Patient selection for ECMO

ECMO is half of "Mechanical supports: ECMO/IABP", an L1 topic of Section 2.1.4 in the second edition of the CICM Syllabus for the Second Part Examination. This chapter is a summary of the relevant indications and contraindications sections from the ELSO guidelines and the Alfred ECMO guidelines. The latter are an excellent place to both start and finish your exam reading, as they compete favourably with official sources in their level of authority, but remain readable, and keep the detail to the sort of minimum that respects the time constraints of the last-muinute cram session. The exam candidate who was somewhat more organised than this could take advantage of the excellent IntensiveBlog "Everything ECMO" series which offers a comprehensive tour of the subject. 

The importance of good patient selection for ECMO

What follows is not unique to any specific ECMO strategy and applies equally well to VV, VA, and some bizarre hybrid VVPa-A quadruple cannulation technique. The term "ECLS", extracorporeal life support, is occasionally used in lieu of ECMO to embrace other extracorporeal strategies, but because this wording could be argued to incorporate such strategies as CVVHDF, we will persist with ECMO for the remainder of this chapter. Of course the following rationale can be applied even more broadly to the decision of commencing any highly invasive and potentially lifesaving therapy in the face of a worsening patient condition (one which occasionally resembles death).

In brief summary:

• ECMO has technical limitations which reduce its applicability to certain patient populations, such as
  • the morbidly obese
  • those with impossible vascular access
  • uncontrolled severe haemorrhage
  • severe vasodilated shock
  • these situations are absolute contraindications, as the initiation or maintenance of ECMO would be unsuccessful in these patient groups on purely technical grounds.
• ECMO is a support strategy rather than a curative therapy, which:
  • does not restore functional independence or organ function
  • does create the risk of considerable dependence
  • has morbidity associated with its use, which on its own can impair organ function
  • It is therefore the responsibility of clinicians to offer ECMO only in those scenarios where it is likely to lead to patient recovery, in accordance with principles of nonmaleficience.
• ECMO [or insert any invasive horrendotherapy] is a highly intrusive treatment which 
  • has the potential to prolong the existing level of patient discomfort or disability
  • has the potential to introduce considerable new discomfort in the form of invasive procedures and ensuing complications
  • introduces ethical challenges around treatment withdrawal by putting patients into "bridge to nowhere" scenarios
  • It is therefore the responsibility of clinicians to offer this option only where the likelihood of benefit outweighs the likelihood of harm, in accordance with the principles of beneficence
• ECMO [lapartomy, VAD, transplant, whathaveyou] is a resource-intense therapy which is:
  • not widely available
  • with resource costs beyond the original initiation (eg. long ICU and hospital stay, rehab, prolonged disability of survivors)
  • It is therefore the responsibility of clinicians to allocate these resources carefully, in accordance with the principles of distributive justice
• ECMO [etc] is often considered for scenarios where consultation with the patient (or shared decisionmaking with their surrogates) is impossible because of the acuity of the situation (eg. ECPR),  and therefore:
  • The patient is unable to exercise their autonomy
  • The clinicians are unable to carry out a comprehensive consent process with surrogates
  • It is therefore the responsibility of professional organisations to develop standards that support clinicians in their decisionmaking where consent is impossible
• On the basis of these concerns, the decision to offer ECMO as a support strategy should have clear objective guidelines with a transparent decisionmaking process behind them.

These are all points designed to answer the question, "why do we even select patients for ECMO". It is "life support", after all, in a lay sense; and we offer that as the default standard wherever there is insufficient information to decide otherwise. It has been pointed out that the ethics of deciding to offer or not offer ECMO, when observed in clinical practice, "closely resembled traditional concerns about the appropriate use of any life-sustaining treatment rather than the novel dilemmas imagined in the current literature" (Courtwright et al, 2016). In short, it is an intervention like any other, and has its indications and contraindications. 

Indications and Contraindications for VV ECMO

The Alfred Patient Selection page is an excellent summary of the most important points, put so succinctly that it is practically dying to become a CICM exam question. These are local and may not be perfectly portable.  ELSO Guideline for Adult Respiratory Failure Managed with Venovenous ECMO are the next best option. Their Table 1 was appended with some of the selection criteria from the CESAR and EOLIA trials:

Indications and Contraindications for VA ECMO

Again, the Alfred guidelines were extensively drawn from, particularly for the indications. ELSO only give "refractory cardiogenic shock with a potentially reversible or surgically correctable cause" as their broad indication for VA ECMO. The Alfred guidelines are more granular, giving a list of favourable and high-risk diagnostic categories.