ASCITIC FLUID MARKERS FOR DIAGNOSING SPONTANEOUS BACTERIAL PERITONITIS IN DECOMPENSATED CIRRHOSIS

Objectives. To synthesize the evidence on the diagnostic value of ascitic fluid markers for spontaneous bacterial peritonitis in decompensated liver cirrhosis and to define their role relative to the current diagnostic criterion. Methods. Critical narrative review based on a structured search of PubMed/MEDLINE, supplemented by the reference lists of the European, American, British and Korean guidelines. The review had no registered protocol, no duplicate selection, no formal risk-of-bias assessment and no original meta-analysis. Results. An ascitic polymorphonuclear neutrophil count ≥250 cells/µL remains the diagnostic criterion for spontaneous bacterial peritonitis, irrespective of culture result and provided a secondary intra-abdominal source is excluded. The serum-ascites albumin gradient characterizes the mechanism of ascites and low ascitic protein is a risk factor; neither diagnoses infection. Leukocyte esterase reagent strips perform variably. Ascitic calprotectin and lactoferrin have the most extensive, but inconsistent, evidence base: in a meta-analysis pooling 12 and 13 studies, respectively, summary sensitivity and specificity were 94.2% and 86.0% for calprotectin and 95.4% and 89.0% for lactoferrin. Specificity is highly heterogeneous, thresholds and platforms differ, no universally validated threshold exists, and a recent prospective study did not reproduce the discriminative capacity of ascitic calprotectin. The remaining markers are exploratory, and molecular technologies increase the yield of pathogen identification without replacing the cellular criterion. Conclusions. No investigational ascitic fluid biomarker can currently replace the polymorphonuclear neutrophil count, and no externally validated universal threshold is available. Calprotectin and lactoferrin remain investigational biomarkers, not recommended by current guidelines for routine triage or treatment decisions. Prospective multicenter validation, assay standardization and platform-specific thresholds are required.