预测高血压病患者合并慢性肾脏病风险的列线图模型构建

Construction of a nomogram model for predicting the risk of chronic kidney disease in patients with hypertension

关键词:  高血压病  慢性肾脏病  风险因素  列线图模型

Key words:  hypertension  chronic kidney disease  risk factor  nomogram model

摘要:

[目的] 分析高血压病患者中慢性肾脏病发生的相关风险因素，依次构建预测高血压病患者合并慢性肾脏病的列线图模型。[方法] 依托天津市基层医疗卫生信息管理系统，采用整群抽样方法，选取天津市北辰区2018年1月1日—2019年12月31日常驻居民健康体检资料，符合高血压病诊断、资料保存完整的4784例高血压病患者为研究对象，其中男2193例（44.67%），平均年龄（66.38±10.10）岁；女2591例（55.33%），平均年龄（67.52±8.80）岁。采用单因素分析及多因素Logistic回归分析方法筛选高血压病患者合并慢性肾脏病的风险因素并建立列线图模型。[结果] 单因素分析结果显示，年龄、性别、白细胞计数、中性粒细胞百分比、淋巴细胞百分比、空腹血糖、糖化血红蛋白、血肌酐、血尿素氮、三酰甘油、吸烟情况、饮食偏好、运动、高血压病程、糖调节受损、2型糖尿病、冠心病、陈旧性心肌梗死、新发房颤、脂肪肝、高尿酸血症、高血压病家族史、中医体质23个因素上存在统计学差异（P<0.05），是高血压病患者合并慢性肾脏病的可疑风险因素；多因素Logistic分析显示，年龄、空腹血糖、血尿素氮、冠心病、新发房颤、高尿酸血症、高血压家族史、高血压病程、中医体质是高血压病患者合并慢性肾脏病的独立风险因素。利用以上9个风险预测指标构建了列线图模型，其一致性指数（C-index）为0.742，诊断灵敏度、特异性、准确度分别为63.18%、72.54%、84.26%，校正曲线显示模型预测效果与实际患病概率基本相同，临床决策曲线分析也显示列线图模型临床效能较好，尤其当阈值概率为0.14~0.63时，列线图模型可为患者带来临床净收益。[结论] 本研究依据高血压病患者合并慢性肾脏病发生的相关风险因素构建了列线图模型，经相关指标证实列线图模型具有较好的预测能力和临床效能，能准确、有效地预测高血压患者慢性肾脏病的发生风险，从而协助临床医师筛选高风险患者，制定针对性的干预措施，降低高血压病患者慢性肾脏病的发生率。

Abstract:

[Objective] To analyze the risk factors related to the occurrence of chronic kidney disease in patients with hypertension,and construct a nomogram model for predicting chronic kidney disease in patients with hypertension. [Methods] Relying on Tianjin's primary medical and health information management system,using a cluster sampling method,selecting the daily resident health checkup data from January 1,2018 to December 31,2019 in Beichen District,Tianjin,which is in line with the diagnosis and data preservation of hypertension. The complete 4 784 patients with hypertension were the subjects of the study,including 2 193 males(44.67%),with an average age of (66.38±10.10) years;2 591 females (55.33%),with an average age of (67.52±8.80) years. Single factor and multivariate logistic analysis methods were used to screen the risk factors of hypertension patients with chronic kidney disease and establish a nomogram model. [Results] Univariate analysis showed that age,gender,white blood cell count,percentage of neutrophils,percentage of lymphocytes,fasting blood glucose,glycosylated hemoglobin,blood creatinine,blood urea nitrogen,triglycerides,smoking status,diet preference,exercise,high are significant differences in 23 factors such as the course of blood pressure,impaired glucose regulation,type 2 diabetes,coronary heart disease,old myocardial infarction,new-onset atrial fibrillation,fatty liver,hyperuricemia,family history of hypertension,and traditional Chinese medicine constitution(P<0.05),it is a suspicious risk factor for chronic kidney disease in patients with hypertension;multivariate logistic analysis showed that age,fasting blood glucose,blood urea nitrogen,coronary heart disease,new-onset atrial fibrillation,hyperuricemia,family history of hypertension,the course of hypertension and the constitution of traditional Chinese medicine are independent risk factors for patients with hypertension and chronic kidney disease. The nomogram model was constructed using the above 9 risk prediction indicators. The consistency index(C-index) was 0.742 and the diagnostic sensitivity,specificity,and accuracy were 63.18%,72.54% and 84.26%,respectively. The calibration curve shows the model prediction. The effect is basically the same as the actual disease probability. The clinical decision curve analysis also shows that the nomogram model has better clinical performance,especially when the threshold probability is 0.14~0.63,the nomogram model can bring net clinical benefits to patients. [Conclusion] This study constructed a nomogram model based on the risk factors associated with the occurrence of chronic kidney disease in patients with hypertension. The nomogram model was confirmed by relevant indicators to have good predictive ability and clinical efficacy,and it can accurately and effectively predict hypertension. Patients with chronic kidney disease are at risk,so as to assist clinicians to screen high-risk patients and formulate targeted intervention measures to reduce the incidence of chronic kidney disease in patients with hypertension.