ROLE OF CHEAPER COMBINED BIOMARKERS LIKE WHITE CELL COUNT, EOSINOPHIL COUNT, PLATELET COUNT AND CRP TOGETHER IN PREDICTING 14 DAY MORTALITY OF ICU PATIENTS WITH SEPSIS

Abstract

Background and Objectives: Sepsis is a common cause of mortality and morbidity especially in resource poor settings in India.[1] It is the need of the hour to devise efcient and cheaper biomarkers to diagnose and predict prognosis in sepsis, so that appropriate antibiotic therapy can be initiated. We conducted the study to nd out the predictability of 14 day mortality in patients with sepsis by combined biomarkers and also to compare the diagnostic validity of the combined biomarkers with individual biomarkers. We enrolled eighty six patients (forty Methodology: three consecutive cases with sepsis and an equal number of age and sex matched controls without sepsis).Convenient sampling was done. Study period was for two months. All patients were followed up for a period of 14 days to assess mortality. Clinical and biochemical parameters were analysed. Fourteen day mortality rate observed was 41.9% (18/43). Area under curve obtained Results: in ROC curves suggested combined bioscore as a signicant predictor of mortality (0.724 ± 0.081). Combined bioscore of ≥ 3 had sensitivity of 77.8 % and specicity of 56 % in predicting mortality. Combination of white cell count, absolute eosinophil count and platelet count was found to be the best predictor [sensitivity of 38.8 %, specicity of 96 %, PPV of 87.5 %, NPV of 68.6 % , ( p = 0.006) ]. In multivariate logistic regression, combined bioscore was found to be an independent predictor of sepsis with a very signicant Odds Ratio of 10.661 ( 95 % CI, 2.179 – 52.165 ). The biomarkers which we Conclusion: had analysed in combination could serve as a valuable predictor of 14 day mortality in sepsis. By selecting the right antibiotic based on severity of sepsis, development of antimicrobial resistance and thus health care cost can be reduced.