The Relationship of Preconception Thyrotropin Levels to Requirements for Increasing the Levothyroxine Dose During Pregnancy in Women with Primary Hypothyroidism

Introduction

C urrent literature indicates that 50%–85% of women with hypothyroidism who are being treated with levothyroxine (LT4) require an increase in their dose of this hormone after they become pregnant (1 –3). The increment of LT4 dose is related to the etiology of their hypothyroidism. Women whose hypothyroidism is due to radioactive iodine ablation, total thyroidectomy, or congenital agenesis of the thyroid gland need an increase that is about 50% greater than women whose hypothyroidism is due to Hashimoto's thyroiditis (4,5). It is likely that euthyroid levels of thyrotropin (TSH) and free-T4 (FT4) should be rapidly achieved, as maternal hypothyroidism, even if subclinical, is associated with complications such as prematurity (6,7), miscarriage (2), gestational hypertension (8), placental abruption (6), and neurointellectual alterations in the offspring (9). Unfortunately, between 24% and 43% of women with hypothyroidism who are taking LT4 before pregnancy have elevated serum TSH concentrations at their first prenatal visit after becoming pregnant (10,11).

Recently published guidelines recommend a serum TSH of <2.5 mIU/L in women who plan to become pregnant (12). The objective of our study was to determine what percent of nonpregnant women whose serum TSH was within the range recommended by the guidelines (12) (i.e., <2.5 mIU/L) would require an increase in their LT4 dose after they became pregnant. We also sought to compare preconception (pre-C) serum TSH concentrations in women who did not require an LT4 dose increase during pregnancy with those who did require an increase in their LT4 dose. We were interested in whether there was an ideal pre-C range for serum TSH that was associated with lack of a need to increase their dose of LT4 during pregnancy but was also within the normal range.

Patients and Methods

We reviewed 460 medical records of pregnant women with hypothyroidism who were seen at our institution from November 2005 to December 2008. There were 53 patients between the ages of 29 and 43 who were taking LT4 before conception, had pre-C serum TSH levels <2.5 mIU/L within the 6 months before pregnancy, and did not have history of thyroid cancer. Patients with thyroid cancer were excluded, because most had suppressed serum TSH (i.e., <0.01 mIU/L). Patients were assigned to two groups based on whether there was (Group 1) or there was not (Group 2) a need to increase their dose of LT4 during pregnancy. The dose of LT4 was increased during pregnancy if the serum TSH was >2.5 mIU/L in the first trimester or >3 mIU/L during the rest of the pregnancy (12).

Serum TSH, FT4, and total T4 (TT4) were measured utilizing the Immulite 2000 chemiluminometric assay (Siemens Healthcare Diagnostics, Deerfield, IL). The samples of venous blood were drawn between 7 and 8 A.M. The normal range for nonpregnant women was TSH from 0.5 to 4.5 mIU/L, FT4 from 0.8 to 1.9 ng/dL, and TT4 from 4.5 to 12 mcg/dL.

Results are reported as mean ± standard deviation or median and range, when appropriate. Significant differences in distribution were calculated by Student t-test for unpaired variables, Mann–Whitney U-test for variables from non-Gaussian distributions, and Fisher's exact test. Statistical significance was considered to be at p < 0.05.

This study was approved by the Ethics Committee of the Durand Hospital, Buenos Aires.

Results

Of the 53 patients, 17 (32.1%) needed to increase their LT4 dose during pregnancy (Group 1) and 36 (67.9%) did not need to do so (Group 2). We did not note any clinical characteristics that could predict which patients would need higher LT4 doses during pregnancy and which would not. Table 1 shows the etiologies of hypothyroidism in these patients. Hashimoto's thyroiditis was the most common etiology diagnosed in more than half of the patients in each group. All in all, hypothyroidism after treatment with ¹³¹I and surgery accounted for slightly >11% in each group. Only three patients in each group had overt hypothyroidism (OH), the others had subclinical hypothyroidism (SCH). The diagnoses were made based on thyroid function tests performed when patients were first diagnosed and they were not taking LT4.

In four patients, the first prenatal visit was in the second trimester. In the rest of the patients, it was in the first trimester. The increase in the LT4 dose in the 17 Group 1 patients, given as a percentage, is shown in Table 2. The data are presented separately for the 14 patients with SCH and the three patients with OH. The degree to which the LT4 dose was increased in patients with SCH was significantly higher (33.97%) than in the three patients with OH (8.1%, p < 0.02). The median pre-C TSH was 1.67 (range = 0.54–2.3) mIU/L in patients with SCH and 1.0 (range = 0.5–1.96) mIU/L in patients with OH. Only four patients required an additional dose increase during the second trimester. The dose increment was 8.77%. The other patients reached adequate doses during the first trimester.

Table 3 presents information regarding serum TSH, FT4, and TT4 before conception in the two groups. The FT4 and TT4 levels were similar in Group 1 and Group 2. Serum TSH was significantly higher (p < 0.005) in Group 1 (1.55 ± 0.62 mIU/L) than in Group 2 (0.98 ± 0.67 mIU/L).

Table 4 shows the number of patients who had to increase the dose, or did not have to, classified by their pre-C serum TSH concentration being or not being higher than 1.2 mIU/L. We choose the cut off value of >1.2 mIU/L based on the 95% confidence intervals (CI) of the mean corresponding to Groups 1 and 2. The 95% CI for Group 1 was 1.22–1.87 mIU/L, whereas for Group 2 it was 95% CI 0.82–1.2 mIU/L. As a result, the 95% CI minimum value (1.22 mIU/L) for Group 1 was similar to the maximum value (1.29 mIU/L) for Group 2.

If the cut off value was a pre-C TSH value of <2.5 mIU/L, as recommended in the guidelines (12), 32.1% of the patients had to increase their LT4 dose during pregnancy. When their pre-C TSH was in the range of 1.2–2.4 mIU/L, 50% of the patients required an increase in their LT4 dose during pregnancy. In contrast, only 17.2% of patients with serum TSH value of <1.20 mIU/L had to increase their dose of LT4 during pregnancy.

Two patients had miscarriages, one in the 8th and one in the 11th week of gestation. Both of these patients were in Group 2 and had normal values for FT4 and TSH at the time of their miscarriages.

Discussion

The mother is the main source of T4 for the fetus during the first trimester and the predominant source during the first half of gestation, a time that the key structures in the fetal brain develop (13). In addition, maternal T4 requirements increase during pregnancy (14). It is, therefore, important that maternal deficits in T4 which may develop in women with hypothyroidism during pregnancy be corrected as rapidly as possible. Most pregnant patients (50%–85%) will have to increase the LT4 dose (1 –5,15,16), though a few studies found that only the minority (21%–44%) needed to do so (17 –20).

Some have found that, of those patients whose dose was increased, only 6% normalized the levels in the first trimester, and almost half (46.2%) achieved normal levels only during the third trimester (21) after the key period for the maternal T4 input had already passed. The etiology of thyroid dysfunction may have a pivotal role in the timing and magnitude of LT4 adjustments (22). To make this more complicated, a higher frequency of miscarriages (26%) has been described in patients who needed to increase the dose versus those who did not (6%) (23), though, in our experience, only two miscarriages occurred in patients who had no need to increase the dose. These miscarriages could not be attributed to hypothyroidism.

An LT4 dose increase of 30%, which is two additional tablets weekly above of the patient's baseline daily LT4 dose, as soon as pregnancy has been confirmed, has been proposed (3,24). However, this increase might be insufficient, as it has been observed by Verga et al. (21) that patients needed to increase their LT4 dose from 45% to 70%.

In the present study, 13/17 of the patients (76.5%) in the group requiring a dose increment achieved adequate hormone levels along the first trimester. This may be due to the fact that most patients are instructed to consult us as soon as pregnancy has been confirmed. Despite this, we had to perform additional dose increases in the remaining four patients (23.5%) during the second trimester.

As it was true for other studies (21), we also observed that within the group needing to increase their LT4 dose during pregnancy, the patients with SCH required a greater increase than those with OH. This may be due to the fact that the mean value of the pre-C serum TSH on LT4 in the patients with SCH was higher than in the patients with OH. However, the number of patients with OH was quite small. Therefore, initial TSH levels in the SCH and OH groups cannot be reliably compared, and the finding that patients with SCH required a greater increase in their LT4 dose than those with OH may not be a reliable observation.

Since it is difficult to achieve rapid normalization of thyroid status in women with hypothyroidism during pregnancy, it would be ideal to use a dose of LT4 before conception that would minimize the need for adjustment of the LT4 dose during pregnancy. The recommendation of the Endocrine Society's Guidelines (12) is that the LT4 dose in women with hypothyroidism be adjusted so as to achieve TSH levels lower than 2.5 mIU/L when the patient indicates she plans to become pregnant or may become pregnant. The two considerations relating to this strategy are to avoid hypothyroidism at the time of conception and to reduce the chances that an increase in the LT4 dose would be required during pregnancy. Some authors suggest that the LT4 dose be increased in women with hypothyroidism who are planning to become pregnant to achieve “partially suppressed TSH levels” (25) so as to anticipate the expected increase in thyroid hormone requirements during pregnancy.

Not surprisingly, we observed that pre-C TSH levels were significantly lower in patients who did not need to increase the LT4 dose during pregnancy than pre-C TSH levels in patients who needed to increase their dose of LT4 during pregnancy. We noted, however, that 32.1% of those patients whose pre-C TSH concentrations were in the range of 0.1–2.4 mIU/L, and therefore satisfactory according to the Endocrine Society's guidelines, required an increase in their LT4 dose during pregnancy. In contrast, only 17.2% of those patients whose pre-C serum TSH was in the range of 0.1–1.2 mIU/L required an increase in their dose of LT4 during pregnancy (p < 0.02). We suggest, therefore, that an ideal serum TSH in women with hypothyroidism who are likely to become pregnant is <1.2 mIU/L. Additional studies of a prospective nature that measure maternal and fetal outcomes are needed to test the validity of this concept.

In summary, it is important to achieve ideal serum TSH and FT4 concentrations in women with hypothyroidism during pregnancy, especially during the first trimester. To best achieve this goal, thyroid status on LT4 needs to be optimized even before the patient becomes pregnant. Therefore, most of them will not need to increase the LT4 dose during gestation, thus diminishing the risks of an inadequate treatment. The close follow-up and correction of hypothyroidism during pregnancy is essential for the gestation continuity and the fetus health.