Role of Initial Albumin Levels Among Other Liver Tests and Prognostic Scores in Predicting In-Hospital Mortality and Illness Severity in Critically Ill Septic Patients

Abstract

SESSION TITLE: Severe Sepsis & Septic Shock SESSION TYPE: Original Investigation Poster PRESENTED ON: Wednesday, November 1, 2017 at 01:30 PM - 02:30 PM PURPOSE: Liver test abnormalities are common in critically ill patients with sepsis, but their prognostic significance is not well defined. In this study, we aim to describe the role of initial liver test evaluation in critically ill septic patients in regards to in-hospital mortality and illness severity and compare how Albumin levels, among other markers, predict these outcomes in conjunction with other severity-of-disease scores. METHODS: We conducted a retrospective cohort study via chart review, in a single-center Intensive Care Unit (ICU) in the United States, which included all consecutive patients admitted with severe sepsis/septic shock from November 2011 to October 2013. Liver tests collected within 24 hours of admission were analyzed to determine their prognostic significance in inpatient mortality and illness severity. Univariate, followed by multivariate logistic regression and Cox regression analysis were used to estimate disease severity and mortality, respectively. RESULTS: The study included 304 severe sepsis/septic shock patients without pre-existing liver disease or coagulopathy admitted directly from the Emergency Department to the ICU. The mean APACHE-II score was 22.42 ± 7.13 and mean SAPS-II 43.53 ± 14.97. Multivariable Cox Regression Analysis showed that hypoalbuminemia (p=0.003) and elevated INR (International Normalized Ratio; p<0.0001) were independent predictors of mortality in septic patients without pre-existing liver disease. Albumin levels and in-hospital mortality were inversely exponentially correlated, with an albumin level of 1g/dL associated with 32% mortality and 4g/dL with 9% mortality. Elevated transaminases and INR were associated with higher illness severity and were independent predictors of higher APACHE-II (transaminases coefficient 1.11, p=0.007; INR coefficient 1.92, p=0.009) and SAPS-II scores (transaminases coefficient 2.59, p=0.003; INR coefficient 3.59, p=0.023). Hypoabluminemia was an independent predictor of mechanical ventilation (p=0.011), vasopressor use (p<0.001) and higher APACHE-II score (p=0.002). Albumin levels were independent predictors of mortality even after adjusting for APACHE-II or SAPS-II scores (Albumin HR 0.64 p=0.027; HR 0.62 p=0.009, respectively). CONCLUSIONS: Liver tests have important prognostic significance in critically ill septic patients. Low Albumin, elevated INR, and transaminase levels are associated with higher illness severity. Low Albumin levels on admission predict higher in-hospital mortality in critically ill patients, regardless of their APACHE-II or SAPS-II scores. CLINICAL IMPLICATIONS: Albumin levels on admission can be a simple and readily available adjunct marker to clinicians for assessing hospitalization prognosis in critically ill patients. DISCLOSURE: The following authors have nothing to disclose: Paris Charilaou, Vaia Florou, Damodar Penigalapati, Haris Rana, Capecomorin S Pitchumoni, Gregoris Komodikis No Product/Research Disclosure Information