Investigating the effect of preconception untreated subclinical hypothyroidism (TSH above normal range and below 6.0 mIU/L) on assisted reproductive technology (ART) outcomes

Subclinical hypothyroidism (SCH) is a biochemical diagnosis with ongoing uncertainty regarding management in reproductive medicine. We performed a retrospective cohort study evaluating fertility and pregnancy outcomes in women undergoing assisted reproductive technology (ART), focusing on untreated preconception SCH with TSH up to 6.0mIU/L. Secondary analysis used data from women screened for thyroid function and thyroid peroxidase antibodies (TPOAb) during the TABLET trial at Birmingham Women's Hospital, UK (June 2012-December 2013). Women with TSH ≥6.0mIU/L received levothyroxine. The primary outcome was live birth. The main comparison was 'total SCH' (TSH >3.63mIU/L) versus 'euthyroid-1' (TSH 0.44-3.63mIU/L), with subgroup analyses comparing 'untreated SCH' (TSH 3.64-5.99mIU/L) to 'euthyroid-1' women, and a stricter euthyroid threshold ('euthyroid-2'; TSH 0.44-2.5mIU/L) versus 'untreated high-normal-mild SCH' (TSH 2.5-5.99mIU/L). Multivariable logistic regression adjusted for key confounders. Of 944 women screened, 560 had outcome data; 60 had SCH. SCH was associated with fewer oocytes retrieved (9.0 vs 10.4, p = 0.02) and mature oocytes (7.4 vs 8.4, p = 0.05), but no difference in live birth was seen across comparisons: 'untreated SCH' vs euthyroid-1 (aOR 1.01, 95% CI 0.54-1.87, p = 1.0) and 'untreated high-normal-mild-SCH' vs euthyroid-2 (aOR 0.96, 95% CI 0.65-1.42, p = 0.8). In women with untreated SCH achieving live birth, TSH normalised spontaneously in 8/15 cases, with no adverse outcomes observed. Untreated TSH <6.0 mIU/L does not appear to adversely affect ART or live birth outcomes, although larger prospective studies are needed.