[Comparison of four early warning scores in predicting the prognosis of critically ill patients in secondary hospitals]
Xiaoqin Su,Hongyan Zhang,Wenjun Yuan,Meng Yi,Chenghao Fu,Jiawei Jiang,Hongmei Gao
DOI: https://doi.org/10.3760/cma.j.cn121430-20230614-00441
Abstract:Objective: To explore the predictive value of acute physiology and chronic health evaluation II (APACHE II), sequential organ failure assessment (SOFA), quick sequential organ failure assessment (qSOFA) and modified early warning score (MEWS) in evaluating the prognosis of patients in intensive care unit (ICU) of secondary hospitals, and to provide guidance for clinical application. Methods: The clinical data of adult critical patients admitted to the ICU of Wanzhou District First People's Hospital from October 2022 to April 2023 were retrospectively analyzed. According to the clinical outcome of ICU, the patients were divided into improvement group and death group. The general information, blood routine, heart, liver and kidney function indicators, coagulation indicators, blood gas analysis, APACHE II score, SOFA score, qSOFA score, MEWS score at the time of admission to the ICU, the number of cases of invasive mechanical ventilation (IMV) and continuous blood purification (CBP) were compared between the two groups. Univariate analysis was performed, and multivariate Logistic regression analysis was used to analyze the related factors of death. Receiver operator characteristic curve (ROC curve) was used to analyze the predictive value of the four scores in ICU patients. Results: A total of 126 patients were included, of which 45 patients died in the ICU and 81 patients improved and transferred out. Univariate analysis of death-related critically ill patients showed that procalcitonin (PCT), serum creatinine (SCr), blood urea nitrogen (BUN), albumin (ALB), prothrombin time (PT), activated partial prothrombin time (APTT), D-dimer, pH value, HCO3-, blood lactic acid (Lac), number of patients treated with IMV and CBP, APACHE II score, SOFA score, qSOFA score and MEWS score were significantly different between the two groups (all P < 0.05). Multivariate Logistic regression analysis showed that the APACHE II score [odds ratio (OR) = 1.115, 95% confidence interval (95%CI) was 1.025-1.213, P = 0.011], SOFA score (OR = 1.204, 95%CI was 1.037-1.398, P = 0.015), MEWS score (OR = 1.464, 95%CI was 1.102-1.946, P = 0.009), and APTT (OR = 1.081, 95%CI was 1.015-1.152, P = 0.016) were independent risk factors affecting the mortality of critically ill patients in the ICU. ROC curve analysis showed that APACHE II, SOFA, qSOFA, and MEWS scores could predict the prognosis of critically ill ICU patients, among which SOFA score had the strongest predictive effect, and the area under the curve (AUC) was 0.808. There was a statistically significant difference in the time required for the four scores (F = 117.333, P < 0.001), among which the MEWS scoring required the shortest time [(1.03±0.39) minutes], and the APACHE II scoring required the longest time [(2.81±1.04) minutes]. Conclusions: APACHE II, SOFA, qSOFA, and MEWS scores can be used to assess the severity of critically ill patients and predict in-hospital mortality. The SOFA score is superior to other scores in predicting severity. The MEWS is preferred because its assessment time is shortest. Early warning score can help secondary hospitals to detect potentially critical patients early and provide help for clinical rapid urgent emergency decision-making.