ROLE OF THE APACHE II SCORE IN PREDICTING SURVIVAL AND LENGTH OF ICU STAY AMONG CRITICALLY ILL PATIENTS

Abstract

Objective:

To determine the predictive ability of the Acute Physiology and Chronic Health Evaluation II (APACHE II) score for ICU mortality and prolonged ICU stay among critically ill adults.

Material & Methods:

This study was a descriptive-observational design. 237 consecutive patients aged 18-65 years were included in the study. APACHE II scores were obtained using the worst physiological data recorded over the preceding 24 hours for each patient in the first 24 hours of their ICU stay. ICU mortality and the extended stay in the university were verified. Data on ICU mortality was collected, along with data on long stays within the ICU. Outcomes analysis of a stay lasting 10 days or longer was categorized as prolonged. Mann-Whitney U test was conducted for non-normally distributed variables. Spearman correlation and receiver operating characteristic analyses were also made. Exploratory logistic regression was performed to assess individual associations between APACHE II scores and study outcomes.

Results:

Average patient age was 43.84 (+12.22) years. Among 237 patients, 140 (59.1%) were male, while ICU mortality occurred in 25 (10.5%) patients. At the ICU, 20.7% (49/234) of patients had a prolonged ICU stay. Higher scores in the APACHE II at admission were associated with non-survival. The median value for non-survivors was higher at 14 [IQR 6-18] versus survivors at 5 [3-7] overall (p<0.001). Patients with longer stays had higher scores. Overall, their median was 13 [10-17] compared to 4 [2-6] in those who did not stay long; p<0.001. For mortality, APACHE II achieved AUC 0.791 (95% CI 0.682-0.899). The best threshold of mortality was computed at 13 points overall. At this cut-off, the sensitivity was 60.0% while the specificity was 92.5% when predicting the mortality of critically ill adults admitted to the ICU. Prolonged-stay discrimination reached AUC 0.906 (95% CI 0.859-0.954). An optimal sensitivity of 77.6% and a specificity of 92.0% for predicting prolonged stay in the intensive care unit (ICU) were obtained with a cut-off of 10. The higher the number of APACHES II points, the greater the chances of outcome. The adjusted mortality odds ratio for each point was 1.219. In exploratory regression, each point was normalized to an OR of 1.529 for long stay.

Conclusion:

APACHE II score was an independent predictor of poor outcome. They were linked with a longer stay in the intensive care unit and death in acutely ill adult patients in general. The score could also be used for early risk stratification and resource planning. However, clinical judgement and local validation are still important.