| Literature DB >> 26608160 |
L Christian Napp1, Christian Kühn2, Marius M Hoeper3, Jens Vogel-Claussen4, Axel Haverich2, Andreas Schäfer5, Johann Bauersachs5.
Abstract
Extracorporeal membrane oxygenation (ECMO) has revolutionized treatment of severe isolated or combined failure of lung and heart. Due to remarkable technical development the frequency of use is growing fast, with increasing adoption by interventional cardiologists independent of cardiac surgery. Nevertheless, ECMO support harbors substantial risk such as bleeding, thromboembolic events and infection. Percutaneous ECMO circuits usually comprise cannulation of two large vessels ('dual' cannulation), either veno-venous for respiratory and veno-arterial for circulatory support. Recently experienced centers apply more advanced strategies by cannulation of three large vessels ('triple' cannulation), resulting in veno-veno-arterial or veno-arterio-venous cannulation. While the former intends to improve drainage and unloading, the latter represents a very potent method to provide circulatory and respiratory support at the same time. As such triple cannulation expands the field of application at the expense of increased complexity of ECMO systems. Here, we review percutaneous dual and triple cannulation strategies for different clinical scenarios of the critically ill. As there is no unifying terminology to date, we propose a nomenclature which uses "A" and all following letters for supplying cannulas and all letters before "A" for draining cannulas. This general and unequivocal code covers both dual and triple ECMO cannulation strategies (VV, VA, VVA, VAV). Notwithstanding the technical evolution, current knowledge of ECMO support is mainly based on observational experience and mostly retrospective studies. Prospective controlled trials are urgently needed to generate evidence on safety and efficacy of ECMO support in different clinical settings.Entities:
Keywords: Cardiogenic shock; ECMO; Extracorporeal circulation; Heart failure; Mechanical circulatory support
Mesh:
Year: 2015 PMID: 26608160 PMCID: PMC4805695 DOI: 10.1007/s00392-015-0941-1
Source DB: PubMed Journal: Clin Res Cardiol ISSN: 1861-0684 Impact factor: 5.460
Fig. 1Veno-venous ECMO (VV). Blood is drained from the right atrium and the inferior vena cava, oxygenated and decarboxylated in an extracorporeal rotor/oxygenator device and returned to the right atrium
Hemodynamic changes during ECMO support depends on the cannulation mode
| Strategy | Right atrial pressure | Left ventricular end-diastolic pressurea | Systemic blood pressure | LV afterload | Catecholamine dosing | |
|---|---|---|---|---|---|---|
| Vasopressors | Inotropes | |||||
| Veno-venous | ↔ | ↔ | ↔ | ↔ | ↔–↓b | ↔ |
| Veno-arterial | ↓–↓↓ | Varies (should decrease) | ↑↑ | ↑↑ | ↓ | ↓ |
| Veno-veno-arterial | ↓↓ | Varies (should decrease) | ↑↑ | ↑↑ | ↓ | ↓ |
| Veno-arterio-venous | Varies | ↑ | ↑ | ↑ | Varies | Varies |
While VV-ECMO is largely neutral in this context, all cannulations with arterial access profoundly influence venous and arterial pressures by modified flow. Much of the information in this table is based on experience and requires formal confirmation by dedicated studies
aEffects vary upon function of the aortic valve
bMay decrease with improvement of metabolic status by enhanced gas exchange
Fig. 2Bicaval dual-lumen cannula. This cannula allows for parallel drainage and supply through one tubing with two lumina during veno-venous ECMO. It thus requires only one large access vein and minimizes recirculation by directed supply towards the tricuspid valve (red arrow), spatially separated from the inflow (blue arrows)
Fig. 3Veno-arterial ECMO (VA). Blood is drained from the right atrium, oxygenated and decarboxylated in the ECMO device and returned to the iliac artery towards the aorta. Note the modified position of the venous cannula tip compared to veno-venous ECMO. Cannulation of the femoral artery requires an additional sheath for perfusion of the leg downstream of the cannulation site (inset)
Fig. 4Watershed phenomenon during veno-arterial ECMO visualized by computed tomography. Antegrade blood flow (low contrast) from the heart competes with retrograde blood flow (high contrast) from the ECMO in the aorta, resulting in a watershed phenomenon (arrowhead). Here computed tomography of a patient with pulmonary embolism and reduced cardiac output demonstrates a rather proximal watershed, leading to perfusion of the right carotid artery with “heart blood” (dark) and the left carotid artery with “ECMO blood” (bright, arrows). Upper panel sagittal oblique maximum intensity projection (MIP), middle panel coronal oblique MIP, lower panel transverse plane
Publications on triple cannulation ECMO support
| Strategy | Patients with triple cannulation | Characteristics | Outcomes |
|---|---|---|---|
|
| |||
| Ford and Atkinson [ |
| A 3000-g 37-week gestation child was born by vaginal delivery and developed respiratory failure from congenital diaphragmal hernia. Veno-arterial ECMO was initiated, but within 24-h hemodynamic support was insufficient due to limited flow through the venous cannula (low bladder pressure, low blood pressure, low central venous oxygenation of 60 %). A third cannula was inserted into the right common iliac vein by cutdown. After veno-veno-arterial ECMO had started central venous saturation increased up to 79 %. Total ECMO support lasted 5 days | The patient underwent surgery for diaphragmal hernia, could be weaned from ECMO and the ventilator and could be discharged home after 31 days in hospital |
| Hou et al. [ | Sheep model | Animal study on the effects of different drainage locations during ECMO support. While veno-arterial ECMO with inferior vena cava drainage was running, acute respiratory failure was initiated. This led to severe upper body hypoxemia, with no significant effect on blood pressure. Repositioning the venous drainage cannula to the superior vena cava strongly increased aortic oxygen saturation from 35 to 75 % and thereby reverted upper body hypoxemia | Drainage from the superior vena cava strongly improved systemic oxygen saturation, strongly suggesting that bicaval drainage is sufficient to disrupt the “two-circulation-syndrome” |
| ELSO [ | Guideline | Guideline for ECMO support in adults of the Extracorporeal Life Support Organization (ELSO). The guideline mentions the option to add a cannula from the superior vena cava for improved venous drainage | |
|
| |||
| Madershahian et al. [ |
| Three patients with veno-arterial ECMO due to ARDS after polytrauma. One of them had persistent upper body hypoxemia and needed conversion to veno-arterio-venous ECMO, which led to an increase of pH from 7.2 to 7.45, lung compliance from 15 to 40 ml/mbar and oxygen saturation from 70 to 95 %. Total ECMO support lasted 4.7 ± 1.1 days | No ECMO-related complications were reported. All patients were successfully weaned from ECMO and later on from ventilation and could be discharged |
| Stöhr et al. [ |
| 30 patients with ARDS from pneumonia ( | Bleeding occured in eight patients (one venous and seven arterial) and hyperperfusion and leg ischemia and wound healing complications in one patient each. 15 patients died during ECMO support, one died after ECMO explantation. Mortality was higher -in the veno-venous cohort (63 %) and the veno-arterial cohort (75 %) than in the veno-arterio-venous cohort (27 %). Overall 30-day mortality rate was 53 %. One patient was bridged to lung transplantation. During a mean follow-up of 21 months three patients died |
| Kustermann et al. [ |
| 30-year-old patient with community-acquired pneumonia who developed ARDS and severe septic cardiomyopathy. Veno-arterial ECMO was initiated, but was expanded to veno-arterio-venous cannulation because of a remaining low Horovitz index of 130 on ECMO support. FiO2 and ventilation pressures could be reduced and 1 day later ECMO was downgraded to veno-venous in the presence of improvement of left ventricular function (LVEF from 10 to 45 %). Total ECMO support lasted for 7 days | No ECMO-related complications were reported. Successful weaning off ECMO was followed by transfer to the referring hospital and complete weaning from ventilation |
| Moravec et al. [ |
| 74-year-old patient with pulmonary hypertension related to pulmonary fibrosis, who developed pneumonia, sepsis and subsequent shock. Initial veno-arterial ECMO was expanded to veno-arterio-venous ECMO with a jugular Shaldon catheter for ARDS. FiO2 decreased from 100 to 45 %, with a nearly doubled PaO2. Total ECMO support lasted 9 days. 59-year-old obese patient with cardiogenic shock, refractory to medical therapy, who was resuscitated during cardiac catheterization and received an IABP. He was stabilized with veno-arterial ECMO, but developed ARDS and a jugular Shaldon catheter as third cannula was implanted for venous preoxygenation. FiO2 decreased from 100 to 40 %, with a more than doubled PaO2. Total ECMO support lasted 13 days. A third patient was reported, who received veno-arterio-venous ECMO with standard ECMO cannulae instead of a Shaldon catheter. In this patient ECMO was withdrawn after 12 days and the patient was discharged from hospital later | No ECMO-related complications were reported. All three patients could successfully be weaned from ECMO support. The first patient died later on from lung fibrosis without the prospect of receiving transplantation, but the second one survived without neurological deficit. The third patient was discharged after weaning from ECMO |
| Chung et al. [ | Review | Excellent review emphasizing the various aspects of monitoring during ECMO support. The authors describe the principle of veno-arterio-venous triple cannulation | |
| Choi et al. [ |
| 39-year-old patient with acute myocardial infarction. Veno-arterial ECMO was inserted during cardiopulmonary resuscitation. 5 days after onset of ECMO secondary respiratory failure and subsequent brain hypoxia (upper body hypoxemia) developed. A third cannula was added for preoxygenating venous blood. PaO2 increased from 39 to 103 mmHg, SO2 from 69 to 89 %. Hemodynamics were not provided in the publication. Duration of ECMO support was 10 days, with 5 days of veno-arterio-venous cannulation | The patient was successfully weaned from ECMO and ventilator and was sent to rehabilitation, with an uneventful recovery at 13-month follow-up |
| Kim et al. [ |
| Nine patients with ECMO after resuscitation for near-drowning. Seven patients received veno-arterial cannulation, one was converted to veno-venous ECMO in the presence of very good hemodynamics and continued ARDS, and one patient initially received veno-arterio-venous ECMO in the presence of severe ARDS and concomitant cardiac dysfunction. Measures for this single patient are not provided. Mean duration of ECMO support was 7.8 days | All patients were weaned from ECMO, and there were no ECMO-related complications reported. Seven patients survived with a favorable neurological outcome, two patients had irreversible hypoxic brain damage and eventually died |
| Biscotti et al. [ |
| 21 patients with veno-arterio-venous ECMO. 11 patients were set at triple cannulation from the beginning for severe combined cardiorespiratory failure, such as pulmonary embolism, terminal lung disease with cardiac failure, ARDS with cardiogenic shock or LVAD failure. Eight patients had veno-venous ECMO, e.g., for ARDS or cystic fibrosis and were switched to veno-arterio-venous cannulation due to new onset of heart failure. One patient had lung transplantation on veno-arterial ECMO and thereafter received veno-arterio-venous ECMO as a bridge to veno-venous ECMO. One patient had ARDS and experienced upper body hypoxemia during veno-arterial ECMO, which was subsequently expanded to veno-arterio-venous ECMO. Mean duration of ECMO support was 6.5 ± 5.5 days | Seven patients had bleeding. Other complications were oxygenator failure ( |
| Ius et al. [ |
| Nine patients with veno-venous ECMO, one patient with veno-arterial ECMO. ECMO was started for ARDS or other forms of respiratory failure. All patients were switched to veno-arterio-venous cannulation for new onset heart failure (right heart failure, pericardial tamponade or mitral regurgitation). Time-to-switch was 2 ± 2.5 days, with a total ECMO support time of 10 ± 4 days | One patient developed pericardial effusion. Three patients had bleeding, and two patients developed leg ischemia. Three patients were successfully bridged to lung transplantation, of which two survived to hospital discharge. Another four were successfully weaned off ECMO, of which three survived to hospital discharge. Three patients died on ECMO support during hospitalization |
| ELSO [ | Guideline | Guideline for ECMO support in adults of the Extracorporeal Life Support Organization (ELSO). The guideline offers to convert veno-arterial to veno-arterio-venous cannulation when severe respiratory failure occurs | |
ARDS denotes acute respiratory distress syndrome
FiO distress syndrome, inspiratory oxygen fraction, LVEF left ventricular ejection fraction, PaO partial oxygen saturation
Fig. 5Veno-veno-arterial ECMO (VVA). When unloading by veno-arterial ECMO is not sufficient, a second draining cannula may be necessary. The draining flows from the two venous cannulas are merged outside the body using a Y-connector (inset)
Fig. 6Veno-arterio-venous ECMO (VAV). When circulatory support with veno-arterial ECMO is complicated by respiratory failure or when respiratory support by veno-venous ECMO is complicated by heart failure, a third cannula may be necessary. Both approaches result in one draining and two supplying cannulae. Flow through the supplying cannulae is balanced using an adjustable clamp (inset, black arrow) and a separate flow sensor (inset, white arrow)
A unified nomenclature for ECMO cannulation
| Strategy | Figures | Draining cannulaa | Supplying cannulaa | Indication |
|---|---|---|---|---|
| VV | 1 | Inferior vena cava | Superior vena cava | ARDS |
| VA | 3 | Right atrium | Common iliac artery | Postcardiotomy cardiogenic shock |
| VVA | 5 | Inferior vena cava | Common iliac artery | Insufficient unloading during VA-ECMO |
| VAV | 6 | Inferior vena cava | Common iliac artery | Respiratory failure during VA-ECMO |
Letters before “A” are draining cannulas, and “A” and all following letters denominate supplying cannulas. The proposed nomenclature does not consider the arterial sheath for distal leg perfusion and does not change upon use of a bicaval dual-lumen cannula
AMI denotes acute myocardial infarction, ARDS acute respiratory distress syndrome, PA pulmonary artery, PCI percutaneous coronary intervention, RV right ventricle
aTypical place of blood supply/drainage (cannula tip), not place of vascular access/puncture