Participants of the current study were enrolled from General Hospital of the eastern theater of the Chinese People’s Liberation Army (Nanjing, Jiangsu Province, China). In this study, pregnant women (within 12 weeks of pregnancy) who registered in our hospital and agreed to participate in this study were recruited in the obstetrics clinic by means of poster advertisements and introduction by outpatient physician. Inclusion criteria are as follows: i) pregnant women within 12 weeks of gestation; ii) registered at our hospital’s obstetrics clinic; iii) willing to participate and provide written informed consent; iv) no depressive symptoms at baseline. Exclusion criteria are as follows: i) any pre-existing psychiatric disorders (assessed by medical records or self-report); ii) voluntary withdrawal from the study; iii) incomplete questionnaire or physical examination data; iv) depressive symptoms at baseline. The flow diagram is shown in Figure 1 . Briefly, a total of 577 Chinese early pregnant were recruited between October 2021 and October 2022. Of these, 18 participants were excluded due to having any psychiatric disorder, or voluntary withdrawal from the study. Among the 559 enrolled participants, 554 completed the questionnaire survey, 557 completed the physical examination and peripheral blood sample collection in the first trimester. Subsequently, 59 participants were excluded due to the following reasons: i) without complete questionnaire or physical examination data; ii) depressive symptoms at the first trimester. Consequently, 500 participants who had no depressive symptoms at the first trimester and complete questionnaire or physical examination data were included. At the second trimester, 500 participants completed the blood sample collection. At the third trimester, 500 participants completed the questionnaire survey and blood sample collection. Blood samples collected during the first, second, and third trimester were used to detect BDNF. Ultimately, a total of 500 participants with complete questionnaire data and blood samples were included in the analysis. This study was approved by ethics committee of Nanjing Jinling Hospital, Affiliated Hospital of Nanjing Medical University (Nanjing, China). All participants provided written informed consent.

The depressive symptoms of pregnant women in the first and third trimester were evaluated using the 9-item Patient Health Questionnaire (PHQ-9), a widely used screening instrument for depressive symptoms (30, 31). The total score of PHQ-9 ranges from 0 to 66, with higher scores indicating more severe depressive symptoms. The PHQ-9 score was used to determine whether the participants had depressive symptoms and the degree of depressive symptoms. Two cut-off values were set in this study: i) no depressive symptoms (PHQ-9 score ≤ 4), depressive symptoms (PHQ-9 score ≥ 5); ii) no depressive symptoms (0 ≤ PHQ-9 score ≤ 4), minor depressive symptoms (5 ≤ PHQ-9 score ≤ 9), mild depressive symptoms (10 ≤ PHQ-9 score ≤ 14), moderate-severe depressive symptoms (15 ≤ PHQ-9 score ≤ 27).

Participants provided fasting blood samples (5 mL) of peripheral blood during the first, second and third trimester of pregnancy. The blood samples were centrifuged in a refrigerated centrifuge at 3000 rpm/min for 8 minutes to collect the serum. Serum was collected after centrifugation and stored in a −80°C freezer until assayed. The levels of serum BDNF in the first, second and third trimester were measured using an AU5400 automatic biochemical analyzer (Olympus Corporation, Japan). Serum BDNF levels were determined with enzyme-linked immunosorbent assay (ELISA) kits (DY248, RD Systems, USA) according to the manufacturer’s instructions.

Covariates were selected based on prior knowledge and literature (18, 32). Information on covariates in the first trimester was obtained as following two ways: i) questionnaire, including age, marital status (married, unmarried, divorced), household economic conditions (poor, fair, good), educational level (high school and below, junior college, undergraduate and above), sleep duration, physical activity (< 1 h/d or ≥ 1 h/d), occupations (worker, clerk, teacher, farmer, medical professional, freelancer, other), and first trimester PHQ-9 score; ii) physical examination, body mass index (BMI) = weight (kg)/height (m) squared.

Sample characteristics were summarized as number (percentage) for categorical variables and as mean ± standard deviation (SD) for continuous variables. Statistical differences in sample characteristics between groups (non-depressive symptoms vs. depressive symptoms) were analyzed using independent-sample t-test for quantitative data and chi-squared test or Fisher exact probability test for categorical data. Differences in PHQ-9 score at the third trimester between different BDNF levels groups (Q1 vs. Q2, Q3, Q4) were tested by one-way analysis of variance (ANOVA). Correlations between BDNF in the first, second and third trimester with the PHQ-9 score in the first and third trimester were examined by the Pearson correlation analysis.

Univariable and multivariable linear regression analyses were conducted to assess the associations between BDNF levels in different trimesters and PHQ-9 score in the third trimester. In addition, binary and multivariable logistic regression models were used to evaluate the association between BDNF levels in different trimesters and depressive symptoms in the third trimester. The β, odds ratio (OR), and 95% confidence interval (CI) were reported. Three models were fitted in multivariable linear regression analyses and logistic regression analyses: model 1 was unadjusted; model 2 was adjusted for the first trimester age, marital status, household economic conditions, educational level, body mass index, sleep duration, physical activity, and occupations; model 3 was additionally adjusted for the first trimester PHQ-9 score. All analyses were completed using SPSS (version 27.0, Chicago, USA).

The characteristics of the study sample are shown in Table 1 . The mean age of the 500 participants was 26.8 ± 2.3 years at the first trimester. At the third trimester follow-up survey, a total of 72 pregnant women (14.4%) developed depressive symptoms. Compared with the non-depressed group, individuals in the depressed group were younger, had higher rates of divorce and unmarried, slept longer, and had lower levels of physical activity (all P-values < 0.05). In addition, BDNF levels in the first, second and third trimester were significantly lower in the depressed group than in the non-depressed group (all P-values < 0.001).

The distribution of PHQ-9 score in the third trimester among different BDNF levels groups is shown in Figure 2 . Compared with the BDNF Q1 group in the first trimester, the PHQ-9 score in the third trimester was significantly lower in the Q2, Q3 and Q4 groups (Q1 vs. Q2 vs. Q3 vs. Q4 = 7.9 ± 6.5 points vs. 2.2 ± 1.9 points vs. 2.3 ± 1.8 points vs. 2.1 ± 2.0 points; all P-values < 0.01). Similarly, compared with the BDNF Q1 group in the second and third trimesters, the PHQ-9 score in the third trimester was significantly lower in the Q2, Q3 and Q4 groups (all P-values < 0.01).

The results of correlation analysis are presented in Supplementary Table S1 . Serum BDNF levels in the first, second and third trimesters were negatively correlated with PHQ-9 score in the first and third trimesters (correlation coefficients between −0.54 and −0.09; all P-values < 0.05).

The results of linear regression analysis are depicted in Figure 3 ; Supplementary Table S2 . In multivariable linear regression models, after adjusting for the first trimester age, marital status, household economic conditions, educational level, body mass index, sleep duration, physical activity, occupations, and the first trimester PHQ-9 score, each 1 pg/mL increase of first trimester BDNF was associated with 0.64 point-decrease in PHQ-9 score in the third trimester (β = −0.64; 95% CI: −0.73, −0.55; P < 0.001); Each 1 pg/mL increase of second trimester BDNF was associated with 0.52 point-decrease in PHQ-9 score in the third trimester (β = −0.52; 95% CI: −0.61, −0.43; P < 0.001); Each 1 pg/mL increase of third trimester BDNF was associated with 0.55 point-decrease in PHQ-9 score in the third trimester (β = −0.55; 95% CI: −0.64, −0.46; P < 0.001). Results from unadjusted and adjusted models were generally consistent ( Figure 3 ; Supplementary Table S2 ).

Table 2 demonstrates the results of binary logistic regression analysis. After adjusting for covariates, each 1 pg/mL increase of first trimester BDNF was associated with a 43% decrease in the risk of depressive symptoms in the third trimester (OR = 0.57, 95% CI: 0.51, 0.65); each 1 pg/mL increase of second trimester BDNF was a 39% decrease in the risk of depressive symptoms in the third trimester (OR = 0.61, 95% CI: 0.54, 0.68); each 1 pg/mL increase of third trimester BDNF was associated with a 36% decrease in the risk of depressive symptoms in the third trimester (OR = 0.64, 95% CI: 0.58, 0.71).

In adjusted multivariable logistic regression models ( Table 3 ), each 1 pg/mL increase of first trimester BDNF was associated with 32% lower risk for minor depressive symptoms in the third trimester (OR = 0.68, 95% CI: 0.59, 0.79), 46% lower risk for mild depressive symptoms in the third trimester (OR = 0.54, 95% CI: 0.45, 0.65), 59% lower risk for moderate-severe depressive symptoms in the third trimester (OR = 0.41, 95% CI: 0.30, 0.56); each 1 pg/mL increase of second trimester BDNF was associated with 24% lower risk for minor depressive symptoms in the third trimester (OR = 0.76, 95% CI: 0.65, 0.90), 40% lower risk for mild depressive symptoms in the third trimester (OR = 0.60, 95% CI: 0.50, 0.72), 76% lower risk for moderate-severe depressive symptoms in the third trimester (OR = 0.24, 95% CI: 0.14, 0.42); each 1 pg/mL increase of third trimester BDNF was associated with 20% lower risk for minor depressive symptoms in the third trimester (OR = 0.80, 95% CI: 0.70, 0.92), 42% lower risk for mild depressive symptoms in the third trimester (OR = 0.58, 95% CI: 0.49, 0.70), 85% lower risk for moderate-severe depressive symptoms in the third trimester (OR = 0.15, 95% CI: 0.07, 0.33).