From the Department of Acute Care, College of Nursing (A.W.A.), and Departments of Neurology (A.W.A., G.T., A.V.A.) and Neurosurgery (A.S.A.), College of Medicine, University of Tennessee Health Science Center at Memphis; Second Department of Neurology (G.T.), National & Kapodistrian University of Athens, School of Medicine, Attikon University Hospital, Greece; Department of Clinical Neuroscience (M.D.H.), Hotchkiss Brain Institute, Cumming School of Medicine, University of Calgary, Canada; Department of Neurology (D.S.L.), University of California at Los Angeles; Department of Neurology and Neurogeriatrics (P.S.), Johannes Wesling Medical Center, Minden, University Hospitals of the Ruhr-University of Bochum, Germany; Department of Neurology (B.O.), Medical University of South Carolina, Charleston; Semmes Murphy Brain and Spine Institute (A.S.A.), Memphis, TN; Stroke Unit (V.C.), Santa Maria Della Misericordia Hospital, University of Perugia, Italy; Marcus Stroke and Neuroscience Center (R.G.N.), Grady Memorial Hospital, School of Medicine, Emory University, Atlanta, GA; Department of Neurology (J.C.H.), University of California at San Francisco; Memorial Hermann Hospital (J.C.G.), Houston, TX; and Department of Neurology (W.H.), University of Heidelberg, Germany.
Neurology. 2018 May 8;90(19):885-889. doi: 10.1212/WNL.0000000000005481. Epub 2018 Apr 11.
To critique the Head Positioning in Stroke Trial (HeadPoST) study methods in relation to preceding research findings in an aim to clarify the potential efficacy of positioning interventions and direction for future research.
Head positioning research prior to the conduct of HeadPoST was reviewed by a team of international stroke experts, and methods and findings were compared to HeadPoST.
Methods used to select HeadPoST patients differ substantially from those used in original head positioning studies, in particular enrollment of all types of stroke. HeadPoST enrolled primarily minor strokes (median NIH Stroke Scale 4, interquartile range [IQR] 2-8) without vascular imaging confirmation of subtype; elapsed time from stroke symptom onset to the initiation of intervention was late (median 14 hours, IQR 5-35), and time from hospital admission to enrollment was delayed (median 7 hours, IQR 2-26). Intervention integrity was not reported, including ability to achieve/maintain 30° head elevation in beds lacking head elevation capabilities. Deterioration or improvement associated with the intervention is unknown as serial assessments were not completed, and the trial's 3-month outcome was powered using unrelated study data.
The design of HeadPoST was suboptimal to measure differences produced by the intervention. Future head positioning trials in discrete patient cohorts (in particular, large vessel occlusion) with endpoints supported by pilot work are required to understand the efficacy of this simple yet potentially important intervention.
对头位在卒中试验（HeadPoST）研究方法进行评价，以澄清定位干预的潜在疗效，并为未来的研究提供方向。
由一组国际卒中专家对 HeadPoST 开展前的头位研究进行了回顾，比较了方法和研究结果。
HeadPoST 患者的选择方法与原始头位研究方法有很大不同，特别是纳入了所有类型的卒中。HeadPoST 主要纳入了轻度卒中（NIH 卒中量表中位数 4 分，四分位距[IQR]2-8），而未对亚型进行血管成像确认；干预开始时间距卒中症状发作时间较晚（中位数 14 小时，IQR 5-35），从入院到入组的时间延迟（中位数 7 小时，IQR 2-26）。干预完整性未报告，包括在缺乏头部抬高能力的床上能否实现/维持 30°头部抬高。由于未完成连续评估，且试验的 3 个月结局是使用不相关的研究数据进行的，因此无法了解与干预相关的恶化或改善情况。
HeadPoST 的设计不利于测量干预产生的差异。需要对头位进行专门的试验，纳入特定患者群体（特别是大血管闭塞），并使用试点工作支持的终点，以了解这一简单但潜在重要的干预措施的疗效。
