Andersen-Ranberg Nina, Poulsen Lone M, Perner Anders, Wetterslev Jørn, Mathiesen Ole, Mortensen Camilla B, Estrup Stine, Ebdrup Bjørn H, Hästbacka Johanna, Citerio Giuseppe, Caballero Jesus, Morgan Matthew P G, Oxenbøll-Collet Marie, Weber Sven-Olaf, Sofie Andreasen Anne, Bestle Morten H, Pedersen Helle B S, Nielsen Louise G, Uslu Bülent, Jensen Troels B, Thee Carsten, Dey Nilanjan, Lange Theis
Department of Anaesthesiology and Intensive Care Medicine, Zealand University Hospital, Koege, Denmark.
Centre for Research in Intensive Care (CRIC), Copenhagen, Denmark.
Acta Anaesthesiol Scand. 2020 Oct;64(9):1357-1364. doi: 10.1111/aas.13661. Epub 2020 Jul 13.
The AID-ICU trial aims to assess the benefits and harms of haloperidol for the treatment of delirium in acutely admitted, adult intensive care unit (ICU) patients. This paper describes the detailed statistical analysis plan for the primary publication of results from the AID-ICU trial.
The AID-ICU trial is an investigator-initiated, pragmatic, international, multicentre, randomized, blinded, parallel-group trial allocating 1000 adult ICU patients with manifest delirium 1:1 to haloperidol or placebo. The primary outcome measure is days alive and out of hospital within 90 days post-randomization. Secondary outcome measures are days alive without delirium or coma, serious adverse reactions (SARs) to haloperidol, use of escape medicine, days alive without mechanical ventilation, and mortality, health-related quality-of-life measures and cognitive function 1-year post-randomization. Statistical analysis will be conducted in accordance with the current pre-specified statistical analysis plan. One formal interim analysis will be performed. The primary outcome will be adjusted for stratification variables (site and delirium motor subtype) and compared between treatment groups using a likelihood ratio test described by Jensen et al A secondary analysis will be conducted with additional adjustment of the primary outcome for prognostic variables at baseline. The primary conclusion of the trial will be based on the intention-to-treat analysis of the primary outcome adjusted for stratification variables.
The AID-ICU trial will provide important, high-quality data on the benefits and harms of treatment with haloperidol in acutely admitted, adult patients with manifest delirium in the ICU.
AID-ICU试验旨在评估氟哌啶醇治疗急性收治的成年重症监护病房（ICU）谵妄患者的益处和危害。本文描述了AID-ICU试验主要结果发表的详细统计分析计划。
AID-ICU试验是一项由研究者发起的、务实的、国际性、多中心、随机、双盲、平行组试验，将1000例有明显谵妄的成年ICU患者按1:1随机分配至氟哌啶醇组或安慰剂组。主要结局指标是随机分组后90天内的存活且出院天数。次要结局指标包括无谵妄或昏迷的存活天数、对氟哌啶醇的严重不良反应（SARs）、急救药物的使用、无机械通气的存活天数、死亡率、健康相关生活质量指标以及随机分组后1年的认知功能。将根据当前预先指定的统计分析计划进行统计分析。将进行一次正式的中期分析。主要结局将针对分层变量（研究地点和谵妄运动亚型）进行调整，并使用Jensen等人描述的似然比检验在治疗组之间进行比较。将进行二次分析，对主要结局进一步根据基线预后变量进行调整。试验的主要结论将基于对经分层变量调整的主要结局的意向性分析。
AID-ICU试验将为氟哌啶醇治疗急性收治的成年ICU明显谵妄患者的益处和危害提供重要的高质量数据。
