Construction of early warning index for critical condition of severe community-acquired pneumonia in emergency department
10.3760/cma.j.cn211501-20200901-03726
- VernacularTitle:急诊重症社区获得性肺炎患者病情危重程度预警指标的构建
- Author:
Xiaoying LIU
1
;
Donglei SHI
;
Fan LI
;
Aihui LIU
;
Liyuan TIAN
Author Information
1. 北京协和医院急诊科 100730
- Keywords:
Emergency;
Severe community-acquired pneumonia;
Early warning index
- From:
Chinese Journal of Practical Nursing
2021;37(28):2221-2228
- CountryChina
- Language:Chinese
-
Abstract:
Objective:Based on the clinical indicators of severe community-acquired pneumonia in emergency department, the early warning indicators of critical condition were constructed.Methods:The general information, vital signs before entering ICU, laboratory indexes and the survival rate in 30 days of 118 patients with community-acquired pneumonia were retrospectively collected in the emergency care unit of Peking Union Medical College Hospital from January to December 2018. The indexes of death patients (24 cases) and alive patients (98 cases) underwent comparative analysis, using ROC curve to predict the clinical outcome and reliable parameters of emergency patients with community-acquired pneumonia. The best cutoff value was determined according to Youden index and then undergoing multiple factors Logistic stepwise regression analysis. Then early warning model of critical degree was finally built.Results:Eleven indicators were used to predict the criticality of patients with acute community-acquired pneumonia, including AVPU (alert, voice, pain, unresponsive) score, Glasgow Coma Scale(GCS), heart rate, pulse oxygen saturation (SpO 2), fraction of inspiration O 2, oxygenation index,potential of hydrogen, blood potassium (K +), bilirubin, urea nitrogen, and C reactive protein, the optimal cutoff values were 2 points, 8 points, 91 times per minute, 0.94, 41%, 81.20%, 7.38, 4.0 mmol/L, 10.90 μmol/L, 2.23 mmol/L, and 41.5 mg/L. Multiple Logistic stepwise regression showed that the independent factors for predicting death were GCS score, SpO 2 and urea nitrogen, and the predictive ability of the early-warning model was 83.7%. Conclusions:GCS score, SpO 2 and urea nitrogen are effective early warning indicators for the severity of severe community-acquired pneumonia in emergency department, which are conducive to the rapid and efficient early identification and treatment of critically ill patients. Therefore, they are worthy of promotion and application in clinical practice.
