Objective: To investigate the diagnostic value of transesophageal real-time three-dimensional ultrasound (RT 3D-TEE) in the diagnosis of paravalvular leak (PVL) after mitral valve replacement and the effect of interventional closure and surgical treatment. Methods: A total of 105 patients with suspected PVL after mitral valve replacement who were diagnosed and treated from May 2018 to June 2020 were selected as the research subjects. A total of 56 patients with PVL after mitral valve replacement were diagnosed by surgery (34 cases of posterior leaflet leakage, 34 cases of anterior leaflet Two-dimensional transesophageal ultrasonography (2D-TEE) and RT 3D-TEE were performed to compare the positive predictive value, negative predictive value, detection rate of PVL leak location, and the length, width and area of the leak. The AUC value, sensitivity and specificity of 2D-TEE and RT 3D-TEE in the diagnosis of PVL; they were divided into interventional occlusion group (n=28) and surgical operation group (n=28) according to the treatment method, and the operation time, hospital stay were compared Cost, length of hospital stay, blood transfusion rate, pulmonary infection rate, new-onset arrhythmia rate, residual PVL rate, left ventricular ejection fraction (LVEF), left ventricular end-diastolic diameter (LVIDd), left ventricular end-systolic diameter (LVIDs), Left atrial diameter (LVD). Results: After 2D-TEE examination of 105 patients with suspected PVL after mitral valve replacement, the positive predictive value was 72.22% and the negative predictive value was 66.67%. Compared with the gold standard, the consistency was poor (Kappa=0.389, P＜0.001); After RT 3D-TEE examination, the positive predictive value was 96.43%, and the negative predictive value was 95.92%, which was consistent with the gold standard (Kappa=0.923, P<0.001). Among them, the detection rate of 2D-TEE was significantly different from that of RT 3D-TEE (χ2=4.102, P<0.05). RT 3D-TEE detected a higher proportion of posterior lobe leak and anterior lobe leak than 2D-TEE, and the leak length, leak width and leak area were higher than those of 2D-TEE (P<0.05). ROC curve analysis showed that the AUC values of 2D-TEE and RT 3D-TEE for the diagnosis of PVL were (0.690, 0.955, P<0.05), respectively. The operation time, hospitalization cost, hospitalization time, blood transfusion rate, pulmonary infection rate, and new-onset arrhythmia rate in the interventional closure group were lower than those in the surgery group (P<0.05). There was no significant difference in LVEF, LVIDd, LVIDs, and LVD between the interventional occlusion group and the surgery group (P>0.05). Conclusion: RT 3D-TEE has a high accuracy in the diagnosis of paravalvular leakage after mitral valve replacement, and the interventional occlusion is effective and safe.