Objective: To investigate the clin. significance of non-invasive prenatal testing technol. (NIPT) in the detection of fetal chromosome copy number variation (CNV). Methods: A total of 22 998 singleton pregnancies who underwent NIPT screening in this hospital from Jan. 2018 to August 2020 and 90 pregnant women with CNV screening results were selected as the research objects. Fetal chromosome karyotype anal. and copy number variation sequencing (CNV-Seq) were performed after interventional prenatal diagnosis. The pregnancy outcome was followed up, the detection efficiency of NIPT for fetal CNV was evaluated, and its clin. significance was discussed. Results: Among the 90 pregnant women with CNV indicated by NIPT, 60 cases received interventional prenatal diagnosis, of which 31 cases (51.67%) did not find pathogenic CNV or other chromosomal abnormalities, 28 cases (46.67%) had CNV (27 cases were consistent with NIPT test results, 1 case was inconsistent), and 1 case only had fetal karyotype anal. and no abnormality was found, but CNV-Seq was not performed. The sensitivity, specificity, pos. predictive value and neg. predictive value of NIPT for CNV detection were 77.14%, 99.85%, 49.23% and 99.90%, resp. Among the 28 pregnant women with abnormal results of interventional prenatal diagnosis, 17 cases had induced labor and 11 cases had vaginal labor. Among the 31 pregnant women with no abnormality in interventional prenatal diagnosis, 1 case had induced labor, 1 case had polydactyly in vaginal delivery, and the rest of the newborns had no abnormal phenotype at birth. Among the 30 pregnant women without interventional prenatal diagnosis, 4 cases had induced labor, 1 case had a U-shaped mouth at birth, 1 case was lost to follow-up, and the remaining 24 newborns had normal phenotype. Conclusion: NIPT has certain clin. significance for fetal CNV detection. A pos. result of NIPT requires interventional prenatal diagnosis, and a neg. result still needs to strengthen ultrasound examination and follow-up, and genetic counseling should be done.