Authors (including presenting author) :
Kwan MK (1), Yau YF (2), Lui TW (1), Ng SM (2), Chan SY (1), Lam SC (1), Yuen LC (1)
Affiliation :
(1)Accident of Emergency Department of Princess Margaret Hospital, (2)Accident of Emergency Department of Alice Ho Miu Ling Nethersole Hospital
Keyword 2: :
Predicting Sudden Cardiac Arrest
Keyword 3: :
Accident and Emergency Department
Introduction :
A crowded Accident and Emergency Department (A&E) is a well-known global challenge. Healthcare providers must manage patients with multiple diseases, critical conditions, and sudden changes in their health status. Given the current situation, early recognition and response are priorities for patient safety. Therefore, it is imperative that a screening tool be implemented to identify high-risk patients, improve monitoring, and provide timely interventions. These measures will result in better resource allocation and risk management, ultimately decreasing the mortality rate in the A&E. The NEWS2 is the standardized track-and-trigger early warning system implemented in the United Kingdom, with increasing adoption at the international level. However, its performance for predicting sudden cardiac arrest (SCA) during A&E stay warrants evaluation.
Objectives :
The objective of the research is the practicality of employing NEWS2 in our A&E, where an effective screening tool is essential for identifying high-risk patients for early mortality, thereby enhancing monitoring and prognostic capabilities.
The primary outcome is assessing the predictive performance of NEWS2 for SCA cases in A&E. The NEWS2 for SCA cases were compared to those of non-SCA cases to identify significant differences. Additionally, the outcome of cases with a NEWS2 ≥5 was compared to assess the suggested threshold.
Methodology :
This retrospective study analyzed adult (≥18 years) A&E attendances triaged as Category 3 to 5 from January 2023 to December 2024, using Clinical Management System (CMS) electronic A&E records. The data retrieval period spanned two years, encompassing all cases of SCA during this timeframe. The non-SCA controls were randomly sampled according to triage-category proportions. It involved the assessment of patients’ physiological parameters at triage using NEWS2. Cases with missing physiological parameters and pregnancy were excluded. Discriminatory performance was evaluated using receiver operating characteristic (ROC) analysis; sensitivity, specificity, likelihood ratios, relative risk (RR), and area under the curve (AUC) were calculated, with NEWS2 ≥5 assessed as a clinically relevant threshold.
This study was approved by the Central Institutional Review Board (IRB) of the Hospital Authority and the Joint CUHK-NTEC Clinical Research Ethics Committee (CREC).
Result & Outcome :
Of 601 recruited cases, 581 were valid for analysis, including 30 SCA cases and 551 non-SCA cases. NEWS2 showed excellent discrimination for SCA during A&E stay (AUC 0.884 [95% CI 0.818–0.950]). Using NEWS2 ≥5, sensitivity was 0.633 and specificity 0.949 (LR+ 12.40; RR 19.6). For the 7-day outcome, discrimination remained excellent (AUC 0.903 [95% CI 0.847–0.959]) with sensitivity 0.694 and specificity 0.960 (LR+ 17.35; RR 25.8). Median NEWS2 was significantly higher in SCA than non-SCA patients (5.5 vs 1; P< 0.001).
