Levothyroxine-replaced women attending our fertility clinic had lower serum free T3 levels than euthyroid women (anti-TPO Ab negative) at baseline, despite adequate replacement (TSH < 2.5 mIU/L) and comparable TSH levels. Ovarian hyperstimulation exaggerated differences in serum FT3 levels, culminating in lower follicular fluid FT3 levels in LT4-replaced women. Oocyte numbers and maturity were similar, but fertilisation rates were reduced in LT4-replaced patients. In general, women with lower follicular fluid FT3 levels were less likely to achieve a pregnancy.
Our prospective study again highlighted the acute, supraphysiological demand COH exerts on thyroid function. This was more apparent in LT4-replaced women, who demonstrated a wider spread of TSH levels throughout ovarian stimulation and beyond into pregnancy. On average, serum TSH increased by 1.4 mIU/L which is comparable to the 1.5 mIU/L reported in a recent meta-analysis (6). The latest ATA guidelines recommend TSH concentration below 2.5 mIU/L pre-pregnancy, in patients treated for clinical hypothyroidism (7). Universal screening of thyroid function post-COH has recently been recommended for LT4-replaced women (6), as a balanced thyroid axis is critical for embryo implantation and placentation, and TSH levels typically remain elevated for months in these women (4). Whether or not a preventive augmentation of LT4 dose before starting COH eliminates the TSH excursion and impacts favourably on pregnancy outcomes, is not known.
Autonomous regulation of thyroid hormone is likely compromised during controlled ovarian hyperstimulation in hypothyroid-treated women. LT4 monotherapy is the mainstay of treatment for hypothyroidism but is not without its critics. As with all medications, LT-4 metabolism, distribution, and absorption is subject to drug interactions and patient compliance. A significant minority of patients (10–15%) experience persistent symptoms related to hypothyroidism despite adequate treatment (8). Notably, LT4 therapy results in a “left shift” in serum T3 compared with euthyroid individuals with comparable TSH levels (9), creating a higher T4:T3 ratio. LT4-replaced women in our study had lower serum FT3 levels compared with matched euthyroid women despite all women recording a TSH < 2.5 mIU/L at baseline, which then persisted through COH. We not only demonstrated lower serum FT3 levels but also highlighted significantly lower FT3 levels within follicular fluid obtained from LT4-replaced women. A prospective study of 133 subjects following total thyroidectomy showed that mildly-suppressed levels of TSH (0.03–0.3mU/L) were required to stabilize FT3 levels to their preoperative levels (10). Whether or not, there is a role for liothyronine sodium (synthetic T3) in addition to LT4 replacement during COH, is not known but widely divergent plasma half-lives of available preparations of T4 (days) and T3 (hours) currently complicates their co-administration (11).
LT4-treated individuals, with normal serum TSH, have been shown to exhibit objective signs of mild reduction in systemic thyroid hormone (TH) signalling (8). To the best of our knowledge, we are first to demonstrate lower FT3 levels in the follicular fluid in LT4-replaced women versus euthyroid women undergoing COH. The role and dynamic control of TH within follicular fluid is not established (12). TH receptors, which bind T3, have been isolated in both granulosa and cumulus cells and the mature oocyte of the human ovarian follicle (3) suggesting a possible role for thyroid hormone in oocyte maturation. A direct link between female infertility and reduced TH receptor signalling was suggested in one small study of human granulosa cells (13). We found no significant difference in numbers of MII oocytes retrieved amongst LT4-replaced women, but fertilisation rates were lower, despite comparable male factor infertility and ICSI rates, as shown by others (2, 14, 15). Although 75% LT4-replaced women were positive for anti-TPO Ab both in the serum and follicular fluid, a causal relationship between thyroid autoimmunity and poorer fertilisation is yet to be determined (16). Hence, we speculate the reduced fertilisation rates are due to inadequate levels of serum and/or follicular FT3.
T3 receptors are also expressed in uterine and placental tissue, namely the extravillous cytotrophoblasts which play a key role in placentation. Thyroid hormones have been shown to play a potential paracrine and intracrine role at the uterine level through thyroid receptors present in these tissues and may influence the endometrial angiogenesis and immune function (17). One study reported no significant difference in the endometrial volume in LT4-replaced compared to euthyroid patients (18). Hypothyroidism has been linked with both decreased decidualisation and decreased trophoblast proliferation and invasion (19). We found no differences in early pregnancy loss between euthyroid and LT4-replaced women however, FF T3 levels were lower in those women who failed to achieve a clinical pregnancy when data was pooled across the entire cohort.
This study had numerous strengths including (i) the baseline TSH < 2.5 mIU/L inclusion criteria for all patients, including euthyroid women, to allow for more robust comparisons (ii) anti-TPO Ab positivity exclusion criteria for euthyroid women to reduce sub-group heterogeneity and risk of sub-clinical hypothyroidism, (iii) prospective, single centre design allowing for consistency within laboratory reporting and (iv) assessment of iodine status at baseline. The main limitation was the small patient numbers which did not allow for pregnancy outcome comparisons between euthyroid and LT4-replaced women.
LT-4 replaced women (TSH < 2.5 mIU/L) exhibit lower serum FT3 levels at baseline which persist through an ART cycle and is a predictor of lower follicular fluid FT3 levels. The inability to augment thyroid hormones production during COH, may have a detrimental impact on the oocyte +/- fertilisation rates in hypothyroid women. Until now, guidelines have centred around optimizing TSH levels in LT4-replaced women undergoing ART but optimizing serum (and follicular) FT3 levels may be pivotal to improving COH outcomes.