Aim

To evaluate the test accuracy of pre-arrest clinical decision tools for in-hospital cardiac arrest survival outcomes.

Methods

We searched Medline, Embase, and Cochrane Library from inception through January 2022 for randomized and non-randomized studies. We used the Quality Assessment of Diagnostic Accuracy Studies framework to evaluate risk of bias, and Grading of Recommendations Assessment, Development and Evaluation methodology to evaluate certainty of evidence. We report sensitivity, specificity, positive predictive outcome, and negative predictive outcome for prediction of survival outcomes. PROSPERO CRD42021268005.

Results

We searched 2517 studies and included 23 studies using 13 different scores: 12 studies investigating 8 different scores assessing survival outcomes and 11 studies using 5 different scores to predict neurological outcomes. All were historical cohorts/ case control designs including adults only. Test accuracy for each score varied greatly. Across the 12 studies investigating 8 different scores assessing survival to hospital discharge/ 30-day survival, the negative predictive values (NPVs) for the prediction of survival varied from 55.6% to 100%. The GO-FAR score was evaluated in 7 studies with NPVs for survival with cerebral performance category (CPC) 1 ranging from 95.0% to 99.2%. Two scores assessed survival with CPC ≤ 2 and these were not externally validated. Across all prediction scores, certainty of evidence was rated as very low.

Conclusions

We identified very low certainty evidence across 23 studies for 13 different pre-arrest prediction scores to outcome following IHCA. No score was sufficiently reliable to support its use in clinical practice. We identified no evidence for children.

中文翻译：

---

院内心脏骤停后生存率的预骤停预测：对诊断测试准确性研究的系统评价

目标

评估用于院内心脏骤停生存结果的骤停前临床决策工具的测试准确性。

方法

我们检索了 Medline、Embase 和 Cochrane Library 从成立到 2022 年 1 月的随机和非随机研究。我们使用诊断准确性研究框架的质量评估来评估偏倚风险，并使用分级推荐评估、制定和评估方法来评估证据的确定性。我们报告了预测生存结果的敏感性、特异性、阳性预测结果和阴性预测结果。PROSPERO CRD42021268005。

结果

我们检索了 2517 项研究，其中 23 项研究使用 13 种不同评分：12 项研究调查 8 种不同评分评估生存结果，11 项研究使用 5 种不同评分来预测神经系统结果。所有都是历史队列/病例对照设计，仅包括成人。每个分数的测试准确性差异很大。在 12 项研究中，调查了 8 种不同评分来评估出院生存率/30 天生存率，预测生存率的阴性预测值 (NPV) 从 55.6% 到 100% 不等。GO-FAR 评分在 7 项研究中进行了评估，其中 NPVs 的脑功能类别 (CPC) 1 的生存率从 95.0% 到 99.2% 不等。两个分数评估了 CPC ≤ 2 的生存率，这些分数未经外部验证。在所有预测分数中，证据的确定性被评为非常低。

结论

我们在 23 项研究中确定了 13 种不同的逮捕前预测评分与 IHCA 后结果的低确定性证据。没有评分足够可靠来支持其在临床实践中的使用。我们没有发现儿童的证据。