According to a report titled “Depression and Other Common Mental Disorders: A Global Health Assessment,” which was released by the World Health Organization, the number of people worldwide suffering from depression has reached 322 million, or 4.4 % of the world's population. The prevalence of depression continues to increase. Major depressive disorder (MDD), associated with increased mortality and the utilization of medical resources, is the leading cause of disability worldwide. Not only that, but MDD imposes a serious economic burden on our society (Friedrich, 2017). The core features of MDD are a variety of emotional, cognitive, and somatic symptoms (Beck and Bredemeier, 2016; Kapfhammer, 2006; Kop, 2012). Studies have shown that patients suffering from MDD often complain about somatic symptoms, and about 69 % of MDD patients demonstrate somatic symptoms (Bekhuis et al., 2015; Haj Kheder et al., 2018; Quick et al., 2010). A study of 3273 patients in China who met the diagnostic criteria of Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) MDD found that the major somatic symptoms in patients with MDD were insomnia, pre-verbal physical complaints, weight loss, low appetite, circulatory system complaints, headache, hyposexuality, gastrointestinal symptom complaints, and respiratory system complaints (Zhao et al., 2018). In fact, somatic symptoms are the main reason many MDD patients seek medical care, and patients often seek medical help for somatic symptoms rather than depressive symptoms (Kop, 2012; Rijavec and Grubic, 2012; Tylee and Gandhi, 2005). Moreover, the majority of these patients do not go to the psychiatric department for medical treatment, but rather to other departments of general hospitals for medical help. This makes it difficult for them to receive a timely and accurate diagnosis, which in turn delays the opportunity for proper treatment of their condition (Bohman et al., 2010). Previous studies have shown that somatic symptoms are associated with poor outcomes in MDD (Gerrits et al., 2012). The remission rate of MDD patients with severe somatic symptoms was half that of those without somatic symptoms (Novick et al., 2013). MDD patients with somatic symptoms also tend to have a longer course of disease and require longer treatment time than MDD patients without somatic symptoms (Bekhuis et al., 2016). Therefore, paying attention to the somatic symptoms of depressive patients and exploring the biological pathogenesis of those with somatic symptoms may help us to improve diagnosis and treatment.

There is increasing evidence that the pathophysiological mechanism of somatic symptoms in patients with depression may be related to both brain structure and function. With the development of neuroimaging technology, resting-state functional magnetic resonance imaging (rs-fMRI) has been widely used in the study of many mental diseases including depression. This is due to the advantages of its non-invasive, accurate localization, in addition to its high temporal and spatial resolution and task-free demands (Jaime et al., 2019; O'Connor and Zeffiro, 2019). Previous studies have shown that the somatic symptoms of depressive patients may be related to abnormalities in multiple brain regions and brain connections, those which involve multiple brain networks such as the limbic cortex and the default mode network (DMN), and that cover the frontal lobe, temporal lobe, insula, thalamus, striatum and other brain regions (R. Yan et al., 2019; Avery et al., 2014; Heinzel et al., 2009; Zhang et al., 2021; Liu et al., 2021a). A study that explored neural activity in the brain during interperceptual attention tasks found that somatic symptoms in MDD patients were associated with abnormal intersensory awareness in the insula (Avery et al., 2014). Some studies used amplitude of low-frequency fluctuation (ALFF) and reagional homogeneity (ReHo) to compare the differences of brain neural activity in patients with and without somatic symptoms of depression, and found significant differences in ReHo and ALFF in the orbitofrontal gyrus, thalamus, hippocampus, middle frontal gyrus, and other brain regions (R. Yan et al., 2019; Avery et al., 2014; Geng et al., 2019). Our group's previous study also showed decreased ReHo and ALFF in the bilateral precentral gyrus, bilateral postcentral gyrus, and left paracentral gyrus, and decreased ALFF in the left superior occipital gyrus and left middle occipital gyrus in the somatic symptoms group (Liu et al., 2021a). Other studies explored the neurobiological mechanism of somatic symptoms in patients with depression using functional connectivity indicators. In patients with somatic symptoms, such studies have located abnormal functional connectivity between the right inferior orbitofrontal gyrus (OFC) and left inferior parietal cortex (R. Yan et al., 2019), and between the ventral anterior insula and right OFC (Zhang et al., 2021).

In conclusion, both previous studies as well as our group's previous study suggested extensive brain dysfunction in depression with somatic symptoms through various brain imaging indicators. Few studies, however, have focused on the brain structure of depression with somatic symptoms. Brain structural neuroimaging provides insights into brain structure and neurocognitive development. T1-weighted imaging is the most commonly acquired MRI sequence of structural neuroimaging, among which the most important structural measures are Gray matter volume (GMV) and gray matter density (GMD). GMV represents the absolute amount of gray matter, whereas GMD represents the relative concentration of gray matter structures in spatially warped images (Mechelli et al., 2005). Gray matter abnormalities have also been widely reported in psychiatric disorders (Brent et al., 2013). Focusing on both absolute GMV and GMD may help our understanding of the mechanism of the brain. In fact, when searching the database for specific somatic symptoms, we found that somatic symptoms such as insomnia, gastrointestinal (GI) symptoms, and chronic pain were associated with a wide range of brain structural abnormalities (Leerssen et al., 2020; Liu et al., 2019, Liu et al., 2021b; Tang et al., 2015). Leerssen's study showed that MDD patients with severe insomnia have a smaller cortical surface area than those without insomnia (Leerssen et al., 2020). The GI symptoms of MDD patients were associated with abnormal GMV and GMD in the thalamus, hippocampus, middle frontal gyrus, precentral gyrus, insula, superior temporal gyrus, superior occipital gyrus, and precentral gyrus (Liu et al., 2019, Liu et al., 2021b). Chronic pain was associated with abnormal GMV in cingulate gyrus, PFC, and insula (Tang et al., 2015). Therefore, exploring brain structural changes in depression with and without somatic symptoms can enrich the understanding of the pathological mechanism of somatic symptoms in depression, and further, can fill a research gap in this particular field. In view of the fact that there is no large sample study on the brain structure changes of somatic symptoms in MDD patients, this study compared the GMV and GMD of MDD patients with and without somatic symptoms by using the large sample data of REST-Meta-MDD project (C.G. Yan et al., 2019), so as to further clarify the neurobiological basis of somatic symptoms in depressive patients.