Qureshi Adnan I, Hutson Alan D, Harbaugh Robert E, Stieg Philip E, Hopkins L Nelson
Department of Neurology and Neurosciences, University of Medicine and Dentistry of New Jersey, H-506, 185 S. Orange Avenue, Newark, NJ 07103, USA.
Neurosurgery. 2004 Feb;54(2):248-64; discussion 264-7. doi: 10.1227/01.neu.0000103446.26057.78.
The results of clinical trials affect the practice of surgery and endovascular therapy for cerebrovascular diseases. The purpose of this report is to review the basic components of the designs and methods for randomized clinical trials and to describe the influence of those components on the interpretation of trial results.
The goal of an optimal clinical trial of a new procedure is to provide the most objective and rigorous evaluation of the safety and effectiveness of that procedure. Anything in the design, performance, or analysis that impairs such an assessment decreases the ability of the trial to achieve its goal and answer the research question. To highlight the components of a clinical trial, this report uses examples of Phase III clinical trials that have influenced the practice of cerebrovascular surgery and endovascular therapy in the past three decades, including the International Cooperative Study of Extracranial/Intracranial Arterial Anastomosis, the North American Symptomatic Carotid Endarterectomy Trial, the Asymptomatic Carotid Atherosclerosis Study, the Prolyse in Acute Cerebral Thromboembolism II study, and the International Subarachnoid Aneurysm Trial.
The research question (objective) of the trial must be clearly defined, with an objective measure of efficacy and a specified quantitative difference to define the superiority of one intervention over another, in a relatively homogeneous patient population. Allocation concealment, randomization with or without stratification, and blinding (or masking) are important strategies to prevent differences in the study populations that could adversely affect the conclusions of the study. The primary end point must correspond to the specific aims of the trial. It should be objectively defined, quantifiable, reliable, and reproducible. Commonly defined end points in surgical trials include changes from baseline illness or disease severity scores, morbidity and mortality rates, and relative risks of reaching an end point with time. The statistical methods used for interim and final analyses are important. The effects of dropouts, crossovers, and missing data should be understood in the context of the final analysis. Additional concepts, such as intention-to-treat analysis and use of actual versus predicted outcomes, are important with respect to interpretation of the final results of the study.
The neurosurgical and neuroendovascular communities are currently planning or conducting several clinical trials to evaluate new procedures for the treatment of cerebrovascular diseases. It is hoped that a better understanding of the components of clinical trials will facilitate the design and implementation of effective studies.
临床试验的结果会影响脑血管疾病的外科手术和血管内治疗实践。本报告的目的是回顾随机临床试验设计和方法的基本组成部分，并描述这些组成部分对试验结果解读的影响。
一项新手术的最佳临床试验目标是对该手术的安全性和有效性进行最客观、最严格的评估。设计、实施或分析过程中任何损害这种评估的因素都会降低试验实现其目标并回答研究问题的能力。为突出临床试验的组成部分，本报告采用了过去三十年中影响脑血管外科手术和血管内治疗实践的Ⅲ期临床试验实例，包括颅外/颅内动脉吻合术国际合作研究、北美症状性颈动脉内膜切除术试验、无症状性颈动脉粥样硬化研究、急性脑栓塞Ⅱ期普罗激酶研究以及国际蛛网膜下腔动脉瘤试验。
试验的研究问题（目标）必须明确界定，在相对同质的患者群体中，要有客观的疗效衡量标准以及规定的定量差异来界定一种干预措施相对于另一种干预措施的优越性。分配隐藏、有或无分层的随机化以及盲法（或遮蔽）是防止研究人群出现可能对研究结论产生不利影响的差异的重要策略。主要终点必须与试验的具体目标相对应。它应该是客观界定的、可量化的、可靠的且可重复的。外科手术试验中常见的界定终点包括基线疾病或病情严重程度评分的变化、发病率和死亡率以及随时间达到终点的相对风险。用于中期和最终分析的统计方法很重要。在最终分析的背景下应理解失访、交叉和缺失数据的影响。其他概念，如意向性治疗分析以及实际结果与预测结果的使用，对于研究最终结果的解读很重要。
神经外科和神经血管内领域目前正在规划或开展多项临床试验，以评估治疗脑血管疾病的新手术。希望对临床试验组成部分有更好的理解将有助于有效研究的设计和实施。
