Purpose

To compare the effects of prevention interventions on delirium occurrence in critically ill adults.

Methods

MEDLINE, Embase, PsychINFO, CINAHL, Web of Science, Cochrane Library, Prospero, and WHO international clinical trial registry were searched from inception to April 8, 2021. Randomized controlled trials of pharmacological, sedation, non-pharmacological, and multi-component interventions enrolling adult critically ill patients were included. We performed conventional pairwise meta-analyses, NMA within Bayesian random effects modeling, and determined surface under the cumulative ranking curve values and mean rank. Reviewer pairs independently extracted data, assessed bias using Cochrane Risk of Bias tool and evidence certainty with GRADE. The primary outcome was delirium occurrence; secondary outcomes were durations of delirium and mechanical ventilation, length of stay, mortality, and adverse effects.

Results

Eighty trials met eligibility criteria: 67.5% pharmacological, 31.3% non-pharmacological and 1.2% mixed pharmacological and non-pharmacological interventions. For delirium occurrence, 11 pharmacological interventions (38 trials, N = 11,993) connected to the evidence network. Compared to placebo, only dexmedetomidine (21/22 alpha₂ agonist trials were dexmedetomidine) probably reduces delirium occurrence (odds ratio (OR) 0.43, 95% Credible Interval (CrI) 0.21–0.85; moderate certainty). Compared to benzodiazepines, dexmedetomidine (OR 0.21, 95% CrI 0.08–0.51; low certainty), sedation interruption (OR 0.21, 95% CrI 0.06–0.69; very low certainty), opioid plus benzodiazepine (OR 0.27, 95% CrI 0.10–0.76; very low certainty), and protocolized sedation (OR 0.27, 95% CrI 0.09–0.80; very low certainty) may reduce delirium occurrence but the evidence is very uncertain. Dexmedetomidine probably reduces ICU length of stay compared to placebo (Ratio of Means (RoM) 0.78, CrI 0.64–0.95; moderate certainty) and compared to antipsychotics (RoM 0.76, CrI 0.61–0.98; low certainty). Sedative interruption, protocolized sedation and opioids may reduce hospital length of stay compared to placebo, but the evidence is very uncertain. No intervention influenced mechanical ventilation duration, mortality, or arrhythmia. Single and multi-component non-pharmacological interventions did not connect to any evidence networks to allow for ranking and comparisons as planned; pairwise comparisons did not detect differences compared to standard care.

Conclusion

Compared to placebo and benzodiazepines, we found dexmedetomidine likely reduced the occurrence of delirium in critically ill adults. Compared to benzodiazepines, sedation-minimization strategies may also reduce delirium occurrence, but the evidence is uncertain.

中文翻译：

---

预防危重患者谵妄的药物和非药物干预：系统评价和网络荟萃分析

目的

比较预防干预措施对危重成人谵妄发生的影响。

方法

从开始到 2021 年 4 月 8 日，搜索了 MEDLINE、Embase、PsychINFO、CINAHL、Web of Science、Cochrane Library、Prospero 和 WHO 国际临床试验注册中心。药物、镇静、非药物和多组分干预的随机对照试验入组的成年重症患者也包括在内。我们进行了传统的成对元分析、贝叶斯随机效应建模中的 NMA，并确定了累积排名曲线值和平均排名下的曲面。审稿人配对独立提取数据，使用 Cochrane 偏倚风险工具评估偏倚，并使用 GRADE 评估证据确定性。主要结局是发生谵妄；次要结果是谵妄和机械通气的持续时间、住院时间、死亡率和不良反应。

结果

80 项试验符合资格标准：67.5% 的药物干预、31.3% 的非药物干预和 1.2% 的混合药物和非药物干预。对于谵妄的发生，11 项药物干预措施（38 项试验，N  = 11,993）与证据网络相关联。与安慰剂相比，只有右美托咪定 (21/22 alpha ₂激动剂试验是右美托咪定）可能会减少谵妄的发生（比值比 (OR) 0.43，95% 可信区间 (CrI) 0.21–0.85；中等确定性）。与苯二氮卓类药物相比，右美托咪定（OR 0.21，95% CrI 0.08–0.51；低确定性），镇静中断（OR 0.21，95% CrI 0.06–0.69；极低确定性），阿片类药物加苯二氮卓类药物（OR 0.27，95% CrI 0.10– 0.76；非常低的确定性）和方案化镇静（OR 0.27，95% CrI 0.09–0.80；非常低的确定性）可能会减少谵妄的发生，但证据非常不确定。与安慰剂相比（均值比 (RoM) 0.78，CrI 0.64–0.95；中等确定性）和与抗精神病药（RoM 0.76，CrI 0.61–0.98；低确定性）相比，右美托咪定可能会缩短 ICU 住院时间。与安慰剂相比，镇静中断、程序化镇静和阿片类药物可能会缩短住院时间，但证据非常不确定。没有干预措施影响机械通气持续时间、死亡率或心律失常。单一和多成分的非药物干预措施没有连接到任何证据网络，无法按计划进行排名和比较；与标准护理相比，成对比较未发现差异。

结论

与安慰剂和苯二氮卓类药物相比，我们发现右美托咪定可能会降低危重患者谵妄的发生率。与苯二氮卓类药物相比，镇静最小化策略也可以减少谵妄的发生，但证据尚不确定。