开发并验证用于预测溺水相关心脏骤停患者结局的临床-生物学评分
本研究基于两个回顾性多中心溺水患者队列,纳入发生溺水相关心脏骤停(DACA)的患者,在推导队列中用logistic回归筛选第28天死亡的危险因素并构建预测评分,再在验证队列中评估。推导队列103例,第28天死亡率51%(53/103);独立早期危险因素包括心肺复苏持续时间>20分钟、ICU入科体温<34°C、需要有创机械通气、乳酸>7 mmol/L。评分AUC 0.91(95%CI 0.86-0.97),最佳截断值4分,敏感度92%、特异度82%、PPV 84%、NPV 91%;验证队列81例,AUC 0.82,同一4分阈值下敏感度81%、特异度67%、PPV 72%、NPV 77%。作者认为DACA评分≥4分者第28天生存可能性显著降低,并需前瞻性验证。
要点
- 推导队列103例溺水相关心脏骤停患者第28天死亡率为51%(53/103),评分AUC达0.91(95%CI 0.86-0.97)。
- 独立早期危险因素:心肺复苏>20分钟(OR 6.40,95%CI 1.88-23.32;p=0.003)、ICU入科体温<34°C(OR 8.84,95%CI 2.66-32.92;p<0.001)。
- 另两项独立危险因素:需要有创机械通气(OR 6.83,95%CI 1.47-40.87;p=0.02)、乳酸>7 mmol/L(OR 3.56,95%CI 1.01-13.07;p=0.04)。
- 评分最佳截断值4分:推导队列敏感度92%(95%CI 82-98%)、特异度82%(95%CI 67-91%)、PPV 84%(95%CI 72-95%)、NPV 91%(95%CI 79-96%)。
- 验证队列81例同一4分阈值下敏感度81%、特异度67%、PPV 72%、NPV 77%,AUC 0.82。
适用人群与场景
- 发生溺水相关心脏骤停(DACA)并收入ICU的患者,来自两个回顾性多中心溺水患者队列。
诊断标准与关键阈值
- DACA评分由心肺复苏持续时间>20分钟、ICU入科体温<34°C、需要有创机械通气、乳酸>7 mmol/L四项变量构成;最佳截断值为4分。
研究设计与关键结果
- 研究类型为基于两个回顾性多中心队列的评分开发与验证研究,推导队列103例、验证队列81例。
- 推导队列第28天死亡率为51%(53/103);独立早期危险因素:心肺复苏持续时间>20分钟(OR 6.40,95%CI 1.88-23.32;p=0.003)、ICU入科体温<34°C(OR 8.84,95%CI 2.66-32.92;p<0.001)、需要有创机械通气(OR 6.83,95%CI 1.47-40.87;p=0.02)、乳酸>7 mmol/L(OR 3.56,95%CI 1.01-13.07;p=0.04)。
- 推导队列评分AUC 0.91(95%CI 0.86-0.97);4分截断值下敏感度92%(95%CI 82-98%)、特异度82%(95%CI 67-91%)、PPV 84%(95%CI 72-95%)、NPV 91%(95%CI 79-96%)。
- 验证队列81例,评分AUC 0.82;同一4分阈值下敏感度81%、特异度67%、PPV 72%、NPV 77%。
证据类型、等级与局限
- 证据类型:基于两个回顾性多中心队列的评分开发与验证研究。
- 局限:作者指出DACA评分需前瞻性验证并评估其用途。
对床旁实践的改变
- 作者认为,在收入ICU的溺水相关心脏骤停患者中,DACA评分≥4分者第28天生存可能性显著降低。
- 作者指出,DACA评分的前瞻性验证及其用途评估需在未来开展。
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: While several scoring systems have been developed to predict short-term outcome in out-of-hospital cardiac arrest patients, there is currently no dedicated prognostic tool for drowning-associated cardiac arrest (DACA) patients.
METHODS: Patients experiencing DACA from two retrospective multicenter cohorts of drowning patients were included in the present study. Among the patients from the development cohort, risk-factors for day-28 mortality were assessed by logistic regression. A prediction score was conceived and assessed in patients from the validation cohort.
RESULTS: Among the 103 included patients from the development cohort, the day-28 mortality rate reached 51% (53/103). Identified independent early risk-factors for day-28 mortality included cardiopulmonary resuscitation duration longer than 20 min (OR 6.40 [95% CI 1.88-23.32]; p = 0.003), temperature at Intensive Care Unit admission <34 °C (OR 8.84 [95% CI 2.66-32.92]; p < 0.001), need for invasive mechanical ventilation (OR 6.83 [95% CI 1.47-40.87]; p = 0.02) and lactate concentration > 7 mmol/L (OR 3.56 [95% CI 1.01-13.07]; p = 0.04). The Area Under the ROC Curve (AUC) of the developed score based on those variables reached 0.91 (95% CI, 0.86-0.97). The optimal cut-off for predicting poor outcomes was 4 points with a sensitivity of 92% (95% CI, 82-98%), a specificity of 82% (95% CI, 67-91%), a positive predictive value (PPV) of 84% (95% CI, 72-95%) and a negative predictive value (NPV) of 91% (95% CI, 79-96%). The assessment of this score on the validation cohort of 81 patients exhibited an AUC of 0.82. Using the same 4 points threshold, sensitivity, specificity, PPV and NPV values of the validation cohort were: 81%, 67%, 72% and 77%, respectively.
CONCLUSION: In patients suffering from drowning induced initial cardiac arrest admitted to ICU with a DACA score ≥ 4, the likelihood of survival at day-28 is significantly lower. Prospective validation of the DACA score and assessment of its usefulness are warranted in the future.
参考来源
- Reizine F, Michelet P, Delbove A, et al. Development and validation of a clinico-biological score to predict outcomes in patients with drowning-associated cardiac arrest. Am J Emerg Med. 2024. PMID: 38670053. doi:10.1016/j.ajem.2024.04.032
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。