부산대학교 도서관

BACKGROUND AND AIM:: The aim of this study is to assess paracentesis utilization and outcomes in hospitalized adults with cirrhosis and ascites. METHODS:: The 2011 Nationwide Inpatient Sample was used to identify adults, non-electively admitted with diagnoses of cirrhosis and ascites. The primary endpoint was in-hospital mortality. Variables included patient and hospital demographics, early (Day 0 or 1) or late (Day 2 or later) paracentesis, hepatic decompensation, and spontaneous bacterial peritonitis. RESULTS:: Out of 8 023 590 admissions, 31 614 met inclusion criteria. Among these hospitalizations, approximately 51% (16 133) underwent paracentesis. The overall in-hospital mortality rate was 7.6%. There was a significantly increased mortality among patients who did not undergo paracentesis (8.9% vs 6.3%, P < 0.001). Patients who did not receive paracentesis died 1.83 times more often in the hospital than those patients who did receive paracentesis (95% confidence interval 1.66–2.02). Patients undergoing early paracentesis showed a trend towards reduction in mortality (5.5% vs 7.5%) compared with those undergoing late paracentesis. Patients admitted on a weekend demonstrated less frequent use of early paracentesis (50% weekend vs 62% weekday) and demonstrated increased mortality (adjusted odds ratio 1.12 95% confidence interval 1.01–1.25). Among patients diagnosed with spontaneous bacterial peritonitis, early paracentesis was associated with shorter length of stay (7.55 vs 11.45 days, P < 0.001) and decreased hospitalization cost ($61 624 vs $107 484, P < 0.001). CONCLUSION:: Paracentesis is under-utilized among cirrhotic patients presenting with ascites and is associated with decreased in-hospital mortality. These data support the use of paracentesis as a key inpatient quality measure among hospitalized adults with cirrhosis. Future studies are needed to investigate the barriers to paracentesis use on admission.