The American Society for Artificial Internal Organs (ASAIO) offers recommendations on selection and treatment of patients with respiratory and circulatory failure associated with coronavirus disease 2019 (COVID-19) using artificial pulmonary and/or cardiac support. These recommendations were published in Circulation: Heart Failure and are based on a review of the epidemiology and pathophysiology of COVID-19, as well as therapeutics for and clinical outcomes of the disease, and should be considered in conjunction with those from the Extracorporeal Life Support Organization (ELSO).
Hypoxic respiratory failure with acute respiratory distress syndrome (ARDS) complications is a leading cause of death in patients with COVID-19.Late-onset cardiac dysfunction, septic shock and multiorgan system failure all have an impact on mortality in those patients. Although only a small subset of critically ill patients with COVID-19 are eligible for and may benefit from artificial support strategies, these approaches, because of associated complexities and resource consumption, demand optimal care pathways to guide management decisions. Despite their increasing use in patients with refractory COVID-19, the precise role and efficacy of modalities such as extracorporeal membrane oxygenation (ECMO) and other mechanical cardiopulmonary support strategies remain largely unclear.
A large proportion of patients in intensive care unit have acute hypoxemic respiratory failure, with many cases complicated by ARDS. Myocardial injury and overt shock are also prevalent in those patients and often accompanied by hepatic and renal end-organ dysfunction. When critically ill patients do not respond to conventional therapies, mechanical pulmonary and/or circulatory support may help stabilize or improve their condition.
When high-flow oxygen yields suboptimal outcomes, noninvasive mechanical ventilation (MV) with continuous or bilevel positive airway pressure may be a useful short-term therapy to correct ventilation/perfusion mismatches in patients without profound hypoxemia, at risk for aerosolization. If a transition to invasive MV is deemed necessary, lung-protective ventilation should be used, regardless of whether the selected MV mode is volume- or pressure-controlled, pressure-supported, or mixed.
Manual or artificial prone positioning for invasive MV should be considered in order to improve basal aeration in patients with COVID-19, despite the fact that it requires more personnel. Clinicians should be aware that high positive end-expiratory pressure settings may result in elevated pulmonary vascular impedance and alveolar overdistention. These phenomena may lead to lower systemic venous return and reduced right ventricular (RV) cardiac output as a result of decreased RV stroke volume.
Rapid assessment and decision making are key when considering extracorporeal gas exchange for patients in whom conventional invasive MV with paralytic agents and prone positioning failed. This approach should be reserved for patients with isolated pulmonary dysfunction who have received invasive MV for ≤7 days. Gas exchange can be optimized with selection of appropriate cannulation.
Although central cannulation provides higher maximal volumetric flow rates due to larger cannulae with lower resistance, this procedure requires open surgery and is associated with greater bleeding risks. Therefore, peripheral cannulation, which is generally percutaneous with a lower risk profile, should be used initially.
Right-sided ECMO, most often achieved with venous-venous (V-V) cannulation, can be performed using single or dual cannulae percutaneously. Although the use of a single cannula is associated with lower bleeding risks, it also requires image guidance. The use of dual vs single cannulae achieves higher flow rates and does not require intra-procedural imaging. Therefore, a dual cannulae bifemoral approach is recommended in most patients with COVID-19, but the decision of which approach to use should be based on team familiarity and comfort level.
When overt shock is obvious during pulmonary arterial catheterization or on echocardiography, left-sided mechanical circulatory support (MCS) with or without pulmonary support, should be selectively considered in patients with COVID-19 who are likely to benefit from this approach. MCS should be considered in patients with systemic arterial hypotension with reduced LV ejection fraction and normal LV end-diastolic volume and heart rate, particularly when inotropic medications have proved inadequate.
For left-sided ECMO, venoarterial (V-A) ECMO is most often used, although the more complex hybrid V-V/V-A approaches may be considered in some cases. Patient- and center specifics should determine the use of central vs peripheral cannulation.
Direct LV unloading — a controversial approach with left-sided V-A ECMO — and better systemic blood flow distribution may prompt the use of LV assist devices (LVADs), such as the Impella (which offers a wide range of flow rates), over V-A ECMO, in select cases. When gas exchangers are added, isolated treatment of pulmonary and cardiac failure is possible. The ability to place an LVAD using an axillary artery approach which can be performed at the bedside affords longer MCS in patients who are slow to recover. ECMO combined with the Impella device have been used in severely compromised patients with COVID-19.
RV support strategies may be necessary in patients in whom V-V ECMO proves insufficient and who have pulmonary failure leading to RV afterload and lower RV output due to cor pulmonale. For patients who need to be in prone position, percutaneous bifemoral right ventricular assist device may be used with an oxygenator. In those patients, peripheral approaches are preferred for device placement, but central cannulation is available for those with suboptimal flow rates.
ELSO recommends the use of ECMO in cases in which conventional management poses a high mortality risk. It is recommended that young patients with COVID-19 who are the most affected, be sent to hospitals with experienced ECMO teams.
Decisions regarding who should receive ECMO during the pandemic must be based on several patient- and center-specific factors, including age, comorbidities, patient load, and hospital capabilities. Patients with the highest priority are those without comorbidities, <50 years of age. Healthcare workers are also given high priority.
V-V ECMO is not an alternative therapeutic option for MV and should only be used in patients for whom invasive MV has not led to good clinical outcomes. Finally, for adult patients on V-A ECMO, the Survival After V-A ECMO scoring system can be used to estimate chances of survival. During the COVID-19 pandemic, when resources are limited, patients who have not demonstrated cardiac or pulmonary improvement after 14 days on ECMO should be taken off ECMO and returned to conventional therapy.
“This document is offered by ASAIO as a starting point of guidance to help our community approach these critically ill patients. This document will evolve as our collective experience grows and as treatment approaches reveal efficacy vs limited success,” noted the authors.
Funding and Conflicts of Interest Disclosures
Dr Slepian has research grants from Abiomed and NIH/NIBIB. Dr Keller has research grants from Abiomed and NIH. Dr Bime has a research grant from NIH/NHLBI. Dr Zwischenberger has research grants from Avalon-Maquet and NIH and is an unpaid consultant for Cytosorbents; Dr El-Banayosy has a research grant from Abbott. The other authors report no conflicts.
Rajagopal K, Keller SP, Akkanti B, et al. Advanced pulmonary and cardiac support of COVID-19 patients. Circ Heart Fail. 2020;13(5). doi:10.1161/circheartfailure.120.007175