[The use and establishment of modified acute physiology and chronic health evaluation II score in emergency patients].

OBJECTIVE

To simplify the acute physiology and chronic health evaluation II (APACHE II) score to make it more suitable for emergency patients.

METHODS

Medical records of all the dead patients in the intensive care unit (ICU) from January 2003 to October 2005 were retrospectively analyzed. Those dead patients whose clinical data were completed for study were chosen as the experimental group (153 cases). Patients who survived in the same ICU at the same period were chosen as cohort for comparison. The worst scores within 24 hours of admission were selected for comparison between survivors and non-survivors to establish the modified APACHE II by linear regression. Then the new score was tested in the emergency patients who needed medical care from January to October 2006, and the accuracy of the prognostic value was evaluated.

RESULTS

Tested by the Lemeshow-Hosmer C-statistic and calibration curve analysis, the modified APACHE II score showed a good clustering data and the high calibration of the potential death of the nonsurvivors (chi2=8.853, P=0.335). If 16 being the forecasting point, the modified APACHE II score showed that sensitivity of the model in forecasting the dead patients from the emergency patients was 86.8% with an exclusivity of 81.6% and accuracy of 81.8%. The scores of the modified APACHE II paralleled with severity of the patients conditions, suggesting that the patients with lower scores were inclined to be discharged early or with a shorter period of hospitalization, while those with higher scores, the possibility of prolonged hospitalization or even death might ensue. In another words, the higher scores the higher risk of death.

CONCLUSION

The modified APACHE II score could be conveniently used in emergency patients with high reliability.