Value Of ¹⁸F-FDG PET-CT In The Diagnosis Of Ascites Of Undetermined Origin

Objective To investigate the value of18F-FDG PET-CT imaging in the identification of benign or malignant ascites.To investigate the role of18F-FDG PET-CT imaging to locate primary tumor of patients with malignant ascites.Materials and Methods Study objective1. Control Group The healthy control group consisted of 20 people who underwent the whole body PET-CT scan.The physical examination, chest X-ray, abdominal B ultrasound and laboratory tests of all were normal.2. Patient Group111 patients with acsites of unknown origin and 102 patients with malignant ascites were enrolled in this study, which from January 2008 to December 2012 in our hospital. All of them underwent the whole body PET-CT scan. The former 68 patients still underwent serum tumor marker detection( including AFP, CEA, CA125 and CA199).Method1. A retrospective analysis were conducted in 111 patients with unexplained ascites. The diagnosis of peritoneal carcinomatosis was established based on the results of multimodality imaging, histopathology examination and clinical follow-up after more than 6 months. Peritoneal SUVmax in benign and malignant ascites groups were compared with the student’s t test. The ROC curve were used to assess the diagnostic abilities ofSUVmax, serum alphafetoprotein(AFP), carcinoembryonic antigen(CEA), CA125 and CA199.2. A total of 102 patients with malignant ascites were analyzed retrospectively. The diagnosis was confirmed by histopathology examination or clinical following-up( more than 6 months). The diagnostic value of PET-CT and CT alone was calculated,respectively. Kappa test were used to measure agreement of the diagnosis using PET-CT and the final diagnosis, and of the diagnosis using CT alone and the final diagnosis,respectively.The diagnostic efficacy of PET-CT and CT alone were tested by x2 test.Results1. The sensitivity,specificity and accuracy for PET-CT to differentiate malignant from benign ascites were 92.4%( 73/79), 87.5%( 28/32) and 91.0%( 101/111), respectively.That of CT alone were 55.7%( 44/79), 81.3%( 26/32) and 63.1%( 70/111), respectively.The agreement between the diagnosis of PET-CT and the histopathology examination was good( Kappa=0.784, P<0.001) whereas than between the result of CT alone and the final diagnosis was poor( Kappa=0.291, P<0.001).2. The average SUVmax of benign peritoneal lesions was( 2. 70 ± 2.32), and that of malignant lesions was( 5.73 ± 3.84). The latter was significantly higher than former.( t=-5.077, P <0. 05).3. The area under the ROC curve of SUVmax, AFP, CEA, CA125 and CA199 were 0.750, 0.583,0.658, 0.608 and 0.678, respectively. That of SUVmax was the largest among the detected five indices.4. The primary decease was correctly identified by PET-CT in 76 patients( 79/102, 77.5%). CT alone detected the primary lesion for 58( 58/102, 56.9%). The agreement between the diagnosis of PET-CT and the histopathology examination was good( Kappa=0.707, P<0.001)whereas than between the result of CT alone and the final diagnosis was poor( Kappa=0.347,P<0.001).There were significanfly different between PET-CT and CT alone( χ2=19.048,P<0.001).Conclusion1.18F-FDG PET-CT has obvious advantages in diagnosing peritoneal metastasis of patients with ascites of unknown causes, it can be used to identify the character of ascites.2. In the differentiating benign and malignant effusion, SUVmax is a useful index, but the detection of serum tumor markers is only useful to clinical auxiliary diagnosis, it can’t be a basis for diagnosis.3. The value of 18F-FDG PET-CT in the detection of primary tumors is significantly better than that of CT alone, it can be used to locate primary tumor in patients with malignant ascites.4. PET-CT is expected to become the preferred method in patients with ascites.