Branched xenopericardial roll graft replacement of an infected aortic arch graft

Abstract Which graft material is the optimal graft material for the treatment of aortic graft infections is still a matter of controversy. We used a branched xenopericardial roll graft to replace an infected aortic arch graft as a “rescue” operation. The patient is alive and well 37 months postoperatively without recurrence of the infection and any surgical complication. This procedure may have the possibility to serve as an option for the treatment of aortic arch graft infection.


| INTRODUCTION
The optimal graft material for the surgical treatment of aortic graft infections is still a matter of controversy. Orthotopic aortic reconstruction with an intraoperatively prepared xenopericardial branched roll graft was successfully performed as a rescue operation to treat a patient with a graft infection.

| Patient profile
A 66-year-old male was urgently transferred to our hospital because of fever, general malaise, and refractory skin to graft fistula after a total aortic arch replacement in 2015. He had a complicated past history of treatment for major cardiac and cerebrovascular disease. He had undergone a percutaneous coronary intervention (PCI) for acute myocardial infarction at 55 years of age, and coronary artery bypass graft surgery x4 using left internal thoracic artery, radial artery, and saphenous vein grafts at 56 years of age. In 2012, at 63 years of age, a triplex 26 mm (Terumo, Tokyo, Japan) had been used to perform a total aortic arch replacement procedure to treat a true aortic arch aneurysm at a nearby hospital. A reversed elephant trunk was inserted to facilitate the distal anastomosis using stepwise distal anastomosis technique. Preoperative coronary angiography showed the following stenosis: proximal left anterior descending artery, 75%; 1st diagonal branch, 75%; 1st left posterolateral branch, 75%; and 2 nd left posterolateral branch, 90%. All coronary artery bypass grafts were occluded. The origin of the left subclavian artery was ligated, and a 3rd graft branch was extended by a graft to the axillary artery via the left thoracic cavity. One month postoperatively, open-chest drainage had been performed to treat left thoracic empyema. Subsequently, an incisional abscess was noted around the site of the left axillary anastomosis, and a graft-branch-to-skin fistula was diagnosed. In 2013, at 64 years of age, the abscess was opened and the graft branch to the axillary artery was partially removed through a subaxillary skin incision. On admission to our hospital, the patient's serum C-reactive protein level was elevated to 22.5 mg/dL, and a blood culture was positive for methicillin-sensitive Staphylococcus aureus. The patient gradually became drowsy, and brain magnetic resonance imaging tomography showed vegetations inside the origin of the graft branch to the brachiocephalic artery and the left common carotid artery ( Figure 1C and D). Graft infection, sepsis, and multiple mycotic brain infarctions were diagnosed, and emergency surgery was planned.
Written informed consent with the approval of our institutional review board was obtained from the patient's family before performing the procedure.

| Surgical procedure
A cardiopulmonary bypass through a re-resternotomy was established by means of two venous cannulas and an inflow cannula into the graft.
The left subclavian abscess cavity communicated with the graft.
Because of the dense adhesions, the left-side of the heart was not icardial roll graft has been described in detail previously 3 ). The felts and threads that had been used to close the orifice of the left subclavian artery were removed, and the stump was re-closed by suturing pericardial patches. One side of the pericardial sheet was anastomosed to the remaining graft with a 4-0 polypropylene single continuous suture and shaped into a cylinder. The third branch was anastomosed to the left common carotid artery. The first branch was used as the inflow route for the cardiopulmonary bypass. The main graft was clamped, and antegrade perfusion was restored. The second branch was anastomosed to the brachiocephalic artery. After completing the proximal anastomosis, the aorta was de-clamped. Although weaning from the cardiopulmonary bypass was achieved without difficulty, sudden ventricular fibrillation due to myocardial ischemia occurred during hemostasis. Direct cardiac massage for 9 min and an intra-aortic balloon pump insertion were required to restore stable vital signs. The   In contrast to simple tube grafts, we developed a branched xenopericardial tube graft. In our first case, even though it was an elective operation, we decided to prepare a branched xenopericardial sheet intraoperatively, because the native aortic infection site appeared to be wider and deeper than expected, rather than being located in just the ascending aorta.
Major concerns in regard to separately reconstructed neck vessels are graft stenosis, calcification, and possible thrombus formation. We routinely administer an anticoagulant or antiplatelet drug to prevent branch-related complications. We have performed six aortic arch reconstructions by using branched xenopericardial sheets thus far. All of the cases except the 1st were treated by "rescue" operations that involved removing the infected graft. The patient in the 1st case died of lung cancer 45 months postoperatively, and the other five patients, reported finding that the technical success rate was 100% in a total of four studies describing 71 cases. Mean 30-day mortality was 25%, and only one death (1.4%) was linked to an operator-prepared pericardial tube graft 8 ).
In conclusion, accumulation of clinical cases and confirmation of the long-term durability of branched xenopericardial grafts may demonstrate their advantage as one of the options for the treatment for native aortic and aortic arch graft infection.