Importance: Abundant evidence suggests thyroid dysfunction is associated with adverse pregnancy outcomes. However, associations of preconception thyrotropin levels outside of reference range with reproductive health outcomes are not well characterized.
Objective: To evaluate the associations of preconception thyrotropin levels with time to pregnancy (TTP) and risk of spontaneous abortion (SA).
Design, setting, and participants: This population-based cohort study used data from the Chinese National Free Prepregnancy Checkups Project. Female participants aged 20 to 49 years who were trying to conceive were enrolled between January 1, 2013, and December 31, 2016, for the analysis of TTP or SA. Data were analyzed between August 1, 2020, and July 5, 2021.
Exposures: Levels of thyrotropin within 1 year prior to pregnancy.
Main outcomes and measures: The main outcomes were TTP, assessed using hazard ratios (HRs), and SA, assessed using odds ratios (ORs), according to preconception thyrotropin levels. Thyrotropin dose-response associations were assessed using restricted cubic spline regression.
Results: Among 11 194 002 female participants (mean [SD] age, 27.56 [5.10] years) in the TTP cohort, 4 739 421 (42.34%) participants became pregnant within 1 year. Both low and high preconception thyrotropin levels were associated with delayed TTP compared with thyrotropin levels of 0.37 to 2.49 mIU/L (thyrotropin <0.10 mIU/L: HR, 0.90; 95% CI, 0.89-0.92; thyrotropin 4.88-9.99 mIU/L: HR, 0.86; 95% CI, 0.86-0.87; thyrotropin ≥10.00 mIU/L: HR, 0.78; 95% CI, 0.77-0.79). In the SA analysis cohort including 4 678 679 pregnancies, 108 064 SA events (2.31%) were documented. High thyrotropin groups showed an increased risk of SA compared with the group with thyrotropin levels of 0.37 to 2.49 mIU/L (thyrotropin 4.88-9.99 mIU/L: OR, 1.33; 95% CI, 1.28-1.38; thyrotropin ≥10.00 mIU/L: OR, 1.25; 95% CI, 1.14-1.36). Preconception thyrotropin levels showed an inverted J-shaped dose-response association with TTP (χ2 = 311.29; nonlinear P < .001) and a J-shaped dose-response association with SA (χ2 = 58.29; nonlinear P < .001).
Conclusions and relevance: In this cohort study, preconception thyrotropin levels outside of reference range were associated with increased risk of reduced fecundity and SA. These findings may provide insights for the implementation of preconception thyroid function screening and the design of future levothyroxine supplementation trials.