Abstract： Objective To analyze the influence of different thyroid stimulating hormone (TSH) cut-offs to diagnose subclinical hypothyroidism (SCH) in the first trimester of gestation.Methods A total of 896 pregnant women were enrolled in Peking University International Hospital between October 2016 and March 2018.Among them,421 pregnant women with single fetus who were conformed to the criteria of National Academy of Clinical Biochemistry (NACB),without adverse pregnancy outcomes and obstetric complications,were selected to establish their self-sequential longitudinal reference ranges of thyroid function.Then,SCH was diagnosed in the first trimester,using different TSH cutoffs,such as the upper limit of the first trimester-specific reference range,4.0 mU/L recommended by the 2017 Guidelines of American Thyroid Association (ATA),5.17 mU/L (Roche reagent) recommended by 2012 Guidelines of Chinese Society of Endocrinology and Chinese Society of Perinatal Medicine,and 2.5 mU/L recommended by 2011 Guidelines of ATA,respectively.Results The TSH reference range was 0.12-4.16 mU/L in the first trimester.Using TSH＞4.16,4.0,5.17 and 2.5 mU/L to diagnose SCH in the first trimester,the prevalence rates were 4.35％ (39/896),5.92％ (53/896),1.56％ (14/896) and 20.87％ (187/896),respectively.There was no statistically significant difference between the prevalence rates of SCH using the TSH upper reference limit of 4.0 mU/L and 4.16 mU/L (P=0.134).When TSH was defined as＞4.0 mU/L to diagnose SCH,the sensitivity,specificity and Youden index was 97.4％,98.2％,and 0.956,respectively.Conclusions The TSH upper reference limit of 4.0 mU/L recommended by 2017 Guidelines of ATA can be used as a cut-off todiagnose SCH in first trimester for the areas without trimester-specific reference ranges for TSH in China.