Invited commentary

Abstract

Fortunately, postcardiotomy mechanical support is required in only 0.5% to 1.0% of cardiac operations. Extracorporeal membrane oxygenation is very versatile, providing both cardiac and pulmonary support, is portable, easy to run, and inexpensive. Vascular access can be obtained through the peripheral circulation. However, a tremendous amount of resources will be utilized to support patients who die 75% of the time. How can information provided in this and other recently published series be used to optimally manage patients? Although peripheral access is often convenient, it is associated with a very high rate of vascular complications of the lower extremity. We have utilized a distal limb perfusion catheter in the superficial femoral artery, while Doll has recommended anastamosing a 6-mm hemashield graft to the common femoral artery. Central aortic cannulation avoids peripheral vascular injuries; however, it often requires the sternum to remain open and can be associated with bleeding from around the cannulas. Attention to detail at this step of the operation will reduce bleeding and many of the complications listed in Table 4. Intraaortic balloon pump counterpulsation should be used in all patients on ECMO support and is associated with an improvement in survival in this series, as well as in our experience. The exact mechanism for this benefit is not known. In addition to decreasing the afterload on the heart, it may improve visceral organ perfusion thus reducing the severity of multisystem organ failure. Survival is poor for patients who are older, specifically those who are over the age of 65, and who have had a preoperative myocardial infarction, diabetes, decompensated congestive heart failure or cardiogenic shock, and, in our experience, thoracic aortic operations. It is noteworthy that 95% of Doll's patients undergoing aortic valve replacement and coronary bypass grafting and 90% of patients listed in the “other” category (pulmonary embolectomy, ventricular septal defect closure, LV and aortic aneurysm repairs) also succumbed after ECMO support. The more complex the procedure, the less likely the patient will survive. If myocardial recovery is not evident within 48–72 hours, the patient should be assessed for more advanced mechanical circulatory support as more than two thirds of patients who do not survive ECMO support have extensive irreversible myocardial injury. This approach can dramatically improve survival. If the patient is not a candidate for advanced support, is elderly, or has preoperative comorbidities, prolongation of ECMO beyond a few days is not justifiable because survival is very unlikely. ECMO can undoubtedly save the lives of patients requiring postcardiotomy support, but device related complications must be minimized and we need to have realistic expectations for survival.