Maternal thyroid-stimulating hormone and thyroid peroxidase antibody levels: Associations with postpartum depression and impacts on adverse pregnancy outcomes

Abstract

BACKGROUND

Postpartum depression (PPD) is closely associated with adverse pregnancy outcomes (APOs); thus, early PPD detection and prevention may help safeguard maternal and neonatal health.

AIM

To clarify the roles played by maternal serum thyroid-stimulating hormone (TSH) and thyroid peroxidase antibody (TPOAb) in PPD risk and APOs.

METHODS

We enrolled 83 women with PPD and 80 healthy pregnant controls from Shenzhen Bao’an District Songgang People’s Hospital (March 2021 to March 2024). Serum TSH and TPOAb levels were compared between groups. In PPD cases, association between thyroid markers and depressive severity, measured using the Edinburgh Postnatal Depression Scale and Self-Rating Depression Scale, were examined using Spearman’s correlations. Based on pregnancy outcomes, APOs (premature membrane rupture, preterm birth, postpartum hemorrhage, macrosomia, hypertensive disorders complicating pregnancy, and gestational diabetes) were contrasted with favorable outcomes. Univariate and multivariable analyses identified APOs predictors.

RESULTS

Compared with controls, PPD cases had higher TSH and TPOAb levels, both positively correlated with Edinburgh Postnatal Depression Scale and Self-Rating Depression Scale. In the final multivariable model, independent predictors of APOs were multiparity ≥ 3 [odds ratio (OR) = 3.782, P = 0.007], PPD (OR = 2.921, P = 0.046), TSH ≥ 3 mIU/L (OR = 3.333, P = 0.019), and TPOAb ≥ 7 U/mL (OR = 3.958, P = 0.008).

CONCLUSION

Serum TSH and TPOAb levels were significantly elevated in mothers with PPD and correlated strongly with depressive severity. Multiparity ≥ 3, PPD, TSH ≥ 3 mIU/L, and TPOAb ≥ 7 U/mL each increase the risk of APOs.

Key Words: Thyroid stimulating hormone; Thyroid peroxidase antibody; Pregnant women; Postpartum depression; Adverse pregnancy outcomes

Core Tip: This study included 83 patients with postpartum depression (PPD) and 80 healthy controls. By comparing serum-stimulating hormone and thyroid peroxidase antibody levels and assessing their associations with depressive symptoms, we identified both makers as positively correlated with PPD severity and as independent contributors to adverse pregnancy outcomes. Multiparity and PPD further heightened adverse pregnancy outcomes risk.

INTRODUCTION

Postpartum depression (PPD) is a common psychiatric disorder occurring within a woman’s first postnatal year and is characterized by persistent anxiety, depression, fatigue, and functional impairment[1]. More than 80% of women experience some degree of emotional disturbance during pregnancy or after childbirth[2]. Beyond affecting maternal well-being, PPD can impair newborn nutrition, growth, and later mental health, and in severe cases, may even lead to infanticide[3]. PPD often begins during pregnancy, with contributing factors such as primiparity, prenatal psychosocial stressors, insufficient prenatal support, and inadequate micronutrient status[4,5]. Because PPD is closely linked to adverse pregnancy outcomes (APOs), early screening and management are essential to reduce these risks and protect maternal and neonatal health[6].

Abnormally elevated serum thyroid-stimulating hormone (TSH) and the presence of thyroid autoantibodies, including thyroid peroxidase antibody (TPOAb), are known to increase the likelihood of PPD and its related symptoms. Prenatal TPOAb titer may help identify individuals at high risk of PPD through mechanisms involving positive feedback between placental corticotropin-releasing hormone and cortisol[7]. Ciolac et al[8] also reported that elevated TSH contributes to recognizing PPD-associated hypothyroidism. Increased TSH levels may adversely impact childbirth outcomes, particularly when accompanied by persistent severe maternal anxiety, which further elevates the risk of delivering a large for gestational age (LGA) infant[9]. Additionally, Wang et al[10], found that TPOAb positivity was inked to higher rates of postpartum thyroiditis and premature rupture of membranes (PROM), suggesting broader associations with adverse maternal and neonatal outcomes. Based on this evidence, we hypothesized that serum TSH and TPOAb levels are associated with maternal PPD and contribute to the development of APOs. The present study therefore analyzed and verified these relationships.

MATERIALS AND METHODS

General data

A total of 163 pregnant women from Shenzhen Bao’an District Songgang People’s Hospital (March 2021 to March 2024) were included, comprising 83 subjects with PPD group and 80 healthy pregnancies. Demographic characteristics were comparable between groups (P > 0.05).

Participant selection criteria

Inclusion criteria: (1) Meeting diagnostic requirements for PPD or an Edinburgh Postnatal Depression Scale (EPDS) score ≥ 13[11,12]; (2) Age 22-45 years; (3) 6-7 weeks postpartum; (4) Normal cognitive, communication, and executive ability; and (5) Delivery at our center with complete clinical data.

Exclusion criteria: (1) Current or past personality or psychiatric disorders; (2) Use of psychotropic agents or sedative-hypnotics within ≤ 4 weeks; (3) Antepartum depression or anxiety; (4) Gestational hypertension or diabetes; (5) Organic brain or systemic disease; (6) Thyroid dysfunction; (7) Major physical illness; and (8) Anxiety-disorders or schizophrenia spectrum disorders.

Detection indicators

Serum indices: All enrolled mothers fasted overnight (> 8 hours). Morning venous blood (3 mL) was collected from resting participants in the supine position via antecubital puncture. After centrifugation, serum TSH and TPOAB concentrations were quantified via chemiluminescence, which measures signal intensity proportional to antigen-antibody complex formation and enables accurate quantitative detection.

Negative emotions: All participants completed the EPDS and the Self-rating Depression Scale (SDS)[13]. The EPDS contains 10 items scored 0-3, yielding a total score of 0-30, with ≥ 13 commonly used as the PPD threshold. The SDS comprises 20 items rated 1-4 and assesses emotional, somatic, psychomotor, and interest-related symptoms. A standard score is derived by multiplying the total by 1.25; scores ≥ 50 indicate depressive symptoms.

APOs: The occurrence of APOs, including PROM, preterm birth, postpartum hemorrhage, LGA infants, hypertensive disorder complicating pregnancy (HDP), and gestational diabetes mellitus (GDM), was recorded in both groups, and total incidence rates were calculated.

Statistical analysis

Statistical analyses were performed using SPSS 22.0. Continuous variables are expressed as mean ± SD and compared using t-tests, whereas categorical variables are shown as n (%) and compared with χ² tests. Spearman correlations assessed associations of TSH and TPOAb concentrations with PPD severity in PPD cases. Predictors of APOs were identified through univariable and multivariable binary logistic regression. Statistical significance was set at P < 0.05.

RESULTS

Baseline data assessment in PPD cases and controls

Table 1 summarizes baseline characteristics for the PPD (n = 83) and control (n = 80) groups. Age, gravidity, gestational age at delivery, neonatal birth weight, and feeding practices did not differ significantly (P > 0.05).

Table 1 Comparative baseline characteristics of study participants, n (%)/mean ± SD.

Indicators	PPD group (n = 83)	Control group (n = 80)	χ2/t	P value
Age (years)	28.37 ± 4.41	29.16 ± 4.19	1.172	0.243
Gravidity (times)	2.10 ± 0.91	2.20 ± 0.89	0.709	0.479
Gestational age at delivery (weeks)	38.72 ± 1.54	38.34 ± 1.55	1.570	0.118
Neonatal birth weight (g)	3110.55 ± 279.95	3161.71 ± 208.20	1.320	0.189
Newborn feeding practices			2.555	0.279
Breast milk	42 (50.60)	43 (53.75)
Milk powder	15 (18.07)	20 (25.00)
Mixed	26 (31.33)	17 (21.25)

PPD: Postpartum depression.

Evaluation of TSH and TPOAb concentrations across groups

As shown in Figure 1, serum TSH and TPOAb concentrations were significantly higher in PPD cases than in controls (P < 0.01).

Figure 1 Serum thyroid-stimulating hormone and thyroid peroxidase antibody measurements in postpartum depression and control groups. A: Serum thyroid-stimulating hormone concentrations in postpartum depression groups vs controls; B: Serum thyroid peroxidase antibody levels in postpartum depression cases vs controls. ^bP < 0.01, vs controls. TSH: Thyroid-stimulating hormone; PPD: Postpartum depression; TPOAb: Thyroid peroxidase antibody.

Correlation of serum TSH and TPOAb levels with EPDS and SDS in PPD cases

Table 2 presents Spearman correlation analyses between serum TSH/TPOAb levels and EPDS/SDS scores in PPD patients (n = 83). TSH levels correlated positively with EPDS (r = 0.500, P < 0.001) and SDS (r = 0.303, P = 0.005). TPOAb levels also showed positive correlations with EPDS (r = 0.347, P = 0.001) and SDS (r = 0.331, P = 0.002).

Table 2 Associations of thyroid-stimulating hormone and thyroid peroxidase antibody concentrations with Edinburgh Postnatal Depression Scale and Self-Rating Depression Scale scales among postpartum depression groups.

Indicators	EPDS (points)		SDS (points)
	r	P value	r	P value
TSH	0.500	< 0.001	0.303	0.005
TPOAb	0.347	0.001	0.331	0.002

TSH: Thyroid-stimulating hormone; TPOAb: Thyroid peroxidase antibody; EPDS: Edinburgh Postnatal Depression Scale; SDS: Self-Rating Depression Scale.

APOs in PPD and control groups

Table 3 compares pregnancy complications between cohorts. Women with PPD exhibited higher incidences of PROM, preterm birth, postpartum hemorrhage, LGA infants, HDP, and GDM than controls (P < 0.001).

Table 3 Adverse pregnancy outcomes in postpartum depression group vs controls, n (%).

Indicators	PPD group (n = 83)	Control group (n = 80)	χ2	P value
PROM	6 (7.23)	1 (1.25)
Preterm birth	5 (6.02)	2 (2.50)
Postpartum hemorrhage	4 (4.82)	1 (1.25)
LGA infants	3 (3.61)	0 (0.00)
HDP	3 (3.61)	1 (1.25)
GDM	2 (2.41)	1 (1.25)
Total	23 (27.71)	6 (7.50)	11.380	< 0.001

PPD: Postpartum depression; PROM: Premature rupture of membranes; LGA: Large for gestational age; HDP: Hypertensive disorder complicating pregnancy; GDM: Gestational diabetes.

Univariable examination of APOs among pregnant women

Univariable comparisons (Table 4), separating adverse (n = 29) from favorable (n = 134) outcomes, showed no significant associations with maternal age, gestational age at delivery, neonatal birth weight, or infant feeding (P > 0.05). Gravidity, PPD, TSH, and TPOAb were significantly associated with APOs (P < 0.05).

Table 4 Univariable examination of adverse pregnancy outcomes among pregnant women.

Indicators	Adverse outcome group (n = 29)	Favorable outcome group (n = 134)	χ2	P value
Age (years)			0.409	0.522
< 30 (n = 93)	15 (51.72)	78 (58.21)
≥ 30 (n = 70)	14 (48.28)	56 (41.79)
Gravidity (times)			9.766	0.002
< 3 (n = 108)	12 (41.38)	96 (71.64)
≥ 3 (n = 55)	17 (58.62)	38 (28.36)
Gestational age at delivery (weeks)			1.754	0.185
< 39 (n = 80)	11 (37.93)	69 (51.49)
≥ 39 (n = 83)	18 (62.07)	65 (48.51)
Neonatal birth weight (g)			0.891	0.345
< 3150 (n = 86)	13 (44.83)	73 (54.48)
≥ 3150 (n = 77)	16 (55.17)	61 (45.52)
Newborn feeding practices			0.585	0.746
Breast milk (n = 85)	15 (51.72)	70 (52.24)
Milk powder (n = 35)	5 (17.24)	30 (22.39)
Mixed (n = 43)	9 (31.03)	34 (25.37)
PPD			11.380	< 0.001
No (n = 80)	6 (20.69)	74 (55.22)
Yes (n = 83)	23 (79.31)	60 (44.78)
TSH (mIU/L)			13.210	< 0.001
< 3 (n = 89)	7 (24.14)	82 (61.19)
≥ 3 (n = 74)	22 (75.86)	52 (38.81)
TPOAb (U/mL)			7.284	0.007
< 7 (n = 82)	8 (27.59)	74 (55.22)
≥ 7 (n = 81)	21 (72.41)	60 (44.78)

PPD: Postpartum depression; TSH: Thyroid-stimulating hormone; TPOAb: Thyroid peroxidase antibody.

Multifactorial evaluation of determinants for APOs in pregnant women

Table 5 presents the multifactorial modeling results, showing that multiparity ≥ 3 [odds ratio (OR) = 3.782, P = 0.007], PPD (OR = 2.921, P = 0.046), TSH ≥ 3 mIU/L (OR = 3.333, P = 0.019), and TPOAb ≥ 7 U/mL (OR = 3.958, P = 0.008) independently predicted APOs.

Table 5 Multifactorial evaluation of determinants for adverse pregnancy outcomes.

Indicators	B	SE	Wald	P value	OR	95%CI
Gravidity (times)	1.330	0.494	7.253	0.007	3.782	1.436-9.960
PPD	1.072	0.538	3.964	0.046	2.921	1.017-8.393
TSH (mIU/L)	1.204	0.513	5.511	0.019	3.333	1.220-9.109
TPOAb (U/mL)	1.376	0.516	7.097	0.008	3.958	1.438-10.889

OR: Odds ratio; CI: Confidence interval; PPD: Postpartum depression; TSH: Thyroid-stimulating hormone; TPOAb: Thyroid peroxidase antibody.

DISCUSSION

PPD is a common perinatal psychiatric disorder affecting up to 30% of mothers and may contribute to postpartum deaths from suicide, with an estimated risk of approximately 20%[14]. Postpartum stress, reduced quality of life, and low marital satisfaction all heighten PPD risk[15]. PPD is also linked to APOs, such as preterm birth, small for gestational age infants, preeclampsia, HDP, and GDM, which in turn increase long-term maternal mortality[16]. To mitigate these risks, the present study analyzed the association between PPD, thyroid function indicators, and APOs.

Our findings show that PPD cases had elevated serum TSH and TPOAb levels, implicating both makers in PPD pathophysiology. Elevated serum TSH level in PPD may partly reflect dysregulation of the microRNA-624-5p-silent information regulator sirtuin 1 axis in neuronal tissue; inhibition of this pathway reduces serum TSH levels and improves depressive symptoms[17]. Experimental in vivo work in rats indicates that TPOAb elevation may down-regulate brain-derived neurotrophic factor and serotonin (5-hydroxytryptamine) levels in the prefrontal cortex, thereby increasing PPD risk[18]. Consistent with our findings, Zhang et al[19] described elevated TSH level as a strong predictor of PPD and identified associations with reduced fetal weight, prenatal depression, and advanced maternal age.

Both serum TSH and TPOAb levels correlated significantly with EPDS and SDS scores, suggesting their contribution to PPD severity. Ciolac et al[8] similarly reported a significant positive correlation between TSH and EPDS scores. Kowalcze et al[20] also suggested that TSH and TPOAb levels may increase with increasing depression severity, supporting a positive correlation between these biomarkers and depression severity, and aligning with the trends observed in this study.

According to the statistical results, the overall incidence of APOs, including PROM, preterm birth, postpartum hemorrhage, LGA infants, HDP, and GDM, was significantly higher in the PPD group, indicating a strong association between PPD and APO occurrence. Multifactorial analysis further identified multiparity ≥ 3 (OR = 3.782, P = 0.007), PPD (OR = 2.921, P = 0.046), TSH ≥ 3 mIU/L (OR = 3.333, P = 0.019), and TPOAb ≥ 7 U/mL (OR = 3.958, P = 0.008) as independent predictors of APOs. Gravidity ≥ 3 may reflect cumulative reproductive tract stress, including potential endometrial injury and reduced myometrium elasticity, thereby increasing risks of PROM, preterm delivery, and postpartum hemorrhage. PPD may elevate APO likelihood by inducing chronic maternal psychological stress, which can contribute to placental dysfunction and inflammatory immune activation that adversely affect fetal development. Elevated TSH (≥ 3 mIU/L) and TPOAb (≥ 7 U/mL) may signal thyroid insufficiency and autoimmune dysregulation, respectively; together, they may exacerbate placental dysfunction and inflammatory pathways, compounding APO risk. Larsen et al[21] tracked pregnant women across 20 prenatal care facilities in western Nias and observed high APO rates linked to prenatal depression and inadequate social support, consistent with our results. Similarly, Wu et al[22] reported that TPOAb not only increased PPD risks but also contributes to GDM, hypertension, anemia, and postpartum thyroiditis. Prior literature also indicates that elevations in TPOAb and TSH levels independently correlate with APOs[23]. Mechanistically, TPOAb may promote Th1/Th2 cytokine imbalance at the maternal-fetal junction, increasing implantation failure or miscarriage risks, while TSH may interact synergistically with TPOAb in predicting adverse outcomes.

This study has several limitations. First, the single-center sample may introduce selection bias, underscoring the need for broader multicenter study to enhance generalizability. Second, maternal TSH and TPOAb levels were not dynamically monitored throughout pregnancy, future research should include longitudinal measurements to clarify their value in predicting PPD and APOs at different gestational stages. Third, although multivariable modeling was conducted, unmeasured confounders, such as psychosocial stress and family support, may still influence outcomes. Subsequent research should incorporate a more comprehensive set of psychosocial variables and extend follow-up durations to examine long-term effects on PPD and APOs.

CONCLUSION

Elevated serum TSH and TPOAb levels were observed in pregnant women with PPD and showed significant positive correlations with depression severity. Both markers independently predict APOs, and multiparity and PPD further amplified these risks. Enhanced clinical surveillance and individualized management are therefore warranted for high-risk populations.