|Home | About | Journals | Submit | Contact Us | Français|
Operative methods for repairing ascending aortic dissections and for implanting left ventricular assist systems have been thoroughly presented in the medical literature. Only a few reports, however, describe the concomitant performance of these procedures in 1 patient. We report the repair of an acute ascending aortic dissection with simultaneous placement of a long-term left ventricular assist system. One week earlier, the patient had undergone emergent coronary artery bypass grafting and short-term postcardiotomy ventricular assistance when he could not be weaned from cardiopulmonary bypass. By creating a graft-to-graft anastomosis on the bench during cooling of the patient on cardiopulmonary bypass, we were able to shorten to 21 minutes the period of hypothermic circulatory arrest required during ascending aortic dissection repair. The procedures were completed successfully. However, the patient developed pneumonia and sepsis during his extended hospital stay and died of multiorgan failure 5 weeks postoperatively.
Standard techniques for insertion of a left ventricular assist system (LVAS)1 and for repair of an acute aortic dissection (AAD) after coronary artery bypass grafting (CABG)2 have been previously described. Today, most LVAS implants require that the outflow graft be anastomosed to the ascending aorta. We report a case in which an AAD was detected after CABG in a patient undergoing placement of a HeartMate® XVE LVAS (Thoratec Corporation; Pleasanton, Calif). The AAD was repaired and the LVAS implanted concomitantly.
A 60-year-old man with a history of hypertension and smoking sustained a cardiac arrest at home. He had previously experienced no symptoms directly attributable to coronary artery disease. He was resuscitated successfully at another medical facility. A large posterior myocardial infarction secondary to left main coronary artery occlusion was diagnosed. The patient remained in fulminant cardiogenic shock and was taken to the operating room for emergent CABG with saphenous vein grafts to the left anterior descending, obtuse marginal, and right coronary arteries. At the end of the operation, he could not be weaned from cardiopulmonary bypass (CPB). A Bio-Medicus® Bio-Pump® (Medtronic Inc.; Minneapolis, Minn) was implanted for temporary ventricular assistance, via a femoral artery approach for outflow cannulation and a left atrial approach for inflow cannulation. The skin was closed, but the sternum was left open. Postoperatively, the patient improved on mechanical circulatory support. His neurologic function recovered, his end-organ function was sustained, and he was weaned from the ventilator and extubated. Seven days after the cardiac arrest, he was transferred to our hospital for further treatment.
Shortly after arrival, the patient required reintubation due to respiratory distress. His initial regimen included high-dose diuretics, afterload reduction, and inotropic support. Transesophageal echocardiography (TEE) showed a small left ventricular apical aneurysm and mild-to-moderate mitral regurgitation but slightly improved left ventricular wall motion. The patient was taken to the operating room for mitral valve repair, plication of the aneurysm, and potential removal of his Bio-Medicus pump.
The chest was reopened, and venous return for CPB was established with bicaval cannulation. The femoral artery cannula from the Bio-Pump was connected to CPB for the return of arterial blood. Once CPB was initiated, the Bio-Pump was removed and the atriotomy closed with a purse-string suture. The left ventricular aneurysm was opened, and an Alfieri mitral valve repair was performed through the left ventriculotomy. The myocardium was then closed with interrupted, pledgeted sutures and Teflon strips.3 An intra-aortic balloon pump was also placed through the other femoral artery. With increased inotropic support, the patient was successfully weaned from CPB. Intraoperative TEE showed marginally improved left ventricular contractility, trace mitral regurgitation, an intact Alfieri repair, and no intracardiac air. In the limited TEE window, the aorta appeared intact and the aortic valve normal. After the skin was closed, the patient was transferred to the cardiac intensive care unit. He did well until the 2nd postoperative day, when his end-organ function started to worsen and his overall status deteriorated.
The patient was immediately returned to the operating room for exploration and placement of a HeartMate XVE LVAS. Cardiopulmonary bypass was instituted via the femoral vessels, and venous return was accomplished with superior vena cava cannulation. The left ventriculotomy was then reopened, and the inflow cannula was inserted into the left ventricle in the usual position.1 During evaluation of the ascending aorta for an appropriate outflow-graft anastomosis site, intraoperative TEE revealed an AAD that seemed to originate from the right coronary artery graft anastomosis site.
The patient was cooled to 18 °C, and circulatory arrest was initiated. During the cooling period, an end-to-side anastomosis between the previously sized LVAS outflow graft and the aortic graft was created on a separate table. With this approach, we were able to shorten the circulatory arrest time to 21 minutes. The brain was protected with retrograde cerebral perfusion via the superior vena cava. On opening the ascending aorta, we confirmed that the dissection originated from the right coronary artery graft anastomosis site and was limited to the proximal portion of the arch.
The 3 vein grafts from the previous CABG were clamped and excised in a Carrel aortic patch. The dissected aortic segment was then excised, and the distal aortic graft anastomosis was performed with the previously created composite graft.
Cardiopulmonary bypass was again initiated. After de-airing of the aorta, both graft limbs were cross-clamped. The proximal aortic anastomosis was then created just above the coronary ostia with running 4–0 polypropylene sutures. The previously excised vein-graft aortic patch was reanastomosed to the ascending aortic graft, in end-to-side fashion, proximal and medial to the LVAS outflow-graft anastomosis (Fig. 1). After de-airing of all the bypass grafts and the heart was accomplished, the aortic-graft cross-clamp and the clamps from the CABG grafts were removed, and the heart resumed sinus rhythm. The LVAS graft remained clamped. While the patient was progressively warmed, the LVAS was de-aired via a vent placed on the LVAS outflow conduit.
After all the cardiac chambers were free of air (seen on TEE) and the LVAS was fully de-aired, the pump's outflow graft was filled with blood in retrograde fashion and was connected to the outflow conduit. Complete de-airing of the LVAS and the heart was reconfirmed with TEE, and CPB flow was slowly reduced. At a flow of 1 L/min, the LVAS was started and CPB was terminated. Then the cannulas were removed, protamine sulfate was given, and the chest was closed. Postoperatively, the patient was neurologically intact, but he acquired pneumonia and became septic after approximately 1 week. He succumbed to multiorgan failure on the 36th postoperative day.
Acute aortic dissection that involves the proximal anastomosis of a coronary artery bypass graft is extremely rare, occurring in only 0.05% of cases.2 In most cases, repair entails graft aortic replacement with the aid of circulatory arrest. Acute aortic dissection after LVAS insertion has been reported only a few times.4,5 In 1 such patient, Naka and coworkers4 reported a radical approach that consisted of closure of the aortic valve and construction of a Cabrol graft to reinstitute coronary flow. In another case,5 an AAD was repaired 18 days after placement of a Novacor® LVAS.
Our case is unique in that the AAD was repaired concomitantly with placement of a HeartMate XVE LVAS, because the dissection was not discovered until LVAS implantation had already begun. The patient had originally been admitted in postcardiotomy cardiogenic shock that required short-term mechanical circulatory support. His clinical status improved sufficiently to allow extubation and, eventually, removal of the assist device. Subsequently, however, the saphenous vein graft to the right coronary artery became dysfunctional because of the AAD, which likely expanded as a result of higher initial systolic pressure after explantation of the Bio-Pump. The patient's condition deteriorated because of right ventricular failure and an absence of flow through the right coronary artery bypass graft.
In patients who are in postcardiotomy cardiogenic shock after CABG, it is necessary to examine the ascending aorta with intraoperative TEE. In our patient, a small AAD was probably present soon after the CABG procedure. Because the patient was supported by a nonpulsatile Bio-Pump, however, the dissection apparently remained stable during the support period while the left ventricle was being decompressed. The goal of the 2nd operation was to implant an LVAS to support the failing heart. After discovering the AAD intraoperatively, however, our only option was to repair the dissection and implant the LVAS concomitantly. By creating an anastomosis between the aortic graft and the LVAS outlet graft on a separate table while the patient was cooling, we were able to shorten the hypothermic circulatory arrest period to 21 minutes.
In summary, our case shows that simultaneous repair of an AAD and implantation of a long-term LVAS is technically feasible. Although it is not generally advisable to combine such procedures, in our case we were forced to proceed, because the dissection had gone unrecognized until we explored the ascending aorta for the outflow graft anastomosis, and by then the device was already in place. The patient's cardiac output improved postoperatively; however, the foregoing factors, along with postoperative pneumonia and sepsis acquired during his extended hospital stay, eventually prevented his recovery.
Address for reprints: Igor D. Gregoric, MD, Department of CardiacTransplantation, Texas Heart Institute, MC 2-114A, P.O. Box 20345, Houston, TX 77225-0345. E-mail: ude.cmt.iht.traeh@cirogergi