| 摘要: |
| 目的应用多元Cox回归模型分析影响脓毒症患者的预后因素并构建判断脓毒症患者死亡危险的函数模型。方法选取在ICU住院治疗的384例脓毒症患者为研究对象，记录其入院时临床指标及住院28d的临床结局，应用多元Cox回归模型分析影响脓毒症患者的预后因素并构建其函数模型。结果384例患者住院28d时病死率61.98%（238/384）。单因素分析显示影响脓毒症预后的危险因素为急性生理与慢性健康评分（APACHEII评分）、急性肾损伤（AKI）、HDL-C、血尿素氮（BUN）、血钙（Ca2+）、外周血淋巴细胞计数（LY）和PLT及序贯器官衰竭（SOFA）评分等32个临床指标，两组比较差异均有统计学意义（均P＜0.05）。多元Cox回归模型分析，结果发现肠道术后[相对危险度（RR）=6.371，95%CI：4.266~8.766，P＜0.01]、APACHEII评分[RR=1.249，95%CI：1.175~1.328，P＜0.01]、血糖[RR=1.008，95%CI：1.045~1.133，P＜0.01]、HDL-C[RR=0.683，95%CI：0.418~0.824，P＜0.01]、BUN[RR=1.068，95%CI：1.009~1.130，P＜0.05]、Ca2+[RR=0.274，95%CI：0.141~0.533，P＜0.01]、LY[RR=0.322，95%CI：0.210~0.495，P＜0.01]、PLT[RR=0.996，95%CI：0.992~1.000，P＜0.05]和SOFA评分[RR=0.894，95%CI：0.795~1.043，P＜0.01]是影响脓毒症患者预后的独立危险因素。根据多元Cox回归模型分析结果拟合判断脓毒症患者死亡的函数模型，即h（t,X）=h0（t）exp（预后指数=6.371X1+1.249X2+1.088X3-0.683X4+1.068X5-0.274X6+0.322X7+0.996X8+0.894X9）。结论肠道术后、APACHEII评分、空腹血糖、HDL-C、BUN、Ca2+、LY、PLT和SOFA评分是影响脓毒症患者预后的独立危险因素。 |
| 关键词: 脓毒症 预后 Cox 回归 危险因素 |
| DOI：10.12056/j.issn.1006-2785.2017.39.9.2017-167 |
| 分类号: |
| 基金项目:浙江省科技厅重症医学科技创新团队项目（2011R50018）；浙江省医药卫生平台计划（2015RCB003） |
|
| Cox regression analysis of risk factors for the prognosis of sepsis |
|
NI Yin, YE Jun, WANG Jinzhu, HAN Fan, WU Aiping, SUN Renhua
|
|
Zhejiang provincial People's Hospital/Hangzhou Medical College
|
| Abstract: |
| Objective To analyze the risk factors for the prognosis of sepsis and to establish a functional model based on Cox regression. Methods A total of 384 patients with sepsis admitted at intensive care uni (ICU) from January 2010 to January 2016 were enrolled in the study. The clinical indicators at admission and outcomes at 28 days after admission were analyzed by Cox regression, and a functional model for prognosis was established. Results The case fatality rate during 28 days after admission was 61.98% (238/384). Univariate analysis showed that acute physiology and chronic health evaluation (APACHEII) score, acute kidney injury (AKI), high density lipoprotein (HDL-C), blood urea nitrogen(BUN) and blood calcium (Ca2+), peripheral blood lymphocyte count (LY), platelet count (Plt) and sequential organ failure assessment (SOFA)score were significantly
associated with the case fatality (all P＜0.05). Cox regression analysis showed that bowel surgery [(RR=6.371, 95%CI: 4.266,
8.536), P＜0.01] , blood glucose [(RR=1.008, 95%CI:1.045,1.133), P＜0.01], HDL-C [(RR=0.683, 95%CI: 0.418, 0.824), P＜0.01],
BUN [(RR=1.068, 95% CI:1.009,1.130), P ＜0.05], Ca2+ [(RR=0.274, 95% CI: 0.141,0.533), P ＜0.01], LY [(RR=0.322, 95% CI: 0.210,0.495), P＜0.01], Plt [(RR=0.996, 95%CI: 0.992,1.000), P＜0.05] and SOFA score [(RR=0.894, 95%CI: 0.795,1.043), P＜
0.01)] were independent risk factors for the prognosis of patients with sepsis. A function model for the prognosis of sepsis was established according to Cox regression analysis: h(t,X)=h0(t)exp(prognostic index=6.371X1+1 . 249X2+1. 088X3-0 . 683X4+ 1 . 068X5- 0 . 274X6+0.322X7+0.996X8+0.894X9). Conclusion Bowel surgery, blood glucose, high density lipoprotein, blood urea nitrogen, blood calcium, blood lymphocyte count and platelet count and SOFA score are independent risk factors for the prognosis of patients with sepsis. |
| Key words: Sepsis Prognosis Cox regression Risk factors |
