Surgical Treatment of Multiple Mycotic Aneurysms in the Ascending Aorta, Aortic Arch, and Descending Aorta

Abstract

We report a clinical case of multiple mycotic aneurysms, in the ascending aorta, aortic arch, and descending aorta. The patient underwent surgery to replace the ascending aorta and aortic arch by means of a highly modified “elephant trunk” technique and with the aid of arterial cannulation from the right subclavian artery, which provided antegrade cerebral perfusion. Samples of purulent material taken from the aneurysmal wall yielded cultures positive for Staphylococcus aureus. The patient was treated with antibiotics for 6 weeks and then underwent a 2nd procedure for the aneurysmal resection of the descending thoracic aorta and the abdominal aorta, through a thoracic laparo-phrenicectomy. We comment on the clinical and surgical aspects of the case. (Tex Heart Inst J 2003;30:225–8)

We report a clinical case of multiple mycotic aneurysms, in the ascending aorta, aortic arch, and descending aorta. Mycotic aneurysms of the aorta account for 0.85% of surgically treated aneurysms. Cases of multiple aneurysms of the aorta, such as the one reported here, are exceptional. We found only 1 similar case in the English medical literature. ¹ There is another published report of a case caused by pancreatitis, but it involved a single, local, retroperitoneal aneurysm. ² This case is also unusual in that we cannulated the right subclavian artery for arch replacement and used the “elephant trunk” technique without subsequent attachment to the distal aorta.

Case Report

A 66-year-old man had a history of arterial hypertension, transient cerebrovascular accident, and claudication in the lower limbs. Four months before we evaluated him in April of 2001, he presented with an episode of pancreatitis complicated by a pancreatic abscess that had to be surgically drained. Staphylococcus aureus grew in the culture taken from the abscess, and this infection was treated with antibiotic agents. The patient progressed satisfactorily; however, he was readmitted a month later with symptoms of thrombophlebitis of the left internal jugular vein and cellulitis of the adjacent subcutaneous tissue. He was treated with anticoagulants.

Two months after his return home, the patient was readmitted with substantial and recurrent hemoptysis, a weight loss of 10 kg during the 2-month interval, and an absence of fever. Fibrobronchoscopy was performed, during which a bronchial hemorrhage was provoked. Arteriography, performed with the aim of achieving selective embolization of the bronchial arteries, revealed saccular aneurysms of the aortic arch, which adhered to the apex pulmonis (Fig. 1). Another possible aneurysm was observed in the greater curvature of the ascending aorta, and yet a 3rd in the descending aorta at the diaphragm.

Fig. 1 Aortography shows the saccular aneurysm at the aortic arch (arrow).

A thoracic computed tomographic (CT) scan revealed a neoformation that affected the upper left pulmonary lobe; this solid mass, surrounded by atelectatic and cicatricial tissue, invaded the visceral pleura and maintained continuity with the aorta at the aortic arch. The aortic arch aneurysm so confirmed lay in the left lateral aortic wall, at the aperture of the left carotid and left subclavian arteries. The CT scan also confirmed the presence of a 2nd aneurysm in the greater curvature of the ascending aorta, adjacent to the superior vena cava, and a 3rd in the descending aorta at the diaphragm, above the celiac trunk (Fig. 2). In addition, CT imaging revealed a 4th lesion—an arteriosclerotic aneurysm of the abdominal aorta, which extended from a point below the renal arteries to the iliac bifurcation, which itself was heavily calcified.

Fig. 2 Thoracoabdominal computed tomographic scan shows another saccular aneurysm, at the vertebral part of the diaphragm and above the aperture of the celiac trunk.

The patient underwent urgent surgery via a median sternotomy. The right subclavian artery was cannulated for arterial perfusion. We found a fibrous, petrous mass, which completely united the aneurysmal wall of the aorta with the pericardium and the superior vena cava. The aneurysm had ruptured, expelling purulent material into the pericardium, which contained the rupture.

We sutured a 30-mm-diameter tube of woven Dacron to the supracoronary aorta with 3-0 Prolene continuous suture material. After lowering the systemic temperature to 20°C, we achieved systemic circulatory arrest (except for the supra-aortic trunks) and then initiated antegrade cerebral perfusion through the right subclavian artery. The aortic arch was opened along its complete length, to a point more than 5 cm distal to the left subclavian artery aperture. The aneurysm was also found to be ruptured in this segment, with the rupture contained by the pleura and by the pulmonary parenchyma, which had fibrosed on the aorta. We sectioned the descending aorta transversely and continuously sutured a 28-mm Dacron tubular prosthesis to it, in the form of an “elephant trunk” (with the prosthesis invaginated in such a manner that about 8 cm of the tube was left hanging in the descending aorta). After performing the anastomosis, we disinvaginated the prosthesis and made a knot in it distally, without attachment to the distal aorta. To this Dacron prosthesis we sutured the 3 supra-aortic trunks, after which we unclamped the brachiocephalic trunk and restored the systemic circulation. The 2 prostheses (for the ascending aorta and the aortic arch) were sutured together by end-to-end anastomosis. The rest of the procedure was completed without incident.

The duration of circulatory arrest was 48 min, of myocardial ischemia was 110 min, and of extracorporeal circulation was 156 min. The postoperative course was favorable, with extubation after 10 hours and normal neurologic and hemodynamic results.

Colonies of methicillin-sensitive S. aureus grew in the sample of pus taken from the aneurysm. The patient was therefore treated continuously for 6 weeks with rifampin, norfloxacin, and cloxacillin (in accordance with antibiogram results), before he underwent surgery on the descending aorta through a thoracic laparo-phrenicectomy. For this procedure, we did not use extracorporeal circulation or spinal cord protection. The distal thoracic aorta and the abdominal aorta were replaced with the use of sequential clamping techniques. After ligating the 2 intercostal branches, we clamped the distal thoracic aorta at a proximal level above the celiac trunk and at a distal level below the renal arteries (also controlling the visceral branches). Upon opening the aorta, we found an aneurysm that had ruptured through the aortic wall but had been contained by the viscera; this aneurysm was approximately 5 cm above the celiac trunk and showed no sign of active infection. We used a 22-mm Dacron tube to replace a segment, about 10 cm in length, of distal thoracic and proximal abdominal aorta, taking care to bevel the aortic ends proximally and distally and to respect the apertures of the visceral arteries. The clamping time was 35 min. The intercostal arteries were not reimplanted.

After that, we clamped the infrarenal aorta and replaced both iliac arteries and the abdominal aorta with a bifurcated 18-mm Dacron tube. The rest of the procedure was performed without incident. Again, the patient's postoperative course was favorable, and he did not suffer any neurologic sequelae. He was kept on antibiotic treatment for 6 weeks. At his 2-year follow-up visit, his laboratory and radiologic results were normal.

Discussion

In many cases of mycotic aortic aneurysm, the cause is not discovered because the infectious agent is never isolated. In the case that we are discussing, the evident source of the infection was the pancreatic abscess that had been drained months earlier, although not before S. aureus had colonized the bloodstream and subsequently some atheromatous aortic plaques. This same sequence was found in an experimental study in animals in which S. aureus was inoculated into different segments of the thoracic aorta in order to yield information about the infection rate and the development of aneurysms in the aortic wall. ³ In only 10% of the cases did mycotic aneurysms develop; in the rest, the infections resolved spontaneously and the sites of infection healed without aneurysm formation.

In this case, the patient did not present with a fever during the evolution of the aortic infection; instead, he presented with episodes of hemoptysis. This form of presentation, accompanied by the patient's weight loss and the apparent neoplasms seen upon CT scanning, made us wonder whether we were facing 2 simultaneous conditions: aneurysms and carcinoma of the lung. The serious prognosis, and the possibility of fatal hemoptysis at any moment, led us to perform urgent surgery.

There is a general impression that the aortic homograft resists infection better than do grafts of manmade material, so it is the graft of choice in repairing mycotic aortic aneurysms. In a series published by Knosalla and colleagues, ⁴ 6 of 8 patients undergoing surgery for these lesions had successful outcomes with cryopreserved homografts. In our patient, the urgency of surgical intervention and the lack of ready access to homografts obliged us to use Dacron. The therapeutic alternative of endovascular stent-grafts was ruled out due to our lack of experience with these devices and to the unknown nature of the disease process in our patient. In a series published by Semba and co-authors, ⁵ complete thrombosis of mycotic thoracic aneurysms was achieved in 3 patients, although 1 patient required a 2nd implantation after migration of the 1st stent-graft and the other 2 required surgery for aneurysms at other levels.

The regimen of antibiotic treatment for mycotic aneurysms varies greatly. Some investigators advocate antibiotic treatment prior to surgery. There are even those who maintain that the antibiotic treatment should be lifelong. In a review by Cina and associates, ⁶ no differences were found in survival or recurrence between clinical series with lifelong antibiotic treatment and clinical series with treatment over a few months. In our patient, we treated the mycotic aneurysms as we would bacterial endocarditis, although the patient had not presented with fever or leukocytosis at any time during the evolution of the infection.

The prognosis for patients with mycotic aneurysms does appear to be substantially worse than that for patients with uninfected aneurysms; and it appears to be worse still in the presence of periaortic infection, S. aureus, aneurysmal rupture, or suprarenal location of the lesion. ⁷ Our patient had all of these variables.

In regard to the surgical treatment, we would like to highlight 2 points of interest. One is our use of arterial cannulation through the right subclavian artery to perform antegrade cerebral perfusion. This technique enables systemic circulatory arrest times to be extended, minimizing the risk of cerebral ischemia. We believe this measure to be indicated in complex surgery of the aortic arch in which circulatory arrest times are going to be long. The 2nd point is our use of the elephant trunk technique without subsequent attachment to the distal aorta, which enabled hemostatic anastomosis of the graft and the descending aorta when surgical access was difficult.