https://scholars.lib.ntu.edu.tw/handle/123456789/192492
標題: | Extracorporeal Membrane Oxygenation Rescue after Heart Transplantation | 作者: | KO, WEN-JE CHEN, YIH-SHARNG |
關鍵字: | extracorporeal membrane oxygenation;heart transplantation;primary graft failure;rejection | 公開日期: | 2000 | 卷: | v.32 | 期: | n.7 | 起(迄)頁: | 2388-2391 | 來源出版物: | TRANSPLANTATION PROCEEDINGS | 摘要: | The mortality of heart transplantation (HTx) had the highest incidence in the first month after the transplantation and sharply declined afterwards .1 Etiologies of the early mortality included primary graft failure, acute rejection, infection, bleeding, multiple organ failure, etc.2, 3 However, cardiac pump dysfunction from primary graft failure or acute rejection was the most important mechanism for the early mortality. Since the primary graft failure and acute rejection were potentially reversible, if there were some mechanical circulatory devices to temporarily support the circulation until recovery of the graft function, some patients could be rescued. Extracorporeal membrane oxygenation (ECMO) could provide excellent mechanical circulatory support (MCS),4 and there were some cases reports of using ECMO to rescue HTx recipients with cardiopulmonary failure in their early post-transplant periods.5-7 We reported our series of using ECMO to rescue HTx recipients with refractory cardiac graft dysfunction in their early post-transplant periods. PATIENTS AND METHODS From Jul. 1987 to Mar. 1999, there were 113 patients undergoing 115 heart transplantations at the National Taiwan University Hospital. Two patients underwent cardiac retransplantation for primary graft failure and chronic rejection, respectively. If circulatory insufficiency occurred after the cardiac implantation, medical treatment with catecholamines and vasodilators were first sought. However, if low cardiac output persisted with maximal medical treatments, MCS would be considered. Because of its relative simplicity and non-invasiveness, intraaortic balloon pumping ( IABP) was the first choice of MCS. If IABP was contraindicated or could not provide enough circulatory support in the patients, ECMO was the next consideration. Femoral veno-arterial route was the preferred route to set up the ECMO support. But if the femoral veno-arterial route was not possible for some reasons, the ECMO route would be through the open sternotomy wound. The whole ECMO apparatus, including centrifugal pump and oxygenator, was from Medtronic Inc. Annahein, CA, USA. The ECMO was primed with normal saline alone. Hemodilution after the hook-up of ECMO to the patients was corrected by packed red blood cells transfusion. Due to MX-2 Tri-optic measurement cells ( Medtronics Inc, Anaheim, CA, USA) attached to the ECMO circuit, hematocrit and blood oxygen saturation in the pre- and postoxygenator circuits could be continuously monitored. Hematocrit was kept between 30% and 35%. Lower hematocrit compromised oxygen delivery, but higher hematocrit increased complications of clot formation and hemolysis by centrifugal pump. Continuous monitoring of postoxygenator blood oxygenation saturation could monitor the gas-exchange function of the oxygenator. Blood oxygen saturation in the preoxygenator circuit could be recognized as the mixed venous blood oxygen saturation and reflected the balance between oxygen supply and demand of that patient . It could guide the manipulation of the patient’s hemodynamics. Usually, the ECMO centrifugal pump speed was set at 2000 rounds per minute, and the ECMO blood blow should be more than 2 L/min. If not, the volume status and venous cannular position would be checked first to improve the ECMO blood blow. Sweep gas flow of the ECMO was initially set at a rate of 2 L/ min, and adjusted subsequently according to the blood gas analysis. Because of heparin-bound Carmeda bioactive surface, systemic heparinization was not needed on the first day of ECMO support, when risk of bleeding was the highest immediately after the HTx operation. Afterwards, heparin infusion was used to keep activated clotting time between 160 and 180 seconds. The ECMO apparatus was changed when oxygenator dysfunction, clot formation, or hemolysis was found. Symptoms and signs of low cardiac output was usually resolved after the ECMO support initiated, catecholamine infusion could be tapered accordingly. Arterial pulse pressure wave contour, serial echocardiography, and blood oxygenation saturation in the preoxygenator circuit were used to monitor the recovery of cardiac allograft. If hemodynamics could be well maintained by reduced ECMO blood flow at 0.5 L/min for one hour, ECMO was removed at bedside. The wound was primarily repaired. Rabbit antithymocyte globulin immunoinduction followed by triple-drug maintenance immunosuppression was used in all our HTx patients, no matter MCS was used or not. RESULTS Mechanical circulatory support was needed in the early post- transplant period in 19 of our 115 HTx operations. The MCS used in these situations included IABP alone (n = 7), IABP followed by ECMO (n = 8 ); and ECMO alone (n = 5). IABP could not be used in 3 girls with body weight less than 25 kg. One patient had bilateral femoral thromboembolism from dilated cardiomyopathy. One patient had the right ventricular failure alone. In these 5 patients, ECMO was directly applied without the previous IABP use. In another 8 HTx, IABP was first tried but failed to support the circulation, then the ECMO support was added. One patient needed twice support of IABP and ECMO. One time was for primary graft failure and MCS was used as a bridge to cardiac retransplantation, the other time was for acute humoral rejection of cardiac graft after the retransplantation. In summary, 13 ECMO supports with or without IABP were used in 12 patients. The sex was male in 7 patients and female in 5 patients. Their ages were 43 ± 19 years old (range: 9 to 65 years old). Table 1 shows the indications, routes, set-up places, duration, and outcomes of the ECMO support. Femoral veno-arterial route was the preferred route for the ECMO support. However, the ECMO support must be through open sternotomy wound for different reasons in 5 patients. The femoral cannulation was excluded in two girls due to small caliber of the femoral vessels. One patient underwent simultaneous cardiac and renal transplantation. His right femoral artery had been used for the IABP, and the left femoral cannulation was excluded due to the renal graft in this side. One patient had bilateral femoral arteries thromboembolism from thrombus in the dilated ventricles. One patient had primary graft failure from hyperacute rejection, and the left heart decompression was needed to prevent the left heart distension. Because venous drainage from both atria was needed, the ECMO route was directly through the sternotomy wound. Nine patients needed the ECMO support for primary cardiac graft failure. Four of these patients could not be weaned from the cardiopulmonary bypass , and directly received the ECMO support in the operation rooms. One patient was fortunate enough to get another donor heart and underwent a successful retransplantation after 8 hours of ECMO and IABP support. One patient was put on the ECMO support for 161 hour and successfully weaned from it, but still died of multiple organ failure one week later. Two patients had high central venous pressure (> 12 cm-H2O) under ECMO support and could not be weaned from the ECMO support, cardiac catheterization was done to search the underlying problems. Anatomic defects of the right atrial twist and stenosis at pulmonary artery anastomosis were found, respectively. Reoperations to correct the anatomic defects were done, and both patients could be weaned off the ECMO support after the reoperations. However, due to complications from the prolonged ECMO support (168 and 216 hours), both patients died of sepsis and multiple organ failure. Five patients received the ECMO support for primary graft failure after they had been transferred to the intensive care units for the postoperative care. These five patients were successfully rescued by the ECMO support. Four patients needed the ECMO support for graft dysfunction from acute rejection. One patient suffered hyperacute rejection. Because the patient could not been weaned from the cardiopulmonary bypass, ECMO support with venous drainage from bilateral atria was set up in the operation room with the hope of retransplantation. However, a donor heart was never found for her, and she died of sepsis after 562 hours of ECMO support. One patient had severe acute rejection 17 days after the transplantation, IABP then ECMO was used for circulatory support. However, complications of pulmonary edema and hemorrhage resulted from distended left heart and elevated left heart filling pressure. Pneumonia and acute respiratory distress syndrome occurred subsequently, and the patient finally died of these complications. One patient was put on the ECMO support 35 hours after the transplantation under the impression of primary graft failure, however, two days later, QRS complex of the ECG widened then rapidly became standstill 33 hours after he was put on the ECMO support. The necropsy revealed acute humoral rejection of the graft. One patient suffered acute humoral rejection 4 days after the transplantation, IABP and ECMO support were needed to support the circulation . However, the graft was successfully rescued by OKT-3, steroid pulse, and plasmapharesis after 85 hours of ECMO support. This was the only survival case in the acute rejection group. The most common complication of the ECMO support was mediastinal bleeding and cardiac tamponade. The bleeding complication was related to the ECMO route. Reexploration for hemostasis and blood colt removal was needed in 1 of 8 patients receiving the ECMO support through the femoral venoarterial route and in 4 of 5 patients receiving the ECMO support through the open sternotomy wound . Subarachnoid hemorrhage occurred in one girl, and caused a neurological sequela of temporal lobe seizure, which required long-term drug treatment. Acute renal failure occurred in 6 patients before initiation of the ECMO support . Continuous hemofiltration was set up on the ECMO circuit for dialysis. Only two patients of them were long-term survivors, but they had complete renal recovery. In comparison, 5 of 7 patients without complications of acute renal failure survived. DISCUSSION Despite the continuous improvement of myocardial protection, the incidence of primary graft failure remained as high as 5% to 7%.8, 9 This etiology explained 52.8% of HTx mortality occurring within the first month after the transplantation.10 The primary graft failure usually resulted from myocardial stunning. The myocardial stunning is defined as a prolonged post-ischemic ventricular dysfunction, but reversible after a prolonged period of time. 11 Acute humoral or cellular rejection was another reason of cardiac graft dysfunction in the early post-transplant period. If there were no mechanical circulatory devices, retransplantation would be the only effective option, since immunorescue treatments of steroid pulse, antithymocyte globulin, or OKT3 would need several days to effectively reverse the rejection. In summary, some HTx patients would need temporary MCS to allow time for stunned heart to recover or anti-rejection therapy to reverse the rejection. When low cardiac output persisted under maximal medical treatment, IABP was the first consideration of MCS due to its relative simplicity and non- invasiveness. However, IABP is useful for heart failure isolated to the left heart alone, and cannot provide support to the right heart or lung dysfunction. IABP support is not feasible in some patients, for example, pediatric patients with smaller femoral arteries, and patients with femoral atherosclerosis or thromboembolism. IABP cannot provide enough circulatory support to patients with profound heart failure. In our series , IABP provide enough support in 7 patients, but 8 patients needed more than IABP support alone and 5 patients directly sought the ECMO support. Ventricular assist devices (VAD) could provide enough circulatory support to patients with profound heart failure. However, VAD was too invasive. It needed thoracotomy and prolonged operations to set up or remove. This excluded its use in critical patients under emergent situations. There was lack of appropriate sized VAD device for small children. VAD was expensive and limitedly availabe. In comparison, ECMO was simpler and more available. It could be quickly set up or removed at bedside under local anesthesia, if femoral veno-arterial route was chosen. This was an important advantage for critical patients. Because of heparin-bound Carmeda bioactive surface in our ECMO, systemic heparinization was not used on the first day, and only low dose of heparin infusion was needed afterward. Less systemic heparin requirements was an important advantage for post- cardiac surgery patients. Bleeding had been the most common complication of ECMO.12 With heparin-bound surface of the ECMO circuits, bleeding complication of the ECMO support still existed; however, it was clinically tolerable even in post-cardiac surgical patients. ECMO could simultaneously support biventricular and respiratory failure. Due to the pre-operative pulmonary hypertension in many HTx patients, the right ventricular failure was more common and severe than the left ventricular failure after the HTx operations. The right ventricular failure was an important source of early mortality after the HTx.13 ECMO decompressed the right heart and directly relieved the right heart failure. Most HTx recipients with profound circulatory insufficiency in their early post- transplant period could rely on the ECMO alone to support the circulation. ECMO could be used in both adult and pediatric patients. In fact, it was the only choice of MCS for children. If patients needed dialysis, it was easy to set up hemofiltration on the ECMO circuit. ECMO returned blood to the aorta and increased the left ventricular afterload. If the left ventricular failure was severe and could not overcome this increased afterload, the left heart would distend and left ventricular filling pressure increased. The increased left ventricular pressure transmitted to pulmonary capillary, and caused hydrostatic pulmonary edema and even hemorrhage. One of our patients died of this complication, and another one needed the left atrial venous drainage to prevent this complication. IABP could decrease the left ventricular afterload and palliated this complication. However, arterial pulse pressure contour and serial echocardiography should be closely followed up. If there was evidence of left ventricular distension and inability to empty its blood, the left heart should be decompressed by some methods.14 Prolonged anticoagulation and bleeding complication, bedridden, infection risk, the limited durability of the oxygenator, etc., precluded the ECMO as a long-term MCS. Severe complications usually developed as time passed by and precluded a patient as a candidate of cardiac retransplantation. Bridge to transplantation by the ECMO support is the exception not the rule, especially when the ECMO support must extend beyond 7 to 10 days. From our experiences, if possible, ECMO support was better though the femoral veno -arterial route than through the open sternotomy wound. The femoral route had fewer complications of bleeding, infection, and sepsis, especially when the ECMO support had to extend beyond several days. Nursing care was much more difficult for patients with open sternotomy wound, and these patients needed more analgesia and sedation. MCS should be used early before secondary organ damage had resulted from profound heart failure. Six patients had shock-induced acute renal failure before initiation of the ECMO support and this complication made difficult the patient care. One patient could be weaned off the ECMO support, but still died of multiple organ failure because of profound shock before the initiation of the ECMO support. Because myocardial stunning should recover gradually under the MCS, if the cardiac graft could not recover as expected, we should diligently search the underlying problems. Two patients, who received the ECMO support for the presumed primary graft failure, were subsequently found to have anatomic defects and underwent reoperation to correct the defect. Both patients could be weaned off the ECMO support after the reoperation, but still died of complications from the prolonged ECMO support. If anatomic defects had been found early, maybe they could have been successfully rescued by the ECMO support. Another patient had unexpected ECG change, and finally died of unsuspected acute humoral rejection. The experiences of these cases suggested if the transplant heart did not recover as expected or had new changes, underlying problems should be searched and resolved to rescue the patients. Since ECMO is not intended for prolonged circulatory support , if graft recovery is not feasible, early re- transplantation or shift to ventricular assist devices should be considered. The fact that none of our patients receiving the ECMO support more than 4 days survived, justified this principle. In conclusion, ECMO could provide temporary MCS and rescued some HTx recipients with profound heart failure in their early post-transplant periods. |
URI: | http://ntur.lib.ntu.edu.tw//handle/246246/96193 |
顯示於: | 醫學系 |
在 IR 系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。