| Objective:To explore the risk factors influencing prognoses of critically-ill elderly patients in the surgical intensive care unit(SICU).Analysis and quantify the relationship between NT-proBNP and prognosis of SICU critical elderly patients. To study the correlation between NT-proBNP level and APACHEⅡscroing system.Methods:This research is an retrospective study on the relationship between NT-proBNP level and prognosis of elderly patients in SICU. We applicated the Medical record management system and set the "age≥65 and surgery" to retrieve 11,168 cases from January 2008 to March 2011 and finally collectted 187 cases who met the inclusion criteria. To review the medical history one by one and collect age, gender, complications, type of surgery, the APACHEⅡscore, NT-proBNP levels and the serum creatinine levelsin the first 24 hours after coming into the SICU, the SICU stay time and the outcome.1.Risk factors for mortalityTo study the impact of the clinical data on mortality, use single factor analysis to identify the possible risk factors on mortality and then apply the Binary Logistic regression analysis to find out the independent factors for mortality2.The relationship between prognosis and NT-proBNP level and APACHEⅡscore(1) Divide 187 critically ill patients into three groups according to APACHEⅡscores:APACHEⅡscore<15 points; APACHEⅡscore≥15 points,<25 points; APACHEⅡscore≥25 points. Compare their NT-proBNP levels, SICU stay time and mortality;(2) Divide 187 critically ill patients into survivors group and nonsurvivors group according to the outcome, compare their NT-proBNP levels and APACHEⅡscore;(3) The relationship between NT-proBNP level and APACHEⅡscore Do partial correlation analysis for NT-proBNP levels and APACHEⅡscore to see their relationship.(4) Compare predictive value of NT-proBNP levels and APACHEⅡscore on prognoses use ROC P<0.05 was considered statistically significant difference, all the statistical analysis were calculated by SPSS 17.0 software.Results:Twenty patients died and the mortality was 39.8%. In 187 patients, the mean plasma NT-proBNP levels(pg/ml) were 533.5(195.5-1265.0), and that in survivor group was significantly lower than non-survivor group[467.2(191.1-1029.0) vs 2497.0(1069.0-5032.0), P<0.05]. The mean NT-proBNP level(pg/ml)in male patients was lower than that in females[409.2 (163.2-1168.0) vs 565.45 (227.55-1643.5) P<0.05]. There was correlation between NT-proBNP levels and APACHEⅡscore(r=0.530, P=<0.001). Binary Logistic regression analysis showed that the NT-proBNP level and the APACHEⅡscore were independently associated SICU mortality[Odds ratio(OR)5.131 [95%confidence interval (95%CI) 1.081-24.347] and 6.450 (95%CI 1.462-28.446).Areas under the receiver operating characteristic curves(ROC curves)of NT-proBNP and APACHEⅡscore were 0.79(95%CI 0.67-0.91) and 0.92(95%CI 0.87-0.96), respectively. The best cut-off level of NT-proBNP for predicting SICU mortality was 957.55pg/ml,with 85.0% sensitivity and 73.7% specificity.Conclusion:The NT-proBNP and APACHEⅡscore can be used as a short-term survival predictor in elderly patients in the SICU.
|
PDF Full Text Request