Egyptian Journal of Chest Disease and Tuberculosis (Oct 2015)

Comparison of the mortality prediction of different ICU scoring systems (APACHE II and III, SAPS II, and SOFA) in a single-center ICU subpopulation with acute respiratory distress syndrome

  • Abdelbaset Saleh,
  • Magda Ahmed,
  • Intessar Sultan,
  • Ahmed Abdel-lateif

DOI
https://doi.org/10.1016/j.ejcdt.2015.05.012
Journal volume & issue
Vol. 64, no. 4
pp. 843 – 848

Abstract

Read online

Background: Scoring systems can be used to define critically ill patients, estimate their prognosis, help in clinical decision making, guide the allocation of resources and estimate the quality of care in the ICU. Purpose: This study compared the predictive accuracy of four predictive scoring systems in the ICU. Methods: A prospective cohort study including consecutively admitted 110 adult ICU patients (88 males) with ARDS from Saudi German Hospital, Madinah, was performed from June 2013 to January 2015. The median age of the patients was 38 years, the median duration of illness before ICU admission was 6 days, and the median duration of ICU admission was 27 days. The APACHE II, APACHE III, SAPS II, and SOFA scores were calculated based on the worst values during the first 24 h of admission. Results: The actual mortality rate (27.3%) was higher than the estimated mortality rates, with the highest predicted rate of 11.3% obtained using the APACHE II. All four severity scores were significantly associated with mortality (F = 62.772, p = 0.000) and explained 83% of its variability (R2 = 0.834). However, after adjustment, only the APACHE III scoring system was a significant predictor (Beta = −0.753, p = 0.000). Three scoring systems were significantly associated with mortality (F = 42.055, p = 0.000) and explained almost 70% of its variability (R2 = 0.712), but after adjustment, only the APACHE II was a significant predictor (Beta = −0.631, p = 0.041). The combination of the severity score and mortality prediction was a significant predictor of mortality (Beta = −1.397, p = 0.000 and Beta = 0.517, p = 0.036, respectively). Conclusion: The accuracy of the studied scoring systems for predicting ICU mortality in ARDS patients is limited. The performance of the APACHE II/III scoring systems was superior to that of other systems in terms of predicting the severity and mortality, and the combination of scores improved the performance. There is a need to develop ARDS-specific scoring systems. Until a new system is developed, it is better to use the updated versions of the APACHE scoring system or a combination of all ICU scoring systems.

Keywords