Department of Biomedical Sciences and Public Health, Università Politecnica delle Marche, Ancona, Italy.
Anesthesia and Intensive Care Unit, Azienda Ospedaliero Universitaria delle Marche, Ancona, Italy.
Crit Care. 2023 Mar 5;27(1):85. doi: 10.1186/s13054-023-04386-w.
Management of bleeding trauma patients is still a difficult challenge. Massive transfusion (MT) requires resources to ensure the safety and timely delivery of blood products. Early prediction of MT need may be useful to shorten the time process of blood product preparation. The primary aim of this study was to assess the accuracy of shock index to predict the need for MT in adult patients with trauma. For the same population, we also assessed the accuracy of SI to predict mortality.
This systematic review and meta-analysis was performed in accordance with the PRISMA guidelines. We performed a systematic search on MEDLINE, Scopus, and Web of Science from inception to March 2022. Studies were included if they reported MT or mortality with SI recorded at arrival in the field or the emergency department. The risk of bias was assessed using the QUADAS-2.
Thirty-five studies were included in the systematic review and meta-analysis, for a total of 670,728 patients. For MT the overall sensibility was 0.68 [0.57; 0.76], the overall specificity was 0.84 [0.79; 0.88] and the AUC was 0.85 [0.81; 0.88]. Positive and Negative Likelihood Ratio (LR+; LR-) were 4.24 [3.18-5.65] and 0.39 [0.29-0.52], respectively. For mortality the overall sensibility was 0.358 [0.238; 0.498] the overall specificity 0.742 [0.656; 0.813] and the AUC 0.553 (confidence region for sensitivity given specificity: [0.4014; 0.6759]; confidence region for specificity given sensitivity: [0.4799; 0.6332]). LR+ and LR- were 1.39 [1.36-1.42] and 0.87 [0.85-0.89], respectively.
Our study demonstrated that SI may have a limited role as the sole tool to predict the need for MT in adult trauma patients. SI is not accurate to predict mortality but may have a role to identify patients with a low risk of mortality.
创伤患者出血的管理仍然是一个具有挑战性的难题。大量输血（MT）需要资源来确保血液制品的安全和及时供应。早期预测 MT 的需求可能有助于缩短血液制品准备的时间过程。本研究的主要目的是评估休克指数（SI）预测成人创伤患者 MT 需求的准确性。对于同一人群，我们还评估了 SI 预测死亡率的准确性。
本系统评价和荟萃分析按照 PRISMA 指南进行。我们在 MEDLINE、Scopus 和 Web of Science 上进行了系统搜索，检索时间从开始到 2022 年 3 月。如果研究报告了 SI 记录在现场或急诊室到达时的 MT 或死亡率，则纳入研究。使用 QUADAS-2 评估偏倚风险。
共纳入 35 项系统评价和荟萃分析，共纳入 670728 例患者。对于 MT，总体敏感性为 0.68 [0.57；0.76]，总体特异性为 0.84 [0.79；0.88]，AUC 为 0.85 [0.81；0.88]。阳性和阴性似然比（LR+；LR-）分别为 4.24 [3.18-5.65] 和 0.39 [0.29-0.52]。对于死亡率，总体敏感性为 0.358 [0.238；0.498]，总体特异性为 0.742 [0.656；0.813]，AUC 为 0.553（特异性给定敏感性的敏感性置信区间：[0.4014；0.6759]；特异性给定敏感性的特异性置信区间：[0.4799；0.6332]）。LR+和 LR-分别为 1.39 [1.36-1.42] 和 0.87 [0.85-0.89]。
我们的研究表明，SI 可能在预测成人创伤患者 MT 需求方面作用有限。SI 不能准确预测死亡率，但可能有助于识别低死亡率风险的患者。
