ObjectiveTo systematically evaluate the association of prenatal thoracic ultrasound abnormalities with copy number variants (CNVs).MethodsChromosomal microarray (CMA) data and clinical characteristics from fetuses with thoracic ultrasound abnormalities were retrieved and analyzed.ResultsThoracic ultrasound findings were mainly isolated except for fetal pleural effusion (FPE) and pulmonary hypoplasia. The diagnostic yield of CMA for thoracic anomaly was 9.66%, and FPE (17/68, 25%), pulmonary hypoplasia (1/8, 12.5%), and congenital diaphragmatic hernia (CDH) (6/79, 7.59%) indicated relatively high pathogenic/likely pathogenic (P/LP) CNV findings. The detection rate for P/LP CNVs was obviously increased in non‐isolated thoracic anomalies (27.91% vs. 1.96%, P < 0.0001), non‐isolated FPE (37.78% vs. 0%, P = 0.0007) and non‐isolated congenital pulmonary airway malformation (CPAM) (27.27% vs. 0%, P < 0.0001), and significantly different among thoracic anomalies. Additionally, the rate of termination of pregnancy in cases with non‐isolated thoracic anomalies (58.49% vs. 12.34%, P < 0.0001) and P/LP CNVs (85.71% vs. 24.15%, P < 0.0001) was obviously increased.ConclusionThe present study expanded phenotype spectrums for particular recurrent CNVs. FPE, CDH, and pulmonary hypoplasia indicated relatively high P/LP CNV findings among common thoracic ultrasound abnormalities, CPAM associated with other ultrasound abnormalities increased the incidence of diagnostic CNVs, while bronchopulmonary sequestration might not be associated with positive CNVs. The present data recommended CMA application for cases with prenatal thoracic ultrasound abnormalities, especially non‐isolated FPE, non‐isolated CPAM, CDH, and pulmonary hypoplasia.