EUS in the detection of ascites and EUS-guided paracentesis

Background. The utility of EUS was evaluated for detection of ascites and EUS-guided FNA of ascites in patients undergoing EUS for diagnosis and staging of GI malignancies. Methods: A series (from March 1994 to October 1997) of 571 consecutive patients who underwent upper EUS for various indications was retrospectively reviewed. Follow-up clinical information was obtained from referring physicians, subsequent CT, and telephone interviews. Results: Eighty-five patients (15% of series) were found to have ascites by EUS. Six did not have CT before EUS. Pre-EUS CT identified ascites in only 14 (18%) of the 79 patients who had pre-EUS CT. Of the patients in whom CT was negative for abdominal fluid (n = 65) and who had clinical follow-up, 13 of 58 (22%) subsequently had ascites develop that were detected by CT or physical examination. Overall, 31 of the 85 patients underwent EUS-guided FNA paracentesis; the mean volume obtained was 7.9 mL (range 1-40 mL). In 5 patients, malignant ascites was diagnosed by EUS-guided FNA; in these patients surgery was avoided. Conclusions: EUS is more sensitive than CT in detecting small amounts of ascites. A significant number (22%) of patients who had ascites by EUS subsequently had ascites develop that was detectable by CT or physical examination. EUS-guided paracentesis appears to be safe and effective and can identify malignant ascites.