Isolated Maternal Hypothyroxinemia: Adverse Maternofetal Outcomes but Uncertainty About Treatment Remains

Isolated hypothyroxemia (IH) is defined as a low maternal free thyroxine (fT4) level with a normal thyrotropin (TSH) concentration. The terminology and concept of IH during pregnancy were first introduced by Vermiglio et al. ¹ and Pop et al. ² in 1999, but diagnostic criteria for this condition have varied. Different cutoffs for fT4 have been used, most frequently derived from the lowest 2.5th to 5th percentile of population-based reference ranges. The reported prevalence of IH during pregnancy has ranged widely across studies, from 0.2% to 31.0%, with a pooled prevalence of 2.1% reported in a recent meta-analysis. ³ A higher prevalence has been seen in iodine-deficient countries and in studies that did not apply pregnancy-specific reference ranges. ⁴

Pop et al. reported impaired psychomotor development in children of mothers with IH diagnosed in the first trimester. ² Subsequently, additional studies have examined the associations between IH and the risk of pregnancy complications and child neuropsychological developmental outcomes, with inconsistent results. ⁴ Whether IH is associated with worse maternal, neonatal, or offspring outcomes requires further elucidation and no trials to date have demonstrated a clear benefit of levothyroxine (LT4) treatment.

Thus, there is currently no consensus on the management of IH. ⁵ American Thyroid Association (ATA) guidelines recommend against routine treatment for IH. ⁶ However, the European Thyroid Association (ETA) recommends considering treatment if IH is detected in the first trimester. ⁷ The American College of Obstetrics and Gynecology ⁸ and the American Society for Reproductive Medicine ⁹ have not addressed IH in their guidelines.

In this issue of Thyroid, Han et al. present the results of a systematic review and meta-analysis on the risk of adverse maternofetal outcomes in pregnant women with IH. ¹⁰ This comprehensive systematic review and meta-analysis summarized and harmonized the heterogeneous results from prior studies. Maternal pregnancy outcomes, including preterm birth, premature rupture of membranes, placental abruption, gestational hypertension, preeclampsia, and gestational diabetes, were assessed. Fetal outcomes such as macrosomia, small for gestational age, low birth weight, fetal distress, neonatal intensive care unit admission, and congenital malformations were also examined.

A total of 19 articles comprising 12,129 pregnant patients with IH were included. Four articles, which described 895 treated pregnant women and 1247 nontreated controls, were pooled to investigate the efficacy of LT4 treatment. The authors found that gestational IH was associated with a significantly higher risk of preterm birth (relative risk [RR] 1.36), premature rupture of membranes (RR 1.41), gestational diabetes (RR 1.34), macrosomia (RR 1.62), and fetal distress (RR 1.72), with study heterogeneity ranging from low to high. However, treatment with LT4 did not result in risk mitigation.

The authors attempted to account for study heterogeneity by delving into different subgroup analyses, including assessments by gestational age when IH was diagnosed, the fT4 cutoff (<2.5th vs. <5th vs. <10th percentile) applied for diagnosis, study design (prospective vs. retrospective), and study country (as a proxy for iodine status). Associations with adverse outcomes were noted when using lower fT4 thresholds (the 2.5th or 5th percentile) to define IH, but no significant associations were found when employing the 10th percentile.

In addition, different adverse outcomes were associated with maternal IH diagnosed in the first trimester versus the second trimester. Maternal thyroid function during the first trimester is particularly crucial in fetal neurocognitive development because it is not until 16–20 weeks of gestation that the fetal thyroid gland begins to function. ^11,12 Overall, Han et al. provide convincing evidence of an association between IH and adverse pregnancy and neonatal outcomes. This systematic review and meta-analysis add to the existing body of data from individual participant-level meta-analyses published by the international thyroid in pregnancy consortium. ^{13 –15}

Study limitations must be considered when interpreting the findings. Using study-level rather than individual participant-level meta-analysis limits the ability to control for important confounding variables such as age, body mass index, and iodine status. The inconsistent definitions of IH with various fT4 cutoffs used across studies also contributed to heterogeneity in the results. In addition, this meta-analysis only examined maternofetal outcomes and did not assess potential effects on children's neurodevelopment, a concern raised in several published studies. ⁴ Although iodine deficiency is detrimental to IH and pregnancy outcomes, the authors could not perform a subgroup analysis for iodine status due to insufficient data.

One major challenge in addressing IH during pregnancy is the lack of evidence for LT4 treatment benefits. Among the four treatment studies included in the Han et al. analysis, only one was a randomized controlled trial. ¹⁶ In this trial, LT4 treatment was started at a mean gestational age of 18 weeks, which may have been too late to see treatment effects. The other three studies were observational studies from China. Two of these studies initiated LT4 treatment within the first trimester. Given the substantial potential for confounding in observational studies, there is currently insufficient evidence to determine whether LT4 treatment mitigates the risk of adverse outcomes during pregnancies with IH.

Accurately establishing the diagnosis of IH and treatment target in clinical practice remains challenging as assay-specific population-based trimester-specific reference ranges for serum TSH and fT4 are not widely available. ¹⁷ In the research setting, thyroid function measurements are done with the same assay, and gestational-age-specific reference ranges can be derived from a healthy pregnant population. However, in the real world, when such ranges are not available, the variable accuracy among different fT4 assays during pregnancy due to increased circulating thyroxine-binding globulin continues to be a barrier for accurate diagnosis of IH and, if treatment is elected, determining therapeutic targets. ¹⁸

Despite challenges in diagnosing IH, some actions can be taken to address the underlying common causes of IH. Optimizing iodine nutrition preconception and during pregnancy would prevent a significant number of IH cases globally. ⁴ In addition, identifying and treating iron deficiency in gestation may also play a role in reducing IH. ¹⁹

Overall, Han et al. provide strong evidence for increased risks of multiple adverse pregnancy and neonatal outcomes in maternal IH. To understand whether there is a treatment benefit for IH, an adequately powered randomized controlled trial implemented in the first trimester with appropriate LT4 dosing titration is needed. Unfortunately, in light of negative trials to date, ^16,20 such a trial may be challenging to fund.

The ATA is currently revising and updating its guideline for managing thyroid disease during pregnancy. A new guideline developed using rigorous methodology, including systematic reviews, is expected in the next one to two years. In the meantime, the question of what can be done to mitigate the risk of adverse outcomes associated with IH during pregnancy remains unanswered.