成人重症监护病房中护士实施及护士可行的拔管后吞咽困难筛查：测试性能、实施与患者结局的范围综述

文献信息

中文摘要

背景： 拔管后吞咽困难（PED）在有创机械通气后常见。护士可提供及时的床旁风险筛查，但工具效度、护士操作中的性能以及实施效果是不同的问题。
目的： 梳理成人重症监护病房（ICU）中直接由护士实施的证据、护士可行工具的可迁移性、护理实施背景以及报告的患者结局。
方法： 在六个数据库中进行文献检索，检索更新至2026年8月。证据根据实际筛查实施者和参考标准评估进行分层。应用条目水平的MMAT和QUADAS-2评估，不使用汇总评分。
结果： 经来源核实的图谱包含34篇报告，代表31个研究家族。由护士或护理助理实施并使用仪器参考标准的证据包括两个完整验证队列：改良容积-黏度吞咽试验对比FEES，在44例拔管患者中（敏感性89.5%，特异性72%）；GUSS-IVA对比FEES，在56例入组患者中的51例中（81.0%，88.9%）。第三项护士筛查在123例患者中的38例进行了可选FEES（86%，21%）。其他护士实施的研究使用临床或替代比较指标，或仅进行阳性验证。由言语语言病理学家实施的FEES研究提供了可迁移性信息，而非护士实施的准确性。实施保真度各异，结局研究均为非随机化。
结论： 护理人员参与结构化PED筛查是可行的，但直接准确性和患者结局证据仍具有异质性，并存在选择偏倚、验证偏倚和混杂偏倚风险。ICU可试点有治理的路径，包括能力评估、记录、专科升级和结局审计；尚无单一筛查被确立为标准照护。
对临床实践的相关性： 床旁筛查是风险分层而非诊断。筛查失败、不一致或高风险者需要及时专科评估，并在有指征时进行FEES或VFSS。

英文摘要

BACKGROUND: Post-extubation dysphagia (PED) is common after invasive mechanical ventilation. Nurses can provide timely bedside risk screening, but instrument validity, performance in nurses' hands and implementation effectiveness are distinct questions.
AIM: To map direct nurse-administered evidence, transferability of nurse-feasible tools, nursing implementation context, and reported patient outcomes in adult intensive care units (ICUs).
METHODS: Literature searches were conducted in six databases, with the search updated in August 2026. Evidence was stratified according to the actual screening administrator and reference-standard assessment. Criterion-level MMAT and QUADAS-2 assessments were applied without summary scores.
RESULTS: The source-verified map comprised 34 reports representing 31 study families. Instrumental-reference evidence with nurse or nursing-assistant administration included two complete-verification cohorts: modified Volume-Viscosity Swallow Test versus FEES in 44 extubated patients (sensitivity 89.5%, specificity 72%) and GUSS-IVA versus FEES in 51 of 56 enrolled patients (81.0%, 88.9%). A third nurse screen had optional FEES in 38 of 123 patients (86%, 21%). Other nurse-administered studies used clinical or proxy comparators or positive-only verification. SLP-administered FEES studies informed transferability, not nurse-administered accuracy. Implementation fidelity varied, and outcome studies were non-randomised.
CONCLUSIONS: Nursing involvement in structured PED screening is feasible, but direct accuracy and patient-outcome evidence remain heterogeneous and at risk of selection, verification and confounding bias. ICUs may pilot governed pathways with competency assessment, documentation, specialist escalation and outcome audit; no single screen is established as standard care.
RELEVANCE TO CLINICAL PRACTICE: Bedside screening is risk stratification rather than diagnosis. Failed, discordant or high-risk screens require prompt specialist assessment and, where indicated, FEES or VFSS.