SAPS II or APACHE II is better to predict mortality in patients admitted at ICU

Selecting patients to admit at the ICU is crucial. The SAPS II and APACHE II scores have been used to choose appropriate patients to the ICU particularly in resource limited setting. However, which score is the best predictor is still debating. To compare both scores in regards of ICU mortality prediction. This was a retrospective cohort study at xx Hospital between Jan 1, 2016 to Jan 1, 2017. The inclusion criteria were consecutive adult patients who admitted and treated at the ICU. Those patients who admitted at the ICU for procedures such as hemodialysis were excluded. Clinical factors including SAPS II and APACHE II were studied. The primary outcome was death at the ICU. Independent factors associated with mortality were analyzed by multivariate logistic regression analysis. For significant scores to predict mortality, a receiver operating characteristic (ROC) curve was executed to calculate diagnostic properties of each cut off points. During the study period, there were 201 eligible patients. Of those, 98 patients (48.76 %) died. Those who died had higher average SAPS II score and APACHE score than those who were survived significantly (49.76 vs 30.61; p value < 0.001; 23.88 vs 20.62; p value 0.016). There were three independent factors associated with mortality including SAPS II score, male sex, and sepsis with adjusted odds ratio (95 % CI) of 1.04 (1.02, 1.07), 2.10 (1.08, 4.06), and 2.33 (1.09, 4.97), respectively. The SAPS II score of more than 27 gave sensitivity of 90.82 %, specificity of 16.50 %, and the area under ROC curve of 73.18 %. SAPS II showed better performance than APACHE II in this population. Further studies are needed to confirm the results of this study.