Bragge Peter, Synnot Anneliese, Maas Andrew I, Menon David K, Cooper D James, Rosenfeld Jeffrey V, Gruen Russell L
1 Centre of Excellence in Traumatic Brain Injury Research, National Trauma Research Institute, Monash University and The Alfred Hospital , Victoria, Australia .
2 BehaviourWorks Australia, Monash Sustainability Institute, Monash University , Victoria, Australia .
J Neurotrauma. 2016 Aug 15;33(16):1461-78. doi: 10.1089/neu.2015.4233. Epub 2016 Mar 18.
Moderate-to-severe traumatic brain injury (TBI) remains a major global challenge, with rising incidence, unchanging mortality and lifelong impairments. State-of-the-science reviews are important for research planning and clinical decision support. This review aimed to identify randomized controlled trials (RCTs) evaluating interventions for acute management of moderate/severe TBI, synthesize key RCT characteristics and findings, and determine their implications on clinical practice and future research. RCTs were identified through comprehensive database and other searches. Key characteristics, outcomes, risk of bias, and analysis approach were extracted. Data were narratively synthesized, with a focus on robust (multi-center, low risk of bias, n > 100) RCTs, and three-dimensional graphical figures also were used to explore relationships between RCT characteristics and findings. A total of 207 RCTs were identified. The 191 completed RCTs enrolled 35,340 participants (median, 66). Most (72%) were single center and enrolled less than 100 participants (69%). There were 26 robust RCTs across 18 different interventions. For 74% of 392 comparisons across all included RCTs, there was no significant difference between groups. Positive findings were broadly distributed with respect to RCT characteristics. Less than one-third of RCTs demonstrated low risk of bias for random sequence generation or allocation concealment, less than one-quarter used covariate adjustment, and only 7% employed an ordinal analysis approach. Considerable investment of resources in producing 191 completed RCTs for acute TBI management has resulted in very little translatable evidence. This may result from broad distribution of research effort, small samples, preponderance of single-center RCTs, and methodological shortcomings. More sophisticated RCT design, large multi-center RCTs in priority areas, increased focus on pre-clinical research, and alternatives to RCTs, such as comparative effectiveness research and precision medicine, are needed to fully realize the potential of acute TBI research to benefit patients.
中重度创伤性脑损伤（TBI）仍然是一项重大的全球性挑战，其发病率不断上升，死亡率居高不下，且会造成终身残疾。科学现状综述对于研究规划和临床决策支持至关重要。本综述旨在识别评估中重度TBI急性处理干预措施的随机对照试验（RCT），综合关键RCT的特征和结果，并确定其对临床实践和未来研究的影响。通过全面的数据库检索和其他搜索方式识别RCT。提取关键特征、结果、偏倚风险和分析方法。对数据进行叙述性综合分析，重点关注可靠的（多中心、低偏倚风险、n>100）RCT，还使用三维图形来探究RCT特征与结果之间的关系。共识别出207项RCT。191项完成的RCT纳入了35340名参与者（中位数为66）。大多数（72%）为单中心研究，纳入的参与者少于100名（69%）。有26项针对18种不同干预措施的可靠RCT。在所有纳入的RCT的392项比较中，74%的组间无显著差异。阳性结果在RCT特征方面分布广泛。不到三分之一的RCT在随机序列生成或分配隐藏方面显示出低偏倚风险，不到四分之一的RCT使用协变量调整，只有7%采用序数分析方法。在为急性TBI管理开展191项完成的RCT过程中投入了大量资源，但可转化的证据却非常少。这可能是由于研究工作分布广泛、样本量小、单中心RCT占主导以及方法学缺陷所致。需要更精密的RCT设计、在优先领域开展大型多中心RCT、更多关注临床前研究以及RCT的替代方法，如比较效果研究和精准医学，以充分发挥急性TBI研究造福患者的潜力。
