ABSTRACT
ObjectiveTo investigate the necessity and indication of magnetic resonance cholangiopancreatography (MRCP) in patients with gallstones diagnosed by color Doppler ultrasound. MethodsA retrospective case-control study was conducted. The clinical data of 1431 patients who were admitted to The Affiliated Hospital of Xuzhou Medical University from January 2013 to December 2017 with the initial diagnosis of gallstones were collected. On admission, abdominal ultrasound found gallstones alone, without other lesions. The patients who were diagnosed with gallstones alone after surgical treatment were enrolled as control group, and those who were diagnosed with gallstones and common bile duct stones were enrolled as experimental group. A logistic regression analysis was performed to establish a predictive model for gallstones with common bile duct stones. The receiver operating characteristic (ROC) curve was plotted for the predictive model and each index, and the area under the ROC curve was calculated. ResultsFour indices were finally screened out by the logistic regression analysis, i.e., bile duct dilation (χ1), gallbladder enlargement (χ2), direct bilirubin (DBIL) (χ3), and gamma-glutamyl transpeptidase (GGT) (χ4), and the logistic regression model was Logi(P) = -4.153 + 3.592χ1 + 1.720χ2 + 1.798χ3 + 1.519χ4. In the diagnosis of common bile duct stones missed by color Doppler ultrasound, this model had a sensitivity of 53.4%, a specificity of 99.6%, an accuracy of 925%, and an area under the ROC curve of 0.904, which was significantly better than each index alone. ConclusionMRCP is not necessary for patients diagnosed with simple gallstones by color Doppler ultrasound. However, enlarged gallbladder, a diameter of common bile duct of ＞0.8 cm, DBil＞7 μmol/L, and GGT＞40 U/L are independent risk factors for gallstones with common bile duct stones. The logistic regression model established in this study has high sensitivity and specificity and can provide a reference for to decision-making among clinicians.