**6. ECMO support for postcardiotomy low cardiac output syndrome**

Postcardiotomy (PC) low cardiac output syndrome is generally defined as a shock state refractory to inotropic support and\or IABP, and it can manifest as an inability to separate from CPB or persistent CS despite maximal use of pharmacological agents and\or IABP in the immediate postoperative period. It is a rare but a detrimental complication after cardiac surgery.

ECMO has been used as a salvage in such complications more than 50 years ago mainly for cardiac surgery in pediatrics but remained quiescent in adult population. However, during the last several years, ECMO is being used more and more in adult patients, particularly for postcardiotomy low cardiac output syndrome. Since its introduction to be used to support PC shock, ECMO has been a lifesaver and an important prognosis changer in such complications. It is reported that from 2007 to 2011, non-percutaneous ECMO cannulation increased 2-fold, while the use of percutaneous ECMO increased by more than 15-fold. In a study that looked at more than 9,000 ECMO patients from the Nationwide Inpatient Sample database in the US from 1998 to 2009, 4,493 cases (approximately 50%) were cannulated for cardiogenic shock in the postoperative period. In the same database, researchers observed that PC-ECMO was the most frequent ECMO indication between the years 2002 and 2011 [47]. The usage of PC-ECMO has increased over the previous ten years, according to data from the Extracorporeal Life Support Organization (ELSO) registry [48]. Despite the growing evidence and the widened use of such expensive, highly debatable yet important surgical armament. Unfortunately, data is unpowered, limited, conflicting, and highly variable.

### **6.1 Indications, contraindications, and cannulation of VA-ECMO**

### *6.1.1 Indications*

Currently, there is no consensus regarding when to initiate extracorporeal life support (ECLS) in the setting of postcardiotomy low cardiac output syndrome. A recent paper, the 2020 EACTS/ELSO/STS/AATS Expert Consensus on PostCardiotomy Extracorporeal Life Support in Adult Patients, represents to date the first comprehensive guideline to provide structured and clinical recommendations about the most relevant issues surrounding its application in this setting [49].

In this joint effort, the authors considered that class I indications of ECMO in the postcardiotomy setting can be summarized as follows:

• ECMO support to be initiated prior to end-organ injury or onset of anaerobic metabolism (lactate level <4mmol/l) in patients with likelihood of myocardial recovery and in the absence of uncontrollable bleeding not amenable to surgical repair [49].

*Venoarterial Extracorporeal Membrane Oxygenation in Cardiac Surgery DOI: http://dx.doi.org/10.5772/intechopen.106823*

• In case the likelihood of myocardial recovery is low, ECMO is recommended in patients who are eligible for long-term mechanical support or heart transplantation (LT-MCS or an HTx) [49].

In addition, timely implantation prior to severe end-organ hypoperfusion and ischemic injury represents one of the most powerful predictors of ECMO outcome, the early use of ECMO after cardiac surgery in a patient with an IABP and optimal medical therapy, with failure to wean from CPB or marginal hemodynamics also has been listed as class 1 recommendation [49]. Scoring systems were developed to prognosticate the outcomes following ECMO patients in general and mainly to help clinicians when best to avoid or consider it and it gives guidance while exploring it as an option for the families and its complications. Of these scoring tools, the survival after venoarterial-ECMO (SAVE) score has been considered one of the best predicting tools for ECMO patients in general due to its independent variable cohort; however, it was not developed to meet the special physiologic milieu of postcardiotomy patients [50].

Recently, a single-center, retrospective study that included 166 postcardiotomy CS patients supported with VA-ECMO after CABG over a 14-year period created a 6-items bedside scoring system; the REMEMBER score has been able to predict the mortality in that study cohort. It was found that older age, left main disease, inotropic score >75, CK-MB >130IU/L, serum creatinine >150 umol/L, and platelet count <100×109/L were identified as pre-ECMO prognosis factors of in-hospital mortality in the REMEMBER score [51]. In this setting, again lack of evidence calls for more powerful multicenter scoring system to accurately predict the prognosis in postcardiac surgery patients requiring ECMO for PC shock.

### *6.1.2 Contraindications*

In General, for patients in whom PC failure is felt to be reversible, all contraindications should be considered relative, except for uncontrollable bleeding not amenable for surgical correction, which is by far the only absolute contraindications for postcardiac surgery patients [49].

#### *6.1.2.1 Relative contraindications*


#### *6.1.3 Cannulation*

It was found that following PC low cardiac output, approximately 40% of ECMO cannulation occurs in the operating room and 60% in the ICU [52]. As these patients' chests are already opened via sternotomy or thoracotomy; central access is an additional modality to cannulate PC patients with ECMO centrifugal pump. However, it was found that peripheral cannulation is more common than central cannulation despite its easiness in terms of the already utilized access via right atrium and aorta, the presumed as well as the gathered evidence showed higher complications in terms of mortality, bleeding, infection, and compression in case of central cannulation in comparison to the peripheral access via the femoral or the axillary sites. In a retrospective multicenter study, Mariscalco et al. compared peripheral and central VA-ECMO in 781 patients with PCS at 19 cardiac surgery centers. Concluded that central cannulation was associated with greater in-hospital mortality than peripheral cannulation, pooled unadjusted risk ratio analysis of these patients showed that patients undergoing peripheral VA-ECMO had a lower in-hospital/30-day mortality than patients undergoing central cannulation, authors stated that results did not alter after cofounders' readjustment [53].

#### *6.1.3.1 Basic principles*

#### *6.1.3.1.1 Peripheral cannulation*

It is the most frequently used access due to less complication rate and it allows sternotomy closure. It is performed via the common femoral artery and vein just below the inguinal ligament and should be above the bifurcations [54]. Arterial cannula should be adequate to supply sufficient flow to meet the patient's needs, sizes larger than 19F cannulas may be considered only when higher flow is needed and is usually rare; keeping in mind the increased vascular complications, including limb ischemia with larger cannula [55]. If feasible, some opinions prefer to place each cannula in different legs as it is thought to reduce the vascular complications associated if both cannulas are placed in the same limb. Also, some experts prefer to insert the venous cannula into the right femoral vein as it is a more direct path to the IVC and right atrium. Nowadays, the Image-guided cannulation, particularly vascular ultrasound is the standard in percutaneous approach. Fluoroscopy can be useful if available. Vascular ultrasound should be started in the short axis and longitudinal views [56].

#### *6.1.3.1.2 Central cannulation*

Although peripheral access is linked to better survival and less complication, in some instances, especially with patients with peripheral vascular disease, the adoption of central cannulation is inevitable. Utilizing the same CPB cannula in the ascending aorta and the right atrium is the most common approach. Other methods have also been described to allow sternotomy closure via tunneling the cannulas through the skin below the sternum to allow the closure, although cardiac compression and kinking of the cannula have been described as complications of this method. Cannulation configuration and strategy can be summarized as follows (**Table 2**) [49].

#### **6.2 Management VA-ECMO**

Management of patients with VA ECMO for postcardiotomy shock is more complicated than for other indications, as surgical patients are usually sicker with many other comorbidities and an already injured heart. Arterial blood gases, lactates, mixed venous oxygen saturation (SvO2), and urine output are all indicators to follow and

**Advantages Disadvantages** Central (aortic\ atrial) • More efficient drainage via antegrade flow • Direct access via established surgical site with possibility of sternotomy closure. • Avoids harlequin syndrome • Opened chest\* • More bleeding • Re-sternotomy is mandatory to decannulate Peripheral Percutaneous femoral artery • Can be done Bedside • Avoids surgical incisions so less bleeding • Less sepsis • Can be switched to VAD implant easily • High limb ischemia complications • LV afterload due to retrograde flow • LV venting cannot be easily achieved • Not suitable for long-lasting support Open femoral artery • Appropriate cannulation sites via Direct visualization of femoral vessels • Less bleeding • Avoids sternotomy • limb ischemia complications • LV afterload due to retrograde flow • LV venting cannot be easily achieved • Not suitable for long-lasting support Pseudo-central Axillary\ Subclavian • Long lasting support • Easy patient mobilization • Avoidance of Harlequin • Time-consuming • Upper limb vascular complications Lower ECMO flow

*Venoarterial Extracorporeal Membrane Oxygenation in Cardiac Surgery DOI: http://dx.doi.org/10.5772/intechopen.106823*

**Table 2.** *Cannulation configuration and strategy summary.*

(North/South) Syndrome

*\*Closed chest is accessible; however, cardiac compression is likely with central approach.*

manage the ECMO patients. Close clinical follow-up using echocardiogram is also crucial to determine the overall cardiac function, right ventricular (RV) function in case RV is not supported, velocity time interval (VTI) are important parameters as well [49].

#### *6.2.1 Sternotomy wound management*

Despite the cannulation site, sternotomy wounds should always be closed to minimize bleeding and also to reduce infections. In case of peripheral cannulation, this can be easily achieved as cannulae are already apart from the wound, but central cannulation might add complexity to the closure. Some have proposed tunneling techniques to divert the cannula away from the wound although it has been shown that it might cause cardiac compression by the cannula themselves in case of subxiphoid tunneling, other tunneling techniques with less compression included externalization through the intercostal spaces, tunneling into the neck to the jugular area, or the anastomosis to prosthetic grafts, which is usually utilized in aortic surgery [57, 58].

## *6.2.2 Leg perfusion*

In case of femoral cannulation, many ways have been adopted to reduce ischemic and vascular complications such as adopting the open technique as possible, using a

smaller cannula, and using vascular graft instead of direct femoral cannulation, but most importantly using distal perfusion cannula to perfuse the cannulated leg. This cannula is then connected with a side way to the arterial cannula and its flow can be monitored using a sensor to ensure optimal leg perfusion. Moreover, continuous daily pulse monitoring should be ensured [49].

#### *6.2.3 Flow management*

Determining how much flow is best to achieve optimal peripheral perfusion with some heart ejection remains unclear. Some have argued that allowing the supported heart to eject is better than full support in terms that it prevents the blood stasis as well as the dilatation [59–61]; however, as mentioned earlier, PC patients are different as the heart is already damaged so allowing the heart to eject might add extra workload [62].

#### *6.2.4 Left ventricular distention*

Although infrequent, LV distention is one of the major issues facing the supported heart while on ECMO regardless of the site of the cannulation as retrograde ECMO flow adds more on the afterload, which can be hazardous for an already dysfunctional ventricle, which is usually the case in PC patients. Another important mechanism, it has been postulated that while on ECMO the aortic valve might exhibit a protracted closure due to the impedance of the forward flow, which causes blood stasis, blood pooling, LV wall tension, and LV workload even in the absence of poor myocardium. For that, several clinical studies have shown that IABP might be beneficial in eliminating the LV distention by restoring AV opening and reducing forward flow impendence [63]. However, in extreme cases of LV dysfunction, IABP might not be enough to alleviate the distention, in such cases more invasive methods should intervene such as direct cannulation of the LV through the apex, surgical or percutaneous cannulation of the pulmonary artery may be considered for indirect LV unloading as well. Trans-aortic devices such as impella and impella RP have also shown great benefit in this setting. Another approach including trans-septal approach surgically or percutaneously has been also used [64]. The true prevalence of LV distention and its clinical impact remains unproven, also the need for LV venting and whether its prophylactic implementation is useful is unknown.

### *6.2.5 Anticoagulation*

The most adopted practice for PC ECMO is to partially reverse with half dose protamine and then wait for 24-48hrs for full heparin administration after excluding major bleeding. As mediastinal collection can be one of the most associated complications after ECMO institution as ECMO itself can exacerbate coagulopathy, management should be directed toward a balance between bleeding management with product transfusion and medication in facing clot formation prevention in the circuit [65–67]. Unfractionated heparin remains an antithrombotic agent of choice for anticoagulation in case of PC ECMO as per the ELSO guidelines. Although monitoring has not been standardized yet, it is recommended that either ACT targeting a level of 180–200s or aPTT up to 50–80s is accepted [68, 69]. In any case of prolonged use of heparin, the possibility of HITT occurrence is likely, in such case direct thrombin inhibitors (DTI) can be used, such as bivalirudin should be used, as an alternative. However, extra caution should be kept given the very short half-life of bivalirudin so the likelihood of developing clots can be life-threatening [70].

*Venoarterial Extracorporeal Membrane Oxygenation in Cardiac Surgery DOI: http://dx.doi.org/10.5772/intechopen.106823*

#### *6.2.6 Intensive care monitoring*

Systematic clinical examination along with physiological and laboratory monitoring with close adjustment of ECMO setting should be implemented. Monitoring of all peripheral arterial saturation should be done for early detection of harlequin syndrome in case of uneven distribution of saturation. Recognizing early signs of infection and early start of empiric antibiotics is crucial to avoid the burden of septic shock occurrence [49].

Timely detection of brain injury is considered an important aspect to consider while in ICU monitoring, and it has been shown that EEG and near-infrared spectroscopy (NIRS) play an important diagnostic and prognostic role in the timely detection of acute brain injury [71, 72]. Confirming the diagnosis with CT is also encouraged despite the complexity of transportation while on ECMO. Nonetheless, transesophageal echocardiographic (TEE) is an important tool to assess the overall cardiac function, cannula positions, and right ventricular dynamics and to guide the suitability of weaning. In our institute, we do not use the swan-Ganz catheter, but it might be useful in few cases to guide management.

#### **6.3 Weaning from VA-ECMO**

The consideration for weaning EMCO exists when absence of specific decompensation factors like supraventricular arrhythmia or severe septic shock could be managed. Recovery of pulsatile arterial waveform for at least 24 h, the patient should be hemodynamically stable, with mean arterial pressure more than 60 mmHg in the absence or reducing doses of inotropes and/or vasopressors [73]. Finally, pulmonary function should be adequate with PaO2/FiO2 more than 200 mmHg [74]. It is unlikely to start weaning trial in the first 72 hours of initiation [75]. Weaning trial starts usually with reducing ECMO blood flow, which eventually causes right ventricular preload increase and LV afterload reduction, therefore myocardial function could be assessed [76]. Patients should have a pulsatile flow with a minimum ECMO flow of 1–1.5 L/min [77]. If mean blood pressure is reduced below 60 mmHg the trial should be abandoned. The echocardiographic criteria favoring successful weaning include LVEF of more than 20–25 %. Patients successfully weaned had aortic velocity time integral above 10 cm, and TDSa of at least 6 cm/s at minimal ECMO flow support [75].

#### **6.4 Complications and early and long-term outcomes**

Despite the exponential increase in ECMO use, PC ECMO is still in the beginning although enormous improvement in ECMO cannulation and management; successful weaning from PC ECMO varies greatly among the published series from 30% to 70%, and the survival to discharge is even much lower [78]. In the most recent ELSO registry, survival for discharge for overall ECMO cases for cardiogenic shock is 50 % [79]. So far, no RCTs have been deployed to illustrate the real survival benefit or even the quality of life in the long term. Based on the most recent report from the ELSO registry, there has been a gradual decline in the survival after PC-ECMO, as low as 15% survival in some analyses [80]. Overall, bleeding is the most frequent complication occurring in up to 90% of patients as described in some series. Other anticoagulationrelated complications also can happen such as heparin-induced thrombocytopenia, intracranial bleeding, and hemolysis. Other complications include high inflammatory markers manifested as inflammatory response that is like that in systemic inflammatory response syndrome, causing an increased risk of thrombosis, infections, sepsis, and end-organ damage thus worsening patient outcomes, steroids have been used as a prophylactic agent and shown to reduce it; However, it did not affect the overall mortality [81, 82]. The prevalence of infection during ECMO is 10% to 12%, with Staphylococcus aureus, Candida, Enterobacteriaceae, and Pseudomonas aeruginosa being the most common bloodstream infective organisms. ECMO site infections are common as well, so special care for ECMO wounds and early recognition are needed. Regular cultures could be done if an infection is suspected, especially with prolonged ECMO use [83]. Limb ischemia is a major vascular complication associated with ECMO, especially the peripherally cannulated; limbs should be frequently monitored by duplex ultrasound also ECMO flow through the distal perfusion cannula should be maintained [84].

#### *6.4.1 Predicting mortality and quality of life*

Predictors for PC ECMO outcome have been studied in many papers, as mentioned earlier scoring systems were also deficient and limited not to authentically predicting outcomes after cardiac surgery. One of the pre-ECMO factors, ECPR was found to have a strong negative predictor of survival in several series [85, 86]. Others have demonstrated that blood lactate level prior to ECMO and up to 48 hours after ECMO initiation is strong predictor value for survival [87]. Early initiation of ECMO has been shown to result in higher survival rates and decrease the dosage of vasoactive drugs by increasing cardiac output and rapidly decreasing arterial lactate levels after cardiovascular surgery [88].

The CESAR trial showed a significant increase in survival without severe disability when ECMO was used instead of conventional ventilation [89].

It is been demonstrated in many series that renal and liver failure, respiratory failure, and the duration of ECMO support are also strong negative predictor factors to affect ECMO outcome [90, 91]. Despite the advances in its use, the ethical and economic implications of ECMO are enormous for both patients and the health system. Psychological distress and memory problems were described in some analyses for post-ECMO survivors. Unfortunately, the long outcome of VA-ECMO survivors remains under investigation. Most studies concentrate on treatment outcomes and survival-to-hospital discharge. The outcomes of 138 patients treated with ECMO for cardiogenic shock are related to acute myocardial infarction. Burrell et al. determined that good long-term survival could be achieved following ECMO, observing 79% survival at 12 months. Survival data are available for only 66% of patients at 24 months [92]. Ørbo et al. identified 30 (41%) of 74 ECMO survivors in Norway and surveyed 23 survivors, with 40% of respondents reporting some degree of restriction in everyday activities and depression in 35% of cases [93]. According to ELSO's data registry, CS was the most common cardiac indication in adult patients with over 2000 runs and with successful ECMO explanations in 56% of cases and an overall 42% survival-to-discharge in 2016 in participating centers. Although not evidenced by powered data, overall long-term outcomes for survivors can be promising especially with improved indications and guidelines.

### **7. Conclusions**

Complex cardiac surgical procedures in high-risk patients may require extending the medical support to a mechanical one. VA-ECMO could offer additional advantages over CBP to support the circulation during CABG surgery in patients with complex

*Venoarterial Extracorporeal Membrane Oxygenation in Cardiac Surgery DOI: http://dx.doi.org/10.5772/intechopen.106823*

coronary anatomy and unstable hemodynamics, with added hemodynamic and economic value.
