Impact of borderline–subclinical hypothyroidism on subsequent pregnancy outcome in women with unexplained recurrent pregnancy loss

Aim Because subclinical hypothyroidism (thyroid-stimulating hormone [TSH] > 4.5 IU/mL) is associated with adverse pregnancy outcome, including early pregnancy loss, TSH is recommended to be titrated to ≤2.5 mIU/L in levothyroxine-treated women before pregnancy. The purpose of this study was to determine whether borderline–subclinical hypothyroidism (borderline-SCH; 2.5 < TSH ≤ 4.5 IU/mL) affects the outcome of subsequent pregnancies in women with unexplained recurrent pregnancy loss (uRPL). Methods After workup for antinuclear antibody (ANA), anti-phospholipid syndrome, thrombophilia, uterine abnormalities, hormone disorders, and/or chromosomal abnormalities, 317 women with a history of uRPL were enrolled. The women were classified into two groups: borderline-SCH, and euthyroidism (0.3 ≤ TSH ≤ 2.5 IU/mL). All women had normal serum free thyroxine (T4) and did not receive levothyroxine before or during the subsequent pregnancy. Results There were no significant differences in age, number of previous pregnancy losses, number of live births, or body mass index between the borderline-SCH (n = 56) and the euthyroid (n = 261) groups, but the rate of ANA positivity differed significantly (53.6% vs 33.7%, respectively; P = 0.005). The subsequent pregnancy rate did not differ between the two groups (55.4%, 31/56 vs 51.3%, 134/261, respectively). The pregnancy loss rate (<22 weeks of gestation) tended to be higher in the borderline-SCH than the euthyroid group (29.0%, 9/31 vs 17.9%, 24/134), although not significantly so (P = 0.16). Conclusions Although some subset of uRPL is though to be due to as-yet-unidentified cause(s), borderline-SCH is unlikely to be involved in uRPL.