الفهرس Only 14 pages are availabe for public view
 |  | from | 114 |  |  |
| | | from | 114 | | |
Abstract
and Conclusion
Our study was aimed to analyze the predictive accuracy and validity of SAPS II and APACHE II in the population of patients admitted to general Suez Canal University ICU, and which one of them was more accurate in predicting patient outcome?
Thirty patients who fulfilled the inclusion criteria were included in the study represented by sixty score sheet of both Apache II and SAPS II.
History ,clinical examination ,laboratory investigation ,parameters of mechanical ventilation were obtained from the day after ICU admission.ready made computer based program was used to automatically calculate the two score ,follow up of the patients untill death or dicharge was done .
variables of SAPS II are Type of admission,Chronic diseases,Glasgow scale, Age (year), Systolic blood pressure (mmHg), Heart rate (beat/minute), Temperature (Co), Urine out put(ml /day), Serum urea, BUN (mmol /l), Serum potassium (mmol /l), Serum HCo3(mEq/l), WBC (x103/mm3/mg/dl) , If MV or CPAP: PaO2 / FiO2(mmHg) , Serum sodium (mmol/l), Bilirubin (mg/dl).
Variables of APACHE II are temperature (oC), mean arterial blood pressure (mmHg), heart rate (beats per minute), respiratory rate (respiration per minutes), if fiO2 >= 0,5 : (A-a) o2 (mmHg), if fio2 < 0,5 : pao2 (mmHg), arterial pH, serum HCO3 ( mmol/l), serum Na (mmol/l),serum K (mmol/l), serum creatinine with renal failure (mg/dl), serum creatinine without renal failure (mg/dl), hematocrit (%), Glasgow coma scale, white blood cells (x103/mm3/mg/dl), age (year), chronic organ insufficiency.
Our results showed that the age group ranged between19-52 years old with mean age of 35.5 years.
Males represented 63.3%of the sample and 46.7% of patients were admitted in ICU due to traumatic head injuries and intra cranial hemorrhage.
Death rate of APACHE II ranged from 2.9-76 with a mean of 32.3 and SAPS II ranged from (1.3-92) with a mean of (33.8) indicating little under estimation of mortality by the two score which was 40% (12 from 30 patients).
For the APACHE II, sensitivity was 58.3%, specificity was 94.4%, accuracy of correct classification was 80%, positive predicted value was 88%, and negative predicted value was 77.3%.
For SAPS II sensitivity was 67%, specificity was 89% and accuracy 80%.
Temperature, WBCs, heart rate, serum creatinine, age and chronic organ insufficiency were the best significant predictor of APACHE II death rate, but surgical causes of admission, temperature, heart rate, age, Glasgow coma score were the best significant predictor of SAPS II death rate.
ROC curves showed that the area under the curve was 0.85 for APACHE II and 0. 84 for SAPS II indicating a very good discrimination of the two models.
Both models showed relative low sensitivity and high specificity. This may be due to small number of the sample so further studies with larger sample size is recommended. It may be also due to the cut of point we choose in the statistics to determine the prediction of mortalitywhich was
0.5. Also because the two scores was not involve any data about pre-admission state and the score is done after 24 hours after admission with the worst values that may be dependent on the treatment given after admission,
This also could explain under estimation of both scores for death rate.
Our results proved that the two scores are fitting to each other and there were a strong correlation between the two score.
Both models show fair caliberation by H statisic test which was 1.39 for APACHE II and 0.77 for SAPS II with p value of 0.25 and 0.41 respectively
Mean APACHE II score was 23.1 ± 9.4, mean SAPS II score was 16.3 ± 6.8.
Mean APACHE II and SAPS II scores of the survived patients were 20.3 and 14.9 respectively , Mean APACHE II and SAPS II scores of the died patients were28.5 and 21.6 respectively with highly significant p value.
Standardized mortality rate was 1.23 for APACHE II and1.18 for SAPS II, this indicating inferior perfoermance in ICU but this may be due to the sample size so this needs more research for calculating it by larger sample size, another cut of point when calculating death rate.
Conclusion
APACHE II and SAPS II scoring systems were valid for general ICU in Suez Canal university hospital and APACHE II was more accurate in predicting patient outcome by high specificity and larger area under the ROC curve and the two models are efficient in determining the predicted death rate by high r2value.