751

Abstract

Introduction: Acute variceal hemorrhage (AVH) is a life threatening condition often requiring admission to intensive care unit and is associated with high mortality. AVH can however complicate admission in a cirrhotic patient who is admitted for other reasons. The outcome of variceal hemorrhage occurring in cirrhotics during inpatient admission is not known. Methods: California state inpatient database is a unique dataset which has information with regards to diagnosis present on admission and diagnoses made during the inpatient admission. The dataset from the year 2005 to 2009 was used for the study. We identified patients admitted with AVH and those who developed AVH during admission using ICD-9-CM codes. Our primary outcomes were in hospital mortality and requirement for blood transfusions, platelets and transjugular intrahepatic portosystemic shunt (TIPS) procedure. Chi square test was used to compare the categorical variables. We developed logistic regression model to examine the odds of mortality associated with AVH developing after admission. The model adjusted for age, gender, race and clinical characteristics. Results: There were 30,009 admissions with AVH from 2005 to 2009. Of these, 3.6% developed AVH during the admission. The mortality of patients presenting with AVH on admission was 10.2%. When AVH developed during admission, mortality was 37.9%. There was no difference in percentage of patients receiving blood (60.9% vs. 70%, p=0.997) but there was significantly higher use of platelets (16.4% vs. 9.65, p<0.001) in those who developed AVH during admission. The use of TIPS was twice more common in those who developed AVH during admission than those who presented with AVH (7.3% vs. 3.3%, p<0.001). The mean number of organ failure was higher in those who developed AVH during admission (2.2 vs. 1.0, p<0.001) On adjusted analysis, the odds of mortality were 2.1 times (95%CI 1.7-2.6) higher in AVH occurring during admission as compared to those who presented with AVH. Conclusions: AVH occurring during admission is associated with 2.1 times higher odds of mortality than those presenting with AVH. These patients are sicker and require more TIPS than those presenting with AVH.