Abstract :
Background: Ascitic fluid PMN count remains an crucial tool in diagnosing spontaneous bacterial
peritonitis (SBP), but may not be done promptly in some situations. A few studies have shown the accuracy of reagent strip test for diagnosing SBP. This is the first study in Thailand to evaluate the use of reagent strip, which is different from that used in previous studies, for diagnosing SBP and determine ascitic fluid protein content.
Patients and Methods: One hundred and seventeen ascitic fluid samples of 72 cirrhotic patients were
analyzed by PMN count and 113 samples used for protein measurement. Leukocyte esterase strip using for diagnosis of SBP, and protein strip using for detection of ascites protein lesser than <1 g/dl compared to the corresponding routine measurement.
Results: Twelve of 117 samples had ?250 PMN/ml, while 7 of these were diagnosed as SBP by reagent
strip. There was no samples diagnosed as SBP by reagent strip in the group which ascitic fluid PMN count was <250 PMN/ml. The sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) for diagnosis of SBP by reagent strip were 58, 100, 100 and 95% respectively. Thirty-three of 113 samples had protein content <1 g/dl of protein, while 18 of these were diagnosed correctly by reagent strip. Of the remaining 80 samples that had protein content ?1 g/dl, 57 samples were diagnosed correctly by reagent strip. The sensitivity, specificity, PPV and NPV for detecting ascites protein <1 g/dl using reagent strip were 55, 71, 44 and 79% respectively.
Conclusions: The reagent strip test can be used for prompt diagnosis of SBP with specificity and PPV
of 100%. The clinical application for protein measurement is not as reliable as conventional laboratory because of low PPV and NPV.
[Thai J Gastroenterol 2005; 6(2): 67-71] |
file/Thai-Journal-of-gastroenterology-vol-6-no-2-598449.pdf