Value Of ¹⁸F-FDGPET/CT In Diagnosing The Undetermined Origin Of Ascites

Background and PurposeThe etiological factors of ascites are various,which involve multiple organs and diseases.Ascites can be divided into malignant ascites and benign ascites,infectious and non-infectious ascites.For clinicians,to determine the primary cause of ascites is of vital importance for the treatment and prognosis.The purpose of this study is to investigate the value of¹⁸F-FDG PET/CT combined with tumor markers in serum or ascites for diagnosing the undetermined origin of ascites,so as to provide reference for the identification and treatment of clinical ascites.MethodsWe retrospectively reviewed patients with ascites of unknown cause who underwent evaluation with¹⁸F-FDG PET/CT and clinical information in The Second Affiliated Hospital of Harbin Medical University from October 2017 and November2019.The final results were confirmed by pathologic diagnosis,cytologic examination of ascites,or clinical follow-up（at least 6 months）.PET/CT scanning with Siemens Biograph m CT 64 PET/CT.All patients measured ascites SUVmax and peritoneal SUVmax.True positive（TP）and true negative（TN）findings indicated that the results of¹⁸F-FDG PET/CT were consistent with the clinical final results.False positive（FP）cases were defined as cases that had no evidence of malignancy detected by the final standard but¹⁸F-FDG PET/CT results that suggested malignancy.In contrast,false negative（FN）cases were defined as cases that had no evidence of malignancy on¹⁸F-FDG PET/CT images but had subsequent proof of malignancy identified using the reference standard.Serum and ascites tumor markers（CEA,AFP,CA125,CA153,CA199）levels were measured within 1 week of the¹⁸F-FDG PET/CT examination.The normal levels of CEA,AFP,CA125,CA153,and CA199 in serum and ascites were less than 5.0 ng/m L,8.78 ng/m L,35 U/m L,31.3 U/m L and 37U/m L,respectively.The sensitivity,specificity,accuracy,positive predictive value（PPV）and negative predictive value（NPV）,were calculated.Ascites patients were divided into four groups:malignant ascites group,benign ascites group,benign tuberculous ascites group and benign non-tuberculous ascites group.Mann-Whitney test was used to compare SUVmax of ascites and SUVmax of peritoneum.The ROC curve was used to assess the sensitivity,specificity and accuracy of ascites SUVmax and peritonel SUVmax in differentiating benign and malignant ascites ability and calculated the cut-off.All statistical analyses were performed using Statistical Product and Service Solutions version 25.0 software,Mann-Whitney test was used for comparison between two groups,and rank sum test was used for comparison between multiple groups.For all analyses,p<0.05 suggested that the difference was statistically significant.All quantitative data were expressed as mean±SD values and counting data is expressed as a percentage.ResultsThis retrospective study included 47 patients（17 men and 30 women）,age58.36±12.51 years（range,20-80 years）who had undergone 18F-FDG PET/CT for the evaluation of ascites of undetermined origin.Malignant ascites was eventually diagnosed in 29 patients,Ovarian Cancer（n=9,31.03%）,Gastric carcinoma（n=8,27.59%）,Peritoneal cancer（n=4,13.79%）,Lymphoma（n=2,6.90%）,Appendix mucinous adenocarcinoma,Multiple myeloma,POEMS syndrome,Pancreatic carcinoma,Gallbladder cancer and Colon carcinoma（n=1,3.45%,respectively）.Benign ascites was diagnosed in the remaining 18 patients,Tuberculous peritoneum（n=8,44.44%）,cirrhosis（n=5,27.78%）,SLE（n=3,16.67%）.The primary lesion was found in 22（22/29,75.86%）patients with malignant ascites by¹⁸F-FDG PET/CT.There were 26 cases of true positive,3 cases of false negative,5 cases of false positive and 13 cases of true negative in the diagnosis of¹⁸F-FDG PET/CT.The results of this study showed that the sensitivity of CA125 in serum or ascites was the highest（89.29%）.The specificity of CEA was the highest at100.00%.The sensitivity,specificity,positive predictive value,negative predictive value and accuracy of¹⁸F-FDG PET/CT alone in the diagnosis of unexplained ascites were 89.66%,72.22%,83.87%,81.25%and 82.98%,respectively.When¹⁸F-FDG PET/CT combined with tumor markers,the sensitivity of diagnosis was increased to96.15%.The results showed that the SUVmax of malignant ascites was higher than that of benign ascites（1.70±0.47 vs 1.28±0.40,P=0.00）.The SUVmax of malignant ascites was higher than that of benign non-tuberculous ascites（1.70±0.47 vs 1.16±0.41,P=0.00）.The thickening of SUVmax in malignant peritoneal lesions was higher than that in benign peritoneal lesions（7.99±4.85 vs 5.40±6.38,P=0.03）.SUVmax of malignant thickening peritoneum was higher than that of benign non-tuberculous peritoneum（7.99±4.85 vs 1.16±0.18,P=0.00）.However,the SUVmax of benign tuberculous ascites and peritoneum were higher.ROC curve analysis of ascites SUVmax showed that when the value of ascites SUVmax was 1.74,the sensitivity,specificity and accuracy of differentiating benign and malignant ascites were 46.40%,93.70%and 63.83%,respectively,and the area under the curve（AUC）was 0.71（P=0.02）.When the SUVmax value of peritoneum was 2.22,the sensitivity,specificity and accuracy of differentiating benign and malignant peritoneum were 96.40%,62.50%and 82.98%,respectively,and the area under the curve was 0.70（P=0.03）.ConclusionsThe results of this study showed that the SUVmax of malignant ascites and tuberculous ascites was higher than that of benign non-tuberculous ascites,but the SUVmax values of malignant peritoneal lesions and tuberculous peritoneal lesions were higher,which could be differentiated by combining tumor markers.We believe that¹⁸F-FDG PET/CT can be used as a powerful imaging tool for the identification of benign and malignant ascites with unknown causes,especially for patients with suspected cancer or cancer at unknown primary sites in clinical practice,and the combination of¹⁸F-FDG PET/CT and tumor markers can improve the sensitivity of diagnosis.