切换至 "中华医学电子期刊资源库"

中华脑科疾病与康复杂志(电子版) ›› 2026, Vol. 16 ›› Issue (02) : 84 -91. doi: 10.3877/cma.j.issn.2095-123X.2026.02.004

临床研究

脑卒中后肢体运动功能恢复风险预测模型的构建与评估
满逸迪1, 李军2,()   
  1. 1250035 济南,山东中医药大学康复医学院
    2100068 北京,中国康复研究中心脊柱与脊髓神经功能重建科
  • 收稿日期:2025-04-24 出版日期:2026-04-15
  • 通信作者: 李军

Key influencing factors and risk prediction nomogram model construction and evaluation of limb motor function recovery after stroke

Yidi Man1, Jun Li2,()   

  1. 1Rehabilitation Medicine College, Shandong University of Traditional Chinese Medicine, Ji'nan 250355, China
    2Surgical Rehabilitation and Treatment of Central Nervous Injuries, China Rehabilitation Research Center, Beijing 100068, China
  • Received:2025-04-24 Published:2026-04-15
  • Corresponding author: Jun Li
引用本文:

满逸迪, 李军. 脑卒中后肢体运动功能恢复风险预测模型的构建与评估[J/OL]. 中华脑科疾病与康复杂志(电子版), 2026, 16(02): 84-91.

Yidi Man, Jun Li. Key influencing factors and risk prediction nomogram model construction and evaluation of limb motor function recovery after stroke[J/OL]. Chinese Journal of Brain Diseases and Rehabilitation(Electronic Edition), 2026, 16(02): 84-91.

目的

分析影响脑卒中患者肢体运动功能恢复的关键因素并构建列线图预测模型。

方法

回顾性分析自2018年8月至2023年8月在中国康复研究中心神经康复科、老年康复科接受康复治疗的289例脑卒中后肢体运动功能障碍患者的临床资料。根据美国国立卫生研究院卒中量表(NIHSS)评分是否改善将所有患者分为改善组(前后差值>0分)和未改善组(前后差值≤0分);采用单因素分析和多因素Logistic回归法分析脑卒中患者肢体运动功能恢复的影响因素;基于多因素Logistic回归分析结果构建列线图预测模型,采用Hosmer-Lemeshow拟合优度检验评价模型拟合度,通过绘制受试者工作特征(ROC)曲线、校准曲线和决策曲线(DCA)评价模型的预测效能。

结果

289例患者中242例纳入改善组,47例纳入未改善组。2组患者的年龄、病程、康复频率及吸烟、高血压病史、高同型半胱氨酸血症占比比较,差异均有统计学意义(P<0.05)。多因素Logistic回归分析结果显示,年龄(OR=8.348)、病程(OR=9.161)、吸烟(OR=7.192)、高血压病史(OR=8.314)、高同型半胱氨酸血症(OR=8.508)及康复频率(OR=0.142)是脑卒中患者肢体运动功能恢复的独立影响因素(P<0.05)。ROC曲线分析结果显示,基于上述6个影响因素所构建的列线图模型预测效能良好[AUC=0.863 (95%CI:0.821~0.905)];校准曲线结果显示,预测模型校准度较好;DCA结果显示,模型的阈值概率在10%~50%时具有较高的正向净获益。

结论

患者的年龄、病程、吸烟、高血压、高同型半胱氨酸血症及康复频率是脑卒中患者运动功能障碍康复疗效的独立影响因素,基于此建立的预测模型具有良好的表现及临床使用价值。

Objective

To analyse the key factors influencing the recovery of lower limb motor function in stroke patients and construct a nomogram prediction model.

Methods

A retrospective analysis was conducted on the clinical data of 289 patients with limb motor dysfunction after stroke, who received rehabilitation treatment in the Department of Neurorehabilitation and Geriatric Rehabilitation of China Rehabilitation Research Center from August 2018 to August 2023. All patients were divided into the improved group (pre-post score difference>0) and the non-improved group (pre-post score difference≤0) according to the changes in the National Institutes of Health stroke scale (NIHSS) scores. Univariate analysis and multivariate Logistic regression analysis were adopted to explore the influencing factors for the recovery of limb motor function in stroke patients. Based on the results of multivariate Logistic regression analysis, a nomogram prediction model was constructed. The Hosmer-Lemeshow goodness-of-fit test was used to evaluate the model fitness. The receiver operating characteristic (ROC) curve, calibration curve and decision curve analysis (DCA) were plotted to assess the predictive performance of the model.

Results

Of the 289 patients, 242 were enrolled in the improved group and 47 in the non-improved group. There were statistically significant differences between the two groups in age, disease course, rehabilitation frequency, as well as the proportions of smoking history, hypertension history and hyperhomocysteinemia (P<0.05). Multivariate Logistic regression analysis indicated that age (OR=8.348), disease course (OR=9.161), smoking (OR=7.192), hypertension (OR=8.314), hyperhomocysteinemia (OR=8.508), and rehabilitation frequency (OR=0.142) were independent influencing factors for the recovery of limb motor function in stroke patients (P<0.05). ROC curve analysis demonstrated that the nomogram model constructed based on the above six risk factors exhibited good predictive efficiency [AUC=0.863 (95%CI: 0.821-0.905)]. The calibration curve suggested favorable calibration of the prediction model. DCA results showed that the model yielded a high positive net clinical benefit within the threshold probability range of 10%-50%.

Conclusions

Age, disease course, smoking, hypertension, hyperhomocysinaemia and the rehabilitation frequency were independent influencing factors for the rehabilitation efficacy of motor dysfunction in stroke patients. The predictive model established on this basis presents good predictive performance and favorable clinical application value.

表1 2组脑卒中后肢体运动功能障碍患者的临床资料比较
Tab.1 Comparison of clinical data between two groups of patients with post-stroke limb motor dysfunction
因素 改善组(n=242) 未改善组(n=47) t/χ2 P 因素 改善组(n=242) 未改善组(n=47) t/χ2 P
性别[例(%)]     2.785 0.095 饮酒[例(%)] 102(42.15) 26(55.32) 2.767 0.096
130(53.72) 19(40.43)     脑卒中类型[例(%)]     2.677 0.102
112(46.28) 28(59.57)     脑梗死 167(69.00) 38(80.85)    
年龄[例(%)]     5.314 0.021 脑出血 75(31.00) 9(19.15)    
<60岁 142(58.68) 19(40.43)     既往病史[例(%)]        
≥60岁 100(41.32) 28(59.57)     高血压 75(30.99) 23(48.94) 5.655 0.017
教育水平[例(%)]     1.512 0.470 糖尿病 19(7.85) 5(10.64) 0.401 0.526
未接受教育 10(4.13) 3(6.38)     冠状动脉疾病 28(11.57) 7(14.89) 0.408 0.523
初中及以下 139(57.44) 30(63.83)     家族病史[例(%)]        
高中及以上 93(38.43) 14(29.79)     高血压 177(73.14) 34(72.34) 0.013 0.910
职业[例(%)]     0.853 0.581 糖尿病 112(46.28) 19(40.43) 0.545 0.461
工人 49(20.25) 6(12.77)     冠状动脉疾病 65(26.86) 9(19.15) 1.288 0.268
农民 30(12.40) 8(17.02)     心房颤动 37(15.29) 9(19.15) 0.438 0.508
职员 45(18.60) 9(19.15)     康复频率[例(%)]     10.314 0.016
自由职业 53(21.90) 9(19.15)     3次/周 17(7.02) 9(19.15)    
其他 65(26.86) 15(31.91)     4次/周 38(15.70) 8(17.02)    
婚姻状况[例(%)]     0.124 0.724 5次/周 56(23.14) 14(29.79)    
未婚/丧偶/离异 61(25.21) 13(27.66)     ≥6次/周 131(54.13) 16(34.04)    
已婚 181(74.79) 34(72.34)     病变部位[例(%)]     6.906 0.075
能否支付医疗费用[例(%)]     0.036 0.850 右脑半球 149(61.57) 30(63.83)    
198(81.82) 39(82.98)     左脑半球 46(19.00) 7(14.89)    
44(18.18) 8(17.02)     脑干 28(11.57) 10(21.28)    
主要照顾者人数[例(%)]     1.059 0.589 双侧脑半球 19(7.85) 0    
1人 70(28.93) 12(25.54)     并发症[例(%)]        
2人 104(42.98) 24(51.06)     肺部感染 37(15.29) 7(14.89) 0.005 0.945
≥3人 68(28.09) 11(23.40)     尿路感染 18(7.44) 7(14.89) 2.768 0.096
入院时NIHSS评分(分,mean±SD) 9.45±2.13 8.82±1.83 1.896 0.059 深静脉血栓 35(14.46) 9(19.15) 0.670 0.413
病程[例(%)]     40.234 <0.001 高脂血症 119(49.17) 22(46.81) 0.088 0.767
≤6个月 22(9.09) 9(19.15)     高同型半胱氨酸血症 9(3.72) 6(12.77) 6.546 0.011
>6~12个月 87(35.95) 12(25.53)     高纤维蛋白原血症 54(22.31) 11(23.40) 0.027 0.870
>12~<24个月 113(46.69) 8(17.02)     低蛋白血症 63(26.03) 8(17.02) 1.725 0.189
≥24个月 20(8.26) 18(38.30)     贫血 56(23.14) 6(12.77) 2.514 0.113
吸烟[例(%)] 74(30.58) 26(55.32) 10.646 0.001 电解质代谢紊乱 19(7.85) 4(8.51) 0.023 0.879
表2 影响脑卒中后肢体运动功能改善的多因素Logistic回归分析自变量赋值
Tab.2 Assignment of independent variables in a multivariate Logistic regression analysis of factors influencing the improvement of limb motor function after stroke
表3 影响脑卒中后肢体运动功能恢复的多因素Logistic回归分析
Tab.3 Multivariable Logistic regression analysis of factors influencing the recovery of limb motor function after stroke
图1 脑卒中患者肢体运动功能恢复的列线图预测模型
Fig.1 Nomogram prediction model for recovery of limb motor function in stroke patients
图2 脑卒中患者肢体运动功能恢复列线图预测模型的ROC曲线分析
Fig.2 ROC curve analysis of the nomogram prediction model for limb motor function recovery in stroke patients
图3 脑卒中患者肢体运动功能恢复列线图预测模型的校准曲线分析
Fig.3 Calibration curves analysis of the nomogram prediction model for limb motor function recovery in stroke patients
图4 脑卒中患者肢体运动功能恢复列线图预测模型的决策曲线分析
Fig.4 Receiver operating characteristic curves analysis of the nomogram prediction model for limb motor function recovery in stroke patients
[1]
Zhang Q, Wang Y, Chen A, et al. Xiaoxuming decoction: a traditional herbal recipe for stroke with emerging therapeutic mechanisms[J]. Front Pharmacol, 2021, 12: 802381. DOI: 10.3389/fphar.2021.802381.
[2]
Alyabsi M, Algarni M, Alshammari K. Trends in colorectal cancer incidence rates in Saudi Arabia (2001-2016) using Saudi national registry: early-versus late-onset disease[J]. Front Oncol, 2021, 11: 730689. DOI: 10.3389/fonc.2021.730689.
[3]
Yaşa İC, Maviş İ, Şalçini C, et al. Comparing the efficiency of speech and language therapy and transcranial magnetic stimulation for treating Broca's aphasia[J]. J Stroke Cerebrovasc Dis, 2023, 32(6): 107108. DOI: 10.1016/j.jstrokecerebrovasdis.2023.107108.
[4]
McCluskey A, Ada L, Kelly PJ, et al. Compliance with Australian stroke guideline recommendations for outdoor mobility and transport training by post-inpatient rehabilitation services: an observational cohort study[J]. BMC Health Serv Res, 2015, 15: 296. DOI: 10.1186/s12913-015-0952-7.
[5]
Zhou Y, Zhang H, Huo H, et al. Efficacy of electrical stimulation for post-stroke motor dysfunction: a protocol for systematic review and network meta-analysis[J]. PLoS One, 2024, 19(6): e0304174. DOI: 10.1371/journal.pone.0304174.
[6]
史慧玲,王清,蒋园园,等.脑卒中患者运动康复体验质性研究的Meta整合[J].中华护理杂志, 2024, 59(11): 1397-1404. DOI: 10.3761/j.issn.0254-1769.2024.11.017.
[7]
Yan Q, Wang X, Zhang Y, et al. Analysis of influencing factors of rehabilitation treatment effect in patients with first-episode stroke[J]. Am J Transl Res, 2021, 13(12): 14046-14056. DOI: 10.3892/ajtr.2021.37487.
[8]
杨雪艳,辉沐吟,李寿蓉.脑卒中患者早期康复时机及方法的研究进展[J].临床护理进展, 2022, 1(3): 49-51. DOI: 10.12208/j.jacn.20220105.
[9]
李良芬.脑卒中患者延续性康复护理效果[J].当代护理, 2024, 5(7): 112-114. DOI: 10.12208/j.cn.20240347.
[10]
Burton CV, Lozano HD, Von Werssowetz OF, et al. Experiences with bobath method of treatment of cerebral palsy[J]. Arch Phys Med Rehabil, 1956, 37(9): 550-554.
[11]
Qian W, Liao X, Ju X, et al. Effects of low frequency repetitive transcranial magnetic stimulation on motor recovery in subacute stroke patients with different motor evoked potential status: a randomized controlled trial[J]. Front Neurol, 2024, 15: 1460925. DOI: 10.3389/fneur.2024.1460925.
[12]
Potter TBH, Tannous J, Vahidy FS. A contemporary review of epidemiology, risk factors, etiology, and outcomes of premature stroke[J]. Curr Atheroscler Rep, 2022, 24(12): 939-948. DOI: 10.1007/s11883-022-01067-x.
[13]
Song R, Park M, Jang T, et al. Effects of a Tai Chi-based stroke rehabilitation program on symptom clusters, physical and cognitive functions, and quality of life: a randomized feasibility study[J]. Int J Environ Res Public Health, 2021, 18(10): 5453. DOI: 10.3390/ijerph18105453.
[14]
胡世莲,王静,方向,等.中国居民主要致死性慢性病的人口死亡趋势及应对策略[J].中国临床保健杂志, 2022, 25(3): 297-302. DOI: 10.3969/J.issn.1672-6790.2022.03.002.
[15]
Lim J, Lim T, Lee J, et al. Patient-specific functional electrical stimulation strategy based on muscle synergy and walking posture analysis for gait rehabilitation of stroke patients[J]. J Int Med Res, 2021, 49(5): 3000605211016782. DOI: 10.1177/03000605211016782.
[16]
Iliescu AM, McIntyre A, Wiener J, et al. Evaluating the effectiveness of aquatic therapy on mobility, balance, and level of functional independence in stroke rehabilitation: a systematic review and meta-analysis[J]. Clin Rehabil, 2020, 34(1): 56-68. DOI: 10.1177/0269215519880955.
[17]
Lv Y, Sun Q, Li J, et al. Disability status and its influencing factors among stroke patients in northeast China: a 3-year follow-up study[J]. Neuropsychiatr Dis Treat, 2021, 17: 2567-2573. DOI: 10.2147/ndt.S320785.
[18]
杜秋明,曹书华,苏玫,等.不同年龄和性别脑卒中患者危险因素及预后[J].中国老年学杂志, 2021, 41(7): 1374-1377. DOI: 10.3969/j.issn.1005-9202.2021.07.009.
[19]
王伟英,桑文文,焉双梅,等.急性缺血性脑卒中患者1年复发危险因素Cox回归分析[J].中华老年心脑血管病杂志, 2016, 18(1): 46-50. DOI: 10.3969/j.issn.1009-0126.2016.01.013.
[20]
Brooks AC, Henderson BJ. Systematic review of nicotine exposure's effects on neural stem and progenitor cells[J]. Brain Sci, 2021, 11(2): 172. DOI: 10.3390/brainsci11020172.
[21]
Sifat AE, Vaidya B, Kaisar MA, et al. Nicotine and electronic cigarette (E-Cig) exposure decreases brain glucose utilization in ischemic stroke[J]. J Neurochem, 2018, 147(2): 204-221. DOI: 10.1111/jnc.14561.
[22]
Pan B, Jin X, Jun L, et al. The relationship between smoking and stroke: a meta-analysis[J]. Medicine (Baltimore), 2019, 98(12): e14872. DOI: 10.1097/md.0000000000014872.
[23]
Epstein KA, Viscoli CM, Spence JD, et al. Smoking cessation and outcome after ischemic stroke or TIA[J]. Neurology, 2017, 89(16): 1723-1729. DOI: 10.1212/wnl.0000000000004524.
[24]
d'Adesky ND, de Rivero Vaccari JP, Bhattacharya P, et al. Nicotine alters estrogen receptor-beta-regulated inflammasome activity and exacerbates ischemic brain damage in female rats[J]. Int J Mol Sci, 2018, 19(5): 1330. DOI: 10.3390/ijms19051330.
[25]
Lackland DT, Carey RM, Conforto AB, et al. Implications of recent clinical trials and hypertension guidelines on stroke and future cerebrovascular research[J]. Stroke, 2018, 49(3): 772-779. DOI: 10.1161/strokeaha.117.019379.
[26]
Li AL, Ji Y, Zhu S, et al. Risk probability and influencing factors of stroke in followed-up hypertension patients[J]. BMC Cardiovasc Disord, 2022, 22(1): 328. DOI: 10.1186/s12872-022-02780-w.
[27]
Kobylańska M, Kowalska J, Neustein J, et al. The role of biopsychosocial factors in the rehabilitation process of individuals with a stroke[J]. Work, 2018, 61(4): 523-535. DOI: 10.3233/wor-162823.
[28]
Ytterberg C, Kristensen HK, Tistad M, et al. Factors related to met needs for rehabilitation 6 years after stroke[J]. PLoS One, 2020, 15(1): e0227867. DOI: 10.1371/journal.pone.0227867.
[29]
Pila O, Duret C, Laborne FX, et al. Pattern of improvement in upper limb pointing task kinematics after a 3-month training program with robotic assistance in stroke[J]. J Neuroeng Rehabil, 2017, 14(1): 105. DOI: 10.1186/s12984-017-0315-1.
[30]
Chaiyawat P, Kulkantrakorn K. Effectiveness of home rehabilitation program for ischemic stroke upon disability and quality of life: a randomized controlled trial[J]. Clin Neurol Neurosurg, 2012, 114(7): 866-870. DOI: 10.1016/j.clineuro.2012.01.018.
[31]
Clark B, Whitall J, Kwakkel G, et al. The effect of time spent in rehabilitation on activity limitation and impairment after stroke[J]. Cochrane Database Syst Rev, 2021, 10(10): CD012612. DOI: 10.1002/14651858.CD012612.pub2.
[32]
Song W, Zhu ZG, Wu Q, et al. A nomogram to predict overall survival for biliary tract cancer[J]. Cancer Manag Res, 2018, 10: 1535-1541. DOI: 10.2147/cmar.S163291.
[33]
Gu Y, Wang F, Gong L, et al. A nomogram incorporating red blood cell indices to predict post-stroke cognitive impairment in the intracerebral hemorrhage population[J]. Front Aging Neurosci, 2022, 14: 985386. DOI: 10.3389/fnagi.2022.985386.
[34]
Tang M, Gao J, Ma N, et al. Radiomics nomogram for predicting stroke recurrence in symptomatic intracranial atherosclerotic stenosis[J]. Front Neurosci, 2022, 16: 851353. DOI: 10.3389/fnins.2022.851353.
[1] 曹喆, 翁桂湖, 刘涛, 张锰钢, 杨刚, 陈浩, 邱江东, 徐建威, 张太平. 新型预后营养炎症评分系统建立以有效预测胰腺癌根治术后患者的长期预后[J/OL]. 中华普通外科学文献(电子版), 2026, 20(02): 85-90.
[2] 贺智恒, 姚德炯, 孙东方. 腹腔镜下胆囊切除术后胆瘘影响因素分析及风险预测模型的构建[J/OL]. 中华普外科手术学杂志(电子版), 2026, 20(02): 175-178.
[3] 邓瑞锋, 程璐, 刘远灵, 郑秋平, 刘溪, 江文聪, 江敏耀, 习明. 基于Logistic回归构建一期输尿管通路鞘置入失败的预测模型[J/OL]. 中华腔镜泌尿外科杂志(电子版), 2026, 20(02): 171-178.
[4] 李坤, 李洪涛, 范瑞芳, 赵丽君. 1990—2021年中国腹股沟疝、股疝和腹部疝的疾病负担分析及未来预测[J/OL]. 中华疝和腹壁外科杂志(电子版), 2026, 20(02): 213-220.
[5] 邢颖, 王峰. 基于机器学习构建肝切除术后肝衰竭预测列线图模型及其预测价值[J/OL]. 中华肝脏外科手术学电子杂志, 2026, 15(02): 190-196.
[6] 王小振, 陈灿辉, 唐善华, 代浩嘉, 丰扬舸, 王恺, 李清平, 李川江. 基于不同机器学习技术构建肝移植术后早期严重并发症预测模型和效能比较[J/OL]. 中华肝脏外科手术学电子杂志, 2026, 15(02): 197-204.
[7] 孙娟华, 白引苗, 孔胜男, 王梦雪, 王文慧, 张红梅. 胰腺癌患者化疗相关性恶心呕吐风险列线图构建及验证[J/OL]. 中华消化病与影像杂志(电子版), 2026, 16(02): 114-119.
[8] 张维娜, 潘亚娟, 徐敏. 晚期消化系统癌症手术患者器官/腔隙感染的风险预测模型的建立[J/OL]. 中华消化病与影像杂志(电子版), 2026, 16(02): 120-124.
[9] 李晨阳, 汪涛, 徐忠瑞, 黄福秀, 李淑玲, 徐衍, 周旋, 赵慧, 张宁宁, 陈超. 急性下消化道出血患者安全出院风险评分的构建与验证[J/OL]. 中华消化病与影像杂志(电子版), 2026, 16(02): 156-161.
[10] 马吉辰, 王少松, 王雪飞, 姫旭, 刘桂林. 铁死亡在中医药治疗缺血性脑卒中中的研究进展[J/OL]. 中华脑血管病杂志(电子版), 2026, 20(02): 209-214.
[11] 尹晓欣, 于立敏. 脑心健康管理师在脑卒中患者健康管理中的实践研究进展[J/OL]. 中华脑血管病杂志(电子版), 2026, 20(02): 215-219.
[12] 宋文迪, 孙蔚, 宋海庆. 脑卒中后平衡障碍的发生机制及治疗的研究进展[J/OL]. 中华脑血管病杂志(电子版), 2026, 20(02): 111-118.
[13] 邓依婷, 李旭子, 罗安玲, 杨玉佳, 陈雨婕, 阳钦, 何俐, 周沐科. 颈动脉漂浮血栓的研究进展[J/OL]. 中华脑血管病杂志(电子版), 2026, 20(02): 119-125.
[14] 于玲, 秦伟, 徐霁华, 胡秀朝, 杨磊, 胡文立. 急性小脑梗死的临床影像学特征[J/OL]. 中华脑血管病杂志(电子版), 2026, 20(02): 133-138.
[15] 曹俊杰, 王芳, 董丽丽, 吕晨晨, 伊晶晶, 侯嘉甜, 熊雁杰, 黄勇华. 基于根因分析法的口腔健康管理方案用于老年脑卒中口腔衰弱患者的效果[J/OL]. 中华脑血管病杂志(电子版), 2026, 20(02): 148-156.
阅读次数
全文


摘要


AI


AI小编
你好!我是《中华医学电子期刊资源库》AI小编,有什么可以帮您的吗?