Study of spectrum of sepsis and prediction of its outcome in patients admitted to ICU using different scoring systems

ABSTRACT

Background: Although sepsis is one of the leading causes of mortality in hospitalized patients, information regarding early predictive factors for mortality and morbidity is limited. The aim was to identify reliable and early prognostic variables predicting mortality in patients admitted to ICU with sepsis.Methods: Patients fulfilling the Surviving Sepsis Campaign 2012 guidelines criteria for sepsis within the ICU were included over two years. Apart from baseline haematological, biochemical and metabolic parameters, APACHE II, SAPS II and SOFA scores were calculated on day 1 of admission. Patients were followed till death or discharge from the ICU. Chi-square test, student t-test, receiver operating curve analyses were done.Results: 100 patients were enrolled during the study period. The overall mortality was 35% (68.6% in males and 31.4% in females). Mortality was 88.6% and 11.4% in patients with septic shock and severe sepsis and none in the sepsis group, respectively. On multivariate analysis, significant predictors of mortality were APACHE II score greater than 27, SAPS II score greater than 43 and SOFA score greater than 11 on day the of admission. On ROC analysis APACHE II had the highest sensitivity (92.3%) and SAPS II had the highest specificity (82.9%).Conclusions: All three scores performed well in predicting the mortality. Overall, APACHE II had highest sensitivity, hence was the best predictor of mortality in critically ill patients. SAPS II had the highest specificity, hence it predicted improvement better than death. SOFA had intermediate sensitivity and specificity.