key: cord-0885187-4pknh85w
authors: Karakike, E.; Giamarellos-Bourboulis, E.; Kyprianou, M.; Fleischmann-Struzek, C.; Pletz, M. W.; Netea, M. G.; Reinhart, K.; Kyriazopoulou, E.
title: COVID-19 as cause of viral sepsis: A Systematic Review and Meta-Analysis
date: 2020-12-04
journal: nan
DOI: 10.1101/2020.12.02.20242354
sha: b92c0be4c0547822694df76ae719bb22d6024698
doc_id: 885187
cord_uid: 4pknh85w

Importance COVID-19 is a heterogenous disease most frequently causing respiratory tract infection but in its severe forms, respiratory failure and multiple organ dysfunction syndrome may occur, resembling sepsis. The prevalence of viral sepsis among COVID-19 patients is still unclear. Objective We aimed to describe this in a systematic review. Data sources MEDLINE(PubMed), Cochrane and Google Scholar databases were searched for studies reporting on patients hospitalized with confirmed COVID-19, diagnosed with sepsis or infection-related organ dysfunctions or receiving organ replacement therapy. Study selection Eligible were full-text English articles of randomized and non-randomized clinical trials and observational studies reporting on patients with confirmed COVID-19, who are diagnosed with sepsis or have infection-related organ dysfunctions. Systematic reviews, editorials, conference abstracts, animal studies, case reports, articles neither in English nor full-text, and studies with fewer than 30 participants were excluded. Data extraction and synthesis All eligible studies were included in a narrative synthesis of results and after reviewing all included studies a meta-analysis was conducted. Separate sensitivity analyses were conducted per adult vs pediatric populations and per Intensive Care Unit (ICU) vs non-ICU populations. Main outcomes and measures Primary endpoint was the prevalence of sepsis using Sepsis-3 criteria among patients with COVID-19 and among secondary, new onset of infection-related organ dysfunction. Outcomes were expressed as proportions with respective 95% confidence interval (CI). Results Of 1,903 articles, 104 were analyzed. The prevalence of sepsis in COVID-19 was 39.9% (95% CI, 35.9-44.1; I2, 99%). In sensitivity analysis, sepsis was present in 25.1% (95% CI, 21.8-28.9; I2 99%) of adult patients hospitalized in non-Intensive-Care-Unit (ICU) wards (40 studies) and in 83.8 (95% CI, 78.1-88.2; I2,91%) of adult patients hospitalized in the ICU (31 studies). Sepsis in children hospitalized with COVID-19 was as high as 7.8% (95% CI, 0.4-64.9; I2, 97%). Acute Respiratory Distress Syndrome was the most common organ dysfunction in adult patients in non-ICU (27.6; 95% CI, 21.6-34.5; I2, 99%) and ICU (88.3%; 95% CI, 79.7-93.5; I2, 97%) Conclusions and relevance Despite the high heterogeneity in reported results, sepsis frequently complicates COVID-19 among hospitalized patients and is significantly higher among those in the ICU. PROSPERO registration number: CRD42020202018. No funding.

COVID-19, caused by the severe acute respiratory syndrome coronavirus (SARS-CoV)-2, is recognized as a pandemic, affecting 230 countries with 54,771,888 confirmed cases and 1,324,249 deaths worldwide, as of 17 November 2020 (1). Several patients are complicated by respiratory insufficiency and they are in need for mechanical ventilation (2). First reports of patients in Wuhan described in addition of lung disease the presence of other organ failures including acute kidney injury, liver dysfunction, coagulopathy, confusion and shock; such an involvement resembles the systemic counterparts of bacterial and viral sepsis (3-5). The current Sepsis-3 definitions define sepsis as a life-threatening organ dysfunction due to the dysregulated host response to an infection. The same definitions introduce the Sequential Organ Failure Assessment (SOFA) score as a measure of organ dysfunction (6). According to World Health Organization (WHO), manifestations of sepsis and septic shock can be also the fatal frequent pathway of infections with highly transmissible pathogens of public health concern such as avian and swine influenza viruses, as well as corona viruses (7). Severe respiratory failure of COVID-19 is accompanied by complex immune dysregulation of the host (8). As a consequence, all elements of the new Sepsis-3 definition may apply for COVID-19

i.e. life-threatening organ dysfunction, dysregulated host response, and viral infection (9).

With this in mind, we systematically reviewed published evidence for COVID-19 as a cause of viral sepsis using the SOFA score as a classification tool. In parallel the prevalence of different organ dysfunctions, the need for ICU admission, the association of sepsis presence with mortality and the presence of alterations of inflammation and coagulation markers were analyzed. 1 0 data on hospitalized patients due to COVID-19 and 35 of them included data for ICU   (20, 25, 33, 36, 37, 41, 44, 45, 48, 52, 53, 56, 58, 66, 72, 73, 81-84, 86 No studies were found that reported sepsis based on Sepsis-3 or Sepsis-1/2 criteria per se in the original publication; zero uncertainty was only achieved when corresponding authors contacted, provided the exact number of patients presenting sepsis based on sepsis-3. Characteristics of the included studies are presented in eTable 1 and quality assessment with MINORS of each in eTable 2.

The pooled estimate for the prevalence of sepsis among COVID-19 patients was 39.9% (95% CI, 35.9-44.1; I 2 , 99%) as assessed from 92 studies reporting data on the primary endpoint (eFigure 1). Due to the high heterogeneity observed, the results of sensitivity analyses are presented in Figure 2 . In sensitivity analysis focusing on ICU versus non-ICU cohorts, among forty studies of a total of 112,796 patients, the pooled estimate for sepsis prevalence was 25.1% (95% CI, 21.8-28.9; I 2 99%) in the general ward ( Figure 2A ). The respective estimate for adult patients in the ICU was 83.8% (95% CI, 78.1-88.2; I 2 91%) among 31 articles assessed including 3,528 patients (p<0.0001) ( Figure 2B ). In sensitivity analysis among pediatric cohorts, sepsis in children hospitalized outside the ICU was as high as 7.8% (95% CI, 0.4-64.9; I 2 , 97%) among 3 studies assessed ( Figure 2C ), whereas only one study provided data for ICU (all 78 children septic) (84). Respective funnel plots of 1 1 the above analyses are provided in eFigure 2. Sensitivity analyses including only articles with zero uncertainty (4 studies outside the ICU and 3 in the ICU, eFigure 3) and high uncertainty (13 studies outside the ICU and 3 in the ICU, eFigure 4) provided similar pooled estimates of sepsis prevalence outside of the ICU, which were 17.1% (95%CI, 6.4-38.4; I 2 , 97%) in zero uncertainty analysis and 21.9% (95%CI, 16.9-27.9; I 2 ,93%) in high uncertainty analysis. However, the pooled estimate in the ICU was 95.7% (95%CI, 75.5-99.4; I 2 , 65%) and 34.7% (95%CI, 9.9-72.0, I 2 , 92%) in zero and high uncertainty analysis respectively (p<0.0001), reducing the heterogeneity and yielding a significant difference between the two groups.

We described the prevalence of different organ dysfunctions, of admission in the ICU, and of organ support and/or replacement among the 104 studies included. A synthesis of the estimates of prevalence of organ dysfunctions in adults is presented in eTables 3 and 4; similar pediatric data are presented in eTable 5. Acute Respiratory Distress Syndrome (ARDS) was the most common organ dysfunction both in non-ICU and ICU patients. ARDS prevalence in non-ICU wards was 27.6% (95% CI, 21.6-34.5; I 2 , 99%) whereas in ICU reached 88.3% (95% CI, 79.7-93.5; I 2 , 97%; p<0.0001). Shock was the second most common dysfunction in ICU patients, followed by renal and liver dysfunction (Table 1, eFigure 5). Coagulopathy was the second most common organ dysfunction among non-ICU patients. Prevalence of all organ dysfunctions was much higher in ICU than non-ICU patients except for coagulopathy and central nervous system dysfunction; the prevalence of these dysfunctions interestingly remained similar. All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted December 4, 2020. ; https://doi.org/10.1101/2020.12.02.20242354 doi: medRxiv preprint Among adult patients hospitalized in the general ward, 16.9% (95% CI, 14.0-20.1; I 2 , 96%) required ICU admission ( Figure 3 ). As expected, need for organ replacement was more likely among ICU than non-ICU patients (Table 2) 

All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted December 4, 2020. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted December 4, 2020. ; https://doi.org/10.1101/2020.12.02.20242354 doi: medRxiv preprint Only studies with zero and low uncertainty have been included. All included studies have been published in 2020.

Abbreviation: CI, confidence interval. Only studies with zero and low uncertainty have been included. All included studies have been published in 2020.

Abbreviation: CI, confidence interval. All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted December 4, 2020. ; https://doi.org/10.1101/2020.12.02.20242354 doi: medRxiv preprint 

Baseline characteristics and outcomes of patients with COVID-19 admitted to intensive care units in Vancouver, Canada: a case series

Clinical characteristics, outcomes, and risk factors for mortality in patients with cancer and COVID-19 in Hubei, China: a multicentre, retrospective, cohort study

Prognostic Accuracy of the SIRS, qSOFA, and NEWS for Early Detection of

Infected Patients

Prognostic Factors for Severe Coronavirus Disease 2019 in Daegu

The Clinical Characteristics and Outcomes of Patients with Moderate-to-Severe Coronavirus Disease 2019 Infection and Diabetes in Daegu, South Korea

A Randomized Clinical Trial of the Efficacy and Safety of Interferon β -1a in Treatment of Severe COVID-19

ICU outcomes of COVID-19 critically ill patients: An international comparative study

The Epidemiological and Clinical Characteristics of 81 Children with COVID-19 in a Pandemic Hospital in Turkey: an Observational Cohort Study

No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity

COVID-19 in patients with thoracic malignancies (TERAVOLT): first results of an international, registrybased, cohort study

Outcomes of COVID-19 in patients with CLL: a multicenter international experience

COVID-19 severity and mortality in patients with chronic lymphocytic leukemia: a joint study by ERIC, the

European Research Initiative on CLL, and CLL Campus

Admissions for Sepsis or Pneumonia in Australia and New Zealand in 2017

SARS-CoV-2 Infection in Children

Extrapulmonary manifestations of COVID-19

Clinical Characteristics of Coronavirus Disease 2019 in China

Multiorgan and Renal Tropism of SARS-CoV-2

No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity