key: cord-0315702-64vd0ta0
authors: Mermiri, M.; Mavrovounis, G.; Laou, E.; Papagiannakis, N.; Pantazopoulos, I.; Chalkias, A.
title: The effect of vasopressors on mortality in critically ill patients with COVID-19: A systematic review and meta-analysis
date: 2022-05-28
journal: nan
DOI: 10.1101/2022.05.27.22275715
sha: 1324bf335888828ecf4df4d66c7e97ca9aa0813d
doc_id: 315702
cord_uid: 64vd0ta0

Purpose: The effect of vasopressors on mortality of critically ill patients with COVID-19 has not been studied extensively. Methods: A systematic search of PubMed, Scopus, and clinicaltrials.gov was conducted for relevant articles until January 2022. Eligibility criteria were randomized controlled and non-randomized trials. The primary outcome was all-cause mortality at 28 days or 30 days. The quality of studies was assessed using the MINORS tool. Paired meta-analysis was used to estimate the pooled risk ratios along with their 95% Confidence Interval. Results: In total, 33 studies were included. Twenty-one studies with a total population of 7900 individuals provided data on mortality. Patients who received vasopressors were statistically significantly more likely to die compared to those who did not receive vasopressor therapy [RR (95%CI): 4.26 (3.15, 5.76); p<0.001]. This result remained statistically significant regardless of the in-hospital setting. In-hospital and 30-day mortality were statistically significantly higher in patients who received vasopressors [RR (95%CI): 4.60 (2.47, 8.55); p<0.001 and RR (95%CI): 2.97 (1.72, 5.14); p<0.001, respectively]. Four studies provided data on specific vasopressors; the highest mortality rate was observed in patients treated with vasopressin or epinephrine, while patients receiving angiotensin-II as a sole or second-line vasopressor agent had the lowest mortality rate. Also, analysis of 10 studies with a total population of 3519 individuals revealed that patients who received vasopressors were statistically significantly more likely to experience acute kidney injury [RR (95%CI): 3.17 (2.21, 4.54); p<0.001]. Conclusion: Vasopressors have detrimental effect on survival of critically ill patients with COVID-19.

Mounting evidence suggest that COVID-19 should be perceived as a new entity with its own characteristics and distinct pathophysiology, including complex immuno-inflammatory, thrombotic, and parenchymal derangements [1] . The cytokine storm and the dysregulation of host response are more severe in COVID-19-related acute respiratory distress syndrome (ARDS) than in ARDS of other causes [2] [3] [4] .

SARS-CoV-2 not only infects the respiratory tract, but also injures the vascular endothelium and epithelium [5, 6] , and thus, COVID-19 can affect many systems of the human body.

Most critically ill patients with COVID-19 need hemodynamic support that is usually guided by the current, non-covid, surviving sepsis campaign guidelines recommending the use of vasopressors to optimize mean arterial pressure (MAP) and cardiac output and provide adequate organ perfusion [7, 8] . Most of these medications improve the hemodynamic function through enhancement of the adrenergic pathway; however, they may have important side-effects due to excessive adrenergic stimulation [9] [10] [11] . Of note, exogenous catecholamines can have a pronounced impact on inflammation and immunosuppression, metabolism, endothelial lesion, platelet activation, and coagulation [12] . As critically ill patients with COVID-19 are characterized by a similar pathophysiological substrate, exogenous vasopressors could further dysregulate their physiological cascades and aggravate outcome [13] . We therefore performed a systematic review and meta-analysis to investigate the effect of vasopressors on mortality of critically ill patients with COVID-19.

The protocol was registered in the PROSPERO international prospective register of systematic reviews on 13 December 2021 (CRD42021297595). This systematic review and meta-analysis were reported according to the preferred reporting items for systematic reviews and meta-analyses (PRISMA) checklist (Appendix A) [14] .

The inclusion criteria of the current systematic review and meta-analysis were:

(1) randomized controlled trials (RCTs) and observational studies; (2) critically ill patients admitted to the intensive care (ICU) or high dependency unit (HDU); (3) adults (≥ 18 years old) hospitalized primarily for COVID-19; (4) SARS-CoV-2 infection confirmed by reverse transcription polymerase chain reaction test of nasopharyngeal or oropharyngeal samples; and (5) vasopressor administration. We excluded animal studies, case reports, review papers, editorials, abstracts, white papers, and non-English literature. We also excluded studies pediatric patients, non-ICU/HDU patients, and patients who did not receive vasopressor therapy.

The primary outcome was all-cause mortality at 28 days or 30 days, or when lacking these data, all-cause mortality as reported by the authors. Secondary outcomes was to investigate (1) the hemodynamic profiles of patients at first measuring point and after six hours [heart rate, MAP, central venous pressure (CVP), urinary output, blood lactate levels, cardiac output or cardiac index, systemic vascular resistance index, central venous oxygen saturation, oxygen delivery index, and oxygen . CC-BY-NC-ND 4.0 International license It is made available under a 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 May 28, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022 consumption index]; (2) the number of participants who achieved the target MAP (≥65 mmHg); (3) time to achieve the target MAP; (4) adverse events including arrhythmia, acute myocardial infarction, cardiac arrest, acute mesenteric ischemia, digital ischemia, acute kidney injury (AKI); (5) vasopressor-free days; (6) ICU length of stay; (7) duration of mechanical ventilation; (8) ventilator free days; (9) hospital length of stay; and (10) all-cause mortality at 90-days.

The search strategy was intended to explore all available published and unpublished studies until January 20 th , 2022. A comprehensive initial search was employed in PubMed (MEDLINE), Scopus, and clinicaltrials.gov databases by two independent investigators (MM, GM) followed by an analysis of the textwords contained in Title/Abstract and indexed terms. A second search was conducted by combining free text words (vasopressor, epinephrine, norepinephrine, phenylephrine, vasopressin, dopamine, angiotensin-II, covid-19, critically ill, intensive care) and indexed terms with Boolean operators. Finally, a third search was conducted with the reference lists of all identified reports and articles for additional studies. Appendix B presents the exact search algorithm used for all databases.

The data from each study were extracted by two independent authors (MM, GM) with a customized format. Any disagreements between the two independent authors were resolved by four other authors (EL, IP, NP, AC). Publication details (authors, year), study information (design, population, department of admission, follow-up, inclusion-exclusion criteria, number of cases/cohort-size, and subgroups), . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 28, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022 hemodynamic profile (heart rate, MAP, CVP, urinary output, blood lactate levels, cardiac output or cardiac index, systemic vascular resistance index, central venous oxygen saturation, oxygen delivery index, oxygen consumption index) at first measuring point and six hours after vasopressor use, the number of participants who achieved the target MAP and time to achieve the target MAP, adverse events, vasopressor-free days, ICU length of stay, hospital length of stay, duration of mechanical ventilation, ventilator-free days, all-cause mortality in all groups at 28 or 30 days, and all-cause mortality at 90 days were extracted in a pre-designed excel spreadsheet. The definition used for AKI and the mortality follow-up timepoints for each study are presented in Appendix C1. Authors of studies with missing data were contacted in an attempt to obtain relevant data.

Articles identified for retrieval were assessed by two independent authors (MM, GM) for methodological quality before inclusion in the review using standardized critical appraisal tools. The quality of the included observational studies was assessed using the MINORS tool [15] , while the Risk of Bias 2.0 (RoB 2.0) tool was used for RCTs [16] . Any disagreements between the authors appraising the articles were resolved through discussion with the other authors.

A paired meta-analysis was used to estimate the pooled risk ratios (RR) along with their 95% Confidence Interval (95% CI). Based on the presence of statistical heterogeneity, the meta-analysis was conducted according to fixed-or random effect models. The statistical heterogeneity was estimated by the use of the Cochran's Q and . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 28, 2022. ; https://doi.org/10.1101 /2022 doi: medRxiv preprint I 2 indices. When I 2 >50% and/or PQ<0.10, the random effects model was used, otherwise the fixed effects model was implemented [17] . Funnel plots as well as the Begg's test were used to determine the existence of publication bias [18, 19] 

Altogether, 809 relevant citations were identified and screened, while 87 studies were included in our final assessment for possible data extraction ( Fig. 1 ). In total, data extraction was possible in 33 studies .

All the 33 included studies were observational in their design . The studies originated from multiple countries, with 13 of coming from institutions located in the USA [20, 21, 23, [32] [33] [34] [35] 38, 42, 43, 45, 48, 51] . Nineteen studies included only patients admitted in the ICU [20, 22, 24, 25, 27, [30] [31] [32] [35] [36] [37] [38] [39] [43] [44] [45] [46] [47] 52] , five studies included patients admitted in a COVID-19-dedicated HDU [28, 29, 33, 34, 42] , eight studies included patients who were admitted in both HDU and ICU [21, 23, 26, 40, 41, 48, 49, 51] , and one study included Emergency Department patients who were later admitted either in the HDU of ICU [50] . Thirty studies included data about patients who received vs. patients who did not receive vasopressors [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] 50, 52] . Moreover, three studies included patients who received angiotensin-II [38, 49, 51] and, out of those, two compared the use of angiotensin-II with other . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) 

Twenty-one out of the 33 included studies provided data on hospital mortality in patients who received vs. patients who did not receive vasopressors, resulting in a total population of 7900 individuals [25, 27, [31] [32] [33] [34] [35] [36] [37] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] 50, 52] . Due to high heterogeneity (I 2 : 94%, PQ<0.001), the random-effects model was implemented.

Patients who received vasopressors were statistically significantly more likely to die (2.51, 12.15 ); p < 0.001], remained statistically significant for higher mortality rates in patients who received vasopressors.

Subgroup analyses were also performed based on the mortality follow-up timepoints. Only the in-hospital and 30-day mortality subgroups had three or more studies that allowed data extraction and analysis. The in-hospital and 30-day mortality were statistically significantly higher in patients who received vasopressors [RR (95%CI): 4.60 (2.47, 8.55 ); p<0.001 and RR (95%CI): 2.97 (1.72, 5 .14); p<0.001, respectively].

. CC-BY-NC-ND 4.0 International license It is made available under a 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 May 28, 2022. ; https://doi.org/10.1101/2022.05.27.22275715 doi: medRxiv preprint

Four studies provided data on mortality based on the specific vasopressor(s) administered [38, 43, 49, 51] . The highest mortality rate was observed in patients treated with vasopressin or epinephrine (78% and 76%, respectively) [43] . Three of those studies investigated the role of angiotensin-II as a sole or second-line vasopressor agent [38, 49, 51] . These studies showed the lowest mortality rate. The relevant data are depicted in Table 2 .

Ten studies provided data on AKI in patients who received vs. patients who did not receive vasopressors, resulting in a total population of 3519 individuals [21] [22] [23] [25] [26] [27] [28] [29] [30] 44 ]. Due to high heterogeneity (I 2 : 92%, PQ<0.001), the random-effects model was implemented. Patients who received vasopressors were statistically significantly more likely to experience AKI compared to those who did not receive Subgroup analyses were performed based on the definition of AKI that was used in the included studies. Only the subgroup with patients at all KDINGO stages included more than three studies, allowing for meta-analysis to be performed.

Specifically, patients who received vasopressors were more likely to experience AKI when compared to those who did not receive vasopressor therapy [RR (95%CI): 2.29 (1.67-3.14); p < 0.001] (Figure 4 ).

No data were identified for the remaining secondary outcomes.

. CC-BY-NC-ND 4.0 International license It is made available under a 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 May 28, 2022. ; https://doi.org/10.1101/2022.05.27.22275715 doi: medRxiv preprint

The overall quality of the studies, as assessed by the MINORS tool, ranged between moderate and high. The exact score for each study is available in Appendix C3. In addition, visual inspection of the funnel plot (Appendix D2) and the Begg's test (p=0.18) did not reveal significant publication bias for the studies included in the AKI analysis. All included studies were observational non-randomized studies, and as a result according to GRADE criteria they have a low grade in quality of evidence.

Many high-quality RCTs have addressed the effects of vasopressors on the outcomes of non-covid patients, yet their impact on mortality in patients with COVID-19 had not been studied so far. The most important finding of this systematic review and meta-analysis is that vasopressors have detrimental effect on survival of critically ill patients with COVID-19. Although these results are based on published non-randomized evidence, they raise significant concerns for the routine management of these individuals.

Universally, the mortality of critically ill patients with COVID-19 is higher than that of other critically ill patients [53, 54] . A main cause for this difference is the characteristics of the SARS-CoV-2 infection, which can rapidly affect other organs including the cardiovascular system [55] . However, there are no specific guidelines on the hemodynamic support of COVID-19 patients. Although administration of vasopressors is a fundamental treatment of hypotension, the traditional (non-covid) hemodynamic management of shock and the adverse effects of vasoactive agents may be associated with complications and poor outcome in patients with COVID-19.

Indeed, the present analysis revealed that critically ill patients with COVID-19 who . CC-BY-NC-ND 4.0 International license It is made available under a 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 May 28, 2022. ; https://doi.org/10.1101/2022.05.27.22275715 doi: medRxiv preprint receive vasopressors are more likely to die in the hospital or at 30 days compared to those who do not receive vasopressor therapy regardless of the in-hospital setting.

Although the association of vasopressors with increased mortality could be coincidental due to the severity of critical illness, catecholamines affect many systems including the immune and hematological systems, the renin-angiotensin-aldosterone system, the cardiovascular and respiratory systems, and others [13], suggesting a complex interplay that can have a detrimental effect on survival.

Of the 33 included studies in this systematic review and meta-analysis, only four studies included data on specific vasopressors. The highest mortality rate was observed in patients receiving vasopressin or epinephrine. Although one can appraise that these patients had severe shock necessitating second-and third-line vasopressors [7] , these observations merit further discussion. Epinephrine is well-known for its adverse effects in non-covid and COVID-19 patients [56], but our observations regarding vasopressin are quite interesting. Although vasopressin infusion reduces total norepinephrine-equivalent dose requirements and may be renal and pulmonary vasculature sparing [57], there is evidence showing a pronounced activation of the vasopressin system in COVID-19 patients and that molecular complexes form between the SARS-CoV-2 spike protein, soluble angiotensin-converting enzyme-2 (ACE2), and vasopressin, facilitating cellular infection and aggravating outcome [58, 59] . However, data from a small clinical cohort did not show a clinically relevant effect of vasopressin infusion on viral mRNA level in critically ill patients with COVID-19 (but who were not treated with corticosteroids or interleukin-6 antagonists) [57] . The findings of the present analysis suggest that arginine vasopressin agonists might not be a good choice for these individuals. Considering that vasopressin is suggested as a second-line vasopressor in the latest international . CC-BY-NC-ND 4.0 International license It is made available under a 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 May 28, 2022. [38, 51] . The rationale for angiotensin-II therapy is based on decreasing the expression of the ACE2 receptors, which can reduce the entry of the COVID-19 virus into cells [60, 61] . However, the progressive loss of ACE2 shifts the system to an overall higher angiotensin level due to the impaired ability of ACE2 to degrade it, which may explain the initial hemodynamic stability of patients with COVID-19 [62] . Therefore, exogenous use of angiotensin-II may be harmful in patients with increased endogenous levels. Taking into consideration its potential favorable effects in critically ill patients with COVID-19, randomized controlled trials are needed to further evaluate angiotensin-II for the treatment of COVID-19-related shock.

In non-covid patients, administration of vasopressors is not associated with increased mortality, while prophylactic administration in patients with vasodilatory shock may improve survival [63] . In addition, a Cochrane systematic review found no evidence of substantial differences in total mortality between several vasopressors [64] . Nevertheless, vasopressors are a heterogeneous class of drugs with powerful and immediate haemodynamic effects, and each drug has advantages and disadvantages.

. CC-BY-NC-ND 4.0 International license It is made available under a 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 May 28, 2022. ; https://doi.org/10. 1101 /2022 These characteristics are particularly important in patients with COVID-19 who are characterized by unique pathophysiological disturbances and different hemodynamic phenotypes that necessitate a thorough understanding of the underlying complex pathophysiology and careful selection and administration of vasoactive agents.

In COVID-19, the progressive hypoxemia initially increases cardiac output and capillary recruitment, which maintain microcirculatory oxygen-extraction capacity by increasing red blood cell availability (silent hypoxia) [65] [66] [67] . However, microcirculatory flow decreases proportionally to the increasing inflammation, hypercoagulation, and thrombosis, resulting in multi-organ failure at later stages [65, [68] [69] [70] . In the study by Mesquida et al., patients showed alterations in systemic microcirculatory status, and the degree of these alterations correlated with the severity of the respiratory disease [24] . The relationship between MAP and organ blood flow may be different in critically ill patients with COVID-19 and improving only macrocirculation might be inadequate to maintain tissue perfusion. In these patients, vasopressor use can overwhelm endogenous receptor-mediated vessel regulation, further contributing to the loss of hemodynamic coherence [66, 71] , and therefore, hemodynamic management should focus on optimizing microcirculatory perfusion instead of attaining a predefined MAP target.

This analysis included patients from various settings, i.e., HDU, ICU, and Emergency Department. Consequently, it may have included heterogeneous groups of patients with COVID-19. Due to the lack of randomized controlled trials, the synthesis of all the available knowledge on the specific outcomes was difficult. The . CC-BY-NC-ND 4.0 International license It is made available under a 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 May 28, 2022. ; https://doi.org/10.1101/2022.05.27.22275715 doi: medRxiv preprint level of heterogeneity was high and the conclusions drawn from this review must be cautious and reserved.

Additionally, no data from the different studies was available to adjust the resulting odds ratios according to age, comorbidities, the presence of septic shock or other known factors that affect ICU mortality. In addition, most of the secondary outcomes could not be assessed. Another limitation is the heterogeneity of definitions of AKI that were used across different studies. Also, many of the included studies were conducted in retrospective fashion. Finally, we did not include non-English publications.

The present systematic review and meta-analysis showed that vasopressor use was associated with an increase in mortality and AKI in critically ill patients with COVID-19, according to published non-randomized evidence. It is worth noting that we found a lower mortality rate in patients receiving angiotensin-II as a sole or second-line vasopressor agent, while the highest mortality rate was observed in patients receiving vasopressin and epinephrine. Further research is required to estimate the correlation of specific vasopressors with adverse effects and mortality in this population.

The results of the present systematic review and meta-analysis suggest for early administration of low-dose vasopressors, with or without inodilator agents, in an effort to avoid excessive doses that could have detrimental effect on survival, especially at later disease stages. An alternative second-line vasopressor may be . CC-BY-NC-ND 4.0 International license It is made available under a 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 May 28, 2022. ; https://doi.org/10.1101/2022.05.27.22275715 doi: medRxiv preprint angiotensin-II. However, further immediate research is recommended to elucidate the effects of angiotensin-II and other vasopressors acting through pathways other than the adrenergic pathway as sole or second-line vasopressor agents. These agents may be associated with a significant increase in survival.

A possible explanation for the association of vasopressors with mortality may lie in the microcirculation [65] [66] [67] [68] [69] [70] [71] . The physiological pulsatile shear stress from normal laminar flow maintains the normal endothelial cell functions and the expression of ACE2s and other anticoagulant/antithrombotic and antioxidant substances [72] . However, dysfunctional endothelium resulting from turbulent flow displays a hypercoagulant/prothrombotic and pro-oxidant state and impairs microcirculatory reactivity and flow [73] . Therapeutic approaches should consider the systemic vascular involvement and include the assessment of microcirculation, allowing individualized, physiology-guided management. Of note, an increased CVP in critically ill patients with COVID-19 may impair venous return and retrogradely increase post-capillary venular pressure which, together with excessive vasopressor doses, impair capillary perfusion and increase the diffusion distance of oxygen [74, 75] . Thus, minimizing fluid administration is also crucial for improving tissue perfusion in this population.

Further research and well-designed trials are necessary to investigate the effect of the type (catecholamines vs. non-catecholamines), time of administration, and infusion rates of vasopressors in order to develop specific treatment strategies and integrate a more individualized approach in patients with COVID-19.

. CC-BY-NC-ND 4.0 International license It is made available under a 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 May 28, 2022. CC-BY-NC-ND 4.0 International license It is made available under a 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 May 28, 2022. . CC-BY-NC-ND 4.0 International license It is made available under a 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 May 28, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022 13. Gubbi S, Nazari MA, Taieb 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 May 28, 2022. . CC-BY-NC-ND 4.0 International license It is made available under a 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 May 28, 2022. . CC-BY-NC-ND 4.0 International license It is made available under a 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 May 28, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022 Mexico COVID-19 Critical Care Collaborative Group (2021) . CC-BY-NC-ND 4.0 International license It is made available under a 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 May 28, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 28, 2022. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 28, 2022. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 28, 2022. ; https://doi.org/10.1101/2022.05.27.22275715 doi: medRxiv preprint Appendix D1: Funnel plot for mortality meta-analysis.

Appendix D2: Funnel plot for AKI meta-analysis.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 28, 2022. ; https://doi.org/10.1101/2022.05.27.22275715 doi: medRxiv preprint 

Identification Screening Included . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 28, 2022. ; https://doi.org/10.1101/2022.05.27.22275715 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 28, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 28, 2022. ;

The COVID-19 puzzle: deciphering pathophysiology and phenotypes of a new disease entity

Factors associated with mortality among moderate and severe patients with COVID-19

Risk factors for mortality in COVID-19 patients in a community teaching hospital

Retrospective Matched Cohort Study

Outcomes for patients with COVID-19 admitted to Australian intensive care units during the first four months of the pandemic

Mortality and Clinical Interventions in Critically ill Patient With Coronavirus Disease 2019: A Systematic Review and Meta-Analysis

Acute complications and mortality in with impaired viral clearance: a pilot study

Soluble ACE2-mediated cell entry of SARS-CoV-2 via interaction with proteins related to the renin-angiotensin system

Activation of Vasopressin System During COVID-19 is Associated With Adverse Clinical Outcomes: An Observational Study

Angiotensin II mediates angiotensin converting enzyme type 2 internalization and degradation through an angiotensin II type I receptor-dependent mechanism

Should we use angiotensin II infusion in COVID-19-associated vasoplegic shock

The Effect of inotropes and vasopressors on mortality: a meta-analysis of randomized clinical trials

Vasopressors for hypotensive shock

Capillary Leukocytes, Microaggregates, and the Response to Hypoxemia in the Microcirculation of Coronavirus Disease

Microcirculatory Alterations in Critically Ill Patients with COVID-19-common in patients hospitalized with COVID-19 infection

In vivo demonstration of microvascular thrombosis in severe COVID-19

Increasing arterial blood pressure with norepinephrine does not improve microcirculatory blood flow: a prospective study

Endothelial KLF2 links local arterial shear stress levels to the expression of vascular tone-regulating genes

The role of endothelial shear stress on haemodynamics, inflammation, coagulation and glycocalyx during sepsis

Prospective ICU + HDU 931 / 3086 (30.2 %) 56 ±

Multicentric study

Year of Publication; SD: Standard Deviation; IQR: Interquartile Range; USA: United States of America; ICU: Intensive Care Unit; HDU: High Dependency Unit; UK: United Kingdom; IRCU: Intensive Respiratory Care Unit; UAE: United Arab Emirates; ED: Emergency Department. * These studies only

**10 African Countries. Table 2. Data on mortality based on major vasopressors Authors Country, YOP Study Design Intervention Group: Deaths / All (%) Comparator Group: Deaths / All (%)

Retrospective Any vasopressor support: 141 / 233 (61%)

Epinephrine  other vasopressors: 19 / 25 (76%) No vasopressor support

Retrospective Angiotensin-II  other vasopressors: 4 / 10 (40%) Other vasopressors: 10 / 19 (53%)

Multicentric study

We would like to thank Dr. George L. Anesi, University of