文献速递神经重症与神经感染认知障碍与痴呆

急性脑损伤无反应患者的语义加工

Semantic Processing in Unresponsive Patients With Acute Brain Injury.
PubMed 文献 · Neurology · 2026-11-11

这项前瞻性两中心队列研究纳入105例急性脑损伤无反应成人患者,用床旁脑电图进行语义范式检测,发现12%患者存在语义加工,其与12个月功能恢复相关,但作为预测标志物特异性高而敏感性低。

要点

  • 急性脑损伤无反应患者中12%可检出语义加工。
  • 语义加工者12个月功能恢复率42% vs 无语义加工者14%(OR 4.5)。
  • 语义加工预测恢复特异性高(94%/92%)但敏感性低(18%/29%)。

核心结论

  • 急性脑损伤无反应患者中部分可检出语义加工的脑电证据,且与良好长期结局相关;语义加工是特异性高但不敏感的恢复标志物,作者认为可能作为保留认知功能的确认性标志物,属假设生成性发现,需在更大规模多中心队列中验证后方可临床实施。

研究设计与样本

  • 前瞻性两中心队列研究,纳入2022年至2025年入住神经重症监护室的105例急性脑损伤无反应成人患者(平均年龄58±19岁;女性21%;西班牙裔/拉丁裔53%),在床旁进行脑电图语义范式检测,比较一致与不一致句尾的事件相关反应。

主要发现

  • 105例患者中常见病因为脑出血45%、创伤性脑损伤43%、蛛网膜下腔出血10%;31%的患者撤除生命维持治疗。
  • 13例(12%)患者检出语义加工。
  • 语义加工患者意识恢复率为77%,无语义加工者为49%(OR 3.5;95%CI 0.9-13.5,p=0.08)。
  • 语义加工患者功能恢复率为42%,无语义加工者为14%(OR 4.5;95%CI 1.2-16.6,p=0.03)。
  • 语义加工预测意识恢复的特异性94%、敏感性18%;预测功能恢复的特异性92%、敏感性29%;排除撤除生命维持治疗或死于非神经科医疗并发症的患者后,特异性分别升至100%和95%。

关键数值与推荐

  • 语义加工定义为中央顶区导联300-500毫秒内事件相关反应差异,采用聚类置换检验并以错误发现率校正。
  • 功能恢复定义为12个月时格拉斯哥结局扩展量表评分≥4分,由盲法访谈员评估。

临床怎么用

  • 原文未据此给出推荐。

局限性

  • 作者指出这些发现属假设生成性,需在更大规模多中心队列中验证后方可临床实施。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

BACKGROUND AND OBJECTIVES: Early prognostication after acute brain injury (ABI) is challenging because behavioral responsiveness may not accurately reflect preserved cognitive function. Neural responses to semantic incongruity can identify language processing despite the absence of overt behavior. We investigated whether bedside EEG could detect semantic processing in unresponsive patients with ABI and whether its presence was associated with long-term outcomes. METHODS: In this prospective 2-center cohort study, unresponsive adults with ABI admitted to neurologic intensive care units between 2022 and 2025 underwent bedside EEG during a language paradigm comparing congruent and incongruent sentence endings (e.g., "my favorite color is green/soon"). Semantic processing was defined by differential event-related responses within centroparietal channels (300-500 milliseconds) using cluster-based permutation testing with false discovery rate correction. Outcomes were assessed for recovery of consciousness and functional recovery (Glasgow Outcome Scale-Extended score ≥4) at 12 months by blinded interviewers. Diagnostic performance was evaluated using sensitivity, specificity, predictive values, accuracy, and likelihood ratios. RESULTS: Among 105 patients (mean age 58 ± 19 years; 21% women; 53% Hispanic/Latino), the most common etiologies were intracerebral hemorrhage (45%), traumatic brain injury (43%), and subarachnoid hemorrhage (10%). Withdrawal of life-sustaining therapy occurred in 31%. Semantic processing was detected in 13 patients (12%). Recovery of consciousness occurred in 77% of patients with semantic processing compared with 49% without semantic processing (odds ratio [OR] 3.5; 95% CI 0.9-13.5, p = 0.08). Functional recovery occurred in 42% with semantic processing compared with 14% without semantic processing (OR 4.5; 95% CI 1.2-16.6, p = 0.03). Semantic processing was highly specific but insensitive for predicting both recovery of consciousness (specificity 94%, sensitivity 18%) and functional recovery (specificity 92%, sensitivity 29%). After excluding patients who underwent withdrawal of life-sustaining therapy or died of non-neurologic medical complications, specificity increased to 100% for recovery of consciousness and 95% for functional recovery. DISCUSSION: EEG evidence of semantic processing can be detected in a subset of unresponsive patients with ABI during the acute phase of injury and is associated with favorable long-term outcomes. Semantic processing was a specific but insensitive biomarker of recovery, suggesting potential utility as a rule-in marker of preserved cognitive function. These findings are hypothesis-generating and require validation in larger multicenter cohorts before clinical implementation.

来源:Neurology,查看原文。版权归原作者所有。

引用

  1. Alkhachroum A, Manolovitz B, McKeown J, et al. Semantic Processing in Unresponsive Patients With Acute Brain Injury. Neurology. 2026. PMID: 43013794. doi:10.1212/WNL.0000000000218680

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。