Infected Pseudoaneurysm of an Outflow Graft After Left Ventricular Assist Device Insertion

Bleeding and infection may complicate LVAD implantation. We present the case of a patient with recurrent fevers after LVAD implantation who was found to have an infected pseudoaneurysm of the outflow graft at the graft-aortic anastomosis, with successful surgical intervention. Patients with infected pseudoaneurysms post LVAD insertion require urgent surgical intervention that may differ from that for regular pseudoaneurysms. Bleeding and infection may complicate LVAD implantation. We present the case of a patient with recurrent fevers after LVAD implantation who was found to have an infected pseudoaneurysm of the outflow graft at the graft-aortic anastomosis, with successful surgical intervention. Patients with infected pseudoaneurysms post LVAD insertion require urgent surgical intervention that may differ from that for regular pseudoaneurysms. A 32-year-old man with non-ischemic cardiomyopathy, recurrent hospitalizations for heart failure, an ejection fraction of 11%, and an implantable cardiac defibrillator, was admitted to the coronary care unit in cardiogenic shock after a recent episode of fever and respiratory symptoms. Echocardiography revealed severe left ventricular dysfunction, moderate to severe right ventricular dysfunction, and moderate to severe tricuspid regurgitation. He required maximal inotropic and pressor support and underwent placement of a transaxillary Impella to prevent hemodynamic collapse. He had persistent high-grade fevers up to 39ºC, with no identified source and no positive blood or sputum cultures. Chest imaging revealed significant mediastinal lymphadenopathy and bilateral pulmonary ground-glass opacities of unclear etiology. He had pulmonary hypertension with systolic pulmonary artery pressures >70 mmHg, and a pulmonary capillary wedge pressure >30 mmHg. As the pulmonary hypertension precluded him from consideration for heart transplantation at that time, it was decided that he should undergo urgent durable LVAD implantation as a bridge to candidacy for transplantation. He eventually defervesced and 48 hours later underwent insertion of a HeartWare HVAD (Medtronic Inc., Minneapolis, MN) with simultaneous intra-operative bronchoscopy, bronchoalveolar lavage and lymph node biopsy. These specimens did not reveal any evidence of bacterial or fungal infection, nor malignancy. On post-operative day 6 he required re-exploration for cardiac tamponade due to a late hemopericardium but his course in hospital was otherwise unremarkable. He was discharged on post-operative day 20. Three days after discharge, he presented with fevers and low flow alarms from his LVAD and purulent discharge from his sternal wound and subcostal chest tube exit sites. Only after multiple sets of cultures from the wound, driveline and blood, a final blood culture grew Candida albicans. A CT scan demonstrated a pericardial collection with pockets of air (Fig. 1A) and an 8x9mm pseudoaneurysm adjacent to the anastomotic site of the outflow cannula. He underwent urgent repair of the presumed mycotic pseudoaneurysm with an uneventful sternal re-entry. Femoral cardiopulmonary bypass was instituted and the LVAD outflow graft was clamped. The patient was cooled to 28º C and the aorta and outflow graft anastomosis were dissected. A large defect was noted at the anastomosis of the outflow graft to the ascending aorta, measuring approximately 8x7 mm, on the left side of the aorta (Fig.1B). Initially a partial occluding clamp was applied to the aorta, the outflow graft to aortic anastomosis was taken down, and the distal 3 cm of the graft was resected. In order to permit adequate resection and reconstruction of the aortic margin, an aortic cross clamp was applied and cold antegrade cardioplegia was given to arrest the heart. A large ellipse of aorta, including a 7-8 mm rim of aorta around the anastomosis and comprising 60% of the aortic circumference, was resected to ensure that the remaining tissue was free from infection and also robust enough to support another graft. Specimens were sent for microbiology. A bovine pericardial patch (3x4 cm) was tailored to reconstruct the aorta (Fig.2B). The aortic cross clamp was removed to allow the heart to re-perfuse and a partial occluding clamp was reapplied to allow the new end of the LVAD outflow graft to be anastomosed to the pericardial patch. There was no visible gross evidence of infection on the outflow graft, bend relief, VAD housing or driveline. The mediastinum was copiously irrigated. He was weaned from cardiopulmonary bypass without difficulty, on inotropic and LVAD support (Fig.2A). Cultures from the mediastinum subsequently grew E. faecalis and Candida albicans. He was treated with sulfamethoxazole/trimethoprim and caspofungin. He was listed for transplantation as a high-status recipient due to a presumed infected LVAD. His hemodynamics at the time of listing included PA systolic pressures of 32/17 with a TPG of 12, indicating rapid resolution of his pulmonary hypertension after LVAD implantation. The patient underwent successful explant of the LVAD and heart transplantation 12 days later. He was initiated on standard immunosuppressants, caspofungin later switched to fluconazole for a total of 6 weeks of antifungal therapy, and discharged home. Two years after his transplant, he remains clinically well with no signs of recurrent infection. Infection following LVAD implantation is difficult to manage and may require temporization until transplantation. The surgical approach in this case involved complete resection of the infected anastomosis and reconstruction of the aorta with a pericardial patch. However, the remainder of the outflow graft, the LVAD housing and the driveline were left in situ. While previous cases of intracerebral pseudoaneurysms have been reported after LVAD implantation, mycotic aortic pseudoaneurysms are uncommon and present a unique challenge. Non-infected pseudoaneurysms may be treated through a variety of means, including direct suture repair, the use of pericardial patches, or even endovascular stent grafting to exclude the aneurysm1Civilini E. Bertoglio L. Rinaldi E. Chiesa R. TEVAR for ruptured mycotic aneurysm in a patient with a left ventricular assist device.J Endovasc Ther. 2012; 19: 370-372Crossref PubMed Scopus (4) Google Scholar,2Combes S. Chabrot P. Camilleri L. Cardiovascular flashlight. Mycotic aneurysm of the ascending aorta: an unusual infectious complication of left ventricular assist device explantation.Eur Heart J. 2009; 30: 1079Crossref PubMed Scopus (1) Google Scholar. However, surgical intervention must remove all infected tissue and synthetic material, which may leave a large defect. To our knowledge, this is the first reported case of infected pseudoaneurysm at the site of an outflow graft anastomosis in a patient dependent on mechanical circulatory support. A case report of an acute non-infected pseudoaneurysm in a LVAD patient treated with endovascular stenting has been described3Tay S. Zaghloul M.S. Shafqat M. et al.Totally percutaneous endovascular repair for ruptured abdominal aortic aneurysms.Front Surg. 2022; 9 (Published 2022 Oct 21)1040929https://doi.org/10.3389/fsurg.2022.1040929Crossref PubMed Scopus (0) Google Scholar. A delayed mycotic pseudoaneurysm at the remnant of an outflow graft in a patient in whom the LVAD was removed at the time of transplant has also been described; this patient was treated with endovascular stenting followed by removal of the infected graft4Yamane K. Bogar L.J. DiMuzio P.J. et al.Contained rupture of a pseudoaneurysm of the descending thoracic aorta related to remnant outflow graft of left ventricular assist device after heart transplantation.Ann Thorac Surg. 2012; 94: 1345-1348https://doi.org/10.1016/j.athoracsur.2012.02.023Abstract Full Text Full Text PDF PubMed Scopus (1) Google Scholar. Compared to our case report, treatment strategies for the above two cases are more straightforward in that decisions about therapies did not involve the potential need for removal of an infected device on which the patient was hemodynamically dependent, nor the potential for urgent transplantation. Given the aggressive infection and the extent of prosthetic material in the mediastinum, excision of as much of the outflow graft as possible and aggressive debridement of all infected aortic tissue and patch repair with bovine pericardium is needed to reduce the risk of reinfection while awaiting transplant. In our case, our strategy was to treat the life-threatening component of his presentation first and urgently list him for transplant. Our team considered other strategies such as full pump exchange or pump removal and bridging with biVAD or extracorporeal life support, but, we felt that these other options had significant drawbacks including some residual hardware/cannulae in the mediastinum or potential time- and oxygenator-related complications in the case of ECLS. Ultimately our heart failure team consensus was that our chosen strategy would be safest, anticipating that the patient had a high likelihood of receiving a heart offer in a reasonable amount of time given his high status, blood group and no significant sensitization. No funding source.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.