Diagnostic parameters for spontaneous bacterial peritonitis in hepatitis B-related cirrhosis: comparison of ascitic fluid polymorphonuclear cell count, lactate and pH values.

BACKGROUND

The clinical value of ascitic fluid pH or lactate in the diagnosis of spontaneous bacterial peritonitis was debated in alcoholic cirrhosis. It was rarely discussed in hepatitis B-related cirrhosis. We investigated the best rapid diagnostic method in spontaneous bacterial peritonitis of patients with hepatitis B-related cirrhosis.

METHODS

Ascitic fluid polymorphonuclear cell count, ascitic fluid pH, arterial-ascitic fluid pH gradient, ascitic fluid lactate, and arterial-ascitic fluid lactate gradient were analyzed in 79 patients with sterile portal hypertension-related ascites, 31 patients with spontaneous bacterial peritonitis, 16 patients with hepatocellular carcinoma, 15 patients with malignancy-related ascites and 9 patients with other miscellaneous diseases.

RESULTS

The sensitivity, specificity and accuracy of the diagnosis of spontaneous bacterial peritonitis were 100, 94 and 95% with the cut-off value of an ascitic fluid polymorphonuclear cell count > or = 250 cells/mm3, and were 86, 98 and 96% with that value > or = 500 cells/mm3, respectively. The sensitivity, specificity and accuracy were 29, 92, 80% using ascitic fluid pH < or = 7.35 as a cut-off value; 38, 91, 82% using arterial-ascitic fluid pH gradient > or = 0.1 as a cut-off value; 52, 92, 85% using ascitic fluid lactate > or = 32 mg/dL as a cut-off value, and 38, 96, 85% using arterial-ascitic fluid lactate gradient > or = 20 mg/dL as a cut-off value.

CONCLUSIONS

The sensitivity of ascitic fluid pH or lactate was too low to aid in a diagnosis of spontaneous bacterial peritonitis in hepatitis B-related cirrhosis in spite of the acceptable specificity. An ascitic fluid polymorphonuclear cell count > or = 250 cells/mm3 or > or = 500 cells/mm3 was the major and appropriate aids in the rapid diagnosis of spontaneous bacterial peritonitis.