Thyrotropin-receptor antibodies (TRAb) are used to diagnose Graves' hyperthyroidism in pregnant women. Bioassays provide a measure of thyrotropin-receptor stimulatory antibodies (TSI) specifically. The objective was to measure TSI in pregnant women for establishment of a pregnancy-specific cut-off and comparison with immunoassay measurements of TRAb. The retrospective Danish study was performed within the North Denmark Region Pregnancy Cohort (2011-2015) that includes stored biobank samples from early pregnancy (median week 10) with immunoassay measurements of thyroid function parameters and TRAb. TSI were measured in the same samples using the Turbo TSI bioassay (Quidel/Ortho-Clinical Diagnostics) with a recommended cut-off of 0.0241 IU/L in non-pregnant adults. A pregnancy-specific TSI cut-off (95-percentile) was established using Regression on Order Statistics. The established TSI cut-off was 0.0418 IU/L (95 % CI: 0.0417-0.0419). Among women with early pregnancy hyperthyroidism (n=438), 43 women (9.8 %) were TSI positive using the established cut-off, and these women had lower TSH (median 0.008 mIU/L) compared to women with TSI levels below 0.0241 (median TSH 0.040 mIU/L) or in the range from 0.0241 to 0.0418 (median TSH 0.033 mIU/L). Among the 438 women with early pregnancy hyperthyroidism, 22 women were positive for TSI and TRAb, 388 were negative for both, and 28 women were positive for either TSI or TRAb. This is the first study on TSI measurements in a large cohort of early pregnant women. A pregnancy-specific cut-off for TSI was established and agreement in the classification with immunoassay measurements of TRAb was seen in 94 % of cases.