Comparison of Apache IV Vs Apache II Scoring System in Predicting the Clinical Outcomes of patients in Intensive Care Unit

Abstract

Background: Prognostication of critically ill patients is an integral part of the quality of care in ICU. The use of scoring system such as Acute Physiology and Chronic Health Evaluation (APACHE) to predict risk of mortality and evaluating outcome in critically ill patients is important in modern evidence-based medicine. the aim of the study was to compare the APACHE II and APACHE IV in predicting the mortality of patients intensive care unit. Methods: A prospective descriptive was among 100 adult patients admitted irrespective of diagnosis and managed for >24hours in the 25 bedded multidisciplinary ICU of a tertiary care hospital. The APACHE II and APACHE IV scores were calculated using the online calculators, based on the worst values in the first 24hours of admission. All the study participants were followed up, to determine the observed mortality rates and length of stay of ICU which were compared with predicted mortality rates obtained from both the APACHE II and APACHE IV scoring systems. The receiver operator characteristic curves (ROC) were used to compare accuracy of the two scores using SPSS 20version. Results: There was no statistically significant difference in the estimated mortality rate of patients in ICU based on APACHE II and APACHE IV scoring system (t= 1.674) at p<0.05 level. There was a significant weak correlation between actual length of stay and estimated length of stay based on APACHE IV score (r = 0.469) at p < 0.01level. Discrimination for APACHE II and APACHE IV models were good with area under the curve of o.965 and 0.760 respectively. APACHE II was more accurate than APACEIV in this regard. Conclusion: Discrimination was more-better for APACHE II than APACHE IV model. there was a moderate correlation observed between the two models in this study.