Role of Percutaneous Cardiopulmonary Support (PCPS) in Patients with Unstable Hemodynamics During the Peri-Coronary-Intervention Period

Coronary artery disease remains the principal cause of death in developed countries, with acute coronary syndrome (ACS) contributing greatly to cardiovascular morbidity and mortality. Optimal revascularization therapy in these patients is of major importance and remains the basis for favorable outcomes. The intervention-based strategy plays a major role in the management of acute-phase patients because of the cardiovascular benefits demonstrated in numerous studies. In serious cases, ACS can be encountered with life threatening episodes such as cardiogenic shock or cardiac arrest in the peri-intervention period. Compared with the intra-aortic balloon pump (IABP), percutaneous cardiopulmonary support (PCPS) can maintain hemodynamic stability in the abscence of an intrinsic cardiac rhythm or effective cardiac output and can provide excellent mechanical circulatory support in these urgent or emergent situations because it can be implanted and stabilized promptly in many different hospital locations, including the cardiac catheterization laboratory, and if necessary, in the emergency room or intensive care unit. Some reports use PCPS to rescue ACS patients with cardiopulmonary failure (Sung, 2006; Grambow, 1994). Herein, we present our experience with the use of PCPS for initial stabilization in patients with unstable hemodynamics before or after coronary intervention and discuss some of the pertinent issues related to this topic.


Introduction
Coronary artery disease remains the principal cause of death in developed countries, with acute coronary syndrome (ACS) contributing greatly to cardiovascular morbidity and mortality. Optimal revascularization therapy in these patients is of major importance and remains the basis for favorable outcomes. The intervention-based strategy plays a major role in the management of acute-phase patients because of the cardiovascular benefits demonstrated in numerous studies. In serious cases, ACS can be encountered with life threatening episodes such as cardiogenic shock or cardiac arrest in the peri-intervention period. Compared with the intra-aortic balloon pump (IABP), percutaneous cardiopulmonary support (PCPS) can maintain hemodynamic stability in the abscence of an intrinsic cardiac rhythm or effective cardiac output and can provide excellent mechanical circulatory support in these urgent or emergent situations because it can be implanted and stabilized promptly in many different hospital locations, including the cardiac catheterization laboratory, and if necessary, in the emergency room or intensive care unit. Some reports use PCPS to rescue ACS patients with cardiopulmonary failure (Sung, 2006;Grambow, 1994). Herein, we present our experience with the use of PCPS for initial stabilization in patients with unstable hemodynamics before or after coronary intervention and discuss some of the pertinent issues related to this topic.

Indications for ECMO
ECMO is indicated for all the reversible pulmonary or cardiac failures (or both) refractory to conventional therapy. ECMO is usually used for initial stabilization and temporary support while awaiting recovery of organs. However, physicians should keep in mind that the results of ECMO support are consistently related to the indication for institution of such therapy. With respect to cardiac failure, ECMO has been used successfully to resuscitate patients with cardiac arrest or cardiogenic shock due to acute coronary syndrome, post-cardiotomy cardiac failure, transplant rejection, intractable dysrhythmia, ruptured coronary artery graft, pericardial tamponade, cardiac trauma, sepsis, and myocarditis. ECMO has also been effective in patients with secondary cardiac failure due to respiratory insufficiency, hypothermic arrest from cold-water submersion, and drug overdose. The indications for ECMO in cardiopulmonary resuscitation are summarized in Table 1. In making the decision to use ECMO, several considerations must be weighed. Most importantly, physicians must consider the possibility of cardiac recovery. If recovery is not expected, consideration must be given to patient eligibility for heart transplantation, other mechanical assist devices to bridge the patient to transplantation, or a definitive mechanical assist device to be inserted as destination therapy. Moreover, patient selection is the most important prognostic indicator. Although some relative contraindications have been proposed, there are few generally accepted absolute contraindications to ECMO. Unwitnessed cardiac arrest, aortic dissection, and terminal illness are considered to be the most important contraindications. The mortality rate after an unwitnessed cardiac arrest approaches 100% and has not changed with the use of ECMO. Early and effective resuscitation can allow for neurologically-intact survivors, but this is less likely to occur in an unwitnessed arrest. Aortic insufficiency is another relative contraindication, particularly if severe. In such situations, aortic valve replacement should be considered. Lesser degrees of aortic insufficiency may be managed with a left ventricular vent. Failure to insert a vent www.intechopen.com Role of Percutaneous Cardiopulmonary Support (PCPS) in Patients with Unstable Hemodynamics During the Peri-Coronary-Intervention Period 133 leads to ventricular distension, compromised subendocardial blood flow with an impact on recovery, and the ability to wean from ECMO. Renal failure, hepatic failure, and significant neurologic disease are relative contraindications to ECMO, depending on the existence of other therapeutic options and the degree of dysfunction (Meurs, 2005). With respect to respiratory failure, the most common indication is due to adult respiratory distress syndrome (ARDS), pneumonia, trauma, or primary graft failure following lung transplantation. ECMO is also used for neonatal and pediatric respiratory support. The use of ECMO in premature neonates is the mainstay of treatment for immature lungs and insufficient surfactant. More recently, indications for ECMO have been expanded to pro-procedural supports, such as major airway intervention/surgery and cardiac intervention, including high-risk coronary angioplasty or arrhythmia ablation (Table 1). In a small and retrospective study, Grambow et al. proposed that patients requiring coronary revascularization by angioplasty who have severe left ventricular dysfunction or target vessels supplying greater than 50% of the myocardium are candidates for elective ECMO (Grambow, 1994

Mode of ECMO
ECMO can be inserted in a veno-venous (VV) mode, which provides oxygenation and CO 2 removal, and thus is used for pulmonary failure not responding to mechanical ventilation, or can be used in a veno-arterial (VA) mode providing sufficient end-organ perfusion and gas exchange. VA ECMO can be instituted peripherally or centrally. VV ECMO refers to blood being drained from the venous system and returned to the venous system. This mode only provides respiratory support and is achieved by peripheral cannulation, usually via the femoral or jugular veins. VA ECMO refers to blood being drained from the venous system and returned to the arterial system. The VA ECMO mode provides cardiac and respiratory support, and can be divided into two groups according to vascular access peripherally or centrally. When using 'central' cannulation, in which the ascending aorta and right atrium are directly cannulated, this is referred to as 'central' ECMO. Central ECMO allows better flow because of better drainage from the right atrium with a bigger cannula and is usually required in patients with a larger body surface area (> 2.0 m 2 ). The other advantage of central ECMO is that the flow directly from the outflow cannula into the aorta provides antegrade flow to the arch vessels, coronary arteries, and the rest of the body (Fig. 1). However, central ECMO also has its drawbacks. Insertion of central ECMO requires a median sternotomy to allow the cannulation. This is a time-consuming procedure and increases the risk of bleeding from the sternum and also of infection. In contrast to central ECMO, it can be achieved by peripheral vascular access and is called 'peripheral' ECMO or percutaneous cardiopulmonary support (PCPS), in which insertion is more prompt via a bed-side procedure than central ECMO because there is no need for a sternotomy and the risk of bleeding is decreased. However, the retrograde aortic flow provided by peripheral ECMO from the femoral artery leads to admixing in the arch (two circulation syndrome). For this reason, monitoring of oxygenation from a right radial arterial line would be prudent. Mechanical ventilation must be continued d u r i n g E C M O s u p p o r t t o m a i n t a i n o x y g e n saturation of blood ejected from the left ventricle to at least above 90%. In patients on PCPS, single access cannulae may not be sufficient to achieve satisfactory perfusion and gas exchange. In these patients a second access cannula is inserted via the right internal jugular vein ,so sufficient flow can be achieved.

Maintenance and Weaning of ECMO
The basic function of VA ECMO in supplying mechanical circulatory support is to drain blood from the venous circulation, oxygenate the blood, and then return the blood to the arterial circulation at physiologic perfusion pressures. Although ECMO efficiently unload the right ventricle, ECMO is not efficient in unloading the left ventricle. Indeed, ECMO increases left ventricular afterload, thus negatively affecting myocardial recovery, even though left ventricular preload is significantly reduced by the diminished return from the lungs. For this reason, attempts to improve left ventricular contractility, reducing left ventricular distension, left ventricular afterload, and clot formation are of utmost importance. Such measures should include inotropic support and may include intra-aortic balloon pumping (IABP), which can decrease left ventricular afterload and therefore reducing wall stress, especially during the critical initial period. Other options include inserting a left ventricular vent to empty the ventricle connecting this line into the venous line of the ECMO circuit, to place an additional drainage catheter in the left atrium that can be placed via the percutaneous route (Aiyagari, 2006), or to perform an atrial septostomy to decompress the left heart (Koenig, 1993). ECMO flow can be volume-dependent and will drop with hypovolemia, cannula malposition, pneumothorax, and pericardial tamponade, and usually manifests as 'kicking' or 'chatter' of the venous tubing as well as a drop in output. Management includes a volume challenge, and exclusion of intra-abdominal distension or compartment syndrome, cardiac tamponade, or pneumothorax. If a volume challenge does not work, a slight reduction in flow may be helpful or there may be a need to insert another venous cannula. Further, centrifugal pumps in contrast to roller pumps are afterload-dependent, and therefore hypertension is another variable that can reduce flow and should be avoided. From a respiratory point of view, hypoxia is treated by increasing the flow rate and FiO 2 of the ECMO circuit, not by altering the FiO 2 and PEEP on the ventilator. Attempts to support and rest the lung should be made to wean the FiO 2 on the ventilator and maintain a PEEP level of 8-15 cmH 2 O. A protective lung ventilation strategy with low plateau pressures and low tidal volumes should be attempted as well as low respiratory rates, unless attempting to wean off the ECMO circuit. PCO 2 control should be via the ECMO fresh gas flow to the oxygenator, not by altering the respiratory rate on the ventilator. From a cardiac point of view, the aim should be to minimize the use of inotropes, and thus rest the heart. However, a low-dose inotrope infusion is often maintained to ensure some contractility and adequate emptying of the left ventricle. It is also important to ensure the patient is not hypovolemic. There are no standardized methods or techniques by which to wean from ECMO. With respect to VV ECMO, weaning and discontinuation are considered when there is evidence of substantial improvement in the underlying disease process and cardiopulmonary function based on radiographs, laboratory parameters, hemodynamic parameters, and gas exchange. While monitoring these parameters, flow through the circuit can gradually be reduced. When a reduction in flow of 75% can be achieved without changes in cardiopulmonary status, membrane gas flow can be stopped. If the patient tolerates those challenges for 12 hours with adequate gas exchange and hemodynamic parameters, VV ECMO can be removed (Gravlee, 2008). With respect to VA ECMO, weaning and discontinuation are considered when there is evidence of cardiac recovery, including increased blood pressure, return of pulsatility or increased pulsatility on the arterial pressure waveform, and falling central venous and/or pulmonary pressures. Weaning is performed under echocardiographic guidance with regular arterial blood gas and lactate measurements. ECMO flow rates are slowly weaned, while inotropes and ventilator support are adjusted after echocardiography confirms adequate ventricular filling and ejection. Blood gases and serum lactate levels are then assayed to confirm adequate gas exchange and oxygen delivery, respectively. Until a reduction in flow of 75% can be achieved or a flow rate is below 1 -1.5 liter/min without changes in cardiopulmonary status over 12 -24 hours, weaning can proceed. If the patient tolerates those challenges for 12 hours with adequate parameters, VA ECMO can be removed.

Complications of ECMO
Bleeding and hemolysis, multi-organ failure, infection, and equipment failure are the common complications that contribute to morbidity and mortality associated with ECMO in adults. The complications that have been reported to the Extracorporeal Life Support Organization (ELSO) Registry in adult patients treated with ECMO are listed in Table 2, and data on complications and events reported in the ELSO Registry are shown in Table 3 (Meurs, 2005). The most common complication associated with the use of ECMO is bleeding from cannulation sites, with typical rates ranging 4-14% (Kurusz, 2002). Excessive bleeding is caused by multiple factors. Hepatic congestion and failure, malnutrition, multiple cannulation sites, low-dose anticoagulation with heparin, decreased platelet function, activation of the coagulation cascade secondary to hemolysis, and hyperfibrinolysis from contact with prosthetic surfaces all may contribute to increased bleeding. The use of h e p a r i n -c o a t e d c i r c u i t s i n a d d i t i o n t o t h e antifibrolytics and vitamin K therapies has contributed to lower bleeding rates. Vena cava tears or rupture with retroperitoneal bleeding, and arterial or aortic injury including dissection and perforation that required surgical exploration and reconstruction have occurred. Other non-cannula-related bleeding complications include pulmonary and gastrointestinal hemorrhage. Ischemic injury to the brain, kidneys, liver, and other end organs has been attributed to prolonged resuscitation and as well as inadequate pump flow. Recovery from a moderate hypoxic-ischemic insult is possible for most organs; however, neurologic recovery is frequently limited. Emergency ECMO and early coronary perfusion using percutaneous transluminal coronary angioplasty (PTCA) can result in good myocardial recovery. By using mild hypothermia in conjunction with ECMO during PTCA, a 2-to 5-fold increase in the rate of good neurologic recovery was achieved in a series reported by Nagao et al (Nagao, 2000). Overall infection rates in temporary mechanical circulatory support have been reported to be as high as 30-40% (Patel, 2003). Immobilization, poor nutritional status, and indwelling www.intechopen.com

Role of Percutaneous Cardiopulmonary Support (PCPS) in Patients
with Unstable Hemodynamics During the Peri-Coronary-Intervention Period 137 catheters and tubes are all likely to contribute to the high incidence of infection. Although infrequent now, severe lower limb ischemia leading to amputation has occurred with femoral artery cannulation. Perfusion in the cannulated leg should be monitored closely for evidence of ischemia. Placement of a smaller, additional femoral artery cannula to perfuse the lower limb distal to the ECMO cannula has reduced the risk of this complication. Limb complications have been reported to occur in as many as 25% of patients prior to the use of the distal cannulation technique (Patel, 2003;Schwarz, 2003  In addition, intracardiac thrombus may form within a poorly contracting, nonejecting left ventricle or atrium because little blood reaches the left atrium with good right atrial drainage (Cohn, 2008). Because intracardiac thrombus may induce systemic embolism, thus serial echocardiogram should be followed up for early detection and ECMO patients do require low level heparinization to prevent this complication.

Outcomes of ECMO
The results of ECMO support are consistently related to the indication for institution of such therapy.
With respect to VV ECMO, Bartlett et al at the University of Michigan were reviewed in 1995 and reported that VV ECMO for respiratory failure provided survival to discharge in 88% of 586 cases of respiratory failure in neonates, 70% for 132 cases of respiratory failure in children and 56% for 146 cases of respiratory failure in adults (Bartlett, 1997). In 2008, the ELSO revealed similar results; specially 972 cases of ECMO support for respiratory failure were reported, of which 53% survived to discharge in the adult population (>18 years).
Unfortunately, good quality randomized controlled trials of ECMO outcomes in the adult population are lacking. The incomplete CESAR (Conventional Ventilation or ECMO for Severe Adult Respiratory Failure) was a national randomized controlled trial in the United Kingdom funded by the National Health Service and the Health Technology Assessment Agency; preliminary results were recently released at the 37th Society of Critical Care Medicine Congress in Honolulu in February 2008. The hypothesis was that ECMO will improve survival without severe disability by 6 months post-randomization for patients with severe, but potentially reversible respiratory failure. Severe disability was defined as confined to bed and unable to dress or wash oneself. Suitable patients can be entered from any of the 96 participating hospitals and were randomized to either transfer to Glenfiele hospital for consideration of ECMO or continued conventional management. The conventional group underwent standard clinical practice in the UK. Recruitment was conducted from July 2001 to August 2006. Of the 90 patients assigned to receive ECMO, 22 did not receive ECMO, most often because they improved without it. Patient characteristics were well matched between groups. Of the patients randomly assigned to receive ECMO, 57 of 90 met the primary endpoint of survival or absence of severe disability at 6 months compared with 41 of 87 patients in the conventional ventilation group. This translated to a relative risk in favor of the ECMO group of 0.69 (95% confidence interval, 0.05-0.97; P = 0.03). A significant number of patients had failure of more than 3 organs in both groups (28 in the ECMO group and 27 in the conventional group). The benefit of ECMO was evident, regardless of age, duration of high-pressure ventilation, primary diagnosis at trial entry, and number of organs failing (Peek, 2006;Hitt, 2008). With respect to VA ECMO, Bartlett et al at the University of Michigan were reviewed in 1995. Their experience with cardiac failure is somewhat smaller with a 33% survival rate in 31 adult patients and 48% survival in 105 pediatric patients (Bartlett, 1997). In 2008, ELSO revealed similar results that 474 patients were supported for cardiac failure of which only 33% survived to discharge. In addition, numerous studies have reported various results according to the disease diagnosis. The data reported to the ELSO registry has shown that myocarditis is associated with a better survival than other diagnosis, with almost three-fifths being successfully weaned from ECMO. Similarly, Acker reported that twenty seven (73.0%) of a group of 37 patients could be weaned from ECMO and 26 patients (70.3%) were discharged from the hospital (Acker, 2001). In this subgroup of patients, ECMO may be used as a bridge to recovery, bridge, or transplantation. Cardiac arrest and shock are the most common indications for ECMO in adults and the benefits of ECMO for cardiac arrest in adults were shown recently (Chen, 2008). Early implementation of ECMO can improve the survival rate to 30-40%, as described in multiple published series and the ELSO registry (Sung, 2006;Meurs, 2005). For acute myocardial infarction, cardiogenic shock is the leading cause of death in hospitalized adults. In these patients, ECMO alone has a limited effect on outcome, but when combined with emergency coronary revascularization by angioplasty, it may improve survival. Also, the concomitant use of an intra-aortic balloon pump with ECMO improves survival in patients with cardiogenic shock. The role of ECMO to provide post-cardiotomy support with severe cardiopulmonary dysfunction after cardiac surgery is well established. Mechanical circulatory support may be required in the post-operative period, either due to the inability to separate from cardiopulmonary bypass, or a progressive low cardiac output syndrome due to a number of factors, such as ventricular dysfunction, pulmonary hypertension, or intractable arrhythmias. The incidence of postcardiotomy myocardial failure is 2-6% in adult cardiac surgical patients. Rousou et al. reported that 56% of a group of 16 patients survived ongoing post-operative ventricular fibrillation or electromechanical dissociation when ECMO was employed (Rousou, 1994). Similarly, 77% of a group of 13 patients described by Kawahito et al. successfully weaned off ECMO after circulatory collapse from ventricular fibrillation or electromechanical dissociation that was refractory to inotropic agents, IABP, and cardiopulmonary resuscitation (Kawahito, 1994). ECMO has been utilized in the cardiac catheterization laboratory for resuscitation during sudden cardiovascular collapse to provide temporary support for interventional procedures. In a small, retrospective study by Grambow et al., rapid initiation of ECMO (within 20 minutes) resulted in the rescue of patients in cardiogenic shock, but all patients who experienced cardiac arrest died despite ECMO and further interventions. Of those in shocks who were initially salvaged with ECMO, approximately 50% survived for 24 hours but only 25% to hospital discharge. Mortality after 24 hours was attributed to sepsis, multi-organ failure, or congestive heart failure (Grambow, 1994 after distal femoral artery puncture to avoid ischemia of the distal lower leg. The PCPS was then started using the Capiox emergency bypass system (EBS; Terumo Inc, Tokyo, Japan), which comprises a centrifugal pump, a polypropylene hollow fiber membrane oxygenator, and a heparin-coated circuit (Fig 2). The most important benefit of this system is its autopriming, which requires less than five minutes to prime the circuit before use and does not require specially trained personnel. Once the femoral cannulation was established, the PCPS system could be run to stabilize the patient. No hypothermia was applied. If possible, we tried to maintain the hematocrit above 35%, the platelet count above 100,000 per μl and the activated clotting time (ACT) at 150 to 200 seconds to minimize complications of bleeding and thromboembolic complications. After weaning the patients from the PCPS, we surgically removed the femoral artery and venous cannulae and repaired the cannulation sites to avoid ischemic limb complications.

Results
The clinical characteristics of the patients are listed in  group. There was no statistical difference in weaning from PCPS and survival discharge between the groups. We divided 79 patients according to the timing of insertion, as follows: group A was instituted before coronary intervention; group B was instituted during intervention; and group C was instituted after intervention. In group A, 25 (65.8%) of 38 patients were weaned from the PCPS and 18 (47.4%) patients survived to discharge; In group B, 8 (61.5%) of 13 patients were weaned from the PCPS and all patients survived to discharge; In group C, 14 (50%) of 28 patients were weaned from the PCPS and 10 (35.7%) patients survived to discharge. There was no statistical difference in weaning from PCPS and survival discharge between the groups.

Conclusion
Our results show that PCPS may be a feasible option as a procedural support during coronary intervention. Moreover, we suggest that rescue PCPS combined with early coronary revascularization by angioplasty may improve survival when acute coronary syndrome results in circulatory collapse, and PCPS should be considered in patients with no definite contraindications. However, further evaluation and investigation are needed because the limitations of the study included the absence of a control group for comparison and selection bias.

Acknowledgment
The author would like to acknowledge Mrs. Nam Kyung Choi for her dedicated assistance of manuscript editing and English revision.