| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2002;73:538-545
© 2002 The Society of Thoracic Surgeons
a Department of Surgery, National Taiwan University Hospital, Taipei, Taiwan
b Department of Pediatrics, Taipei Veteran General Hospital, Taipei, Taiwan
Accepted for publication September 22, 2001.
* Address reprint requests to Dr Yih-Sharng Chen, Department of Surgery, National Taiwan University Hospital, 7, Chung-Shan S. Road, Taipei, Taiwan 100
e-mail: wenje{at}ha.mc.ntu.edu.tw
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
Methods. We retrospectively reviewed the medical records of ECMO patients.
Results. From August 1994 to May 2000, 76 adult patients (48 men, 28 women; mean age, 56.8 ± 15.9 years) received ECMO support for postcardiotomy cardiogenic shock at the National Taiwan University Hospital. The mean ECMO blood flow was 2.53 ± 0.84 L/min. The cardiac operations included coronary artery bypass grafting (n = 37), coronary artery bypass grafting and valvular operation (n = 6), valvular operation alone (n = 14), heart transplantation (n = 12), correction of congenital heart defects (n = 3), implantation of a left ventricular assist device (n = 2), and aortic operations (n = 2). Fifty-four patients received ECMO support after intraaortic balloon pumping, but 22 patients directly received ECMO support. Two patients were bridged to heart transplantation and two bridged to ventricular assist devices. Thirty patients died on ECMO support. The causes of mortality included brain death (n = 3), refractory arrhythmia (n = 2), near motionless heart (n = 2), acute graft rejection (n = 1), primary graft failure (n = 1), uncontrolled bleeding (n = 5), and multiple organ failure (n = 16). Twenty-two patients were weaned off ECMO support but presented intrahospital mortality. The cause of mortality included brain death (n = 1), sudden death (n = 4), and multiple organ failure (n = 17). Twenty patients were weaned off ECMO support and survived to hospital discharge. During the follow-up of 33 ± 22 months, all were in New York Heart Association functional status I or II except two cases of late deaths. Among the ECMO-weaned patients, "dialysis for acute renal failure" was a significant factor in reducing the chance of survival.
Conclusions. The ECMO provided a satisfactory partial cardiopulmonary support to patients with postcardiotomy cardiogenic shock, and allowed time for clinicians to assess the patients and make appropriate decisions.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
|
Patients and methods |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
16 years old) underwent open heart operations at the hospital. The operations included coronary bypass grafting (CABG) (n = 1,318), heart valvular operations (n = 1,194), and other operations (n = 400, including correction of congenital heart disease, heart transplantation, and aortic operations). Beating heart CABG began at the hospital in February 2000, but the number was negligible during the period of data collection. The Intraaortic balloon pump (IABP) was the first choice and ECMO support was the second, whenever MCS was required for PCS. The ECMO system (catalog no. CB2505, Medtronic Inc, Anaheim, CA) consisted of a centrifugal pump and a hollow fiber microporous membrane oxygenator with an integrated heat exchanger. The entire ECMO circuit had a heparin-bound carmeda-bioactive surface. For simplicity the ECMO circuit was primed only with normal saline containing heparin at a concentration of 2 U/mL. An additional 5,000 U of heparin were injected intravenously during cannulation. Heparin was not again used in the first 24 hours of ECMO support. For patients bridged to ECMO support directly from cardiopulmonary bypass (CPB), half the regular dose of protamine was used, and heparin was not added to the priming solution. After 24 hours of ECMO support, heparin infusion was started to keep the activated clotting time in the range of 160 to 180 seconds, depending on clinical judgment for the risk of bleeding. The dilution anemia occurring after the patient was connected to the ECMO was corrected by packed red blood cell transfusion. The hematocrit was maintained at 30% to 35%. Although lower hematocrit reduced blood oxygen-carrying capacity, a higher hematocrit increased the risk of clot formation in the ECMO. Platelets were transfused when the platelet count was less than 50 x 103/mm3, unless a bleeding complication was present.
The femoral route was preferred to the open sternotomy route for ECMO support because the presence of an open sternotomy wound increased the risk of bleeding and infection, and made nursing care more difficult. The centrifugal pump was typically set at a speed of 2,000 revolutions per minute. The blood flow would be more than 2 L/min. A pulse oximeter was used to monitor the patient’s arterial blood oxygen saturation. MX-2 tri-optic measurement cell (Medtronic) was attached to the pre- and postoxygenator circuit to continuously monitor blood oxygen saturation and hematocrit. Continuous monitoring by the pulse oximeter and the MX-2 monitor made frequent blood gas examination unnecessary; therefore, blood gas was checked once a day. Hematology and blood biochemistry were checked once a day and when clinically indicated. Inotropic agents infusion was only slowly tapered after initiation of ECMO support to prevent left heart distention, because ECMO increased the afterload of the left ventricle. Arterial pulse pressure was measured to monitor the function of the left heart and was used as a guide to taper inotropic agents infusion. A flattened arterial pressure wave indicated left heart drainage.
Sedation by infusion of midazolam and fentanyl was routinely used. Diuretics maintained fluid balance. A hemofilter (FH66, Gambro, Hechingen, Germany) was connected between the ECMO arterial and venous circuits for hemofiltration, if dialysis was required.
Neither microporous membrane oxygenator nor centrifugal pump was intended for prolonged use. The indications of ECMO replacement included severe plasma leakage from the oxygenator, deterioration of the oxygenator gas exchange function, decreased ECMO blood flow, or hemoglobinuria. A policy of replacing the entire ECMO circuits rather than the centrifugal pump or the oxygenator alone was followed for safety and simplicity.
Weaning off ECMO support was usually not attempted in the first 48 hours. The criteria for ECMO weaning included mixed venous oxygen saturation (SvO2) greater than or equal to 70%, stable hemodynamics and inotropic agents score less than or equal to 10 (see note on Table 1), and echocardiographic determination of the absence of tamponade, the absence of left heart distention, and a left ventricular ejection fraction of greater than or equal to 35%. The ECMO blood flow was slowed to 0.5 L/min for a while and the vital signs was observed. If hemodynamics remained stable, ECMO was removed at bedside under intravenous anesthesia, and the vessels were primarily repaired. When the ECMO was being weaned off, inotropic agents infusion and ventilator setting were increased as necessary.
|
Data were analyzed by nonparametric methods. A significant difference was defined as a p value less than 0.05. The variables of groups of patients with different outcomes were compared using Fisher’s exact test for categorical variables and Kruskal-Wallis test for continuous variables. Where the groups of patients differed, every pair of groups were further compared; again Fisher’s exact test for categorical and Mann-Whitney U test for continuous variables. The statistical software was SAS 8.0 (SAS Institute Inc, Cary, NC)
|
Results |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
30 kg) (n = 2), repair of dissecting aortic aneurysm (n = 1), a modified Blalock-Taussig shunt in a patient with single ventricle (n = 1), and severe femoral artery arteriosclerosis (n = 1). The ECMO cannulation was through the right atrium–aorta route (n = 15, 19.7%), through the femoral vessels by the percutaneous Seldinger technique (n = 17, 22.3%), or through the femoral vessels by the open method (n = 44, 57.9%). Although we favored the femoral venoarterial route, 15 patients received ECMO support through the open sternotomy wound. The femoral route was not selected, because (1) circulatory collapse occurred when the sternotomy wound was open, and emergent ECMO support was directly set up through the right atrium–aorta route (n = 6); (2) small body weight patients with small femoral vessels (n = 2); (3) femoral artery arteriosclerosis was evident (n = 1); (4) patients could not be weaned off CPB, and were directly bridged to ECMO through the existing right atrium and aorta cannula (n = 3); (5) the renal graft on one side and IABP on the other side precluded the ECMO of the femoral route (n = 1); and (6) left heart decompression by the left atrial drain was required (n = 2). Eight of these 15 sternotomy ECMO patients directly received ECMO support without the prior IABP support.
Outcome of ECMO support
The patients were divided into four groups depending on their outcomes.
Bridge to heart transplantation (n = 2) and ventricular assist devices (n = 2)
One heart recipient was put on the ECMO support directly from CPB because of primary graft failure; the patient was fortunate to get another donor heart 8 hours later. However, he died of a stroke 3 months after the heart retransplantation. One patient underwent emergent CABG for myocardial infarction complicated by cardiogenic shock, and needed ECMO support for failure to be weaned from CPB. After 31 hours on ECMO support, he underwent successful heart transplantation, but he died suddenly 21 months later. Another 2 patients used ECMO as a bridge to VAD. One patient underwent tricuspid valve replacement for Ebstein anomaly, but could not be weaned off CPB and was put on ECMO support. Finally, the ECMO was shifted to Thoratec right VAD after 117 hours of ECMO support. The patient underwent heart transplantation after 47 days of VAD support and was well at 4-year follow-up. The other patient received open cardiac massage for persistent ventricular tachycardia/fibrillation on the first postoperative day after CABG, and he was finally put on ECMO support with drainage from both atria. The ECMO shifted to centrifugal pump left VAD after 130 hours of support. The patient died of sepsis and multiple organ failure after he was on left VAD for another 7 days.
Death on the ECMO support (n = 30)
Thirty patients depended on ECMO support until their deaths. Brain death was noted in 3 CABG patients. Their ages were 56, 81, and 70 years, and their preoperative left ventricular ejection fraction were 70%, 48%, and 42%. Their failure to be weaned from CPB had no obvious causes. Intraaortic balloon pump followed by femoral ECMO was required to wean the patients from CPB, but no resuscitation was performed during the operations. The patients did not wake after the operations, and brain death was noted on the first postoperative day. The ECMO supports were terminated at 31, 44, and 46 hours and the patients died. Although a head computed tomographic scan was not performed, intraoperative stroke was assumed to be the cause of brain death. One patient had a refractory arrhythmia immediately after CABG operation. The patient died of the refractory ventricular arrhythmia 4 hours later, despite of the ECMO support. One patient had been stable on the femoral ECMO for 187 hours but died of ventricular fibrillation, possibly from a new myocardial infarction. Two patients had PCS with almost motionless hearts and died of circulatory shock even with ECMO support. One patient underwent heart retransplantation for chronic graft rejection; however, he required ECMO support for postoperative primary graft failure. The graft function recovered gradually with ECMO support. However, 2 days later, while the patient remained on the ECMO support, the electrocardiographic tracing revealed that the QRS wave widened and its amplitude decreased. The heart function deteriorated rapidly, and electrocardiographic tracing indicated cardiac standstill a few hours later. Necropsy revealed complement and immunoglobulin G deposition on the vascular endothelium, and the final diagnosis was acute humoral rejection of the graft heart.
Five patients died of uncontrolled bleeding soon after the operation, while on ECMO support. Their ECMO durations were 3, 7, 12, 23, and 24 hours. One patient had primary graft failure and could not been weaned off CPB after heart transplantation. She was put on ECMO support for 548 hours because a second donor heart was not available for heart retransplantation, and the ECMO was finally terminated at the family’s request. The remaining 16 patients could not be weaned off the ECMO support, and eventually died of multiple organ failure while still on ECMO support.
Intrahospital mortality after weaning off ECMO support (n = 22)
A 67-year-old CABG patient suffered ventricular fibrillation while the sternotomy wound was being closed. The ECMO support was applied after prolonged open cardiac massage. Brain death was noted postoperatively, and the ECMO support was removed at 69 hours. The patient survived for another 3 days before cardiac standstill. Four patients suffered sudden death after ECMO had been removed for a while. Two of them were IABP dependent when ventricular fibrillation occurred; the other 2 patients had been removed from IABP support but remained bedridden. All 4 patients suffered no other organ dysfunction except for persistent heart failure.
The remaining 17 patients were weaned off the ECMO support, but still died of multiple organ failure in the intensive care unit. The periods from ECMO removal to death were 24 ± 49 days, with a median of 8.2 days.
Weaning off ECMO and survival to hospital discharge (n = 20)
Twenty patients were weaned off ECMO support and survived to be discharged from the hospital. The intervals of ECMO support were 99 ± 32 hours. The intervals from ECMO removal to extubation and intensive care unit discharge were 15 ± 18 days and 20 ± 20 days, respectively. These patients were followed up for 33 ± 22 months, and two late death occurred. One patient died suddenly at 11 months and the other died of pneumonia at 12 months. The other 18 patients are now in New York Heart Association functional status II or I.
Factors affecting outcomes of patients
Figure 1
displays the length of time on ECMO support (does not include the 4 patients bridged to heart transplantation or VAD). Only 2 wean-and-survive patients required ECMO support beyond 6 days, actually 7 and 8 days. All survivors needed ECMO support for less than 8 days. Beyond 8 days, either patients were unlikely to recover from the postcardiotomy heart failure or the ECMO complications would supervene and preclude the chance of survival.
|
Outcomes and complications of ECMO support through different routes
The patients were divided into three groups depending on how the ECMO support was set up (Table 2).
The chance of survival to discharge was the same for all three groups. Underlying diseases rather than the type of ECMO support decided the outcome. The sternotomy group required the most transfusions, especially in the first day of ECMO support. Four patients receiving the open femoral EMCO support were brain dead after the operation. The causes were assumed to be intraoperative stroke in 3 patients and prolonged resuscitation after the operation in 1 patient. Two patients suffered severe hypoxic encephalopathy. The femoral arterial cannula of 1 patient accidentally slipped out, causing hypovolemic shock from severe bleeding. The other patient underwent prolonged resuscitation during the percutaneous transluminal coronary angioplasty before emergent CABG. All 6 patients with brain death or hypoxic encephalopathy eventually died. Three patients in the open femoral ECMO group suffered hemiplegia/hemiparesis after the operation. No evidence indicated that these neurologic complications resulted from ECMO support. No neurologic complications were found in patients who received percutaneous femoral ECMO or sternotomy ECMO support.
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
Successful use of HeartMate VAD [4] and Thoratec VAD [5] for PCS has been reported. However, due to technical complexity, these VADs are not suitable for critical patients in emergencies. Most institutions refuse to administer this expensive therapy in patients with only a moderate chance of survival, because of high cost of the VAD support. Strict criteria of patient selection deprived many critical patients of the opportunity to receive VAD support. A centrifugal pump can provide a cheaper and simpler VAD, but its use requires sternotomy. Left atrial drainage is not easy [6]. Biventricular failure was common in patients with PCS [7, 8], and biventricular assistance was necessary in up to 50% of postcardiotomy patients who required MCS [9]. Bilateral centrifugal pump support was complex and predisposed to lung edema [6].
In comparison, ECMO is a better MCS choice for PCS not amenable to IABP support alone. The ECMO support has several advantages. It is composed of a microporous membrane oxygenator and centrifugal pump, which allows rapid priming. The ECMO support through the femoral route can be emergently set up at bedside under local anesthesia. It can be easily applied through the cannula also used for CPB. It can support both right and left heart failure, and substitute for lung function. All these advantages make ECMO support an ideal means of cardiopulmonary rescue for critical patients with an uncertain diagnosis. The ECMO is much cheaper than the currently available pulsatile VAD, and can provide temporary support, giving clinicians time to decide whether patients can benefit from further aggressive treatment. Four patients suffered brain death from intraoperative stroke or resuscitation, and 1 patient suffered hypoxic encephalopathy from preoperative resuscitation. Applying expensive VAD in these patients would be a waste. Myocardial stunning from ischemic–reperfusion injury is the most important cause of PCS [10]. In theory, patients can recover from myocardial stunning regardless of the severity of cardiac dysfunction. However, no feasible methods distinguish irreversible infarction from reversible stunning in the immediate postoperative period. Seventeen patients were weaned off ECMO support, but later died of multiple organ failure. Multiple organ failure was attributed to severe secondary organ damage, which had occurred before initiation of ECMO support. The severity of secondary organ damage was unknown when MCS was used for PCS. The ECMO support allowed time for medical decision. Then, if necessary, the patients could be bridged to heart transplantation or other more permanent VAD. The indication of ECMO support could be less strict, because of its relative simplicity and lower cost. More patients could be evaluated. Patients could recover rapidly from myocardial stunning caused by ischemia reperfusion injury. Therefore, VAD, which could provide long-term MCS, was usually unnecessary for most PCS. Nearly 50% (25 of 51 patients) of Thoratec-VAD-supported postcardiotomy survivors were supported for more than 7 days, and the mean duration of support was 12.8 days [11]. Major operations involved in VAD implantation might delay patients’ subsequent weaning off VAD. In comparison, mean duration of ECMO support for postcardiotomy survivors were 37.3 ± 12.7 hours in the report by Magovern and colleagues [6], 55.6 ± 26.2 hours in the one by Muehrcke and associates [12], and 99 ± 33 hours in the present study. The ECMO support could be easily weaned off and removed; this may explain why ECMO support duration for postcardiotomy survivors was much shorter than that of VAD patients.
However, the effectiveness of ECMO support was limited in postcardiotomy patients. The ECMO can provide only partial cardiopulmonary support; 2.53 ± 0.84 L/min blood flow in this series. Bilateral VAD or total artificial heart is a better choice of MCS for patients with very poor heart function. Support was not a solution for uncontrolled bleeding, and reliable hemostasis was a prerequirement of successful ECMO support. The ECMO support was not intended for long-term MCS as a bridge to heart transplantation. Complications usually developed during prolonged ECMO support that precluded heart transplantation [13]. Two of our patients were successfully bridged to heart transplantation after 8 and 31 hours of ECMO support, respectively. This was an exception not a rule. If a donor heart was not immediately available, patients should be bridged to VAD for long-term support. Four patients suffered sudden cardiac death some time after ECMO removal. Two remained IABP dependent, and the other two were in New York Heart Association functional status IV. Ventricular arrhythmia should be carefully prevented in ECMO-weaned patients with compromised heart function. Seventeen ECMO patients were weaned off the ECMO support, but later died of multiple organs failure. These wean-but-die patients had higher inotropic agents scores and blood lactate levels before the ECMO support, and higher creatine kinase, bilirubin, and glutamic oxaloacetic transaminase levels in the first week of ECMO support than did the wean-and-survive patients. The finding shows that wean-but-die patients had more severe shock damage before ECMO support than did the wean-and-survive patients. Earlier ECMO support might have saved some of them.
Kawahito and colleagues [14] reported that ECMO support for PCS with a mean ECMO blood flow of 2.2 ± 0.5 L/min and mean support time of 27.4 ± 26.7 hours yielded 77% weaning and 39% long-term survival. Fiser and associates [15] reported 31% weaning and 16% survival. Extracorporeal Life Support Organization (ELSO) registry revealed a 37% survival rate in patients receiving ECMO for PCS [16]. The ECMO support for PCS in this study yielded 58% weaning and 28% survival to hospital discharge, if patients bridged to VAD or heart transplantation were excluded. Better case selection to exclude patients with uncontrollable bleeding, very poor cardiac contractility, and refractory ventricular arrhythmia, would have improved the result. Applying ECMO earlier, when it was indicated, would have decreased shock damage and prevented some patients from dying of multiple organ failure.
Bleeding was the most important complication in ECMO support for PCS [6]. The heparin-bound carmeda bioactive surface allowed an ECMO support with minimal or no systemic heparinization [17]. Lazzara and colleagues [18] used the same heparin-bound ECMO support in their postcardiotomy patients; however, the mean packed red blood cells transfusion requirements were 24 ± 9 U for a mean support duration of 47.9 hours. All of these patients received ECMO support through open sternotomy wounds. In contrast, most of our patients received ECMO support through the femoral route, and their blood transfusion requirements were much less, especially when the percutaneous insertion technique was used (see Table 2). The ECMO support through the open sternotomy wounds provided higher blood flow due to the larger size of the cannula. However, the ECMO blood flow was only marginally increased in comparison to the ECMO flow through the femoral route. The femoral ECMO had fewer bleeding complications than the sternotomy ECMO and allowed for a much easier nursing care. Magovern and coworkers [6] reported that 2 patients suffered strokes after the removal of ECMO support due to a clot at the tip of the cannula. A thromboembolism would go to legs rather than to the brain, if the femoral route were chosen for ECMO support.
In conclusion, IABP is the first consideration if MCS is indicated for PCS. ECMO is the second choice, when IABP support is insufficient or not possible. Extracorporeal membrane oxygenation support through the femoral route is preferred because of fewer related complications and easier nursing care. That support allows for a reasonable time to evaluate the patients and decide on the next therapeutic step, and prevents the misuse of expensive VAD.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
