文献速递心脑血管急症重症综合管理与镇静镇痛

丹麦认知筛查工具(DANcSCA)对心脏骤停幸存者的预测价值——一项数字化认知筛查

Predictive values of the DANish cognitive screen for cardiac arrest survivors (DANcSCA) - A digital cognitive screening.
PubMed 文献 · Resuscitation · 2025-09-23
一句话结论前瞻性多中心验证显示,数字化DANcSCA可将院外心脏骤停幸存者分为绿/黄/红三级,红/非红分类AUC为0.89,绿/非绿AUC为0.77。

该前瞻性多中心研究在丹麦医院和康复与姑息治疗知识中心纳入成年院外心脏骤停幸存者,以四项神经心理学测验为参考标准,评估数字化DANcSCA(基于剑桥神经心理学测试自动化组合的子测验)在心脏骤停后6–8周或康复期间(3个月至3年)的分类预测效度。175名参与者中,DANcSCA将23%归为绿色(低风险)、59%黄色(中风险)、18%红色(高风险),假阳性率9%、假阴性率5%。绿/非绿分类的AUC为0.77,红/非红分类的AUC为0.89。结论认为数字化DANcSCA似乎适用于OHCA后的认知筛查,但绿/非绿分类的AUC中等且可能存在过拟合,需在新样本中验证。

要点

  • 在175名成年OHCA幸存者中,DANcSCA将23%分类为绿色、59%为黄色、18%为红色,假阳性9%、假阴性5%。
  • 以神经心理学测验组合为参考标准,DANcSCA区分红/非红的AUC为0.89,绿/非绿的AUC为0.77。
  • 绿/非绿ROC的阈值设定为固定敏感度0.90,红/非红ROC的阈值设定为固定特异度0.90。
  • 认知评估在心脏骤停后6–8周或康复期间(3个月至3年)进行,DANcSCA使用CANTAB子测验。
  • 作者认为数字化DANcSCA似乎适用于OHCA后认知筛查,但绿/非绿AUC中等且可能过拟合,需在新样本中验证。

适用人群与场景

  • 丹麦医院和康复与姑息治疗知识中心纳入的成年院外心脏骤停(OHCA)幸存者。
  • 认知评估在心脏骤停后6–8周或康复期间(3个月至3年)进行。

诊断标准与关键阈值

  • 参考标准为四项神经心理学测验:Rey听觉词语学习测验、符号数字模态测验、连线测验和词语流畅性测验。
  • DANcSCA指数测试使用剑桥神经心理学测试自动化组合(CANTAB)的子测验。
  • 绿/非绿ROC的阈值设定为固定敏感度0.90;红/非红ROC的阈值设定为固定特异度0.90。

研究设计与关键结果

  • 研究类型为前瞻性、多中心DANcSCA验证研究,样本量175名成年OHCA幸存者。
  • DANcSCA分类:23%绿色(低风险)、59%黄色(中风险)、18%红色(高风险),假阳性9%、假阴性5%。
  • 绿/非绿分类的AUC为0.77;红/非红分类的AUC为0.89。

证据类型、等级与局限

  • 证据类型:前瞻性多中心验证研究。
  • 局限:绿/非绿分类的AUC中等且可能存在过拟合,需在新样本中验证。

对床旁实践的改变

  • 作者认为数字化DANcSCA似乎适用于OHCA后的认知筛查。
  • 作者指出绿/非绿分类的AUC中等且可能存在过拟合,需在新样本中验证。

提炼说明

  • 本条依据论文摘要整理;摘要未涉及「处置流程与药物剂量」「红旗征与不能漏的情况」「指南或共识出处」,故未列出这些栏目;全文请点「阅读原文」。
英文原文(摘要)· 点击展开

BACKGROUND: Persistent cognitive deficits affect up to 50 % of out-of-hospital cardiac arrest (OHCA) survivors. Despite this prevalence, no consensus exists on the optimal tool for identifying deficits. This study assessed the predictive validity of the digital DANcSCA by comparing its classification of OHCA survivors into risk classes-green (low), yellow (moderate), and red (high)-against a neuropsychological test battery as the reference standard.

METHODS: The prospective, multicentre DANcSCA validation study was conducted at Danish hospitals and the Knowledge Centre for Rehabilitation and Palliative Care. Adult OHCA survivors were included. The reference standard comprised four neuropsychological tests: Rey Auditory Verbal Learning Test, Symbol Digit Modalities Test, Trail Making Test, and Verbal Fluency. The DANcSCA index test used subtests from the Cambridge Neuropsychological Test Automated Battery. Cognitive assessments occurred six to eight weeks post-arrest or during rehabilitation three months to three years post-arrest. ROC curves were constructed for binary classifications of green versus non-green and red versus non-red. Thresholds were set by fixing sensitivity at 0.90 for the green/non-green ROC and specificity at 0.90 for the red/non-red ROC.

RESULTS: Among 175 participants, the DANcSCA classified 23 % as green, 59 % as yellow, and 18 % as red, with 9 % false positives and 5 % false negatives. The ROC curve for distinguishing green from non-green had an AUC of 0.77, and for distinguishing red from non-red an AUC of 0.89.

CONCLUSION: The digital DANcSCA appears suitable for cognitive screening following OHCA, although the modest AUC for green/non-green and potential overfitting warrant validation in new samples.

来源:Resuscitation,查看原文。版权归原作者所有。

参考来源

  1. Wieghorst A, Borregaard B, Zwisler AD, et al. Predictive values of the DANish cognitive screen for cardiac arrest survivors (DANcSCA) - A digital cognitive screening. Resuscitation. 2025. PMID: 40997999. doi:10.1016/j.resuscitation.2025.110842

供执业医师抢救与诊疗参考,不替代临床判断;剂量请结合体重、肝肾功能和本院药品规格复核,以指南原文为准。