Meta-analysis of the effects of levothyroxine therapy for subclinical hypothyroidism during pregnancy on offspring outcomes.

OBJECTIVE

To conduct a systematic evaluation of the impact of levothyroxine (L-T4) therapy on birth outcomes in pregnancies complicated by subclinical hypothyroidism (SCH).

METHODS

A thorough literature review was conducted across several databases, including PubMed, Embase, Cochrane Library, Web of Science, Sinomed, Wanfang Data Knowledge Service Platform, China National Knowledge Internet (CNKI), and VIP Chinese Science and Technology Journal Database (VIP), to investigate the impact of L-T4 treatment of SCH during pregnancy on birth outcomes in offspring. Based on predefined inclusion and exclusion criteria, two researchers were appointed to extract data and assess the quality of the literature. Subsequently, meta-analysis was performed using RevMan 5.4 and Stata 14 software.

RESULTS

The study incorporated a total of thirty randomized controlled trials (RCTs) and cohort studies, encompassing four countries and regions. The sample included 18,568 pregnant women with SCH and 5,578 pregnant women undergoing L-T4 treatment. The meta-analysis indicated that L-T4 treatment for SCH during pregnancy may reduce the incidence of preterm birth (RCTs (RR = 0.56, 95% CI = 0.41-0.77), cohort studies (RR = 0.71, 95% CI = 0.51-0.99)) and low birth weight infants (LBWI) (RCTs (RR = 0.56, 95% CI = 0.35-0.89), cohort studies (RR = 0.71, 95% CI = 0.58-0.88)), while it does not significantly affect the risk of macrosomia (RCTs (RR = 0.29, 95% CI = 0.06-1.38), cohort studies (RR = 0.70, 95% CI = 0.30-1.62)), small for gestational age (SGA) infants (RCTs (RR = 1.18, 95% CI = 0.74-1.90)), or congenital hypothyroidism (CH) (cohort studies (RR = 1.27, 95% CI = 0.16-10.07)) in children. No significant difference in birth weight (RCTs (RR = 0.10, 95% CI = -0.04-0.24), cohort studies (RR = 0.10, 95% CI = -0.08-0.28)) was observed between the L-T4 treatment group and the non L-T4 treatment group. Regarding neonatal cord blood thyroid function, the TSH levels in the L-T4 treatment group were lower compared to the non L-T4 group (RCTs (RR = -2.48, 95% CI = -4.51--0.45), cohort studies (RR = -3.53, 95% CI = -4.27 - -2.79)); however, no significant differences were found in FT3 (RCTs (RR = 0.06, 95% CI = -0.24-0.36), cohort studies (RR = 0.08, 95% CI = -0.72-0.88)) and FT4 levels (RCTs (RR = 0.07, 95% CI = -0.41-0.56), cohort studies (RR = 0.03, 95% CI = -1.18-1.24)) between the two groups.

CONCLUSION

L-T4 treatment appears to reduce the incidence of preterm birth and LBWI in pregnant mothers with SCH, but it does not significantly affect the incidence of macrosomia, SGA, CH, or birth weight. Regarding the thyroid function in neonatal umbilical cord blood, L-T4 treatment in SCH pregnant women can reduce TSH levels in the umbilical cord blood of their newborns, while having no significant effect on FT3 and FT4 levels.

SYSTEMATIC REVIEW REGISTRATION

https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD420251016450, PROSPERO CRD420251016450.