Introduction

Preeclampsia is a pregnancy‑related multisystem condition affecting about 4.6% of pregnancies globally, with a wide variation across different regions. The prevalence rises significantly in women with type 1 diabetes mellitus (T1DM), reaching 9%–20%.^1,2 Although the exact cause of preeclampsia remains unclear, the condition is characterized by placental dysfunction associated with abnormal trophoblast invasion and impaired spiral artery remodeling. The resulting placental hypoxia stimulates the release of antiangiogenic factors, particularly soluble fms‑like tyrosine kinase‑1 and soluble endoglin, as well as proinflammatory cytokines. These circulating factors disrupt maternal endothelial function, promote oxidative stress, and trigger a widespread systemic inflammatory response.³

The bidirectional relationship between changes in thyroid function and preeclampsia implies that thyroid disorders contribute to the development of preeclampsia. Specifically, hypothyroidism might be directly implicated in the pathophysiology of preeclampsia. Hypothyroidism can significantly impact the contraction of smooth muscles in both renal and systemic arteries, resulting in increased peripheral vascular resistance, elevated diastolic blood pressure (BP), and decreased tissue perfusion.⁴ Consequently, appropriate identification and management of thyroid abnormalities can have a significant impact on the incidence of preeclampsia. It has been suggested that assessing thyroid function can help predict the occurrence of preeclampsia, with thyroid‑stimulating hormone (TSH) playing a pivotal role in the screening and diagnosis of various thyroid disorders. On the other hand, it is essential to note that preeclampsia can potentially play a role in the pathogenesis of hypothyroidism.⁵

T1DM is closely associated with an increased prevalence of thyroid dysfunction, particularly autoimmune thyroid disease (AITD), which includes both hypo- and hyperthyroidism. AITD is observed in approximately 17% to 30% of individuals diagnosed with T1DM.⁶ Moreover, the presence of thyroid peroxidase antibodies (TPOAbs) in T1DM patients further elevates the likelihood of developing hypothyroidism. TPOAb positivity is associated with endothelial dysfunction, characterized by impaired endothelium‑dependent arterial dilation. This dysfunction is observed in patients with Hashimoto thyroiditis, even in the euthyroid state, suggesting that autoimmune activity may contribute to vascular abnormalities.^7,8 In pregnancy, elevated TPOAb levels, particularly in the first trimester, are associated with an increased risk of hypertensive disorders, such as preeclampsia and eclampsia, as well as higher systolic and diastolic BP in the later stages of pregnancy. The presence of TPOAbs may reflect immune dysregulation, potentially disrupting vascular function and contributing to the development of hypertension during pregnancy.⁹

Our study aimed to assess the association between maternal thyroid function and the risk of preeclampsia in women with T1DM.

Patients and methods

This prospective study was conducted at the Gynecologic and Obstetric University Hospital in Poznań, Poland, and involved pregnant women with T1DM who were admitted between June 2012 and March 2015. All participants were hospitalized as part of routine care for pregnant women with T1DM in a tertiary referral center. Hospitalizations occurred for scheduled pregnancy monitoring and for delivery. Data were collected during 3 planned inpatient visits: in the first trimester (<⁠12 weeks of gestation), midpregnancy (20–24 weeks), and late pregnancy (34–39 weeks). For women diagnosed with hypertensive disorders, the therapeutic target was BP below 135/85 mm Hg.¹⁰

The participants were divided into 3 groups based on maternal hypertensive status during pregnancy: normotensive, gestational hypertension, and preeclampsia. Blood samples were collected after overnight fasting and analyzed in the hospital’s central laboratory. Glycated hemoglobin (HbA_1c) levels were measured using a turbidimetric inhibition immunoassay (Tina‑quant Hemoglobin A_1c II test on a Cobas c311 analyzer, Roche Diagnostics, Basel, Switzerland), with a standard range of 4.8%–6% (29–42 mmol/mol) for nonpregnant individuals. Serum TSH concentrations were measured using a chemiluminescent immunoassay. TSH assessments were performed during routine study visits, with first‑trimester values used as the primary analytical time point in accordance with clinical guidelines.¹¹ Anthropometric measurements (height, weight, and waist / hip circumference) and BP readings were taken at the study onset. BP was measured 4 times daily during the first hospitalization, with the mean systolic and diastolic values recorded. Insulin resistance (IR) was assessed using the estimated glucose disposal rate (eGDR; mg/kg/min), calculated as follows: eGDR = 24.31 – (12.22 × WHR) – (3.29 × HTN) – (0.57 × HbA_1c [%]), where WHR is the waist‑to‑hip ratio and HTN is hypertensive status (0 = no; 1 = yes). A lower eGDR indicates higher IR. Preeclampsia and gestational hypertension were diagnosed based on the American College of Obstetricians and Gynecologists criteria.¹² The diagnosis and treatment of hypothyroidism in pregnancy were based on the national guidelines.¹¹ In line with these guidelines, TSH levels should be kept below 2.5 mIU/l throughout the preconception period and during pregnancy. De novo hypothyroidism was defined as hypothyroidism diagnosed for the first time during pregnancy in women without a documented history of thyroid disease prior to conception, in line with previously published clinical definitions.¹¹

The study protocol was approved by the Institutional Ethical Committee of Poznan University of Medical Sciences (673/12). Written informed consent was obtained from each participant before inclusion in the study.¹³

Results

The study included 197 pregnant women aged 17 to 41 years (median [IQR], 30 [26–33] y). All patients had been diagnosed with T1DM, with a median disease duration of 10.5 (4–16) years. Diabetic complications were reported in 27.4% of the patients (n = 54). According to the White classification, most patients presented with class B or C (41.1% and 25.3%, respectively), followed by class D, R, or RF (17.9%, 6.8%, and 4.7%, respectively). Of the enrolled patients, 8.4% demonstrated preeclampsia (n = 16), while gestational hypertension was diagnosed in 17.7% (n = 35). A total of 5 patients were diagnosed with chronic hypertension, 3 of whom subsequently developed preeclampsia. Due to a limited sample size, the 2 patients with chronic hypertension who did not develop preeclampsia were excluded from further analysis. Detailed characteristics of the study group are presented in Supplementary material, Table S1.

Hypothyroidism was diagnosed in 71 patients (including 18 diagnosed de novo), and 46 women had elevated TSH levels (≥2.5 mIU/l) during the first trimester of pregnancy.

A similar prevalence of preeclampsia was observed in the patients with and without hypothyroidism (8.5% vs 8.4%, respectively; P = 0.99). Also, there were no differences in the incidence of preeclampsia between the women with levothyroxine (LT4)-treated hypothyroidism and those without hypothyroidism (11.3% and 8.4%, respectively; P = 0.75). None of the patients with de novo hypothyroidism developed preeclampsia, so it was apparently more frequently observed in the patients already treated with LT4, but the difference was not significant (0% vs 11.3%, respectively; P = 0.32). The rate of preeclampsia was insignificantly higher among the patients with than those without hypertension (14.3% vs 7.1%, respectively; P = 0.3).

As a complementary analysis supporting the primary study objective, TSH levels were additionally compared in the women with and without preeclampsia, showing similar values (median, 1.41 [1.05–1.76] vs 1.66 [1.06–2.54] mIU/l, respectively; P = 0.22).

A multidimensional correspondence analysis was performed to investigate the relationship between preeclampsia, thyroid function, and hypertension (Figure 1), showing an association between these 3 conditions during pregnancy (Supplementary material, Table S2). The women with elevated TSH levels (≥2.5 mIU/l) and hypothyroidism were more likely to develop hypertension and preeclampsia, forming a distinct clinical group on the 2‑dimensional plot. In contrast, the women without thyroid dysfunction, hypertension, or preeclampsia clustered separately, indicating a markedly different clinical profile. These findings suggest that impaired thyroid function may play a significant role in the development of hypertensive disorders of pregnancy.

Discussion

This study was primarily designed to assess thyroid dysfunction as a potential risk factor for hypertensive disorders of pregnancy, rather than to evaluate preeclampsia as a determinant of thyroid function.

In our study, less than 10% of pregnant women with diabetes had preeclampsia. Despite a comparable prevalence of preeclampsia in the patients with and without hypothyroidism, the presence of both hypothyroidism and a TSH level equal to or greater than 2.5 mIU/l was associated with a higher risk of developing preeclampsia.

T1DM is strongly associated with thyroid dysfunction, particularly autoimmune hypothyroidism, due to shared genetic and autoimmune features. The prevalence of thyroid autoantibodies in individuals with T1DM ranges from 7% to 35%, depending on the country and ethnic group, and affects both sexes. However, the presence of thyroid autoantibodies is significantly more frequent in women with T1DM, who are approximately 3 times more likely to develop autoimmune disorders than men.¹⁴ Pregnant women with T1DM and hypothyroidism, particularly those with elevated TPOAb levels, are at an increased risk of developing hypertensive disorders, including preeclampsia and eclampsia. Elevated TSH levels and TPOAb positivity, especially during the first trimester, are significantly associated with higher systolic and diastolic BP in the later stages of pregnancy.⁹ Similarly, a study by Yang et al¹⁵ indicated that TPOAb positivity in early pregnancy was linked to an increased risk of hypertensive disorders, including preeclampsia. The women with positive TPOAb in the first trimester had a significantly higher likelihood of developing these conditions, suggesting that thyroid autoimmunity may play a role in the development of pregnancy‑related hypertension.

A meta‑analysis by Hajifoghaha et al⁵ indicated that women with preeclampsia were at a higher risk of abnormal thyroid function test results. The authors reported that the preeclamptic women had significantly higher TSH levels than those without preeclampsia. However, several studies showed an opposite trend, with decreased TSH levels observed in preeclamptic patients.^16,17 Additionally, total thyroxine (T4), total triiodothyronine (T3), and free T3 levels were reported to be lower in preeclamptic women, but there was no significant difference in free T4 levels between the groups.¹⁷ Similarly, Berg et al¹⁸ reported a significant association between maternal hypothyroidism and an increased risk of preeclampsia. They found that the severity of hypothyroidism played a crucial role in determining the risk, with more severe cases of hypothyroidism being associated with a higher likelihood of developing preeclampsia. Interestingly, their findings did not support a direct link between thyroid autoimmunity and preeclampsia. In another meta‑analysis, Toloza et al¹⁹ found that, as compared with the women with normal thyroid function, those with subclinical hypothyroidism were at a higher risk of developing preeclampsia. However, there was no significant association between subclinical hyperthyroidism, isolated hypothyroxinemia, or TPOAb positivity with gestational hypertension or preeclampsia. Both high and low levels of TSH were associated with an increased risk of preeclampsia, indicating a U‑shaped relationship.

Torres‑Torres et al²⁰ observed no significant differences in the prevalence of hypothyroidism between the control and preeclampsia groups. However, it is worth noting that in the majority of patients the condition was effectively managed with LT4 therapy. Additionally, previous studies reported a reduction in mean arterial pressure with the use of LT4 therapy in individuals with hypothyroidism, which supports the findings of the current study.²¹ A meta‑analysis by Goodarzi‑Khoigani et al²² indicated that after LT4 therapy, the incidence rates of gestational hypertension and preeclampsia in all forms of hypothyroidism were not significantly different from those observed in the control group. Similarly, in our study, the incidence of preeclampsia did not differ significantly between the patients with hypothyroidism treated with LT4 and those without hypothyroidism.