Methods and Design Considerations for Randomized Clinical Trials Evaluating Surgical or Endovascular Treatments for Cerebrovascular Diseases

Abstract

Abstract OBJECTIVE 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. METHODS 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. RESULTS 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. CONCLUSION 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.