The ongoing outbreak of severe acute respiratory syndrome coronavirus-2 [SARS-CoV-2, or coronavirus disease 2019 (COVID-19)] was declared a pandemic by the World Health Organization on March 11, 2020. Worldwide, more than 65 million people have been infected with this SARS-CoV-2 virus, and over 1.5 million people have died due to the viral illness. Although a tremendous amount of medical progress has been made since its inception, there continues to be ongoing research regarding the pathophysiology, treatments, and vaccines. While a vast majority of those infected develop only mild to moderate symptoms, about 5% of people have severe forms of infection resulting in respiratory failure, myocarditis, septic shock, or multi-organ failure. Despite maximal cardiopulmonary support and invasive mechanical ventilation, mortality remains high. Extracorporeal membrane oxygenation (ECMO) remains a valid treatment option when maximal conventional strategies fail. Utilization of ECMO in the pandemic is challenging from both resource allocation and ethical standpoints. This article reviews the rationale behind its use, current status of utilization, and future considerations for ECMO in critically ill COVID-19 patients.

On December 31, 2019, the World Health Organization (WHO) was alerted of cases of pneumonia with an unknown etiology detected in Wuhan City, Hubei Provence, China. With rising fear of a potential endemic in the overpopulated city of Wuhan, Chinese national authorities along with the Wuhan Municipal Health Commission began a quest to identify all cases, amongst the 19 million occupants, as early as possible, as well as to trace potential sources through retrospective investigation. Initial investigations revealed the source of the first 27 confirmed cases of the novel coronavirus, severe acute respiratory syndrome coronavirus-2 [SARS-CoV-2, coronavirus disease 2019 (COVID-19)], was the Huanan seafood market[1]. The market was immediately shut down, but the virus had already spread beyond what was anticipated. Not long after, reports of human-to-human transmission were documented and surrounding areas including Hong Kong, Taiwan and Macau took the drastic step of shutting down borders with their long-time allies.

Chinese scientists continued to study this unidentified pathogen until, finally, on the 7^th of January 2020, the novel coronavirus was isolated from a single patient and gene sequencing was successfully performed and made available to the WHO five days later. This facilitated the ability for laboratories worldwide to produce diagnostic PCR tests to detect this new virus.

The novel coronavirus continued to spread to neighboring countries despite valiant efforts to subdue the spread. Today, COVID-19 has spread to over 200 countries, spread over six continents, infected over 65.8 million, and taken the lives of 1.5 million people worldwide to date[2]. On the 11^th of March 2020, the WHO officially declared the COVID-19 outbreak a global pandemic, as what began as a simple case of viral pneumonia subsequently became one of the most devastating pandemics of the twenty-first century.

The main method of person-person transmission of SARS-CoV-2 is by respiratory droplets, which is similar to the spread of influenza[3]. With droplet transmission, the virus can be spread by coughing, sneezing, or conversing up to six feet away[3]. The virus has also been shown to be able to linger on surfaces for hours and in the air under experimental conditions[3]. Upon exposure to the virus, the incubation period has been shown to be within 14 d, with most cases presenting 4-5 d after exposure[3].

In order to fully comprehend the pathophysiology of SARS-CoV-2, the basic viral structure must first be understood. Coronaviruses, including SARS-CoV-2, are positive single stranded RNA viruses of approximately 30 kb in length[4]. They are composed of four main structural proteins: Membrane (M), spike (S), envelope (E) and nucleocapsid (N)[5]. The spike protein dictates host tropism and has been found to have an affinity to angiotensin converting enzyme-2 (ACE-2) receptors[5]. Epithelial cells in the lungs have a high concentration of ACE-2 receptors likely explaining the high incidence of respiratory symptoms associated with SARS-CoV-2[5]. Other organs with increased ACE-2 expression include the heart, ileum, kidneys and urinary bladder. Following virus binding to host cells via previously mentioned receptors, spike protein is cleaved and subsequently activated leading to irreversible membrane fusion[5]. After cell invasion, the positive stranded RNA released by the virus leads to the production of peptides and proteins by translation in the host, and RNA-dependent RNA polymerase which can further replicate viral RNA[5].

Based on degree of invasion and inflammatory response, symptoms can range from mild to severe respiratory distress and multiorgan failure[4]. Typically, initial symptoms are respiratory in nature given that the inhaled droplets easily invade lung epithelial cells expressing ACE-2 receptors. Surrounding the epithelial cells of the lungs are dendritic cells and macrophages, also known as antigen presenting cells, which present viral antigen to neighboring T-cells initiating a T-cell mediated response[4]. Cytotoxic T-cells (CD8+) play a role in killing the viral antigen while helper T-cells (CD4+) activate B-lymphocytes promoting antibody formation. However, with time, T-cell exhaustion is being observed in patients infected with SARS-CoV-2 which can partially explain clinical deterioration over time[4].

Activation of lymphocytes along with destruction of infected cells leads to the release of cytokines and inflammatory mediators, leading to what is commonly known as the cytokine storm[4]. The most important cytokines released include interleukins 6 and 8 (IL-6, IL-8). IL-6 interacts with the hypothalamus leading to high grade fever; IL-8 is a well-known chemoattractant for T-cells and neutrophils leading to an influx of inflammatory cells into the lungs, or other infected areas, and subsequent diffuse alveolar damage and pulmonary infiltration[4]. These cytokines can also cause vasodilation and increased vessel permeability that leads to hypoxemia, increased work of breathing, and acute respiratory distress syndrome (ARDS)[4]. This increased inflammatory response has also been found to cause significant endothelial damage, creating a hypercoagulable status with the end result ranging from capillary microthrombi to diffuse pulmonary emboli[4].

Unfortunately, the destructive effects of SARS-CoV-2 are not limited only to the pulmonary system. The inflammatory cascade it triggers can involve other organ systems as well, most prominently the cardiovascular system, as evidenced by increasing reports of myocarditis in the younger population[4]. This overwhelming inflammatory state can progress into critical disease, multiorgan failure, and eventually death. Most often, cytokine storm causes systemic vasodilation, which in turn leads to hemodynamic instability and suboptimal peripheral perfusion[4]. Compromised perfusion can cause renal failure, seen by elevation in blood urea nitrogen and creatinine, liver damage, noted by increased liver enzymes, myocardial infarction, and other organ dysfunction. Endothelial injury can also lead to prothrombotic states[4]. It is any combination of ARDS, multi-organ dysfunction from poor perfusion, hemodynamic cardiovascular collapse, and hypercoagulability that puts a patient at risk of shock and ultimately death[4] (Figure 1).

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Pathophysiology of coronavirus disease 2019 infection. Viral binding and invasion of angiotensin converting enzyme 2 receptor-rich cells triggers destruction of infected cells with release of cytokines (mainly interleukin-6, interleukin-8 and tumor necrosis factor) and chemo-attractants, as well as activation of neighboring antigen presenting cells (APCs). Cytokine surge and APC activation triggers a T-cell mediated response and further release of cytokines. Activation of T-cells along with ongoing destruction of infected cells leads to cytokine storm. Symptoms developed range from mild respiratory symptoms to multiorgan failure and death based upon host response. ACE-2: Angiotensin converting enzyme 2; APCs: Antigen presenting cells; IL: Interleukin; TNF: Tumor necrosis factor; ARDS: Acute respiratory distress syndrome.

Extracorporeal membrane oxygenation (ECMO) is often used as a last resort in patients with critical pulmonary or cardiovascular compromise, requiring mechanical support (Figure 2). It has various configurations based on the patient’s initial requirement (pulmonary support, cardiovascular support, or both) and can be adjusted according to complication. Cardiac indications for ECMO include cardiogenic shock from a myocardial infarction, arrhythmia, pulmonary embolism, etc., as well as post heart transplant, or as a bridge to longer term ventricular assist device (VAD) placement[6]. Respiratory indications for ECMO include ARDS secondary to pneumonia, aspiration, etc., in addition to lung transplant (as a bridge before the procedure or after if evidence of graft failure), and pulmonary hemorrhage[6]. The potential for ECMO use in COVID-19 patients has been a topic of discussion recently. Previous success with ECMO in critically-ill patients diagnosed with Middle Eastern respiratory syndrome (MERS) encouraged physicians to try ECMO in an attempt to treat COVID-19 patients[7].

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Respiratory management in coronavirus disease patients with pulmonary compromise . Extracorporeal membrane oxygenation is reserved as a final resort when all other noninvasive and invasive ventilation options fail. SpO₂: Saturation of oxygen via pulse oximetry; HFNC: High flow nasal cannula; BiPAP: Bilevel positive airway pressure; ET: Endotracheal; NMBA: Neuromuscular blockade agent; NIV: Noninvasive ventilation; IMV: Invasive mechanical ventilation; ECMO: Extracorporeal membrane oxygenation.

The most common clinical scenario in which patients with COVID-19 require ECMO is ARDS refractory to standard lung-protective ventilation and pronation[8,9]. In this situation, gas exchange is compromised given the underlying alveolar inflammation and edema; hence, patients require assistance with oxygenation. Veno-venous ECMO (V-V ECMO) is the modality used in such cases where blood is typically drained from a large peripheral vein, oxygenated via a synthetic lung, and returned to the circulation via a large peripheral vein[8]. Afterwards, newly oxygenated blood flows through the normal circulatory pathway to provide oxygenation to the remainder of the organ systems. With these ECMO settings, the native heart is required to function appropriately to ensure adequate blood distribution[8]. Initiation of ECMO in COVID-19 patients presents unique challenges as these patients are on maximal ventilatory support and are often in a prone position. Therefore, very specific criteria for ECMO initiation have been suggested: PaO₂/FiO₂ < 60 mmHg for > 6 h, PaO₂/FiO₂ < 50 mmHg for > 3 h or PCO₂ > 80 mmHg for > 6 h and arterial pH < 7.2[8].

In situations where the patient’s cardiovascular function may be severely compromised in addition to respiratory compromise, such as in severe myocarditis, veno-arterial ECMO (V-A ECMO) is the optimal configuration used. In V-A ECMO, venous blood is drained, oxygenated via synthetic lung, then returned with force via a large peripheral artery towards the aorta. This increase in aortic blood flow enhances peripheral perfusion[10].

Superimposed sepsis or multi-organ dysfunction may develop in patients on V-V/V-A ECMO, requiring further calibration of ECMO settings to enhance cardiac output and support bodily functions. The main concept behind modifying V-V/V-A ECMO in such difficult circumstances is through the addition of an extra lumen, converting double lumen to triple lumen ECMO. The addition of this third lumen can help optimize settings based on a patient’s requirements. For instance, if a patient on V-V ECMO (pulmonary support only) develops cardiac complications leading to compromise of cardiovascular function, the addition of an arterial output lumen, veno-venoarterial ECMO (V-VA ECMO), will allow for the addition of cardiac support to pre-existing pulmonary support[11]. In other circumstances, patients with both pulmonary and cardiovascular compromise may be inadequately oxygenating despite V-A ECMO; this is typically seen in larger patients or if a lumen with a small diameter is used[11]. The addition of a venous drainage lumen, venovenous-Arterial ECMO (VV-A ECMO), will allow more blood to be drained and oxygenated at a faster rate, thus improving oxygen supply[11]. Furthermore, it is not uncommon for patients on ECMO to develop acute kidney injuries within the first 48 h. The ECMO circuit itself can create an inflammatory reaction leading to capillary leak and subsequent pre-renal azotemia or even acute tubular necrosis[6]. In these instances, if pre-renal oliguria does not resolve after 72 h, continuous renal replacement therapy simultaneously with ECMO can be used to manage fluid status and maintain renal function[6]. Lastly, electrolytes and blood counts should be monitored very closely, as platelet consumption and potassium, magnesium, and phosphorous shifts have been observed in patients on ECMO and should be replaced accordingly[6].

Given the above, it would theoretically be rational to use ECMO for pulmonary and/or cardiovascular support in patients with COVID-19 refractory ARDS and certain other COVID-19-related complications; yet given the lack of clinical trials and prospective studies, questions regarding the true validity in the clinical setting remain unanswered. The two main factors that should be taken into consideration are its effectiveness and feasibility.

With regards to effectiveness, proof of ECMO success in patients with COVID-19 is scarce. Even prior to COVID-19, ECMO was shown to not lower 60-d mortality in patients with severe ARDS (from other non-COVID-19 conditions) vs other invasive ventilation techniques[9]. While trialing of ECMO in COVID-19 patients has increased during the pandemic, there are very limited reports of clinical outcomes. Furthermore, the handful of cases that have been published report inconsistent results. In a retrospective multicenter study by Ruan et al[12] that included 137 patients with COVID-19, seven patients required ECMO and there was 100% mortality despite ECMO use. These findings were supported by Yang et al[13] and Zhou et al[14], who reported 83% (5 out of 6) and 100% (3 out of 3) mortality rates in patients with COVID-19 who required ECMO at their respective centers. However, Wu et al[15] and Shen et al[16] each reported one patient on ECMO who survived. Although there are not any other official publications regarding ECMO support, the Extracorporeal Life Support Organization (ELSO) is performing real-time tracking of all COVID-19 cases on ECMO worldwide, and there is currently insufficient data for the ELSO to recommend either for or against ECMO in patients with COVID-19[17].

With regards to feasibility, ECMO is complex, especially when designing a referral system. ECMO is expensive to incorporate, and there is a complexity of management associated with its use that requires an individually trained critical care team, often only available in highly specialized centers. In addition, increasing healthcare worker exposure with such a high-risk procedure, particularly with lack of clinical trial evidence to prove its efficacy, raises ethical concerns. Most smaller healthcare centers, both inside and outside of the United States, lack access to ECMO devices and the training required to operate them. Therefore, most authors are supportive of ECMO use in critically ill patients, but only in experienced centers with the necessary resources. For other less-equipped areas, ensuring availability of more basic equipment such as noninvasive and invasive mechanical ventilation with adequate direction for referral to centers with ECMO expertise is of higher priority, and is projected to save more lives in the current pandemic[18]. However, this approach comes at a cost to the critically ill that may benefit from ECMO in less-equipped areas, and is an ethical dilemma worth mentioning.

Currently, the ELSO requires a set of guidelines to proceed with establishment of ECMO as a viable treatment option. These guidelines mandate ECMO be administered at a tertiary care center or greater with available facilities of a tertiary level Neonatal Intensive Care Unit, Pediatric Intensive Care Unit, and/or Adult Intensive Care Unit[17]. The location of service should also cover a geographic area that can provide a minimum of 6 ECMO patients per year[17]. The center should be actively participating in the ELSO registry[17]. The structure of the center should have a hierarchy including an ECMO program director, multiple associate directors assigned to a specific focus pertinent to ECMO care, an ECMO coordinator, and a multi-disciplinary team responsible for annual internal ECMO evaluation for quality improvement[17]. Every ECMO center should have its set of policies and procedures established with comprehensible indications and contraindications. Moreover, there should be distinct guidelines for clinical management, equipment maintenance, termination of therapy, and follow up of ECMO patients[17].

Currently, ECMO is used for respiratory support in 63% of cases, cardiac support in 29% of cases, and both in 8% of cases. The four categories that the ELSO registry considers in its recording of ECMO as it pertains to the pandemic are as follows: COVID-19 confirmed by testing, COVID-19 suspected but no testing confirmation, no clinical suspicion of COVID-19 (and no testing), and COVID-19 confirmed negative[17]. On June 26, 2020, the ELSO registry reported 1619 suspected or confirmed cases of COVID-19 patients on ECMO and specifically listed 1604 confirmed cases of COVID-19 patients on ECMO[17]. The discharged alive rate at 90 d from ECMO was reported at 541/975 patients (55%), and included discharges to rehabilitation facilities and long-term care facilities, indicating a possible lengthier recovery[17]. This rate is not far off from non-COVID-19 ARDS patients on ECMO, where 52% survived to hospital discharge[6]. For reference, in patients who require ECMO for cardiac support due to cardiac arrest or cardiogenic shock, survival rates range from only 20%-30% to hospital discharge[6]. The predominant form of ECMO utilized was VV, which was reported to be used 95% of the time. VA and other configurations were used in 5% cases. The utilization of ECMO as per various ELSO chapters can be seen in Figure 3, where North America demonstrated the highest use of ECMO followed by Europe[17].

Open in New Tab   Full Size Figure   Download Figure

Figure 3 Utilization of extracorporeal membrane oxygenation. Various Extracorporeal Life Support Organization chapter uses were reported. Data reported was based on reports from June 26, 2020. ECMO: Extracorporeal membrane oxygenation; ELSO: Extracorporeal Life Support Organization.

ECMO remains a valid treatment option for patients when other conventional treatment strategies fail. In patients diagnosed with COVID-19, therapy is guided largely from experience with previous coronavirus pandemics such as MERS. North America is the largest geographical region to utilize ECMO in the treatment of COVID-19, and it is without question necessary to have the personnel and infrastructure in place in order to safely treat patients with ECMO. In recent months, new literature continue to demonstrate more clear indications and contraindications for ECMO use, however, much research is still needed to demonstrate clear mortality benefit. Ethical dilemmas also need to be considered, such as ECMO use in the setting of CPR, and modes of expansion need to be examined in order to minimize the treatment availability gap between patients with access ECMO centers and those without access.