脑血管病前沿进展·限制平均生存时间在脑血管病研究中的方法学应用

doi:10.3969/j.issn.1006-7795.2026.01.004

首都医科大学学报 ›› 2026, Vol. 47 ›› Issue (1): 30-36.doi: 10.3969/j.issn.1006-7795.2026.01.004

脑血管病前沿进展·限制平均生存时间在脑血管病研究中的方法学应用

吕宗政^1,2,3，胥芹^1,2,3，张怡君^1,3，孟霞^1,3，王安心^1,2,3*

1.首都医科大学附属北京天坛医院国家神经系统疾病临床医学研究中心，北京 100070；2.首都医科大学附属北京天坛医院北京市神经外科研究所流行病学研究室，北京 100070；3.首都医科大学临床流行病学与临床试验学系，北京 100070

收稿日期:2025-10-14 修回日期:2025-11-06 出版日期:2026-02-21 发布日期:2026-02-02
通讯作者: 王安心 E-mail:wanganxin@bjtth.org
基金资助:
国家重点研发计划项目（2022YFC3501100），国家自然科学基金项目（82504483）。

Methodological applications of restricted mean survival time in cerebrovascular disease research

Lyu Zongzheng^1,2,3, Xu Qin^1,2,3, Zhang Yijun^1,3, Meng Xia^1,3, Wang Anxin^1,2,3*

1.National Clinical Research Center for Neurological Disease, Beijing Tiantan Hospital, Capital Medical University, Beijing 100070, China; 2.Department of Epidemiology, Beijing Neurosurgical Institute, Beijing Tiantan Hospital, Capital Medical University, Beijing 100070, China; 3.Department of Clinical Epidemiology and Clinical Trial, Capital Medical University, Beijing 100070, China

Received:2025-10-14 Revised:2025-11-06 Online:2026-02-21 Published:2026-02-02
Supported by:
This study was supported by National Key Research and Development Program of China (2022YFC3501100), National Natural Science Foundation of China (82504483).

摘要/Abstract

摘要： 目的探讨限制平均生存时间（restricted mean survival time, RMST）的原理及其在脑血管病研究中的应用。方法 RMST是一种稳健的生存分析方法，在满足比例风险（proportional hazards，PH）假设时可作为Cox比例风险模型的有益补充，在不满足PH假设时可作为替代方法。根据PH假设检验结果，本研究对主要结局分别采用Cox比例风险模型和RMST方法进行分析。以氯吡格雷用于急性非致残性脑血管事件高危人群的疗效研究-Ⅱ（Clopidogrel with Aspirin in High-Risk Patients with Acute Nondisabling Cerebrovascular Events Ⅱ，CHANCE-2）为例，计算替格瑞洛－阿司匹林组相比于氯吡格雷－阿司匹林组1年卒中复发的RMST差值；同时在中国国家卒中登记研究-Ⅲ（China National Stroke Registry-Ⅲ，CNSR-Ⅲ）中，探讨降压治疗对脑血管病合并高血压患者5年生存时间的影响。结果在CHANCE-2研究中，PH假设成立，相比于氯吡格雷－阿司匹林组，Cox回归提示替格瑞洛－阿司匹林组1年卒中复发风险下降约20%（HR=0.805, 95% CI: 0.682~0.949），RMST分析进一步显示，替格瑞洛－阿司匹林组在1年随访期内平均无卒中复发时间可延长约6.646 d（341.316 d vs 334.755 d，P < 0.05）。在CNSR-Ⅲ研究中，PH假设不成立，多变量调整的RMST分析表明降压治疗可在5年随访期内为脑血管病合并高血压患者平均延长17.455 d生存时间，倾向性评分匹配后观察到类似结果（ΔRMST=22.883 d，P =0.011）。结论 RMST不仅能在比例风险假设不满足时提供稳健的效应估计，还能以直观的时间尺度呈现临床获益，在脑血管病临床研究中具有重要的应用价值。

关键词: 限制平均生存时间, 生存分析, 比例风险假设, 脑血管病, CHANCE-2, CNSR-Ⅲ

Abstract: Objective To explore the principle of restricted mean survival time (RMST) and its application in cerebrovascular disease research. Methods RMST is employed as a robust method for survival analysis, providing an alternative to Cox proportional hazards （PH） model when the PH assumption is violated and a complement when it was satisfied. In this study, outcomes were analyzed by Cox PH model or RMST method according to the results of the PH assumption test. The Clopidogrel with Aspirin in High-Risk Patients with Acute Nondisabling Cerebrovascular Events Ⅱ (CHANCE-2) trial was used to assess the difference in 1-year RMST for stroke recurrence between the ticagrelor-aspirin and clopidogrel-aspirin groups. The China National Stroke Registry-Ⅲ (CNSR-Ⅲ) was used to examine the effect of antihypertensive therapy on 5-year survival among patients with cerebrovascular disease and hypertension. Results In CHANCE-2, the PH assumption was satisfied. Compared with clopidogrel-aspirin, Cox regression showed that ticagrelor-aspirin reduced the risk of recurrent stroke by 20% (HR=0.805, 95% CI: 0.682－0.949), and RMST analysis further indicated an approximate 6.646－d prolongation in the mean stroke-free survival time over one year (341.316 vs 334.755 d, P < 0.05). In CNSR-Ⅲ, the PH assumption was violated. Multivariable-adjusted RMST analysis demonstrated that antihypertensive treatment extended overall survival by 17.455 d over five years in patients with cerebrovascular disease and hypertension, with similar findings after propensity score matching (ΔRMST=22.883 d, P < 0.05). Conclusion RMST yield robust estimates and interpretable effect measures beyond the PH assumption, highlighting its significance in cerebrovascular disease research.

Key words: restricted mean survival time, survival analysis, proportional hazards assumption, cerebrovascular disease, CHANCE-2, CNSR-Ⅲ

中图分类号:

R181
R743

吕宗政, 胥芹, 张怡君, 孟霞, 王安心. 脑血管病前沿进展·限制平均生存时间在脑血管病研究中的方法学应用[J]. 首都医科大学学报, 2026, 47(1): 30-36.

Lyu Zongzheng, Xu Qin, Zhang Yijun, Meng Xia, Wang Anxin. Methodological applications of restricted mean survival time in cerebrovascular disease research[J]. Journal of Capital Medical University, 2026, 47(1): 30-36.

参考文献

［1］Hernán M A. The hazards of hazard ratios［J］. Epidemiology, 2010, 21(1): 13－15.
［2］Han K, Jung I. Restricted mean survival time for survival analysis: a quick guide for clinical researchers［J］. Korean J Radiol, 2022, 23(5): 495－499.
［3］Royston P, Parmar M K B. Restricted mean survival time: an alternative to the hazard ratio for the design and analysis of randomized trials with a time-to-event outcome［J］. BMC Med Res Methodol, 2013, 13: 152.
［4］Weir I R, Marshall G D, Schneider J I, et al. Interpretation of time-to-event outcomes in randomized trials: an online randomized experiment［J］. Ann Oncol, 2019, 30(1): 96－102.
［5］Trinquart L, Jacot J, Conner S C, et al. Comparison of treatment effects measured by the hazard ratio and by the ratio of restricted mean survival times in oncology randomized controlled trials［J］. J Clin Oncol, 2016, 34(15): 1813－1819.
［6］Chen H S, Yang Y, Ni J, et al. Effects of cinepazide maleate injection on blood pressure in patients with acute ischemic stroke and hypertension［J］. Chin J Obstet Gynecol, 2022, 61(8): 916－920.
［7］Uno H, Claggett B, Tian L, et al. Moving beyond the hazard ratio in quantifying the between-group difference in survival analysis［J］. J Clin Oncol, 2014, 32(22): 2380－2385.
［8］Manner D H, Battioui C, Hantel S, et al. Restricted mean survival time for the analysis of cardiovascular outcome trials assessing non-inferiority: case studies from antihyperglycemic drug development［J］. Am Heart J, 2019, 215: 178－186.
［9］Han L. Breaking free from the hazard ratio: embracing the restricted mean survival time in clinical trials［J］. NEJM Evid, 2023, 2(8): EVIDe2300142.
［10］Klein J P, Gerster M, Andersen P K, et al. SAS and R functions to compute pseudo-values for censored data regression［J］. Comput Methods Programs Biomed, 2008, 89(3): 289－300.
［11］Tian L, Zhao L H, Wei L J. Predicting the restricted mean event time with the subject's baseline covariates in survival analysis［J］. Biostatistics, 2014, 15(2): 222－233.
［12］Horiguchi M, Cronin A M, Takeuchi M, et al. A flexible and coherent test/estimation procedure based on restricted mean survival times for censored time-to-event data in randomized clinical trials［J］. Stat Med, 2018, 37(15): 2307－2320.
［13］Liao J J Z, Liu G F, Wu W C. Dynamic RMST curves for survival analysis in clinical trials［J］. BMC Med Res Methodol, 2020, 20(1): 218.
［14］Vandenbroucke J P, Von Elm E, Altman D G, et al. Strengthening the reporting of observational studies in epidemiology (STROBE): explanation and elaboration［J］. Int J Surg, 2014, 12(12): 1500－1524.
［15］Hopewell S, Chan A W, Collins G S, et al. CONSORT 2025 statement: updated guideline for reporting randomised trials［J/OL］. Lancet, 2025: S0140－6736(25)00672－5. (2025－04－14)［2025－10－06］. https://pubmed.ncbi.nlm.nih.gov/40245901/.
［16］Ananthakrishnan R, Green S, Previtali A, et al. Critical review of oncology clinical trial design under non-proportional hazards［J］. Crit Rev Oncol Hematol, 2021, 162: 103350.
［17］Guimarães H P, Lopes R D, De Barros E Silva P G M, et al. Rivaroxaban in patients with atrial fibrillation and a bioprosthetic mitral valve［J］. N Engl J Med, 2020, 383(22): 2117－2126.［18］Freidlin B, Hu C, Korn E L. Are restricted mean survival time methods especially useful for noninferiority trials?［J］. Clin Trials, 2021, 18(2): 188－196.
［19］Lane M, Miao T, Turgeon R D. Clinician’s approach to advanced statistical methods: win ratios, restricted mean survival time, responder analyses, and standardized mean differences［J］. J Gen Intern Med, 2024, 39(7): 1196－1203.
［20］李宣, 倪颖, 于涵, 等. 限制平均生存时间回归分析及其SAS实现［J］. 中国卫生统计, 2024, 41(6): 927－931.
［21］Kloecker D E, Davies M J, Khunti K, et al. Uses and limitations of the restricted mean survival time: illustrative examples from cardiovascular outcomes and mortality trials in type 2 diabetes［J］. Ann Intern Med, 2020, 172(8): 541－552.

[1]	王拥军, 刘涛, 刘子阳, 熊云云, 荆京, 谢雪微, 李子孝. 人工智能在脑血管病领域中的应用[J]. 首都医科大学学报, 2025, 46(2): 177-183.
[2]	田若冰, 胥芹, 周全, 田雪, 李欣亚, 郑曼琪, 王安心. 胜出率法在脑血管病临床试验中的应用[J]. 首都医科大学学报, 2025, 46(2): 191-196.
[3]	张婉婉, 董晓, 于尔澜, 吉训明, 吴川杰. 人工智能在脑血管病诊疗中的应用[J]. 首都医科大学学报, 2025, 46(1): 6-10.
[4]	郭怡兵, 孟然. 脂蛋白a在脑血管病中的研究新进展[J]. 首都医科大学学报, 2025, 46(1): 71-75.

脑血管病前沿进展·限制平均生存时间在脑血管病研究中的方法学应用

Methodological applications of restricted mean survival time in cerebrovascular disease research

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 4

编辑推荐

Metrics

本文评价