Predicting Death and ICU Admission in COVID-19 Patients in ED

Overview

INTRODUCTION. The novel coronavirus designated SARS-CoV-2, has determined an international outbreak of respiratory illness named Covid-19. Patients with Covid-19 present primarily with fever, myalgia or fatigue, and dry cough. Based on available data from 5% to 10% among hospitalized patients will require ICU admission. In this context of overflow of critically ill patients, it is mandatory to establish clear and objective criteria to assess and predict a Covid-19 patient's need for ICU admission, and potentially predict death occurrence. Early Warning Scores (EWS) are used in hospitalized patients to predict clinical deterioration. Several study demonstrate the utility of EWS in ED to predict patient outcome. AIM. The objective of this study is to evaluate five EWSs, to predict the need for ICU admission and the mortality in patients admitted in ED with COVID-19. METHODS. This is a single-center, retrospective observational study. We will review the clinical records of all the patients consecutively admitted to our ED for Covid-19 over a three-weeks period (March 1 to 21, 2020). We will exclude from study cohort patients aged \<18 years old and pregnant women, and patients already on oro-tracheal intubation at ED arrival. Based on clinical records five EWS will be calculated: NEWS, NEWS2, qSOFA, MEWS, REMS. Study endpoints. The primary study endpoints will be death at 7 days, and need for ICU at 7 days, since ED admission. As secondary endpoints we will evaluate need for ICU and death at 24 and 48 hours since ED admission. Statistical Analysis Receiver operating characteristic (ROC) curve analysis will be used to evaluate the overall performance of the selected EWSs in predicting the defined adverse outcomes. According to Youden's index we will estimate the optimal cut-off points and corresponding sensitivity and specificity at selected score threshold values. The comparison between the ROC AUCs will be made according to DeLong method.

INTRODUCTION After the first cases identified in Wuhan city (China) on December 2019, the novel coronavirus designated SARS-CoV-2 has caused a global epidemic of respiratory illness named COVID-191. To date, more than, 3,000,000 cases have been reported worldwide, including more than 300,000 deaths. Typical COVID-19 patients present with fever, myalgia or fatigue, and dry cough. Severe cases progress to severe dyspnoea and hypoxemia within one week after the onset of symptoms 3-5. In hospitalized COVID-19 patients, the prevalence of hypoxemic respiratory failure is around 20%, and more than 25% of them may require intensive care treatment. The increasing number of COVID-19 cases has challenged the healthcare systems worldwide. Given the current overflow of critically ill patients in the Emergency Departments (EDs), an early identification of patients who need admittance to an intensive care unit (ICU) because of an increased risk of unfavourable outcome is necessary. Although general guidelines for ICU admission and triage exist, only limited guidance is available for the specific setting of COVID-19 patients. In this context of overwhelming demand for medical assessment and triage in ED, early warning scores (EWS) may be useful. EWS are based on a rapid and quantitative assessment of changes in vital signs, and were developed to identify and track patients at risk of deterioration in non-critical areas of the hospital in order to ensure an early stabilisation and ICU transfer when appropriate and prevent avoidable cardiac arrest. However, in recent years these scores have been used in ED to predict ICU admission and mortality. Use of EWS has recently been proposed for the triage of COVID-19 patients in ED. However, their usefulness has not been demonstrated yet. METHODS Study design This is a retrospective observational study conducted in the ED of the largest urban teaching hospital in Rome, a referral center for COVID-19 in central Italy. The participants reviewed the electronic medical records (EMR) of all adults (\>18y) patients admitted to ED for suspected COVID-19 over three consecutive weeks from March 1 to March 21, 2020, tested for COVID-19 according to the WHO interim guidance. Inclusion and exclusion criteria Study will include only patients whose diagnosis is confirmed with real-time reverse-transcriptase-polymerase-chain-reaction assay of nasal and pharyngeal swab specimens. Will be excluded from the study cohort pregnant women, patients discharged from ED with normal chest x-ray findings, and patients who were already mechanically ventilated on ED arrival. For patients with more than one access to our ED, only the latest access will be included in the analysis. Study variables The following information will be extracted from digital clinical records: age, sex, clinical history and presentation, temperature, heart rate (HR), respiratory rate (RR), arterial blood pressure (BP), Glasgow Coma Scale (GCS) score, oxygen therapy, peripheral oxygen saturation (SpO2), and chest x-ray findings.Clinical signs, including SpO2, will be assessed on arrival at the ED. National Early Warning Score (NEWS), National Early Warning Score 2 (NEWS2), Quick Sepsis Related Organ Failure Assessment (qSOFA), Modified Early Warning Score (MEWS), and Rapid Emergency Medicine Score (REMS) will be calculated for each patient. For NEWS2 calculation, patient will be considered at risk of type 2 respiratory failure if had a confirmed history of chronic obstructive pulmonary disease (COPD). Study endpoints The primary study endpoints considered will be death or ICU admission within 7 days from arrival at the ED. Secondary endpoints will be death or ICU admission within 24 and 48 hours from ED arrival. Criteria for ICU admission The criteria for ICU admission of COVID-19 patients in include need for invasive respiratory support, or extra-pulmonary organ failure such as circulatory shock requiring vasopressors, or renal failure. These criteria are expected to be consistent throughout the study period. Statistical Analysis and sample size Continuous variables will be reported as median \[interquartile range\], and will be compared at univariate analysis by Mann-Whitney U test. Categorical variables will be reported as absolute number (percentage), and will be compared by Chi-square test (with Fisher's test if appropriate). Receiver operating characteristic (ROC) curve analysis will be used to estimate the overall performance of the evaluated scores in predicting the defined adverse outcomes. For each score threshold values will be calculated sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio (+LR), and negative likelihood ratio (-LR). The Youden index will be used to estimate the optimal cut-off points for sensitivity and specificity. The comparison between the ROC AUCs will be made according to DeLong method. A p value ≤ 0.05 will be regarded as significant. Data will be analyzed by IBM SPSS statistics v25® (IBM, IL, USA). Sample size. For a correct estimation of ROC Curve a minimum of 50 patients for each endpoint should be included in the analysis. Given the estimate flow of at least 300 COVID-19 confirmed patients in the study period, and the actual rate of ICU admission in these patients (about 20%), the sample should be adequate for statistical estimation. Aim of the study The objective of the present study is to assess the ability of EWS to predict ICU admission and mortality in COVID-19 patients in the emergency department. Ethical considerations All the patients accessing the "COVID" ED signs a comprehensive ethical agreement for collection of blood samples and clinical data, for bio-bank and research purposes (Informative mod 147 25/06/2019).