key: cord-0758822-bpj7ha3j
authors: Mehraeen, Esmaeil; Karimi, Amirali; Barzegary, Alireza; Vahedi, Farzin; Afsahi, Amir Masoud; Dadras, Omid; Moradmand-Badie, Banafsheh; Alinaghi, Seyed Ahmad Seyed; Jahanfar, Shayesteh
title: Predictors of mortality in patients with COVID-19 – a systematic review
date: 2020-10-17
journal: Eur J Integr Med
DOI: 10.1016/j.eujim.2020.101226
sha: da5e2d5798c58f60df58cf8f81723883d4e6de75
doc_id: 758822
cord_uid: bpj7ha3j

INTRODUCTION: In the current COVID-19 pandemic, disease diagnosis is essential for optimal management and timely isolation of infected cases in order to prevent further spread. The aim of this study was to systematically review the assessment of risk and model the predictors of mortality in COVID-19 patients. METHODS: A systematic search was conducted of PubMed, Scopus, Embase, Google Scholar, and Web of Science databases. Variables associated with hospital mortality using bivariate analysis were included as potential independent predictors associated with mortality at the p <0.05levels. RESULTS: We included 114 studies accounting for 310,494 patients from various parts of the world. For the purpose of this analysis, we set a cutoff point of 10% for the mortality percentages. High mortality rates were defined as higher than 10% of confirmed positive cases and were given a score of two, while low mortality percentage (<10%) was assigned to the score of one. We then analyzed the associations between 72 variables and the observed mortality rates. These variables included a large range of related conditions such as demographics, signs and symptoms and related morbidities, vital signs, laboratory findings, imaging studies, underlying diseases, and the status of countries' income based on United Nation's classifications. CONCLUSION: Findings suggest that older age, hypertension, and diabetes mellitus conferred a significant increased risk of mortality among patients with COVID-19. In the multivariate analysis, only diabetes mellitus demonstrated an independent relationship with increased mortality. Further studies are needed to ascertain the relationship between possible risk factors with COVID-19 mortality.

Coronavirus Disease 2019 (COVID- 19) attracted worldwide attention as an international public health emergency and the first pandemic caused by a coronavirus (1, 2) . The global number of cases and deaths has reached almost 11,000,000 and 404,396 (3), imposing an unavoidable burden and pressure on the healthcare systems in all countries as well as their economies (4) (5) (6) .

This inflicted pressure requires careful strategies as well as their implementation. Actions should be guided by scientific facts to minimize the imposed harms, and this has created an urgent need to examine studies and model outcomes (7) .

Miscellaneous COVID-19 mortality rates are reported so far, but an accurate mortality rate determination is still a challenge and might not be available (8) . The mortality rate shows an increase in older populations having underlying diseases (9, 10) . Characteristic signs and symptoms raise clinical suspicions and are vital for detecting an infected individual (11, 12) . In the current pandemic settings, diagnosing the disease is essential for providing the best management to the infected people and avoiding further spreading the disease to others through timely isolations (13, 14) . All this outlines the vast importance of understanding signs and symptoms and their role in the disease's pathogenesis and clinical manifestations. Estimating the patterns of signs, symptoms, comorbidities, and other variables, and their association with mortality rates might be the key to the management of COVID-19. Understanding these issues helping us to provide adequate and in-time personalized care based on individual's conditions. This article aimed to provide evidence-based modeling of COVID-19 by addressing variables that might be related to an alteration in Severe Acute Respiratory Syndrome-Coronavirus-2 (SARS-CoV-2) mortality rates. Variables were divided into six major categories: demographics, signs and symptoms and related morbidities, laboratory findings and vital signs, imaging studies, underlying conditions, and countries' income. The focus was on increasing knowledge about the disease for better prevention, diagnosis, and treatment.

This study was a systematic review conducted in 2020 was conducted in order to provide a risk analysis and model predictors of mortality in COVID-19 patients.

A systematic review was conducted using PubMed, Scopus, Embase, Google Scholar, and Web of Science between January 1, 2020, and June27, 2020. A literature search was done using the keywords in combination on the following search strategy: 

For study selection, we followed the PRISMA guidelines ( Figure 1 ). The selection of the studies was performed for relevance by titles and abstracts by three independent investigators. The full texts were reviewed for the eligibility criteria. The English-written peer-reviewed original papers published from January 1, 2020, to June 25, were included. The exclusion criteria were as follows:

− Different types of studies, such as ongoing protocols, abstracts, reports, and letters to the editor.

− No access to the full-text document.

− Duplicated results in databases.

− Papers addressing non-human studies, or discussing COVID-19 in general, without reference to the disease's mortality.

The title and abstract of each manuscript were evaluated, and the most relevant manuscripts were selected based on the previously mentioned inclusion and exclusion criteria. Two independent investigators conducted the screening, and disagreement was resolved through discussion.

We used the data extraction forms, including information on the authors, year of publication, country, sample size, age, gender, clinical symptoms (e.g., fever, chills). This information was obtained independently by two investigators, and disagreement was resolved through discussion.

To ensure the quality assessment of selected articles, a checklist (Table 1 ) with 15 items was developed based on the relevant studies (15) (16) (17) . The quality of articles was evaluated by the scoring of checklist items and estimated the mean score and rated on a three-point scale: low quality (0-5), medium quality (6) (7) (8) (9) (10) , and high quality (11) (12) (13) (14) (15) . The full text of selected articles was then thoroughly studied to extract the essential findings. The qualified full-text articles were included, and their results were discussed to make the final selection. After reading the full text of all eligible papers, the researchers decided to include/exclude each study. Does the study provide any theoretical framework for the evaluation method? 3

Does the theoretical framework of the study include any health promotion theory? 4

Does the study provide a timeframe for the data collection? 5

Does the study identify the country where the search was conducted? 6

Does the study mention that the reviewed current evidence was downloaded for evaluation? 7

Does the study discuss the selection criteria for current evidence to be included or excluded for review? 8

Does the study provide a clear description of the evaluation method? 9

Are there at least two independent data extractors with a consensus procedure in place in case of disagreement? 10

Is a list of the reviews current evidence provided? 11 Does the study discuss the findings of the evaluation? 12

Does the study look at the reviewed evidence to promote or enable behavioral change? 13 Does the study discuss any limitations? 14 Does the study provide any future recommendations in general? 15 Does the study state any conflict of interest?

A quantitative synthesis of the identified studies was carried out according to the search strategy and the identified characteristics; these were later analyzed according to frequencies (n) to facilitate data interpretation. We used SPSS software (version 26) for data analysis and data extraction. The data was analyzed and categorized using the following variables listed in Table  2 . Thesefactors were extracted after thorough and careful reading of the articles to include them as efficient as possible. Variables associated with hospital mortality in the bivariate analysis were included as potential independent predictors at the p <0.05 levels. The final model retained those variables associated with the mortality at the p<0.05 level.

In this study, using the applied systematic search strategies, 751 sources were identified and retrieved. After an initial review of retrieved articles, 148 duplicates were removed, and the title and abstract of the remaining 603 articles were reviewed. Applying the selection criteria, 489 articles were excluded, and only 114 articles met inclusion criteria and were included in the final review ( Figure 1) . The mean quality score of the selected articles was 13 (= range: 11 to 15),

indicating the high quality of these articles. We included 114 studies accounting for 310,494 patients from various parts of the world. For this analysis, we set a cutoff point of 10% for the mortality percentages. High mortality rates were defined as higher than 10% of confirmed positive cases and were given a score of two, while low mortality percentage (<10%) was assigned to the score of one. We then analyzed the associations between 72 variables and the observed mortality rates. These variables included a vast range of related conditions such as demographics, signs and symptoms and related morbidities, vital signs, laboratory findings, imaging studies, underlying diseases, and the status of countries' income based on United Nation's classifications. To categorize countries based on their incomes, we used the United Nation's classification of least developed countries, developing countries, and developed countries and gave them one, two, and three points, respectively. For instance, China, the country reporting the first cases of COVID-19, scores two in our classification. Iran also scored two in this classification. USA and Western European countries were among those achieving a score of three. Table 3 .

In the bivariate analysis applying a cut point of P <0.05 for significance, three variables, mean age (P<0.001, OR of 1.18; 95% CI: 1.08-1. The detailed relationship of each variable with mortality rates is thoroughly presented in Table   3 .

Taking a step forward, we then conducted a multivariate regression analysis of variables that were found to be significantly associated with COVID-19, with a similar threshold of significance (P <0.05). Independent association with the mortality was observed with Diabetes 

and currently is the greatest challenge of the health care system all over the world. Preliminary molecular studies have shown that bats can be the potential reservoir of the virus (18) . To date, however, no specific treatment has been approved for this disease, and the core of current therapies is symptomatic and supportive care (19) . Therefore, it is crucial to study the rate and to affect factors of mortality in this disease. During this short time-period, many studies have been conducted in this field. Based on these studies, the predictors of mortality can be divided into five categories (20) .

Among the demographic factors, age is one of the most important factors affecting mortality, and in our study, age was significantly associated with increased mortality. Studies have shown that the age-related defects in immune cell function and increased production of inflammatory cytokines may play a role (21) . Studies have shown that the male gender is also a risk factor for severity and higher mortality (22) .

According to the findings, the first ten most common observed signs and symptoms with the highest mean percentage affecting COVID-19 mortality were fever, cough, olfactory dysfunction, postnasal drip, gustatory dysfunction, face painor heaviness, dyspnea, malaise, arthralgia, and nasal obstruction. Our large sample of 310,494 patients included from 114 studies showed that fever is the most common disease sign. Our findings were consistent with that of the others (20, 23) .

This study does not support the findings of recent studies that have highlighted the role of laboratory results in predicting mortality. For example, a recent study of 485 patients in Wuhan, China, noted the role of LDH and CRP in increasing mortality. The study, conducted on 4659 patients, also found CRP, LDH, Troponin, Creatinine, and Albumin as predictors of mortality (23) . Nevertheless, we only found the increased level of LDH as a relative predictor of mortality.

So it seems important to investigate more in this field.

Imaging plays a key role in the diagnosis of COVID-19. Among these, we can mention chest X-Ray and Chest CT Scan. Common findings of Chest X-Ray include multifocal peripheral consolidation and multifocal rounded opacities and nodules (23) , and of chest CT are groundglass opacities (86% frequency percentage), consolidation (29%), and Crazy-paving (19%) (24) .

In contrast, this study does not show any significant correlation between radiologic findings and mortality rate.

Endothelial dysfunction is one of the very first changes in diseases such as hypertension, diabetes, Coronary heart disease (CHD), and CKD. Numerous studies have also shown that SARS-COV2 tends to bind to the host Angiotensin-converting enzyme-2 (ACE2) receptor in vascular endothelial cells, which could well justify the role of underlying diseases in increasing mortality (23) (24) (25) . Infection-related demand Ischemia and direct viral infection of the myocardium have also been reported in studies among the etiologies of increased mortality in patients with COVID-19 with a prior history of cardiovascular disease. The role of underlying diseases alongside age in increasing mortality has been strongly suggested in related articles, which has also been shown in our study, especially in diabetes and hypertension (26) . The pathogenesis of increased mortality of COVID-19 in patients with diabetes is still unknown.

Immune dysregulation in diabetic patients such as phagocytic cell dysfunction, inhibition of neutrophil chemotaxis and impaired T-cell mediated immune response can be one of probable mechanism (27). In addition, type 2 diabetes mellitus and coronavirus infection have common pathogenic pathways. So that two receptors of coronavirus, ACE2 and Dipeptidy-l Peptidase-4 (DPP4), have also a role in regulating glucose homeostasis (28) . Diabetes was also a strong predictor of mortality among patients suffering Middle-Eastern Respiratory Syndrome (MERS) and SARS in previous studies (29) . Therefore, all diabetic patients with COVID-19, should be taken as high risk, even though he or she may present only mild or no symptoms. These patients will need extra monitoring, and their threshold for hospitalization and ICU admission also are lowered. This finding is unlike several previous studies that showed association between smoking and high mortality rate in patients with COVID-19 (31) .

Our findings support that older age, hypertension and diabetes mellitus might increase the risk of mortality among patients with COVID-19. In the multivariate analysis, only Diabetes Mellitus demonstrated an independent relationship with increased mortality. Further studies may be needed to ascertain the relationship of possible risk factors with COVID-19 mortality.

This is an extensive systematic review of COVID-19 mortality and its associated factors. We screened a large number of available articles in several databases, assessed their quality, and extracted relevant data to run regression analysis. However, we were not able to obtain adequate information to run weighted analysis and draw forest plots. As many of our studies comprised of those with small populations, it was not feasible to analyze according the population density.

Given that the studies are of case series, cross-sectional design, it was also not possible to pool the data together to estimate the heterogeneity between the studies. We attempted to categorize the countries based on income and present a map, demonstrating the differences between countries in presented cases in the literature. Owing to the circumstances of the pandemic, published data on some potentially suitable factors were limited. Therefore, these factors were not included in our study. Stability and availability of medical service and the role of lack of facility, treatment protocols, economical situations, and ethnicity of the patients are some of these missing variables.

Health-care providers should pay special attention to comorbidities such as diabetes and hypertension because these conditions, if not controlled, can increase mortality in patients with COVID-19.

By doing further and more complete studies on patients with COVID-19, a more appropriate and complete model can be found for factors related to patient mortality. Also, based on the sample size, different weights can be given to these studies to consider the effect of sample size.

(1) The conception and design of the study:Esmaeil Mehraeen, SeyedAhmad SeyedAlinaghi 

Rolling updates on coronavirus disease (COVID-19) 2020

A Systematic Review of Current Treatments and Therapeutic Options for COVID-19 Patients

COVID-19) Situation Report-141 2020

How Will COVID-19 Affect the Health Care Economy?

COVID-19 and economy

Transmission Modes of COVID-19: A Narrative Review of Current Evidence

Exit strategy to control covid-19 and relaunch the economy

Estimating excess 1-year mortality associated with the COVID-19 pandemic according to underlying conditions and age: a population-based cohort study

Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study

Epitope-based peptide vaccines predicted against novel coronavirus disease caused by SARS-CoV-2. Virus research

Characteristics of and Important Lessons From the Coronavirus Disease 2019 (COVID-19) Outbreak in China: Summary of a Report of 72314 Cases From the Chinese Center for Disease Control and Prevention

Zika virus: A review of literature

Wuhan novel coronavirus (COVID-19): why global control is challenging? Public Health

Isolation, quarantine, social distancing and community containment: pivotal role for old-style public health measures in the novel coronavirus (2019-nCoV) outbreak

A systematic review of quality assessment methods for smartphone health apps

Empirical studies on usability of mHealth apps: a systematic literature review

Evaluating mobile phone applications for health behaviour change: a systematic review

Coronaviruses: a paradigm of new emerging zoonotic diseases. Pathogens and disease

Epidemiology and Clinical Characteristics of COVID-19. Archives of Iranian medicine

SARS-CoV-2: Virology, epidemiology, immunology and vaccine development

The immunopathogenesis of sepsis in elderly patients

Gender differences in patients with COVID-19: Focus on severity and mortality

Predictors of mortality in hospitalized COVID-19 patients: A systematic review and meta-analysis

With Myocardial Injury and Mortality JAMA Cardiol

Cardiovascular implications of fatal outcomes of patients with coronavirus disease 2019 (COVID-19)

Commentary: COVID-19 and diabetes mellitus: What we know, how our patients should be treated now, and what should happen next

COVID-19 and diabetes: Can DPP4 inhibition play a role? Diabetes research and clinical practice

Is diabetes mellitus associated with mortality and severity of COVID-19? A meta-analysis. Diabetes & metabolic syndrome

Hypertension is associated with increased mortality and severity of disease in COVID-19 pneumonia: A systematic review, meta-analysis and metaregression

Severity and Mortality associated with COPD and Smoking in patients with COVID-19: A Rapid Systematic Review and Meta-Analysis

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

The authors declare that there is no conflict of interest regarding the publication of this manuscript.

The present study was conducted in collaboration with Khalkhal University of Medical Sciences, Iranian Institute for Reduction of High-Risk Behaviors, Tehran University of Medical Sciences, and Department of Global Health and Socioepidemiology, Kyoto University.

The authors state that all information provided in this article can be obtained from the author on request.

Authors We confirm that the manuscript has been read and approved by all named authors and that there are no other persons who satisfied the criteria for authorship but are not listed. We further confirm that the order of authors listed in the manuscript has been approved by all of us.

We understand that the Corresponding Author is the sole contact for the Editorial process. He/she is responsible for communicating with the other authors about progress, submissions of revisions and final approval of proofs.

We declare that this manuscript is original, has not been published before and is not currently being considered for publication elsewhere.

We also declare that the study was performed according to the international, national and instutional rules considering animal experiments, clinical studies and biodiversity rights.Financial Disclosure All affiliations with, or financial involvement in any entity with a financial interest in, or in competition with, the manuscript's subject matter are disclosed. This includes stock ownership, employment, consultancies, honoraria, grants, patents and royalties.