糖尿病患者维持性血液透析生存分析及预测模型构建

Survival analysis and establishment of predictive model for diabetic patients with maintenance hemodialysis

目的:探讨合并糖尿病的维持性血液透析(MHD)患者生存时间的影响因素,构建透析后生存时间的预测列线图.方法:回顾性分析2013年~2016年186例合并糖尿病的MHD患者信息,应用COX模型筛选影响预后的危险因素,并根据COX回归结果构建死亡风险列线图.结果:186例患者1年、2年、3年累积生存率分别为88.7%,74.2%,64.5%.COX回归分析结果显示,糖尿病病程、心血管疾病、血管通路、估算的肾小球滤过率(eGFR)为影响合并糖尿病MHD患者生存时间的独立危险因素,据此建立死亡预测列线图,一致性系数为0.670(95%CI 0.604~0.737).结论:基于糖尿病病程、心血管疾病、血管通路、eGFR四个因素建立的糖尿病MHD患者透析后生存预测的列线图有助于个体化预测预后,针对性地治疗预后较差的糖尿病透析患者.

Objective:To investigate factors affect the survival time and build the nomogram to predict the survival rate for diabetic maintenance hemodialysis(MHD)patients.Methodology:Between January 2013 and December 2016,the information of 186 diabetic MHD patients was retrospectively analyzed.The independent risk factors were screened by the Cox model and the nomogram was established according to the screened factors.Results:1-year,2-year and 3-year cumulative survival rate of the 186 cases was 88.7%,74.2%,64.5%,results of the COX regression analysis showed that duration of diabetes,cardiovascular diseases,vascular access and estimated glomerular filtration rate are the independent risk factors,nomogram was established in accordance with the prediction.The consistency was 0.670(95%CI 0.604~0.737).Conclusion:The nomogram established for patients of diabetic MHD contribute to the individual prediction of survival rate and targeted treatment for patients with poor prognosis.