Li Anxin, Zhang Yuchen, Zhang Xiaoshi, Duan Zixiao, Chen Yan, Jiang Xiaoyan, Deng Wuquan
Chongqing University Central Hospital, Chongqing Emergency Medical Centre, School of Medicine, Department of Endocrinology, Chongqing, People's Republic of China.
Chongqing Medical University the Second Affiliated Hospital, Department of Hematology, Chongqing, People's Republic of China.
West J Emerg Med. 2025 Mar 22;26(3):657-666. doi: 10.5811/westjem.23666.
Heatstroke (HS) is a severe condition associated with significant morbidity and mortality. In this study we aimed to identify early risk factors that impacted the 30-day mortality of HS patients and establish a predictive model to assist clinicians in identifying the risk of death.
We conducted a retrospective cohort study, analyzing the clinical data of 203 HS patients between May 2016-September 2024. The patients were divided into two groups: those who had died within 30 days of symptom onset; and those who had survived. We analyzed the risk factors affecting 30-day mortality. A nomogram was drawn to visualize the clinical model. We used the receiver operating characteristic (ROC) curve and calibration curve to verify the accuracy of the nomogram. A decision curve analysis was also performed to evaluate the clinical usefulness of the nomogram.
Within a 30-day period, 57 patients (28.08%) died. The APACHE II score, the ratio of lactate-to-albumin (LAR), and the core temperature at 30 minutes after admission were independent risk factors for death of HS patients at 30 days. The area under the ROC curve (AUC) for predicting mortality based on the APACHE II score was 0.867, with a sensitivity of 96.5% and a specificity of 61.6%. Moreover, the AUC for predicting mortality based on the LAR was 0.874, with a sensitivity of 93.0% and a specificity of 77.4%. The AUC based on the core temperature at 30 minutes after admission was 0.774, with a sensitivity of 70.2% and a specificity of 78.8%. Finally, the AUC for predicting death due to HS using the combination of these three factors was 0.928, with a sensitivity of 82.5% and a specificity of 91.8%. The calibration curve and the decision-curve analysis showed that the new nomogram had better accuracy and potential application value in predicting the prognosis of HS patients.
A nomogram with these three indicators in combination-APACHE II score, lactate-to-albumin ratio, and core temperature at 30 minutes after admission-can be used to predict 30-day mortality of heatstroke patients.
中暑（HS）是一种与高发病率和死亡率相关的严重病症。在本研究中，我们旨在确定影响中暑患者30天死亡率的早期风险因素，并建立一个预测模型，以帮助临床医生识别死亡风险。
我们进行了一项回顾性队列研究，分析了2016年5月至2024年9月期间203例中暑患者的临床数据。患者被分为两组：症状发作后30天内死亡的患者；以及存活的患者。我们分析了影响30天死亡率的风险因素。绘制了列线图以直观显示临床模型。我们使用受试者工作特征（ROC）曲线和校准曲线来验证列线图的准确性。还进行了决策曲线分析以评估列线图的临床实用性。
在30天内，57例患者（28.08%）死亡。急性生理与慢性健康状况评分系统（APACHE II）评分、乳酸与白蛋白比值（LAR）以及入院30分钟时的核心体温是中暑患者30天死亡的独立风险因素。基于APACHE II评分预测死亡率的ROC曲线下面积（AUC）为0.867，敏感性为96.5%，特异性为61.6%。此外，基于LAR预测死亡率的AUC为0.874，敏感性为93.0%，特异性为77.4%。基于入院30分钟时核心体温的AUC为0.774，敏感性为70.2%，特异性为78.8%。最后，使用这三个因素联合预测中暑死亡的AUC为0.928，敏感性为82.5%，特异性为91.8%。校准曲线和决策曲线分析表明，新的列线图在预测中暑患者预后方面具有更好的准确性和潜在应用价值。
结合这三个指标——APACHE II评分、乳酸与白蛋白比值以及入院30分钟时的核心体温——的列线图可用于预测中暑患者的30天死亡率。
