Logistic regression analysis of the prognostic factors of patients with sepsis
- VernacularTitle:脓毒症患者预后危险因素的Logistic回归分析
- Author:
Quan FAN
;
Hong ZHANG
- Publication Type:Journal Article
- Keywords:
sepsis;
prognosis;
risk factors;
Logistic regression
- From:
Acta Universitatis Medicinalis Anhui
2014;(10):1479-1481,1501
- CountryChina
- Language:Chinese
-
Abstract:
Objective To investigate the prognostic factors in patients with sepsis and to provide the theoretical basis for clinical diagnosis and treatment. Methods A total of 141 septic patients were analyzed retrospectively. Patients were divided into two groups:those who were alive at time of ICU discharge ( survivors;n=83 ) and those who died or to give up before terminal stage ( nonsurvivors;n=58 ) . Then the clinical data were statistically ana-lyzed. Results In the death group: age, APACHE II score, blood urea nitrogen, serum creatinine, C-reactive protein were significantly higher than the survival group (P<0. 01) while albumin, pre albumin were lower than that in the survival group (P<0.01). The Logistic regression analysis illustrated: age ( > 60 years), APACHEIIscore ( >20 score) and blood urea nitrogen ( > 9 mmol/L) , serum creatinine ( > 176 μmol/L) , C-reactive protein level were independent predictors of death in patients with sepsis ( P<0 . 05 ) . Then drew the receiver oper-ating characteristic curve ( ROC curve) and analyzed the respective risk factor for predicting prognosis value of sep-sis with the help of ROC curve. Conclusion Age, high APACHE II score in early sepsis, increased urea nitro-gen, elevated serum creatinine and increased C-reactive protein levels were regarded as prognostic value indicators for adverse clinical outcomes in patients with sepsis.
