ABSTRACT
OBJECTIVE: To evaluate the utility of B-mode gray-scale histogram analysis in the management of patients with infected cirrhotic ascites. METHODS: A total of 97 patients (mean(SD) age : 66.8(14.2) years, 50.5% were males) diagnosed with cirrhotic ascites were included in this non-interventional study. Paracentesis for ascitic fluid analysis [culture tests, white blood cell count, albumin and protein levels, serum ascites albumin gradient (SAAG)] and gray-scale histogram analysis for ascites/subcutaneous echogenicity ratio (ASER) were performed at baseline in each patient and on Day 2 and Day 5 of treatment in patients with infected ascites. Receiver operating characteristics (ROC) curve was plotted to determine performance of ASER in identification of antibiotic resistance with calculation of area under curve (AUC) and ideal cut-off value of % change in ASER to detect antibiotic resistance. RESULTS: Treatment was associated with a significant decrease in median (min-max) ASER [from 0.005(0.0002-0.02) at baseline to 0.003(0.0001-0.01) on Day 2 and 0.0005(0.0001-0.009) on Day 5] and ascitic fluid polymorphonuclear leukocyte (PMNL) count [from 600(300-2200) at baseline to 350(50-1250) on Day 2 and 100(50-1100) on Day 5] (p<0.001 for each). ROC analysis revealed that less than 38% reduction in ASER [AUC: 0.923, 95% CI (0.797-0.982), p<0.001] was a potential marker of antibiotic resistance with a sensitivity of 90.9% and a specificity of 95.0%. CONCLUSIONS: In conclusion, our findings emphasize potential utility of gray-scale histogram based quantitative analysis of ascitic fluid echogenicity as an adjunct non-invasive method in the assessment of treatment response and early recognition of treatment failure in patients with infected ascites.
