined whether CT affects mood-related aspects by mediating changes in WM. Participants’ levels of depression and anxiety were assessed using depression and anxiety scales. Based on this, we further investigated the correlation between WM structure and the severity of depression, anxiety symptoms, and the extent of CT. We hypothesize that CT may have long-term and sustained impacts, associated with structural changes in the brains of adult patients with MDD.

This study aimed to identify neuroimaging evidence of structural brain changes in MDD patients who have experienced CT. We examined the effect of CT on MDD and the associations between CT, brain structural changes, and MDD. Our findings indicate a considerable main effect of diagnosis, with the MDD group exhibiting notable structural differences in the right IFOF, PLIC, and bilateral SLF compared to the HC group. CT did not show a significant main effect. Simple effects analysis revealed notable differences between the MDD-CT and HC-CT groups in the right IFOF, PLIC, and bilateral SLF, while the MDD-nCT and HC-nCT groups differed only in the right IFOF and PLIC. However, there were no interactive effects of CT × diagnosis. Moreover, a pronounced main effect of CT on anxiety symptoms was observed, with the CT group displaying heightened anxiety symptoms relative to the nCT group. Lastly, in the patient group, the right SLF was negatively correlated with EA, PA, PN, and total scores. This study is the first to comprehensively explore WM structural changes in drug-naive first-episode MDD patients who have experienced CT using the TBSS method.

Initially, our study found a reduction in FA in the right IFOF and PLIC in the MDD group compared to HCs. Our findings are supported by several MRI studies (30–32). Reduced FA is considered indicative of decreased WM organization, reduced axonal density, and myelination (33). Changes in these regions might be involved in the pathophysiology of depression. However, these differential changes were unrelated to the presence of CT. There were no significant differences in brain regions between the CT and nCT groups. This might indicate that the reduction in FA in the right IFOF and PLIC might be a specific change in the WM regions of MDD patients, independent of CT. These results differ from previous studies related to CT (15–17, 34, 35). However, Lim et al.’s study found that specific WM changes in the CT group were rectified in the subgroup analysis of unmedicated participants (34). Many studies have shown that medication can impact brain structure (30, 36). Therefore, the difference in our findings might be attributed to the inclusion of unmedicated participants. The IFOF, one of the longest association tracts in the human brain, links the occipital, frontal, and temporal lobes, and plays a role in emotional evaluation and visual perception regulation (37). The internal capsule, comprising multiple fiber tracts, connects the posterior limb with the parietal, temporal, occipital, and sensorimotor brain regions (38), and is associated with brain reward and self-stimulation processes (39). Structural changes in the IFOF and PLIC might be involved in the neurobiological mechanisms of MDD. We did not find a relationship between the IFOF, PLIC, and CT, which warrants further exploration in future research.

Secondly, we noted a marked decrease in FA in the bilateral SLF in the MDD group compared to HCs. Our study revealed structural changes in the SLF only between the MDD-CT and HC-CT groups, with no significant differences between MDD-nCT and HC-nCT groups. The SLF is an extensive fiber tract connecting almost all cortical areas of the lateral hemisphere, especially the frontal lobe with other cognitive or executive function brain regions. The SLF is associated with various cognitive and sensory functions, including language processing, attention control, working memory, executive function, visual and spatial processing (40, 41). Our findings in the SLF only validate prior research on MDD, suggesting that changes in SLF structure could be closely linked to the pathogenesis and progression of MDD. However, our study offers new insights regarding CT: the specific brain structural changes in MDD patients appear to be unrelated to CT, which differs from previous studies (15, 24, 42). We speculate that this might be due to related studies treating CT or MDD as isolated factors rather than examining them in conjunction. Additionally, clinical heterogeneity, including medication status, illness duration, and sample size, needs to be considered. These factors might collectively mediate the heterogeneity observed in our results. It is worth noting that CT and MDD are not entirely independent factors; CT, as a significant risk factor, has been proven to be involved in the development of MDD (43). Secondly, CT is an influencing factor in the pathogenesis of MDD but does not play a decisive role, which might partially explain the results we obtained after comparing the four groups. Finally, the significant differences between brain regions in MDD-CT and MDD-nCT might be due to differing levels of resilience among individuals with CT. Evidence suggests that CT can have varying impacts on an individual’s resilience (44), so those with experiences of CT might develop different outcomes (depression or non-depression) based on their resilience.

Although we did not identify significant brain structural changes related to CT experiences, our correlational analysis revealed a significant association between CT and FA values, as well as clinical symptoms. Our study found a notable main effect of CT on levels of anxiety. This suggests that CT might affect anxiety symptoms in both depressed and non-depressed individuals, with this impact being persistent and latent. Early life traumatic experiences might lead to higher levels of anxiety in adulthood. Our findings are supported by previous research indicating that CT experiences increase symptoms of depression, anxiety, and negative life event appraisal in adults (45). Furthermore, our exploration of the correlation between CT and WM fiber bundle FA values revealed negative correlations in the SLF with all trauma types except for the SA and EN. This suggests that CT may have a latent impact on the progression of MDD in patients. Adverse experiences in childhood, processed through individual sensory perceptions such as vision and hearing, could potentially affect the SLF, as it is involved in the processing of visual and auditory information. Consequently, we speculate that brain regions responsible for information processing may be susceptible to the effects of CT. This is supported by related research, which has found that CT predominantly affects several brain regions involved in cognitive and executive functions and information processing, such as areas for visual and auditory perception and language comprehension (46). Our findings of the correlation between CTQ scores and FA values in the SLF may indicate that CT subtly influences the developmental process of the SLF in the brains of MDD patients, thereby impairing cognitive and emotional functions and leading to more severe clinical symptoms.

Previous research has suggested that specific trauma types might have more severe impacts on individual brain structures: emotional neglect (35, 47), physical neglect (16, 35, 47), emotional abuse (8, 24), physical abuse (8, 13, 24), and sexual abuse (8, 13, 24). Our findings that EA, PA, PN and total scores predict a decrease in regional FA in the SLF suggests that the right SLF is susceptible to early adverse experiences. Future studies may concentrate on the association between different trauma types and the SLF. Moreover, the left SLF is not significantly correlated with CT scores. These differential results in the bilateral SLF suggest that we need to consider the effects of laterality. A recent study focusing on the SLF proposed that the SLF might play different roles in different hemispheres: the left SLF has a greater involvement in processing language, while the right SLF is closely related to visual spatial processing (41). A previous study supporting this view suggested that the right SLF facilitates rapid, prioritized visual-spatial processing in the right hemisphere (48). The greater correlation of the right SLF with CT might indicate a more substantial impact of CT experiences on individual visual processing-related brain regions. Additionally, in most individuals, the left hemisphere is notably less dominant in processing negative emotions (49). Therefore, CT might have a more widespread impact on the right hemisphere, known for its dominance in emotional processing, by affecting emotion regulation-related brain regions. Integrating these results, we speculate that CT experiences might subtly influence the right SLF in MDD patients, potentially altering functional networks. This alteration could impact the emotional and cognitive regulation functions of patients, potentially forming the basis for psychiatric symptoms. While these conclusions need further confirmation in future research, our findings might provide new insights into how and to what extent CT may influence the pathogenesis and progression of MDD.

Our study included drug-naive first-episode MDD patients and healthy individuals, and divided the patient and control groups into different subgroups based on CT experiences to explore the separate and interactive effects of CT and MDD diagnosis on individual brain WM fiber bundles. Our research provides new insights into the impact of CT experiences on MDD patients. However, this study has some limitations. Firstly, participants who experienced SA were quite rare in our study, which might limit our ability to detect brain changes related to this subscale. The reason for this phenomenon might be that many participants are unwilling to disclose. Future research should identify SA experiences or specifically study the impact of SA experiences on individual brain WM structure. Additionally, as the MDD-nCT patient group and HC-CT group had relatively small sample sizes, these results should be analyzed and interpreted cautiously, and future studies should include more samples. Then, our study focused solely on the FA. Although FA is widely used due to its sensitivity to microstructural changes (50), other DTI metrics such as mean diffusivity, radial diffusivity, and axial diffusivity also provide unique and complementary information about brain tissue characteristics. Future research could benefit from an integrated analysis incorporating these DTI metrics, which could lead to more comprehensive results. Lastly, the CTQ is a retrospective questionnaire, which may be subject to recall bias. Although previous research has shown that self-recollection can provide reliable information (51, 52), in MDD patients, due to their more negative self-evaluation and cognition, recollections of childhood experiences might be more negative. Future research can avoid this flaw through prospective studies or structured interviews.

This study explored the specific changes in brain WM structure in drug-naive first-episode MDD patients who have experienced CT. MDD individuals exhibited reduced FA in the right IFOF, PLIC, and bilateral SLF compared to HCs. Notably, the right SLF showed a significant reduction in FA and negative correlation with CT. Our findings suggest that the brain development of MDD patients might be latently influenced by CT, potentially linked to individual differences in resilience, with the SLF playing a pivotal role. These findings help us understand the role of the SLF in the pathophysiology of MDD and the neurobiological mechanisms by which CT influences MDD.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by the Ethics Committee of the Renmin Hospital of Wuhan University (WDRY2020-K236). The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

HZ: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Validation, Visualization, Writing – original draft, Writing – review & editing. BR: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Validation, Writing – review & editing. GG: Conceptualization, Data curation, Investigation, Methodology, Validation, Writing – review & editing. MZ: Data curation, Investigation, Methodology, Validation, Writing – review & editing. JH: Data curation, Investigation, Validation, Writing – review & editing. NT: Data curation, Investigation, Resources, Writing – review & editing. LB: Data curation, Investigation, Resources, Writing – review & editing. LX: Conceptualization, Methodology, Resources, Supervision, Validation, Writing – review & editing. GW: Conceptualization, Funding acquisition, Project administration, Resources, Supervision, Validation, Writing – review & editing.