天津医药 ›› 2023, Vol. 51 ›› Issue (6): 658-661.doi: 10.11958/20221288

• 应用研究 • 上一篇    下一篇

帕金森病患者冻结步态列线图预测模型的构建

季莉莉1(), 许元丰1, 史红娟1, 陈文亚1,, 刘良卿2   

  1. 1 江苏大学附属武进医院/常州市武进人民医院神经内科(邮编213000)
    2 江苏大学附属武进医院/常州市武进人民医院影像科
  • 收稿日期:2022-08-17 修回日期:2022-12-20 出版日期:2023-06-15 发布日期:2023-06-20
  • 通讯作者: E-mail:wmyjsk.love@163.com
  • 作者简介:季莉莉(1980),女,副主任医师,主要从事神经变性疾病的诊疗方面研究。E-mail:jill188cz@163.com
  • 基金资助:
    第十五批“六大人才高峰”高层次人才选拔培养资助项目(YY-027);常州市科技计划项目(CJ20210011)

Construction of a nomogram predictive model for freezing of gait in patients with Parkinson's disease

JI Lili1(), XU Yuanfeng1, SHI Hongjuan1, CHEN Wenya1,, LIU Liangqing2   

  1. 1 Department of Neurology, Wujin Hospital Affiliated to Jiangsu University / Changzhou Wujin People's Hospital, Changzhou 213000, China
    2 Department of Imaging, Wujin Hospital Affiliated to Jiangsu University / Changzhou Wujin People's Hospital, Changzhou 213000, China
  • Received:2022-08-17 Revised:2022-12-20 Published:2023-06-15 Online:2023-06-20
  • Contact: E-mail:wmyjsk.love@163.com

摘要:

目的 分析帕金森病(PD)患者冻结步态(FOG)的发生率和危险因素,构建定量列线图预测模型并进行验证。方法 208例PD患者根据是否存在FOG分为FOG组(98例)和无FOG组(110例),分别采用单因素和多因素Logistic回归分析筛选FOG的危险因素,并建立列线图模型。结果 FOG组患者年龄、PD首发年龄、PD病程、帕金森病问卷(PDQ39)评分、汉密尔顿抑郁量表17项(HAMD-17)和焦虑量表14项(HAMA-14)评分、快速眼球运动睡眠行为障碍筛查问卷(RBDSQ)和匹兹堡睡眠质量指数(PSQI)评分明显大于无FOG组,视空间功能障碍比例和改良Hoehn-Yahr(H-Y)分级严重程度高于无FOG组,而简易精神状态检查(MMSE)评分低于无FOG组(P<0.05)。多因素Logistic回归分析结果显示,较高的RBDSQ评分(OR=2.724,95%CI:1.458~5.090)、PDQ39评分(OR=7.358,95%CI:2.448~22.078)、H-Y分级(OR=4.272,95%CI:1.886~9.673)和有视空间功能障碍(OR=2.134,95%CI:1.349~3.376)是PD患者发生FOG的独立危险因素。R软件建立列线图模型,总分120分。受试者工作特征(ROC)曲线计算模型预测FOG的曲线下面积(AUC)为0.867(95%CI:0.810~0.935),提示模型预测效能较好。Hosmer-Lemeshow检验显示,模型拟合优度较好(χ2=2.635,P=0.642)。校准曲线显示模型预测概率与实际发生率有较好的一致性。临床决策曲线显示模型的获益性尚可。结论 PD患者有较高的FOG发生率,RBDSQ评分、PDQ39评分、视空间功能障碍和H-Y分级是其独立影响因素;建立的列线图模型预测FOG的效能较好。

关键词: 帕金森病, 冻结步态, 危险因素, 列线图

Abstract:

Objective To investigate and analyze the incidence and risk factors of freezing of gait (FOG) in patients with Parkinson's disease (PD), construct and verify the quantitative nomogram predictive model. Methods A total of 208 PD patients were retrospectively summarized. According to clinical symptoms, 98 patients were diagnosed as the FOG group and 110 patients were diagnosed as the non-FOG group. Univariate and multivariate Logistic regression analysis were used to screen risk factors of FOG, and the nomogram model was established. Results Univariate comparison showed that age, initial age of PD, course of PD, scores of PD questionnaire (PDQ39), Hamilton depression scale (HAMD-17) and anxiety scale (HAMA-14), rapid eye movement sleep behavior disorder screening questionnaire (RBDSQ) and Pittsburgh sleep quality index (PSQI) were significantly higher in the FOG group than those in the non-FOG group. The visuospatial dysfunction and modified Hoehn-Yahr (H-Y) grade were significantly higher in the FOG group than those in the non-FOG group, while the score of Mini Mental State Examination (MMSE) was less in the FOG group (P<0.05). Multivariate Logistic regression analysis showed that higher RBDSQ score (OR=2.724, 95%CI: 1.458-5.090), PDQ39 score (OR=7.358, 95%CI: 2.448-22.078), visuospatial dysfunction (OR=2.134, 95%CI: 1.349-3.376) and H-Y grade (OR=4.272, 95%CI: 1.886-9.673) were independent risk factors to FOG in PD patients. R software was used to establish the nomogram model, and total score was 120. The area under the curve (AUC) of the model for predicting FOG by receiver operating curve (ROC) was 0.867 (95%CI: 0.810-0.935, P<0.001), suggesting that the predictive efficiency of the model was good. Hosmer - lemeshow test showed that the goodness of fit of the model was good (χ2=2.635, P=0.642). The calibration curve showed that the predictive probability of the model was in good agreement with the actual incidence. The decision curve showed that the benefit of the model was acceptable. Conclusion Patients with PD have a high incidence of FOG. RBDSQ score, PDQ 39 score, visuospatial dysfunction and H-Y grade are independent risk factors. The established nomogram model is effective in predicting FOG.

Key words: Parkinson's disease, freezing of gait, risk factors, nomogram

中图分类号: