精神分裂症长期住院患者QTc间期延长危险因素分析及基于Lasso回归建立风险预测模型  

Analysis of risk factors for prolonged QTc interval in long-term hospitalized patients with schizophrenia and establishment of a risk prediction model based on Lasso regression

作　　者：郭蕊 韩勇[2] 刘涛[1] Guo Rui(The Second Affiliated Hospital of Xinxiang Medical University,Xinxiang 453002,Henan,China)

机构地区：[1]新乡医学院第二附属医院(河南省精神病医院)电生理科,河南新乡453002 [2]新乡医学院第二附属医院(河南省精神病医院)中心实验室,河南新乡453002

出　　处：《牡丹江医学院学报》2024年第5期76-80,75,共6页Journal of Mudanjiang Medical University

基　　金：河南省科技发展计划项目(232102311031)。

摘　　要：目的基于Lasso回归建立风险预测模型分析精神分裂症长期住院患者QTc间期延长的危险因素。方法回顾性收集本院2023年1月至2023年12月期间收治的精神分裂症长期住院患者256例的临床资料,根据QTc间期情况将其分为延长组(n=46)和非延长组(n=210),分析两组的一般资料,通过Lasso回归、Logistic回归模型分析影响精神分裂症长期住院患者QTc延长的危险因素,并根据危险因素构建预测精神分裂症长期住院患者QTc间期延长发生的模型;绘制受试者工作特征(ROC)曲线分析危险因素及综合指数预测精神分裂症长期住院患者QTc间期延长的AUC、敏感度及特异度。结果延长组的女性比例高于非延长组,BMI高于非延长组,空腹血糖(FPQ)高于非延长组,血钾(K)低于非延长组,差异有统计学意义(P<0.05)。性别、BMI、FPG、K为影响精神分裂症长期住院患者QTc间期延长发生的独立危险因素(P<0.05)。根据Logisitc回归结果,构建相关预测模型:模型公式=1.175×年龄-0.568×BMI-0.715×FPG+2.312×K-10.144。建立ROC曲线分析该模型预测精神分裂症长期住院患者QTc间期延长发生的价值,结果显示,性别、BMI、FPG、K、联合预测的AUC分别为0.599、0.741、0.701、0.898、0.954,敏感度分别为0.674、0.739、0.652、0.805、0.862,特异性分别为0.524、0.652、0.657、0.915、0.935。成对Z检验显示,联合的AUC高于单个指标(P<0.05)。结论女性精神分裂症患者,尤其是那些BMI指数偏高、FPG水平较高以及K水平较低的患者,在长期住院治疗期间,出现QTc间期延长的风险会显著增加。在预测这类患者QTc间期延长发生风险时,性别、BMI、FPG和K这四个指标均表现出良好的预测效能。而且,当这四个指标被联合使用时,它们的预测效能能够达到最佳水平。Objective To establish a risk prediction model based on Lasso regression,and to analyze the risk factors for prolonged QTc interval in long-term hospitalized patients with schizophrenia.Methods A retrospective collection of clinical data was conducted on 256 long-term inpatients with schizophrenia admitted to our hospital from January 2023 to December 2023.They were divided into prolonged group(n=46)and non-prolonged group(n=210)based on the QTc interval.The general information of the two groups was analyzed.Lasso regression and logistic regression models were used to analyze the risk factors affecting QTc prolongation in long-term inpatients with schizophrenia and a predictive model for the occurrence of QTc interval extension was constructed based on the risk factors.Receiver operating characteristic(ROC)curves were plotted,and the AUC,sensitivity,and specificity of the risk factors and composite were analyzed in predicting QTc interval prolongation.Results The proportion of females,BMI,fasting plasma glucose(FPG),and potassium(K)levels were significantly higher in the prolonged group compared to the non-prolonged group(P<0.05).Gender,BMI,FPG and K levels were independent risk factors for the occurrence of QTc interval prolongation in long-term hospitalized patients with schizophrenia(P<0.05).Base on the Logistic regression results,a relevant prediction model was constructed:model formula=1.175×age-0.568×BMI-0.715×FPG+2.312×K-10.144.The established ROC curve was used to analyze the value of QTc interval prolongation in long-term hospitalized patients with schizophrenia,the results showed that gender,BMI,FPG,K levels and combined prediction of AUC were 0.599,0.741,0.701,0.898 and 0.954,respectively,with sensitivity of 0.674,0.739,0.652,0.805,and 0.862 and specificity of 0.524,0.652,0.657,0.915 and 0.935,respectively.Paired Z-tests showed that the combined AUC was superior to a single indicator(P<0.05).Conclusion Female patients with schizophrenia,especially those with high BMI,high FPG levels and low K level

关 键 词：精神分裂症 QTC间期 危险因素 Lasso回归 ROC曲线分析 

分 类 号：R749.3[医药卫生—神经病学与精神病学]