Journal Pre-proof Early predictors of mortality for moderate to severely ill patients with Covid-19

Gökhan Aksel, Mehmet Muzaffer Islam, Abdullah Algin, Serkan Emre Eroğlu, Gökselin Beleli Yaşar, Enis Ademoğlu, Ümit Can Dölek PII:

S0735-6757(20)30770-1

DOI:

https://doi.org/10.1016/j.ajem.2020.08.076

Reference:

YAJEM 159344

To appear in:

American Journal of Emergency Medicine

Received date:

16 July 2020

Revised date:

21 August 2020

Accepted date:

23 August 2020

Please cite this article as: G. Aksel, M.M. Islam, A. Algin, et al., Early predictors of mortality for moderate to severely ill patients with Covid-19, American Journal of Emergency Medicine (2020), https://doi.org/10.1016/j.ajem.2020.08.076

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

© 2020 Published by Elsevier.

Journal Pre-proof TITLE PAGE Title of the article: Early predictors of mortality for moderate to severely ill patients with Covid-19 The running head: Predictors of mortality for Covid-19 Type of article: Original article Word counts: Abstract: 339

of

Main text: 2121

ro

References: 16

-p

Number of figures: 2

re

Number of tables: 4

lP

AUTHORS

1. Gökhan AKSEL, MD, Emergency Medicine Department, University of Health Sciences, Umraniye

na

Training and Research Hospital, Istanbul, Turkey.

Jo ur

E-mail: [email protected] (Corresponding Author) Orcid ID: 0000-0002-5580-3201

2. Mehmet Muzaffer ĠSLAM, MD, Emergency Medicine Department, University of Health Sciences, Umraniye Training and Research Hospital, Istanbul, Turkey. E-mail: [email protected] Orcid ID: 0000-0001-6928-2307 3. Abdullah ALGIN, MD, Emergency Medicine Department, University of Health Sciences, Umraniye Training and Research Hospital, Istanbul, Turkey. E-mail: [email protected]

Journal Pre-proof Orcid ID: 0000-0002-9016-9701 4. Serkan Emre EROĞLU, MD, Emergency Medicine Department, University of Health Sciences, Umraniye Training and Research Hospital, Istanbul, Turkey. E-mail: [email protected] Orcid ID: 0000-0002-3183-3713 5- Gökselin Beleli YAġAR, MD, Emergency Medicine Department, University of Health Sciences,

of

Umraniye Training and Research Hospital, Istanbul, Turkey.

ro

E-mail: [email protected]

-p

Orcid ID: 0000-0001-5974-8695

re

6-Enis ADEMOĞLU, MD, Emergency Medicine Department, University of Health Sciences,

E-mail: [email protected]

na

Orcid ID: 0000-0002-6330-666X

lP

Umraniye Training and Research Hospital, Istanbul, Turkey.

Jo ur

7-Ümit Can DÖLEK, MD, Emergency Medicine Department, University of Health Sciences, Umraniye Training and Research Hospital, Istanbul, Turkey. E-mail: [email protected] Orcid ID: 0000-0003-2895-7472

Corresponding Author: Name : Gökhan AKSEL, MD, Associated Professor. Mail address: Ümraniye Eğitim ve AraĢtırma Hastanesi, Acil Tıp Kliniği, Elmalıkent Mahallesi Adem Yavuz Cad. No:1 Ümraniye / Ġstanbul / Türkiye

Journal Pre-proof Phone numbers: +90 505 350 86 90 E-mail address: [email protected]

Keywords: 2019 novel coronavirus disease, SARS-CoV-2 infection, COVID-19 pandemic, mortality This article has not been previously presented at any event (congress, symposium etc.) Source(s) of support / Funding: No funding.

of

Conflict(s) of Interest : The authors declare no conflict of interest.

Jo ur

na

lP

re

-p

ro

Acknowledgements: None

Journal Pre-proof TITLE: Early predictors of mortality for moderate to severely ill patients with Covid-19 1. INTRODUCTION Since December 2019, the world has been facing an unprecedented coronavirus (Covid-19) outbreak that started in China and rapidly spread internationally. Although outbreaks of the bird flu (H5N1), severe acute respiratory syndrome (SARS), swine flu (H1N1), Middle East respiratory syndrome (MERS), Ebola virus, and Zika virus have occurred in recent history, the most similar pandemic to the Covid-19 outbreak in terms of the number of people affected was the Spanish flu,

of

which occurred about a century ago (1–4). The Spanish flu is thought to have been responsible for 17–

ro

50 million deaths (5). According to the World Health Organization‘s (WHO) most recent report dated

-p

17 August 2020, 21,294,845 people have been infected by Covid-19 and 761,779 have died (6).

re

Treatment guidelines have been constantly updated in the past few months. It is clear that better disease management and treatment algorithms for pandemics are needed, given that there is not

lP

yet an effective treatment or vaccine for this virus and the outbreak has not been fully controlled. More descriptive studies are needed to determine an effective approach for prevention and treatment of the

na

disease. In particular, to develop patient treatment algorithms, it is crucial to determine the factors that

Jo ur

affect the mortality and clinical conditions of patients. Thus, in this study, we aimed to determine the parameters that can predict the mortality of moderate to severely ill patients with laboratory-confirmed Covid-19.

Journal Pre-proof 2. METHODS 2.1. Study design and setting

This prospective observational study was carried out in the emergency department (ED) of a tertiary care teaching hospital between 16 April 2020 and 16 June 2020 following approval by the local ethical committee.

of

2.2. Selection of participants

Patients older than 18 years who met the criteria for moderate to severe Covid-19 were

ro

consecutively included in the study. Moderate to severe Covid-19 was defined according to the

-p

COVID-19 Outbreak Management and Working Guideline created and published by the Turkish

re

Ministry of Health. It is recommended that these patients be hospitalized, and patients who do not meet these criteria should undergo home isolation (7). This definition included patients with symptoms

lP

of fever, muscle ache, cough, dyspnea, sore throat or nasal congestion, and at least one of the following: O2 saturation below 93% in room air; tachypnea (>22/min); poor prognostic criteria in

na

blood tests (blood lymphocyte count 1000 ng/ml, ferritin>500ng/ml, or C-reactive

Jo ur

protein [CRP] >40 mg/L); bilateral diffuse parenchymal infiltration on chest x-ray or computed tomography; age of >50 years; and presence of comorbid diseases (diabetes mellitus, hypertension, malignancy, chronic lung disease, immunodeficiency disorders). Finally, at least one Covid-19 polymerase chain reaction test result was required to be positive. Patients with all negative results were excluded from the study.

2.3. Measurements

The baseline characteristics of the included patients—age, gender, complaint at the time of admission, comorbid diseases, angiotensin-converting enzyme inhibitors (ACEI), angiotensin II receptor blockers (ARBs), prophylactic hydroxychloroquine use, smoking, clinical features (such as vital signs), laboratory test results (white blood cell count, neutrophil count, lymphocyte count, CRP level, paO2 level), and fiO2 level—were recorded on a form created by a research assistant. The

Journal Pre-proof patients were followed up to determine the treatment they received, the duration of treatment, and the outcome.

2.4. Outcome Measures In this study, we aimed to determine the factors that affect mortality in moderate to severe Covid19 patients. The primary outcome was 30-day mortality rate. The secondary outcome was determination of the factors that affect the development of acute respiratory distress syndrome

of

(ARDS).

ro

2.5. Statistical Analysis

-p

SPSS 26 (IBM Corp. Released 2019. IBM SPSS Statistics for Windows, Version 26.0. Armonk, NY: IBM Corp.) was used for statistical analyses. The Shapiro–Wilk test was used to determine the

re

distribution of normality for the continuous variables. Normally distributed variables were expressed

lP

as the mean and standard deviation, while non-normally distributed variables were expressed with the median and 25th and 75th percentiles. Categorical data were represented by frequency and percentage.

na

Comparison between the groups were performed by a Student‘s t-test or Mann–Whitney U test according to the normal distribution of the continuous variables. Comparison of the categorical data

Jo ur

was performed with a Chi-square test and Fisher‘s exact test when necessary. A receiver operating characteristic (ROC) curve was drawn for the laboratory values that were statistically associated with mortality, and the value with the highest sum of sensitivity and specificity was accepted as the cut-off value. Positive and negative likelihood ratios were determined according to this cut-off value. To determine the predictive value of several variables for both mortality and ARDS development, a multivariate regression model was created using variables whose p values were