老年慢性心力衰竭患者发生低三碘甲状腺原氨酸综合征风险列线图模型的建立

Developing the Risk Nomogram Model of Low Triiodothyronine Syndrome in Elderly Patients with Chronic Heart Failure

【目的】本研究探讨老年慢性心力衰竭(CHF)患者发生低三碘甲状腺原氨酸综合征(LTS)的临床特点及可能的危险因素,建立预测LTS的列线图模型。【方法】回顾性分析老年CHF患者291例,根据患者入院后检测的甲状腺功能,将其分为LTS组107例和非LTS组184例。收集所有研究对象的一般信息资料和生化指标,同时对可能引起LTS的危险因素中的数值变量(除年龄)均采用中位数进行分组。采用卡方检验对各变量进行统计分析,多因素回归模型分析老年CHF患者合并LTS的独立危险因素并建立列线图模型。采用C指数、校准曲线及受试者工作特征(ROC)曲线对预测模型进行评估。【结果】血肌酐(Scr)、C-反应蛋白(CRP)、白介素-6(IL-6)及纽约心脏病协会(NYHA)心功能分级与LTS发生呈正相关(OR值分别为1.893、2.356、1.021、1.815,P<0.05),而血清白蛋白(Alb)与LTS呈负相关(OR=0.412,P<0.05),即血清Alb水平越低,LTS越容易发生,将上述具有统计学意义的变量引入列线图后,其C指数为0.807[95%CI=(0.757,0.856)];内部验证的校准曲线表明此预测模型的校准度良好;ROC曲线分析显示此预测模型的区分度良好。【结论】老年CHF患者血清Scr、CRP、IL-6、Alb及心功能分级可能是LTS发病的危险因素;血清Alb可能是LTS发病的保护因素。基于以上多种危险因素构建的预测老年CHF患者发生LTS的列线图模型,具有较好的区分度和准确度,可为临床个体化防治提供指导。

【Objective】The clinical characteristics and the possible risk factors were explored for the low triiodothyronine syndrome(LTS)in elderly patients with chronic heart failure(CHF),and the risk prediction model of LTS was established.【Methods】For this research,291 elderly patients with CHF were selected as sample and retrospectively reviewed was used as analytical method.According to the thyroid function of admission patients,two groups including LTS group(n=107)and normal thyroid function group(n=184)were divided.The general information and biochemical indicators of patients were collected and recorded,and the risk factors for LTS were assigned scores,in which numerical variables(except age)were grouped by median.The chi-squared test was used for statistical analysis of each variable,and multivariate regression model was used to analyze the independent risk factors of LTS in elderly patients with CHF,then the nomog ram model for LTS was established based on the results from the final regression analysis.Furthermore,the prediction model was evaluated using C-index,calibration curve and receiver operating characteristic(ROC)curve.【Result】Serum creatinine(Scr),C-reactive protein(CRP),interleukin-6(IL-6)and the cardiac functional grading according to the New York Heart Association(NYHA)were positively correlated with LTS(OR values were 1.893,2.356,1.021 and 1.815,respectively,P<0.05),serum Albumin(Alb)was negatively correlated with LTS(OR=0.412,P<0.05).This means that,the LTS was easily occurred as the serum Alb level declined.When introduce the above statistically significant variables into nomogram obtained the C index with 0.807[95%CI=(0.757,0.856)].The calibration curve verified by internal verification showed that the calibration degree of this prediction model was well calibrated.ROC curve analysis showed that the prediction model was well differentiated.【Conclusion】The Scr,CRP,IL 6,Alb and cardiac functional grading in elderly CHF patients may be risk factors for incidence of LTS,while serum Alb may be a protective factor for LTS.Based on the above risk factors,the nomogram model for predicting the occurrence of LTS in elderly CHF patients had good differentiation and accuracy,and can provide guidance for clinical individualized prevention and treatment.