Now we are at the point where we can start to apply our discussion of physiology to specific extracorporeal membrane oxygenation (ECMO) configurations. As you will see, even though they both use the same pump, circuit, membrane oxygenator, and cannulas, our two flavors of ECMO, veno-venous ECMO and veno-arterial ECMO, provide very different, support. You will get a sense of the subtleties related to these modalities in the chapters to follow.
In this chapter, we will focus on veno-venous ECMO, otherwise known as respiratory ECMO or VV ECMO.
VV ECMO: the basics
Let’s start this discussion by reminding ourselves of the configuration of VV ECMO. When we refer to VV ECMO, we are draining deoxygenated blood from the venous system, running it through our membrane oxygenator, and returning the oxygenated blood to the venous circulation where it gets circulated to the body through the native right ventricle ( Fig. 12.1 ).
The mechanism allows for improvement of oxygenation and ventilation/CO 2 removal parameters in patients with respiratory failure. We will further our understanding of this mechanism and physiology, exploring the rationale for the beneficial and the potential adverse hemodynamic and respiratory effects of VV ECMO.
Mechanism of hemodynamic/respiratory benefits of VV ECMO
Let’s start by reviewing the physiologic rationale for VV ECMO. Remember that ECMO does not cure any of the inciting etiologies. Rather, it is only effective to the extent that it mitigates the toxicity related to the support that is required in response to an insult. Thus, even if the physiologic rationale is intriguing, it is only relevant in as much as it translates to an improvement in the outcome of the patient.
With that in mind, we will now explore the two primary mechanisms of benefit related to VV ECMO support, improvement of shunt and improvement of right ventricular hemodynamics.
Let’s review each.
Effect of VV ECMO on shunt physiology
As discussed in Chapter 4 , while there are a variety of mechanisms for hypoxia (hypoventilation, diffusion defect, low inspired oxygen levels, intracardiac shunt), shunt physiology plays an important role. When there is worsening mismatch at the alveolar/blood vessel interface leading to V/Q of less than 1, blood shunts across the pulmonary circulation without participating in gas exchange ( Fig. 12.2 ).
This is especially important because as the percentage of shunt increases, the lungs become less responsive to administration of supplemental oxygen, and positive pressure is required to decrease the shunted alveoli.
The effect of positive pressure is relevant to our discussion with ECMO because as respiratory failure worsens, the increasing shunt and decreasing pulmonary compliance combines to require escalating ventilator pressures, and worsens the potential for damage to the lungs due to the ventilator in the form of barotrauma/volutrauma.
Shunt in VV ECMO behaves much differently.
In this case, even though blood is still being shunted across the pulmonary circulation without participating in gas exchange in the lungs, some of this blood is now already oxygenated from the ECMO circuit, and the shunted blood is oxygenated ( Fig. 12.3 ).
The result is that shunt fraction is improved even if V:Q matching is not. In fact, VV ECMO may even worsen V:Q matching by decreasing hypoxic vasoconstriction and increasing perfusion to damaged potions of the lungs.
This may seem like an esoteric point. You may be asking – why is this important to the patient?
Since shunt fraction is less of a priority while on ECMO, you do not need to improve shunt fraction with higher and potentially damaging airway pressures.
The extent that you can reduce elevated/damaging ventilator pressures/volumes with ECMO will determine the extent that it is able to benefit your patient.
Hemodynamic effects of VV ECMO
To better understand the hemodynamic effects of VV ECMO, let’s start with the hemodynamic effects of hypoxia. If oxygen levels precipitously drop, how does this affect the cardiac output and blood pressure? To help answer, let’s imagine a patient in respiratory failure who has a cardiac arrest due to a respiratory arrest. What is the typical presentation?
Usually, if you were to review the telemetry and monitor, the presentation is sinus rhythm, followed by a bradycardia, followed by hypotension, followed by arrest. Why this pattern?
The answer is that hypoxia causes hypoxic vasoconstriction as an adaptive mechanism to overcome V/Q mismatch. This vasoconstriction raises right ventricular afterload, which will drop right ventricular output leading to conduction abnormalities (bradycardia) followed by cessation of right ventricular output altogether (arrest).
You may also recall from Chapter 2 the other causes of elevated pulmonary vascular constriction that can worsen right ventricular afterload and precipitate right heart failure:
↓O 2
↓pH
↓temperature
↓pulmonary compliance
↑CO 2
↑α-adrenergic tone
Now let’s return to what happens during VV ECMO – where blood is returned to the right atrium with high levels of O 2 , low levels of CO 2 , and is temperature controlled. The effect can be a rapid improvement in hemodynamics as the oxygenated blood reduces pulmonary vasoconstriction, decreases right ventricular afterload and eventually increases right ventricular output, leading to improved left ventricular filling, and improvement in overall cardiac output.
This hemodynamic effect of VV ECMO can lead to an improvement in DO 2 that is just as significant as the infusion of oxygenated blood via the circuit.
The degree that VV ECMO improves hemodynamics depends on the degree to which cardiac output is limited by right ventricular afterload due to hypoxia/hypercapnea. Persistent hypotension following initiation of VV ECMO should raise suspicion and prompt investigation for other causes of shock, namely vasodilation and sepsis.
It is difficult to predict the hemodynamic effect on right ventricular cardiac output that can be anticipated from the initiation of VV ECMO. At this point, it is important to be aware of the potential for improvement and to factor the anticipation of this improvement into the prediction of how ECMO will improve the patient with hypoxic respiratory failure. Understanding the physiology behind patients who respond and those who do not respond will be an essential part of selection for ECMO as well as management while on ECMO.
This is a nuanced way of thinking about VV ECMO support, but will ultimately equip you with a better sense of our sweet spot (patients who will do well with ECMO who would do poorly without ECMO) rather than just selecting patients based on specific respiratory parameters (P:F, OI, ventilator requirements, etc.).
Mechanism of Adverse Respiratory/Hemodynamic Effect on VV ECMO
When considering ECMO and the physiologic effects of ECMO we must always consider the harmful effect of ECMO support. Usually we do this by thinking about the adverse events that are often reported for ECMO – bleeding, clotting, stroke, damage to underlying structures. However, we should also consider the physiologic risks of support, particularly what limits the effectiveness of support and the degree of toxicity of conventional support that may need to be incorporated into the care as a result.
Limitations to the Respiratory Effects of VV ECMO
Said another way, what are the physiologic mechanisms behind hypoxia for patients on VV ECMO.
Wait, hypoxia on VV ECMO?
You read that right. Although VV ECMO is designed to mitigate hypoxia, hypoxia can be quite common. Let’s explore why.
For a patient who is not on ECMO, as the lungs become progressively more compromised, and shunt is increasing despite all interventions, you can anticipate more blood to be shunted across the lungs, such that the saturation of blood entering the left atrium approximates the saturation of the blood leaving the right ventricle ( Fig. 12.4 ).
This will then lead to a decrease in the saturation of the venous blood returning to the heart, which in turn drops the saturation of the blood returning from the lungs, leading to a rapid deterioration.
ECMO can mitigate this downward spiral by draining the venous blood that would be going across the shunted pulmonary blood vessels and replacing it with oxygenated blood. To the extent that the drainage/return of this blood is greater than the native cardiac output, this will result in a patient oxygen saturation of 100% ( Fig. 12.5 ).
However, let’s consider this same clinical situation, except now, where the native cardiac output picks up and is greater than the ECMO blood flow by a factor of 2:1 (with a right ventricular output say of 8 L compared to a maximum effective achievable ECMO blood flow of 4 L). In this case, half of the blood that is being circulated across the shunted pulmonary circulation is 100% oxygenated while the other half is shunted venous blood that is 70% saturated.
In this case, the expected saturation of blood returning to the left heart and consequently the expected patient saturation would be 85%, the average between the blood returning from the oxygenator (100% saturated) and the blood that is not flowing through the oxygenator (70% saturated) ( Fig. 12.6 ).
