Introduction: Using the disease severity scoring system can guide the physician in objective assessment of disease outcome and estimation of treatment success. We studied the APACHE II disease severity scoring system. Using this system, the patients can be prioritized for receiving appropriate care and also more clarity is provided for the physician in describing the final disease outcome for the patient's relatives. In the current study, we aimed to determine the relation between the APACHE II score of the patients in surgical Intensive Care Unit (ICU) of Hazrat-Rasul hospital with patients' mortality rate.

Methods: In this study, 300 patient records were assessed and the APACHE II scores calculated. There was no exclusion criteria based on age, sex or history of any disease. The study was cross-sectional and retrospective. Patients' survival was recorded. Statistical analysis was performed with SPSS software version 11.5. Clinical measures were calculated as mean ± standard deviation (SD) or percentage.

Results: In the survivors, the average (±SD) APACHE II score was 11.75±6.9, and in non-survivors it was 22.06±7.9, and the difference between them was statistically significant. Therefore, the rise in APACHE II score was related with increase in mortality. The mortality rate in patients with scores less than 15 was 8%, in those with scores of 16-19 was 39%, in scores of 20-30 was 59%, and in patients with scores higher than 30 was 100%.

Conclusion: The APACHE II scoring system has been successful in predicting the mortality rate of the patients in this ICU. Comparison of our findings with the previous studies' results showed that in patients with scores less than 15, the observed mortality rate was comparable with the standard range mentioned in the textbooks, but in patients with scores higher than 15, the observed mortality was significantly higher than the standard limit. This difference was about 20% for patients with scores of 16-19, about 39% in those with scores of 20-30, and about 25% for patients with scores more than 30.

Hakim Research Journal 2008 11(1): 22- 28.