Chapter 044. Abdominal Swelling and Ascites (Part 3) A gradient <1.1 g/dL (low gradient) suggests that the ascites is not due to portal hypertension with >97% accuracy and mandates a search for other causes such as peritoneal carcinomatosis, tuberculous peritonitis, pancreatitis, serositis, pyogenic peritonitis, and nephrotic syndrome (Table 44-1). Table 44-1 presents some of the disease states that produce high-SAAG and low-SAAG ascites. Although there is variability of the ascitic fluid in any given disease state, some features are sufficiently characteristic to suggest certain diagnostic possibilities. For example, blood-stained fluid with >25 g/L protein is unusual in uncomplicated cirrhosis but is consistent with tuberculous peritonitis or neoplasm. Cloudy fluid with a predominance of polymorphonuclear cells (>250/µL) and a positive Gram's stain are characteristic of bacterial peritonitis, which requires antibiotic therapy; if most cells are lymphocytes, tuberculosis should be suspected. The complete examination of each fluid is most important, for occasionally only one finding may be abnormal. For example, if the fluid is a typical transudate but contains >250 white blood cells per microliter, the finding should be recognized as atypical for cirrhosis and should warrant a search for tumor or infection. This is especially true in the evaluation of cirrhotic ascites where occult peritoneal infection may be present with only minor elevations in the white blood cell count of the peritoneal fluid (300–500/µL). Since Gram's stain of the fluid may be negative in a high proportion of such cases, careful culture of the peritoneal fluid is mandatory. Bedside inoculation of blood culture flasks with ascitic fluid results in a dramatically increased incidence of positive cultures when bacterial infection is present (90 vs. 40% positivity with conventional cultures done by the laboratory). Direct visualization of the peritoneum (laparoscopy) may disclose peritoneal deposits of tumor, tuberculosis, or metastatic disease of the liver. Biopsies are taken under direct vision, often adding to the diagnostic accuracy of the procedure. Table 44-1 Characteristics of Ascitic Fluid in Various Disease States Cell Count Cond ition G ross Appeara Pro tein, g/L Se rum- Ascites Albumin Re d Blood White Blood Cells, Oth er Tests nce Gradient , g/dL Cells, >10,000/µ L per µL Cirrh osis St raw- colored or bile- stained <25 (95%) >1 .1 1% <250 (90%) a ; predominantly mesothelial Neopl asm St raw- colored, hemorrha gic, mucinous , or chylous >25 (75%) <1 .1 20 % >1000 (50%); variable cell types Cyt ology, cell block, peritoneal biopsy Tuber culous Cl ear, >25 <1 7% >1000 (70%); Peri toneal peritonitis turbid, hemorrha gic, chylous (50%) .1 usually >70% lymphocytes biopsy, stain and culture for acid-fast bacilli Pyoge nic peritonitis Tu rbid or purulent If purulent, >25 <1 .1 Un usual Predom inantly polymorphon uclear leukocytes Posi tive Gram's stain, culture Cong estive heart failure St raw- colored Var iable, 15– 53 >1 .1 10 % <1000 (90%); usually mesothelial, mononuclear Nephr osis St raw- colored <25 (100%) <1 .1 Un usual <250; mesothelial, mononuclear If chylous, ether or chylous extraction, Sudan staining Pancr eatic ascites (pancreatitis, pseudocyst) Tu rbid, hemorrha gic, or chylous Var iable, often >25 <1 .1 Var iabl e, may be blood- stained Variabl e Incr eased amylase in ascitic fluid and serum a Because the conditions of examining fluid and selecting patients were not identical in each series, the percentage figures (in parentheses) should be taken as an indication of the order of magnitude rather than as the precise incidence of any abnormal finding. . Chapter 044. Abdominal Swelling and Ascites (Part 3) A gradient <1.1 g/dL (low gradient) suggests that the ascites is not due to portal hypertension with >97% accuracy and mandates. pancreatitis, serositis, pyogenic peritonitis, and nephrotic syndrome (Table 44-1). Table 44-1 presents some of the disease states that produce high-SAAG and low-SAAG ascites. Although there is variability. be recognized as atypical for cirrhosis and should warrant a search for tumor or infection. This is especially true in the evaluation of cirrhotic ascites where occult peritoneal infection