Major depression disorder (MDD) has a high incidence and low treatment efficiency, which is the leading cause of global disease burden. The survey showed that about 67 percent of mental illness starts before the age of 21，Children experience depression might have a higher risk of impaired function and depression onset in the future life. Therefore, the related characteristics of neural development is crucial. To describe the developmental trend and influencing factors in the school-age stage can help us understand the brain mechanism of depression from the perspective of development and intervene early.
This paper includes three studies:
Study one reviewed the resting state functional magnetic resonance imaging researches of MDD according to different age groups and different medicine status, in order to determine the abnormal patterns of brain activity associated with adolescent depression. The results found that adolescent, adult and the eldely MDD patients showed different resting state brain activity characteristics: adolescents MDD mainly shows abnormal connectivity in the default network, rarely involves the control network. Adult MDD showed a decease of function connectivity in the whole brain on the condition of unmedicated, after antidepressant treatment, the control network and the emotion network showed a recovery of function connectivity, but the sensorimotor network still decline. Different analysis methods such as seed-bassed function connection analysis (SCA), regional homogeneity (ReHo) and low frequency amplitude (ALFF) showed consistent results.
According to the results of study one, study two choose the core regions of default network一posterior cingulate cortex(PCC) and medial prefrontal cortex (MPFC) as regions of interest and investigated their structure development among three measurements in five years.We are also interested in the influence of depression on the development of brain structure. Subjects were from "China color nest project (CCNP)", 81 children were included. Latent variable grow curve model (LGCM) were used to model the three times of brain structure index, which include surface area，cortical thickness, and volume. The results found that in the index of surface area，PCC showed a linearly decline trend, while MPFC did not present a linearly increase or decrease trend. In the cortical thickness index, the MPFC regions showed a trend of linear decline, while the PCC did not show any trend. In the cortical volume index, both PCC and MPFC regions showed a trend of linear decline. This study also found the subregion of PCC一right PCV had significant individual differences at the initial level. Latent variable growth model analysis shows that children's first depression scores has positive prediction effect on the initial volume of PCV, children with high depression scores has larger initial volume.
Study three firstly used ALFF and ReHo to explore the developmental tendency and predictive effects of local function of brain in school-age children. The results showed that the development trend of PCC and MPFC on different indexes is different. In ALFF index, the development of MPFC is linearly decline, while PCC does not show this trend. On the ReHo index, the PCC showed a linearly decline trend, and the MPFC did not show this trend. The study also found a prediction of gender,depression scores and their interactions on the initial level and development speed. In the subregion of PCC一31 pd, the initial values of ALFF and ReHo are higher in girls than in boys, girls with high depression scores growing fast on ALFF value, at the same time, girls with high depression scores decrease slowly on ReHo value.
Study three then take the PCC_ 31pd as the seed region to calculate the correlation of the region with the other region of the whole brain. Each individual has three times correlation values. LGCV analysis showed that the PCC showed a downward connectivity trend with multiple regions, including auditory cortex, the visual cortex, sensorimotor cortex, temporal association cortices. While the connectivity between 31pd and other regions did not show significant decline,including lateral temporal lobe cortex, the medial prefrontal cortex, posterior cingulate cortex and dorsolateral prefrontal cortex. Further, we found the interaction of gender and depression could predict the connectivity between right 31pd and left PGi area. Girls with high depression scores had higher connectivity between the right side of 31pd and the left PGi, while boys did not show significant differnece. PGi also belongs to the default network, which showed that gender and depression scores can prediction the default network long-distance connection.
Based on the review of abnormal patterns of depression brain activity according to different age and medication status, we choose the core regions of default mode network一PCC and MPFC as regions of interest to investigate the development of brain structure and function in school一age children and the effect of depression on brain development. The results showed that depression had a prediction on both the initial level and rate of development, which suggest that the onset of depression may be a continuous process from quantitative change to qualitative change and it is important to detect early brain imaging indicators of depression.