Low Absolute Eosinophil Count Predicts In-Hospital Mortality in Cirrhosis With Systemic Inflammatory Response Syndrome.

Introduction Chronic liver disease (CLD) or Cirrhosis is one of the most common causes of morbidity as well as mortality. Child-Turcotte-Pugh (CTP) score and the model for end-stage liver disease (MELD) are useful to assess the long-term prognosis of a patient with CLD. When a patient with CLD is admitted with an acute illness leading to systemic inflammatory response syndrome (SIRS), these scores may not be reliable to predict the short-term prognosis and survival. Absolute eosinophils count (AEC) allows the rapid identification of patients at increased risk for sepsis-related mortality in patients. Methods This was a cross-sectional study conducted among patients in a tertiary care hospital in South India during a period of one and a half years between October 2018 and April 2020. Cirrhotic patients with SIRS aged between 16 years and 80 years were included in the study. AEC was measured as a part of automated complete blood counts. Patient demographics, lab parameters, and outcomes in terms of mortality were studied. Continuous variables were expressed as mean ± SD/median and categorical variables were expressed in frequency. Receiver operating characteristic (ROC) curve analysis was used to find an ideal cutoff for AEC in predicting hospital mortality. Multi-variate Cox regression analysis was performed to find predictors of mortality. Results A total of 100 patients who fit the pre-determined criteria for cirrhosis with SIRS were enrolled in the study. Sixteen (16%) patients died at the end of the study while 84 (84%) were alive. Using a ROC curve, the area under the curve (AUC) was 0.716 with 95% CI of AUC (0.564-0.867), the p-value was found to 0.006, a cut-off of eosinophil count of 198.5 cells/uL was found to be the cut-off for the prediction of in-hospital mortality in this subset of patients with cirrhosis and sepsis with SIRS, with a sensitivity of 75% and specificity of 38.1%. In a multi-variate Cox regression analysis, only age (hazard ratio {HR}: 1.175, 95%CI, 1.084 to 1.275, p<0.001) , CRP (HR : 1.008, 95%CI, 1.00 to 1.015, p=0.042) values, total leukocyte counts (TLC) (HR: 1.226, 95%CI, 1.116 to 1.346, p<0.001) and AEC (HR: 0.993, 95%CI, 0.987 to 0.999, p=0.016) were found to be statistically significant independent predictors of mortality. Conclusions The presence of eosinopenia may be considered as an in-expensive warning biomarker for poorer clinical outcomes in the form of in-hospital mortality in hospitalized cirrhotic patients. Other biomarkers such as CRP and TLC could also play a role both independently and in conjunction with AEC to predict outcomes and mortality in cirrhotic patients with sepsis and SIRS.