To evaluate the efficacy of non-invasive prenatal screening (NIPT) for detecting fetal sex chromosome abnormalities, a total of 639 women carrying sex chromosome abnormalities were selected from 222,107 pregnant women who participated in free NIPT from April 2018 to December 2020. The clinical data, prenatal diagnosis results, and follow-up pregnancy outcomes of participants were collected. The positive predictive value (PPV) was used to analyze the performance of NIPT. Around 235 cases were confirmed with sex chromosome abnormalities, including 229 cases with sex chromosome aneuploidy (45, X (n = 37), 47, XXX (n = 37), 47, XXY (n = 110), 47, XYY (n = 42)) and 6 cases with structural abnormalities. The total incidence rate was 0.11% (235/222,107). The PPV of NIPT was 45.37% (235/518). NIPT accuracy for detecting sex chromosome polysomes was higher than that for sex chromosome monomers. The termination of pregnancy rate for fetal diagnosis of 45, X, and 47, XXY was higher than that of 47, XXX, and 47, XYY. The detection rate of fetal sex chromosome abnormalities was higher in 2018–2020 than in 2010–2012 (χ2 = 69.708, P < 2.2 × 10−16), indicating that NIPT is greatly efficient to detect fetal sex chromosome abnormalities.