Background
Hypothyroidism during pregnancy, even when subtle, is considered to increase the risk of miscarriage, gestational hypertension, placental abruption, preeclampsia, and postpartum hemorrhage, as well as psychoneurological disorders in the child [
1‐
4]. Therefore, levothyroxine (LT4) replacement is recommended to maintain the serum concentration of thyroid stimulating hormone (TSH) at < 2.5 μIU/ml (1st trimester) and later at < 3 μIU/ml (2nd, 3rd trimester) in women with hypothyroidism during pregnancy [
5]. The clinical practice guidelines of the Endocrine Society recommend starting LT4 dosage at 50 μg/day or more in pregnant women with hypothyroidism if the serum TSH level is in a range of 2.5–10 μIU/ml [
6]. However, not all pregnant women with hypothyroidism are required to increase LT4 dosage to maintain the serum level of TSH [
4,
7,
8], and wide inter-individual differences in regard to increased LT4 dosage have been reported [
5]. Although serum TSH level prior to pregnancy is known to be a predictor of the requirement for increasing LT4 dosage [
9], there are no other known predictors of that in women with primary hypothyroidism.
Importantly, pregnant women with hypothyroidism acquired due to radioablation or surgery require a greater increase in LT4 dosage as compared to those with primary hypothyroidism [
10], indicating that thyroid function reserve is closely associated with an increase in LT4 dosage during pregnancy. However, the clinical significance of inferior thyroid artery peak systolic velocity (ITA-PSV) has not been shown in patients with euthyroid Hashimoto’s disease, though we previously reported that ITA-PSV reflects thyroid gland capacity to produce thyroid hormone in not only untreated Graves’ disease [
11] but also euthyroid Graves’ disease [
12,
13]. Based on these findings, we postulated that ITA-PSV might reflect thyroid function reserve in patients with Hashimoto’s thyroiditis. Here, we examined the relationship of ITA-PSV with an increase in LT4 dosage in women with Hashimoto’s thyroiditis during pregnancy.
Discussion
The present findings demonstrated that ITA-PSV is significantly associated with an increase in LT4 dosage during pregnancy independent of serum TSH, while FT4, FT3, baseline LT4 administration, and thyroid volume showed no such association. Furthermore, the combination of ITA-PSV with TSH predicted an LT4 dosage increase ≥50 μg during pregnancy more precisely than TSH alone in pregnant women with Hashimoto’s thyroiditis. Together, our results suggest that ITA-PSV independent of TSH is useful to reflect thyroid function reserve to increase thyroid hormone production during pregnancy.
Previous investigations have shown that 50 to 85% of women with hypothyroidism require an increase in LT4 dosage during pregnancy [
4,
7,
8], though there are wide inter-individual differences in regard to such an increase given to compensate for the increased demand for thyroid hormone production while pregnant [
5]. Consistent with those reports, 81.8% of the present pregnant patients required an LT4 dosage increase for maintaining serum TSH level within the target range, with that increase level varying from 12.5 to 100 μg in the present study. Abalovich M et al. reported that a higher TSH level prior to pregnancy is predictive of a requirement for increased LT4 dose during pregnancy in women with primary hypothyroidism, in whom serum TSH prior to pregnancy was < 2.5 μIU/ml [
9], suggesting that serum TSH level might reflect the current status of thyroid hormone production even in women with euthyroid Hashimoto’s thyroiditis. In agreement with that report, the present study found that a high serum level of TSH was independently associated with an increase in LT4 dose and also had a tendency to be associated with an increase in dosage ≥50 μg. However, pregnant women with hypothyroidism due to undergoing radioablation therapy or a total thyroidectomy were previously reported to require a greater increase in LT4 dosage during pregnancy as compared to those with primary hypothyroidism, even when baseline TSH normalized with LT4 replacement therapy did not differ significantly prior to pregnancy between those two patient groups [
10]. Those findings reflected a lack of residual thyroid function to increase hormone production in response to an increased hormone demand during pregnancy in the former group of patients in that study.
Although thyroid blood flow was shown to be increased in patients with clinically overt hypothyroidism as compared to healthy individuals [
17], that level was reported to be lower as compared to patients with a TSH-producing tumor [
18], suggesting that a rise in ITA-PSV in response to an increase in TSH is weak in patients with Hashimoto’s thyroiditis due to destruction of the thyroid gland. Importantly, we previously showed that ITA-PSV predicts Graves’ disease relapse after anti-thyroid drug (ATD) withdrawal [
12] as well as after normal delivery in patients with euthyroid Graves’ disease [
13], while a maintenance dosage of ATD is required to keep serum TSH level within a normal range in those patients who are otherwise untreated [
11]. Interestingly, in patients with euthyroid Graves’ disease, ITA-PSV was shown to be correlated with serum VEGF level, known to be a marker of intra-thyroid angiogenesis, whereas neither serum F-T4 nor F-T3 was found to have a relationship [
12], suggesting that ITA-PSV reflects thyroid hormone production in euthyroid Graves’ disease. In contrast, we previously reported that ITA-PSV was not correlated with the level of F-T4 or F-T3 in healthy subjects [
11]. In addition, though ITA-PSV was not significantly correlated with serum F-T4 (ρ = 0.183,
p = 0.414), F-T3 (ρ = 0.091,
p = 0.687), or TSH (ρ = − 0.262,
p = 0.239) in the present study, it was independently associated with an increase in LT4 dosage, suggesting that ITA-PSV reflects residual thyroid capability to increase thyroid hormone production, but not current thyroid hormone production capability, in patients with euthyroid or subclinical Hashimoto’s thyroiditis.
This study has some important limitations. First, it was retrospective in design. Second, the patient cohort was nearly exclusively Japanese individuals, thus it is unclear whether these findings can be generalized for other ethnic groups. Third, the number of subjects investigated was low. Therefore, the power of the present findings is low, and we were not able to fully estimate the association of gestational age with evaluation or sensitivity of ITA-PSV. Our findings should be confirmed in a large-scale study. Nevertheless, to the best of our knowledge, no previously reported study has performed such an examination of predictors of increased LT4 outside of TSH level.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.