Correlations Between Thyroid Hormone Levels of Women With Twin Pregnancies in the Third Trimester and Pregnancy Complications and Pregnancy Outcomes
Abstract
Purpose: To explore changes in thyroid hormone levels of women with twin pregnancies in the third trimester under different pregnancy complications and pregnancy outcomes and to analyze their associated correlations.
Methods: A total of 646 women with twin pregnancies who attended their first prenatal visit and received subsequent prenatal care in our hospital were enrolled in this study. Their thyroid hormone levels—free thyroxine (FT4), thyroid-stimulating hormone (TSH), and thyroid peroxidase antibody (TPO-Ab)—in the third trimester were collected for analysis of correlations with pregnancy complications and outcomes, using Spearman’s correlation and a logistic regression model.
Results: Adverse events, such as premature birth, premature rupture of membranes, placental abruption, fetal distress, fetal growth restriction, neonatal asphyxia, pregnancy-induced hypertension, gestational diabetes mellitus, and postpartum hemorrhage, occurred among the 646 women with twin pregnancies in the third trimester. The highest proportion of pregnancy outcomes was premature birth, accounting for 42.26%, and the top two pregnancy complications were gestational diabetes mellitus and pregnancy-induced hypertension, accounting for 31.42% and 20.28%, respectively. Assisted reproduction and twin type correlated with fetal distress during the third trimester of pregnancy and neonatal asphyxia. Maternal age correlated with gestational diabetes mellitus in the third trimester of pregnancy and postpartum hemorrhage. FT4 levels varied in pregnant women with placental abruption, TSH levels varied in pregnant women with pregnancy-induced hypertension, and TPO-Ab levels varied in pregnant women with premature birth, placental abruption, and postpartum hemorrhage.
Study Limitations: This study only analyzed thyroid hormone levels at a single timepoint (third trimester) during pregnancy. Dynamic changes in hormone levels across different gestational stages—such as TSH suppression patterns in the first trimester due to hCG elevation or FT4 adjustments in the second trimester—were not tracked. This limitation may hinder the understanding of how temporal hormonal variations contribute to complications like gestational diabetes mellitus or placental abruption. For instance, transient TSH fluctuations in early pregnancy or mid-trimester FT4 instability might differently influence outcomes. Future studies should incorporate longitudinal monitoring at multiple intervals (e.g., first, second, and third trimesters) to capture hormonal trajectories and their phase-specific associations with adverse outcomes. Additionally, the exclusion of women with preexisting thyroid conditions or medication use limits the generalizability of findings to high-risk populations who may benefit most from thyroid monitoring. The single-timepoint design also precludes causal inference—while correlations between third-trimester hormones and complications are observed, it remains unclear whether these abnormalities initiate pathological processes (e.g., placental dysfunction) or arise as secondary consequences of existing complications. Furthermore, the homogeneity of the cohort (e.g., predominantly dichorionic twins) may underestimate risks in monochorionic pregnancies, which inherently carry higher placental vulnerability.
Conclusion: Abnormal changes in thyroid hormones during pregnancy may correlate with adverse events such as pregnancy-induced hypertension and placental abruption. Clinicians should prioritize routine third-trimester monitoring of FT4, TSH, and TPO-Ab levels in twin pregnancies, particularly for women with assisted reproduction or monochorionic placentation, to enable early risk stratification and tailored interventions.
1. Introduction
Twin pregnancy poses tougher challenges to the physiological adaptations of the mother’s body, especially in the third trimester when maternal thyroid hormone levels closely correlate with healthy fetal development. These challenges arise from the doubled metabolic demands of supporting two fetuses, which require significantly increased thyroid hormone production. Twin pregnancies are also characterized by elevated placental mass and human chorionic gonadotropin (hCG) levels, leading to amplified thyroid-stimulating hormone (TSH) suppression and transient free thyroxine (FT4) fluctuations compared to singleton pregnancies. Additionally, immunological adaptations in twin gestations may elevate thyroid peroxidase antibody (TPO-Ab) titers, further complicating placental function. Thyroid hormones are not only critical to the mother’s metabolism but also significantly impact the neurodevelopment and overall growth of the fetuses [1–5]. However, abnormal changes in thyroid hormone levels may correlate with pregnancy complications and adverse pregnancy outcomes. This is particularly prominent in twin pregnancies. During pregnancy, TSH levels are suppressed due to elevated hCG, while FT4 increases to meet fetal metabolic demands. In twin pregnancies, higher hCG levels and greater placental mass may amplify these changes, leading to more pronounced TSH suppression and FT4 fluctuations compared to singleton pregnancies. Additionally, twin pregnancies may exhibit higher TPO-Ab positivity due to immunological adaptations specific to multiple gestations. This study aims to explore the correlations between thyroid hormone levels—including FT4, TSH, and TPO-Ab—in women with twin pregnancies during the third trimester and pregnancy complications and outcomes. Twin pregnancies were specifically selected for this study due to their unique physiological demands. Compared to singleton pregnancies, twins require a doubled maternal metabolic output, leading to amplified thyroid hormone fluctuations. For example, elevated placental mass in twins increases hCG production, which suppresses TSH and induces transient hyperthyroxinemia (higher FT4 variability). Additionally, twin pregnancies exhibit heightened immunological stress, which may elevate TPO-Ab titers and exacerbate autoimmune-mediated placental dysfunction. These dynamic hormonal interactions are particularly pronounced in the third trimester, when fetal growth demands peak and placental efficiency declines, making twin pregnancies a critical model to study thyroid-driven complications. We aimed to investigate the potential correlations between thyroid function abnormalities and pregnancy complications such as gestational diabetes mellitus, pregnancy-induced hypertension, and others [6, 7] and to evaluate the specific impacts of these hormone levels on pregnancy outcomes. In addition, factors including the application of assisted reproductive technology, type of twins (such as dichorionic or monochorionic), and ages of the pregnant women are taken into consideration. These factors may affect thyroid hormone levels and pregnancy outcomes. The all-around analysis of these complicated relationships can provide clinicians with more comprehensive strategies for thyroid function evaluation and management, thus optimizing maternal and fetal care for twin pregnancies and improving pregnancy outcomes.
2. Materials and Methods
2.1. Study Materials
The enrollment criteria were (1) pregnant women, age ≥ 18 years, and premenopausal; (2) twin pregnancy; (3) in the third trimester; and (4) informed consent of the enrolled pregnant women. The exclusion criteria were (1) a history of thyroid diseases such as thyroiditis and thyroid cancer, as well as a history of diabetes and hypertension; (2) a history of thyroid hormone medication; (3) a history of comorbid coagulation disorders and anticoagulant or hemostatic drug use; and (4) presence of vital organ dysfunction. This study was approved by the hospital’s ethics committee. Ethical approval for this study was granted by the Ethics Committee of Maternal and Child Health Hospital of Hubei Province (Approval No. 2023IEC013).
2.2. Study Methods
Questionnaires on prenatal tests completed by enrolled pregnant women included demographic characteristics, pregnancy and childbirth history, and clinical test data (including the type of twins). The questionnaires were completed with the assistance of trained professionals and collected by them.
All pregnant women accepted tests of thyroid hormone levels from the 28th to 30th gestational week. In the early morning, 3 mL of fasting blood was drawn from the median cubital vein to test the levels of FT4, TSH, and TPO-Ab on a fully automated chemiluminescence immunoassay analyzer.
All pregnant women received follow-up until the postpartum period, and pregnancy-related complications and outcomes were recorded. Relevant data were collected, verified, and analyzed by professionals.
2.3. Statistical Methods
The data in this study were processed using SPSS Statistics 22.0. Count data were described as percentages using the chi-square test. Measurement data subject to skewed distribution were described as M (IQR) and analyzed using the Mann–Whitney U test. Spearman’s correlation and logistic regression model were adopted to analyze the correlations between thyroid hormone levels, pregnancy complications, and pregnancy outcomes. Statistical significance was set as p < 0.05.
3. Results
3.1. Baseline Materials
A total of 646 women with twin pregnancies who attended their first prenatal visits in the hospital from May 2020 to May 2023 were enrolled in this study. Their median age was 31 (±4.00) years. Of these women, 345 conceived via assisted reproductive technology, accounting for 53.56%. The proportion of dichorionic twins was 76.78%.
3.2. Distribution of Pregnancy Complications and Pregnancy Outcomes
Among the 646 women with twin pregnancies in the third trimester, the highest proportion of pregnancy outcomes was premature birth, accounting for 42.26%, and the top two pregnancy complications were gestational diabetes mellitus and pregnancy-induced hypertension, accounting for 31.42% and 20.28%, respectively. This is shown in Table 1.
| Item | n (%) |
|---|---|
| Premature rupture of membranes | |
| Yes | 74 (11.46) |
| No | 572 (88.54) |
| Premature birth | |
| Yes | 273 (42.26) |
| No | 373 (57.74) |
| Placental abruption | |
| Yes | 4 (0.62) |
| No | 642 (99.38) |
| Fetal distress | |
| Yes | 17 (2.63) |
| No | 629 (97.37) |
| Fetal growth restriction | |
| Yes | 80 (12.38) |
| No | 566 (87.62) |
| Neonatal asphyxia | |
| Yes | 17 (2.63) |
| No | 629 (97.37) |
| Postpartum hemorrhage | |
| Yes | 51 (7.89) |
| No | 595 (92.11) |
| Pregnancy-induced hypertension | |
| Yes | 131 (20.28) |
| No | 515 (79.72) |
| Gestational diabetes mellitus | |
| Yes | 203 (31.42) |
| No | 443 (68.58) |
3.3. Correlations Between Assisted Reproduction and Pregnancy Complications and Pregnancy Outcomes
As shown in Table 2, assisted reproduction correlated with fetal distress in the third trimester of pregnancy and neonatal asphyxia (p < 0.05).
| Item | Assisted reproduction | χ2 | p | |
|---|---|---|---|---|
| Yes (n = 346) | No (n = 300) | |||
| Premature rupture of membranes | 0.81 | 0.37 | ||
| Yes (n = 74) | 36 (10.40) | 38 (12.67) | ||
| No (n = 572) | 310 (89.60) | 262 (87.33) | ||
| Premature birth | ||||
| Yes (n = 273) | 139 (40.17) | 134 (44.67) | 1.33 | 0.25 |
| No (n = 373) | 207 (59.83) | 166 (55.33) | ||
| Placental abruption | 0.02 | 0.89 | ||
| Yes (n = 4) | 2 (0.58) | 2 (0.67) | ||
| No (n = 642) | 344 (99.42) | 298 (99.33) | ||
| Fetal distress | 6.33 | 0.01 | ||
| Yes (n = 17) | 4 (1.16) | 13 (4.33) | ||
| No (n = 629) | 342 (98.84) | 287 (95.67) | ||
| Fetal growth restriction | 1.35 | 0.25 | ||
| Yes (n = 80) | 38 (10.98) | 42 (14.00) | ||
| No (n = 566) | 308 (89.02) | 258 (86.00) | ||
| Neonatal asphyxia | 12.26 | < 0.00 | ||
| Yes (n = 17) | 2 (0.58) | 15 (5.00) | ||
| No (n = 629) | 344 (99.42) | 285 (95.00) | ||
| Pregnancy-induced hypertension | 0.90 | 0.34 | ||
| Yes (n = 131) | 75 (21.68) | 56 (18.67) | ||
| No (n = 515) | 271 (78.32) | 244 (81.33) | ||
| Gestational diabetes mellitus | 0.80 | 0.37 | ||
| Yes (n = 203) | 114 (32.95) | 89 (29.67) | ||
| No (n = 443) | 232 (67.05) | 211 (70.33) | ||
| Postpartum hemorrhage | 0.04 | 0.84 | ||
| Yes (n = 51) | 28 (8.09) | 23 (7.67) | ||
| No (n = 595) | 318 (91.91) | 277 (92.33) | ||
3.4. Correlations Between Type of Twins, Pregnancy Complications, and Pregnancy Outcomes
As shown in Table 3, the type of twins correlated with fetal distress in the third trimester of pregnancy and neonatal asphyxia (p < 0.05).
| Item | Type of twins | χ2 | p | |||
|---|---|---|---|---|---|---|
| Dichorionic diamniotic (DCDA) (of different genders) (n = 215) | Dichorionic diamniotic (DCDA) (of the same gender) (n = 281) | Monochorionic diamniotic (MCDA) (n = 144) | Monochorionic monoamniotic (MCMA) (n = 6) | |||
| Premature rupture of membranes | 2.14 | 0.55 | ||||
| Yes (n = 74) | 23 (10.70) | 37 (13.17) | 14 (9.72) | 0 (0) | ||
| No (n = 572) | 192 (89.30) | 244 (86.83) | 130 (90.28) | 6 (100.00) | ||
| Premature birth | 7.66 | 0.05 | ||||
| Yes (n = 273) | 80 (37.21) | 128 (45.55) | 60 (41.67) | 5 (83.33) | ||
| No (n = 373) | 135 (62.79) | 153 (54.45) | 84 (58.33) | 1 (16.67) | ||
| Placental abruption | 3.36 | 0.34 | ||||
| Yes (n = 4) | 3 (1.40) | 1 (0.36) | 0 (0) | 0 (0) | ||
| No (n = 642) | 212 (98.60) | 280 (99.64) | 144 (100.00) | 6 (100.00) | ||
| Fetal distress | 26.94 | < 0.00 | ||||
| Yes (n = 17) | 3 (1.40) | 5 (1.78) | 7 (4.86) | 2 (33.33) | ||
| No (n = 629) | 212 (98.60) | 276 (98.22) | 137 (95.14) | 4 (66.67) | ||
| Fetal growth restriction | 3.48 | 0.32 | ||||
| Yes (n = 80) | 22 (10.23) | 35 (12.46) | 23 (15.97) | 0 (0) | ||
| No (n = 566) | 193 (89.77) | 246 (87.54) | 121 (84.03) | 6 (100.00) | ||
| Neonatal asphyxia | 9.86 | 0.02 | ||||
| Yes (n = 17) | 2 (0.93) | 7 (2.49) | 7 (4.86) | 1 (16.67) | ||
| No (n = 629) | 213 (99.07) | 274 (97.51) | 137 (95.14) | 5 (83.33) | ||
| Pregnancy-induced hypertension | 3.83 | 0.28 | ||||
| Yes (n = 131) | 47 (21.86) | 62 (22.06) | 21 (14.58) | 1 (16.67) | ||
| No (n = 515) | 168 (78.14) | 219 (77.94) | 123 (85.42) | 5 (83.33) | ||
| Gestational diabetes mellitus | 3.41 | 0.33 | ||||
| Yes (n = 203) | 61 (28.37) | 99 (35.23) | 41 (28.47) | 2 (33.33) | ||
| No (n = 443) | 154 (71.63) | 182 (64.77) | 103 (71.53) | 4 (66.67) | ||
| Postpartum hemorrhage | 0.84 | 0.84 | ||||
| Yes (n = 51) | 17 (7.91) | 21 (7.47) | 13 (9.03) | 0 (0) | ||
| No (n = 595) | 198 (92.09) | 260 (92.53) | 131 (90.97) | 6 (100.00) | ||
3.5. Correlations Between Maternal Age, Pregnancy Complications, and Pregnancy Outcomes
As shown in Table 4, maternal age correlated with gestational diabetes mellitus in the third trimester of pregnancy and postpartum hemorrhage (p < 0.05).
| Item | Age | Z | p |
|---|---|---|---|
| Premature rupture of membranes | −0.457 | 0.648 | |
| Yes (n = 74) | 30.00 (4.25) | ||
| No (n = 572) | 31.00 (4.00) | ||
| Premature birth | −0.097 | 0.923 | |
| Yes (n = 273) | 31.00 (5.00) | ||
| No (n = 373) | 31.00 (4.00) | ||
| Placental abruption | −0.553 | 0.580 | |
| Yes (n = 4) | 29.50 (5.50) | ||
| No (n = 642) | 31.00 (4.00) | ||
| Fetal distress | −0.397 | 0.691 | |
| Yes (n = 17) | 31.00 (6.50) | ||
| No (n = 629) | 31.00 (4.00) | ||
| Fetal growth restriction | −1.451 | 0.147 | |
| Yes (n = 80) | 30.00 (5.00) | ||
| No (n = 566) | 31.00 (4.00) | ||
| Neonatal asphyxia | −1.154 | 0.248 | |
| Yes (n = 17) | 30.00 (7.00) | ||
| No (n = 629) | 31.00 (4.00) | ||
| Pregnancy-induced hypertension | −0.067 | 0.946 | |
| Yes (n = 131) | 31.00 (4.00) | ||
| No (n = 515) | 31.00 (5.00) | ||
| Gestational diabetes mellitus | −3.001 | 0.003 | |
| Yes (n = 203) | 31.00 (5.00) | ||
| No (n = 443) | 30.00 (5.00) | ||
| Postpartum hemorrhage | −3.004 | 0.003 | |
| Yes (n = 51) | 32.00 (5.00) | ||
| No (n = 595) | 31.00 (5.00) | ||
3.6. Thyroid Hormones of Patients With Different Pregnancy Complications and Pregnancy Outcomes
As shown in Table 5, FT4 levels varied in pregnant women with placental abruption (p < 0.05), TSH levels varied in pregnant women with pregnancy-induced hypertension (p < 0.05), and TPO-Ab levels varied in pregnant women with premature birth, placental abruption, and postpartum hemorrhage (p < 0.05).
| Item | FT4 (ng/L) | Z, p | TSH (mIU/L) | Z, p | TPO-Ab (%) | Z, p | |
|---|---|---|---|---|---|---|---|
| Premature rupture of membranes | Yes (n = 74) | 1.00 (0.14) | −1.78, 0.07 | 2.57 (2.49) | −0.18, 0.86 | 9.32 (5.70) | −0.16, 0.87 |
| No (n = 572) | 0.96 (0.20) | 2.60 (1.73) | 9.01 (5.02) | ||||
| Premature birth | Yes (n = 273) | 0.97 (0.17) | −0.43, 0.67 | 2.53 (2.14) | −0.06, 0.95 | 8.37 (4.67) | −3.39, 0.00 |
| No (n = 373) | 0.95 (0.21) | 2.62 (1.58) | 9.59 (5.35) | ||||
| Placental abruption | Yes (n = 4) | 1.18 (0.18) | −2.55, 0.01 | 2.65 (3.19) | −0.12, 0.91 | 21.73 (245.93) | −2.15, 0.03 |
| No (n = 642) | 0.96 (0.20) | 2.60 (1.80) | 9.03 (5.10) | ||||
| Fetal distress | Yes (n = 17) | 0.99 (0.24) | −0.06, 0.95 | 2.89 (3.29) | −1.17, 0.24 | 8.71 (11.48) | −0.72, 0.47 |
| No (n = 629) | 0.96 (0.20) | 2.57 (1.80) | 9.05 (5.14) | ||||
| Fetal growth restriction | Yes (n = 80) | 0.93 (0.20) | −1.12, 0.24 | 2.83 (1.79) | −1.25, 0.21 | 9.33 (4.65) | −0.88, 0.38 |
| No (n = 566) | 0.96 (0.20) | 2.57 (1.83) | 8.96 (5.18) | ||||
| Neonatal asphyxia | Yes (n = 17) | 1.01 (0.25) | −1.73, 0.08 | 2.53 (2.62) | −0.34, 0.73 | 8.82 (5.42) | −0.34, 0.74 |
| No (n = 629) | 0.96 (0.20) | 2.60 (1.79) | 9.05 (5.15) | ||||
| Gestational diabetes mellitus | Yes (n = 203) | 0.97 (0.19) | −1.52, 0.13 | 2.60 (1.93) | −0.63, 0.53 | 8.92 (5.68) | −0.83, 0.41 |
| No (n = 443) | 0.95 (0.19) | 2.58 (1.80) | 9.14 (5.00) | ||||
| Pregnancy-induced hypertension | Yes (n = 131) | 0.95 (0.24) | −0.88, 0.38 | 3.10 (2.44) | −4.23, < 0.00 | 8.85 (4.53) | −0.05, 0.96 |
| No (n = 515) | 0.96 (0.19) | 2.48 (1.68) | 9.14 (5.19) | ||||
| Postpartum hemorrhage | Yes (n = 51) | 0.93 (0.26) | −1.01, 0.31 | 3.01 (2.30) | −1.19, 0.23 | 8.35 (3.17) | −2.66, 0.00 |
3.7. Correlations Between Thyroid Hormone Levels, Pregnancy Complications, and Pregnancy Outcomes
As shown in Table 6, the Spearman’s correlation revealed the following: FT4 positively correlated with pregnancy with placental abruption (r = 0.100, p = 0.011); TSH positively correlated with pregnancy-induced hypertension (r = 1.167, p < 0.001); TPO-Ab negatively correlated with premature birth and postpartum hemorrhage (r = −0.134, 0.084, −0.105, p = 0.001, 0.032, 0.008) and positively correlated with placental abruption (r = 0.084, p = 0.032).
| Item | Placental abruption | Premature birth | Pregnancy-induced hypertension | Postpartum hemorrhage | ||||
|---|---|---|---|---|---|---|---|---|
| r | p | r | p | r | p | r | p | |
| FT4 | 0.10 | 0.01 | — | — | — | — | — | — |
| TSH | — | — | — | — | 0.17 | < 0.00 | — | — |
| TPO-Ab | 0.08 | 0.03 | −0.13 | 0.00 | — | — | −0.11 | 0.01 |
3.8. Regression Analysis of Multiple Factors Affecting Pregnancy Complications and Pregnancy Outcomes
The occurrence of placental abruption was the dependent variable (occurred = 1, not occurred = 0), and assisted production, type of twins (DCDA (of different genders) = 3, DCDA (of the same gender) = 2, MCDA = 1, MCMA = 0), age, FT4, TSH, and TPO-Ab were the independent variables of the regression model. The results revealed that TPO-Ab levels positively correlated with the occurrence of placental abruption during pregnancy (β = 0.014, SE = 0.006, Wald χ2 = 5.979, OR = 1.015, 95% CI: 1.003–1.026, p < 0.05).
The occurrence of premature birth was the dependent variable (occurred = 1, not occurred = 0), and assisted production, type of twins (DCDA (of different genders) = 3, DCDA (of the same gender) = 2, MCDA = 1, MCMA = 0), pregnancy-induced hypertension (yes = 1, no = 0), gestational diabetes mellitus (yes = 1, no = 0), age, FT4, TSH, and TPO-Ab were the independent variables of the regression model. The results are shown in Table 7: gestational diabetes mellitus positively correlated with premature birth (p < 0.05), and TPO-Ab negatively correlated with premature birth (p < 0.05).
| Item | β | SE | Wald χ2 | OR | 95% CI | p |
|---|---|---|---|---|---|---|
| Gestational diabetes mellitus | 0.454 | 0.176 | 6.617 | 1.574 | 1.114–2.225 | 0.010 |
| TPO-Ab | −0.014 | 0.005 | 6.638 | 0.986 | 0.976–0.997 | 0.010 |
The occurrence of pregnancy-induced hypertension was the dependent variable (occurred = 1, not occurred = 0), and assisted production, type of twins (DCDA (of different genders) = 3, DCDA (of the same gender) = 2, MCDA = 1, MCMA = 0), gestational diabetes mellitus (yes = 1, no = 0), age, FT4, TSH, and TPO-Ab were the independent variables of the regression model. The results revealed that TSH positively correlated with the occurrence of pregnancy-induced hypertension (β = 0.240, SE = 0.052, Wald χ2 = 21.175, OR = 1.271, 95% CI: 1.148–1.408, p < 0.05).
The occurrence of postpartum hemorrhage was the dependent variable (occurred = 1, not occurred = 0), and assisted production, type of twins (DCDA (of different genders) = 3, DCDA (of the same gender) = 2, MCDA = 1, MCMA = 0), pregnancy-induced hypertension (yes = 1, no = 0), gestational diabetes mellitus (yes = 1, no = 0), age, FT4, TSH, and TPO-Ab were the independent variables of the regression model. The results revealed that age positively correlated with the occurrence of postpartum hemorrhage (β = 0.097, SE = 0.036, Wald χ2 = 7.178, OR = 1.102, 95% CI: 1.026–1.183, p = 0.007).
4. Discussion
Pregnancy complications and adverse pregnancy outcomes result from a group of complicated clinicopathologic correlations. The influencing factors are complicated, including age, assisted reproduction, number of fetuses, and abnormalities in blood lipid and hormone levels, all of which increase the risks for high-risk pregnant women. In particular, women with twin pregnancies face higher risks [8, 9]. A survey on twin pregnancy outcomes found that age, conception method, and the type of chorionic membrane were risky factors resulting in adverse pregnancy outcomes. In particular, advanced maternal age, conception via assisted reproductive technology, and monochorionic twins caused adverse pregnancy results more easily. Clinicians should pay high attention to such pregnant women [10]. The data in this paper showed high percentages of twin pregnancies with gestational diabetes mellitus, pregnancy-induced hypertension, and premature labor, especially premature labor, which was as high as 42.26% of cases. Assisted reproduction and type of twins correlated with the occurrence of fetal distress in the third trimester of pregnancy and neonatal asphyxia, and that age correlated with the occurrence of gestational diabetes mellitus in the third trimester of pregnancy and postpartum hemorrhage. An epidemiological study found that premature birth occurred in 65.1% of multiple pregnancies, with a significantly higher rate of premature birth in monochorionic twins and a higher risk of neonatal asphyxia [11].
Some clinical studies have explored the correlations between changes in thyroid hormone levels and pregnancy complications and outcomes. The finding of this study also found that FT4 levels varied in pregnant women with placental abruption, TSH levels varied in pregnant women with pregnancy-induced hypertension, TPO-Ab levels varied in pregnant women with premature birth, placental abruption, and postpartum hemorrhage, and there were some correlations among these factors. However, the low incidence of placental abruption (0.62%, n = 4) in this cohort raises caution in interpreting the statistical significance of its correlation with FT4 and TPO-Ab. While the observed associations (e.g., elevated FT4 and TPO-Ab in placental abruption cases) may reflect a biological trend, the small sample size limits generalizability and increases the risk of Type I error. Future studies with larger populations are needed to validate whether these hormonal markers are true predictors of placental abruption or artifacts of limited event rates. Clinically, these findings should be viewed as preliminary, warranting cautious application in risk stratification. In a study conducted by Long et al. [12], it was also found that serum TSH, FT4, and TPO-Ab levels correlated with placental abruption. As pointed out in the study, placental abruption caused by TSH and FT4 abnormalities might correlate with abnormalities of lipid metabolism and vascular functions caused by hypothyroidism. TPO-Ab is a manifestation of thyroid autoimmune abnormalities, which can lead to abnormalities in the local immunoregulation of the placenta-decidua interface, thereby affecting placental function. Additionally, hyperthyroidism caused by these abnormalities might enhance myocardial contractility, raise lateral pressure on the blood vessel wall, increase cardiac output, and prompt the increase of blood pressure. Barišić et al. [13] also found a high level of TSH in pregnant women with preeclampsia, with an average of (2.5 ± 1.54 mIU/L), which was significantly higher than that of normal pregnant women, whereas TSH levels in women with gestational diabetes mellitus were not significantly different from those of normal pregnant women. This is similar to the result in this study, where TSH level correlated with gestational diabetes mellitus. It is thus clear that abnormal thyroid hormone levels not only increase the risk of pregnancy complications but also the risk of adverse pregnancy outcomes.
As pointed out in the study conducted by Dhillon-Smith and Coomarasamy [14], TPO-Ab levels correlated with adverse maternal and fetal outcomes, especially miscarriage and premature birth. The data in this paper also revealed that it significantly correlated with premature birth, and the study noted that this affected maternal and fetal outcomes in the same way even in the absence of thyroid dysfunction, as TPO-Ab titers were also elevated in approximately 10% of women with normal biochemical functions of the thyroid. In addition, TOP-Ab titers change during the entire pregnancy. The titers of antithyroid antibodies decrease during pregnancy due to immunosuppression and rebound in the postpartum period. These changes lead to an increased risk of relevant maternal and fetal complications [15, 16]. Moreover, TPO-Ab titers might have some impacts on pregnancy outcomes. Previous studies have shown that damage to the thyroid caused by TPO-Ab could lead to hypothyroidism by preventing the thyroid from producing enough hormones. Compared to TPO-Ab-negative pregnant women, the risk of premature birth in those with mere hypothyroidism was 46% higher, and that in TOP-Ab-positive pregnant women could increase to 78% [17, 18]. Another study found that TPO-Ab levels were elevated during pregnancy compared to nonpregnant women but showed a certain decreasing trend with the increase of gestational weeks. In particular, the TPO-Ab positivity rate in the third trimester of pregnancy might be lower than that in the first trimester, likely due to the development of maternal immunosuppression [19]. It is thus clear that abnormal changes in TSH, FT4, and TPO-Ab levels correlated with adverse events such as placental abruption, premature birth, pregnancy-induced hypertension, and postpartum hemorrhage. However, the data were collected in the third trimester of pregnancy, which was close to delivery. Further studies should be conducted on whether dynamic testing of TPO-Ab changes in the entire pregnancy is necessary to evaluate and analyze their impact on pregnancy outcomes.
The physiological uniqueness of twin pregnancies may amplify the mechanisms linking thyroid hormone abnormalities to complications. Twin pregnancies impose a doubled metabolic demand, requiring higher maternal thyroid output to sustain fetal growth. Elevated FT4 levels in placental abruption cases might reflect a compensatory hyperthyroid state attempting to meet this demand, but excessive thyroid activity could destabilize placental vascular function. Similarly, the increased placental mass in twins enhances hCG production, which suppresses TSH and creates a state of transient hyperthyroxinemia. This imbalance may impair placental adaptation to oxidative stress, exacerbating risks of hypertension or preterm birth. Furthermore, the immunological strain of supporting two fetuses may elevate TPO-Ab titers, triggering autoimmune-mediated placental injury—a mechanism less prominent in singleton pregnancies.
While this study excluded participants with thyroid medication use, clinical guidelines emphasize that pharmacological interventions (e.g., levothyroxine for hypothyroidism or antithyroid drugs for hyperthyroidism) are critical for managing thyroid dysfunction in pregnancy. Such treatments may mitigate risks of complications like placental abruption or preterm birth by restoring thyroid homeostasis [12, 14]. However, the exclusion of medicated cases in our cohort limits direct analysis of drug effects. Future studies should include treated populations to evaluate whether thyroid hormone normalization reduces adverse outcomes in twin pregnancies, particularly those linked to FT4 variability (e.g., placental abruption) or TPO-Ab positivity (e.g., preterm birth).
In conclusion, TSH, FT4, and TPO-Ab levels in women with twin pregnancies in the third trimester may correlate with adverse events such as placental abruption, pregnancy-induced hypertension, and premature birth. These findings can help clinicians to evaluate relevant complications and adverse pregnancy outcomes in the third trimester.
Conflicts of Interest
The authors declare no conflicts of interest.
Funding
This work was supported by Wu Jieping Medical Foundation Special Fund for Clinical Research (Grant Number: 320.6750.2023-06-6).
Open Research
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
