Diagnosing ascitic etiology on a biochemical basis.

BACKGROUND/AIMS: The analysis of ascitic fluid has been complicated by several new tests. To simplify its assessment, we evaluated ascitic fluid pH with six other parameters prospectively and simultaneously in blood and ascitic fluid in sixty patients with ascites.

METHODOLOGY: Sixty patients were selected on whom abdominal paracentesis was performed in order to determine which tests were the most reliable for the etiological diagnosis of ascites. Their ascitic fluid pH, cholesterol, albumin, and total proteins were analyzed to distinguish between malignant and cirrhotic ascites. The diagnosis of cirrhosis was based on histological findings, and/or clinical, biochemical, and endoscopic/ultrasonographic findings in whom percutaneous liver biopsy was contraindicated.

RESULTS: The serum-ascites albumin gradient was higher (p<0.001) in cirrhotic as compared to tubercular, malignant or cardiac ascites. We discovered that ascitic fluid lactate dehydrogenase and cholesterol were best for diagnosing malignant ascites, ascitic fluid glucose and ascitic fluid/blood glucose ratio were perfect for diagnosing tubercular ascites, and ascitic fluid pH and lactate dehydrogenase were suitable for diagnosing spontaneous bacterial peritonitis. Correspondingly, ascitic fluid albumin, ascitic fluid/serum albumin, total proteins, and serum ascites albumin gradient were best for diagnosing sterile cirrhotic ascites.

CONCLUSIONS: Diagnostic paracentesis is a useful procedure. The practice of ordering a battery of tests on every ascitic fluid specimen should be abandoned. Rather, an algorithm approach should be adopted in which the results of initial analysis guide us to further relent tests that help in arriving at the etiology of ascites.