脑深部电刺激改善帕金森病患者睡眠状态预测模型的构建及其预测价值

Construction and prediction value of model of deep brain stimulation for improvement of sleep state in patients with Parkinson's disease

目的分析帕金森病(PD)患者脑深部电刺激(DBS)术后1年睡眠障碍改善的影响因素,构建可有效预测患者术后睡眠质量改善的列线图模型。方法回顾性分析2020年1月至2022年8月于中国科学技术大学附属第一医院(安徽省立医院)神经外科接受以丘脑底核(STN)为靶点DBS(STN-DBS)治疗的333例PD患者的临床资料和问卷调查结果。根据术后1年PD睡眠障碍量表(PDSS)评分是否改善,将333例患者分为睡眠障碍改善组和未改善组。采用单因素和多因素logistic回归分析筛选影响术后1年PD患者睡眠障碍改善的因素,并建立预测患者睡眠障碍改善的列线图模型。采用受试者工作特征曲线和决策曲线分析评估预测模型的区分度和拟合性能。结果与术后睡眠障碍未改善组患者(97例)相比,睡眠障碍改善组患者(236例)术前的药物改善率、简易精神状态检查量表(MMSE)评分、汉密尔顿焦虑量表(HAMA)评分均更佳,而PDSS评分更差(均P<0.05)。进一步行单因素和多因素logistic回归分析显示,术前药物改善率(OR=0.00,95%CI:0.00~0.00,P=0.002)、术前MMSE评分(OR=0.78,95%CI:0.71~0.86,P<0.001)、术前HAMA评分(OR=1.15,95%CI:1.06~1.25,P=0.001)和术前PDSS评分(OR=1.03,95%CI:1.00~1.06,P=0.004)为PD患者术后睡眠障碍改善的独立危险因素。基于独立危险因素建立列线图模型,其受试者工作特征曲线下面积为0.78(95%CI:0.69~0.88);决策曲线分析结果显示,列线图预测值与实际观察值之间的拟合良好(C指数为0.784)。结论本研究建立的列线图模型在预测PD患者STN-DBS术后1年的睡眠质量改善方面有一定的价值。

Objective To analyze the influencing factors of sleep disorder improvement in patients with Parkinson's disease(PD)1 year after deep brain stimulation(DBS),and to construct a nomogram model that can effectively predict the improvement of patients'sleep quality after surgery.MethodsA retrospective analysis was performed on the clinical data and questionnaire survey results of 333 PD patients who received DBS(STN-DBS)targeting the subthalamic nucleus(STN)at the Neurosurgery Department of the First Affiliated Hospital of University of Science and Technology of China(Anhui Provincial Hospital)from January 2020 to August 2022.According to whether the PD Sleep Scale(PDSS)score improved 1 year after surgery,333 patients were divided into a sleep disorder improvement group and a non-improvement group.Univariate and multivariate logistic regression analyses were used to screen factors affecting the improvement of patients'sleep disorders 1 year after surgery,and a nomogram model was established to predict the improvement of patients'sleep disorders.Receiver operating characteristic(ROC)curve and decision curve analysis(DCA)were used to evaluate the discrimination and fitting performance of the prediction model.Results Compared with the patients in the group whose sleep disorder did not improve after surgery,the patients in the group whose sleep disorder improved had better preoperative drug improvement rate,Mini-Mental State Examination(MMSE)score,and Hamilton Anxiety Rating Scale(HAMA)score,while they had worse PDSS score(all P<0.05).Further univariate and multivariate logistic regression analyses showed that the improvement rate of preoperative medication(OR=0.00,95%CI:0.00-0.00,P=0.002),preoperative MMSE score(OR=0.78,95%CI:0.71-0.86,P<0.001),preoperative HAMA score(OR=1.15,95%CI:1.06-1.25,P=0.001)and preoperative PDSS score(OR=1.03,95%CI:1.00-1.06,P=-0.004)were independent risk factors for postoperative sleep disorder improvement in PD patients.A nomogram model was established based on independent risk factors,and the area under the ROC curve was 0.78(95%CI:0.69-0.88);the DCA results showed good fit between the predicted values of the nomogram and the actual observed values.Conclusion The nomogram model established in this study has certain value in predicting the improvement of sleep quality in PD patients 1 year after STN-DBS surgery.