IDDF2020-ABS-0221 Ascitic fluid cholesterol is a useful marker to rule out malignant ascites

Background Diagnosis of malignant ascites is a clinical challenge. The gold standard method for diagnosis is with cytology; however, it has low sensitivity, is observer-dependent, and may require repeated paracentesis. Recently, some studies showed that high ascitic fluid cholesterol level could be a marker of malignant ascites and thus can differentiate between benign and malignant ascites. This study aimed to assess and evaluate the diagnostic accuracy of ascitic fluid cholesterol level in differentiating malignant from non-malignant ascites. Methods We prospectively studied patients of ascites, who were admitted in from February 2019 to July 2019. Ascitic fluid cholesterol of all patients was sent along with cytology for malignant cells. Malignancy was proven by tissue diagnosis (peritoneal biopsy or EUS guided abdominal lymph node FNA or cytology of ascitic fluid). Patients with spontaneous bacterial peritonitis, tubercular ascites and pancreatic ascites were excluded. Results A total of 50 patients of ascites were included in the study, and of these 11 patients had malignant ascites as proved by peritoneal biopsy or EUS guided abdominal lymph node biopsy or by ascitic fluid cytology. On taking a cut-off of 30 mg/dL of ascitic fluid cholesterol level, it was found that 9 of 11 patients (81%) of malignant ascites had high cholesterol values; however, only 8 of 39 patients (20.5%) of non-malignant ascites had high cholesterol values (p<0.05).The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of ascitic fluid cholesterol for diagnosing malignant ascites were 81.81%, 79.48%, 52.94%, 94%, and 80% respectively. Conclusions Ascitic fluid cholesterol has high negative predictive value for ruling out malignancy-related ascites and thus can be used as an excellent biochemical investigation to rule out malignant ascites.