IDDF2018-ABS-0235 Reappraisal of the accuracy of diagnostic criteria for spontaneous bacterial peritonitis in cirrhotic patients with or without hepatocellular carcinoma: a preliminary result

Background Since 1976, the laboratory diagnosis of spontaneous bacterial peritonitis (SBP) has been established by ascetic-fluid (AF) polymorphonuclear-leucocyte (PMN) count ≥250/mm3 with or without the AF culture result in cirrhotic patients. We aimed to reevaluate whether the current cutoff count of PMN would be still optimal to diagnose SBP in cirrhotics having a hepatocellular carcinoma (HCC) or not. Methods This preliminary study included 136 consecutive patients having cirrhosis with (n=60) or without HCC (n=76) and a positive AF culture at the first exploratory paracentesis between 2007–2016 as the SBP group; and 202 cirrhotics with neither bacterascites on the first AF specimen nor any typical symptoms suggestive of peritoneal infection as the non-SBP control group. Among the SBP group, none had a clinical suspicion of secondary peritonitis and malignant ascites. Results Fever (28%) and E. coli (22%) were the most common symptom and isolated pathogen, respectively in the SBP group. Ascitic white-blood-cell and PMN counts, and lactate-dehydrogenase were significantly higher in the SBP group, whereas glucose and serum-ascites-albumin-gradient were higher in the control group (Ps Conclusions Based on our results, update or revision of the old threshold count for ascetic-PMN may be taken into consideration to be timely treated with antibiotics in not only HCC patients but cirrhotics who have established or suspected SBP, leading to prevention of undertreatment of serious infection and overtreatment of sterile ascites. Our further study of the extended cohort is ongoing.