|Home | About | Journals | Submit | Contact Us | Français|
The presence of prohibitive risk may preclude usual surgical management. Such was the case for a critically ill, 60-year-old woman who presented with concomitant, life-threatening conditions.
The patient presented with acute central cord syndrome and lower-extremity paraplegia after completing a 6-week course of intravenous antibiotics for methicillin-sensitive Staphylococcus aureus bacteremia and osteomyelitis of the thoracic spine. Radiologic examination revealed bony destruction of thoracic vertebrae T4 through T6, impingement on the spinal cord and canal by an inflammatory mass, and a separate 2.5-cm mycotic aneurysm of the infrarenal aorta.
The clinical and radiologic findings warranted immediate decompression and stabilization of the spinal cord, aneurysmectomy, and vascular reconstruction. However, the severely debilitated patient could not tolerate 2 simultaneous open procedures. She underwent emergent endovascular exclusion of the mycotic aneurysm with a stent-graft, followed immediately by laminectomy and stabilization of the thoracic spine.
Intraoperative microbiology specimens showed no growth. The patient was maintained on prophylactic antibiotic therapy for 6 months. Fourteen months postoperatively, her neurologic function was near full recovery, and neither surveillance blood cultures nor radiologic examinations showed a recurrence of infection or aneurysm.
Although the long-term outcome of endovascular stent-grafts in the treatment of culture-negative mycotic aneurysms is unknown, the use of these grafts in severely debilitated patients can reduce operative risk and enable recovery in the short term.
Aneurysmectomy and wide débridement of infected tissue, combined with either extra-anatomic bypass or in situ graft placement, can be an effective treatment for mycotic aortic aneurysms.1–4 However, the presence of unfavorable anatomy, a hostile operative field after previous surgery, a contained rupture, or a physiologically debilitated state can elevate operative risk to a prohibitive level. Herein, we describe our rationale for the endovascular exclusion of a mycotic aortic aneurysm in a debilitated patient for whom open surgery posed an unacceptably high risk.
In July 2005, a 60-year-old woman with a history of tobacco use and hypertension presented at our hospital with acute central cord syndrome and bilateral lower-extremity paraplegia. She had recently completed a full course of antibiotic therapy as treatment for osteomyelitis of the thoracic spine and for methicillin-sensitive Staphylococcus aureus bacteremia, which was thought to have been caused by incomplete treatment of a previously diagnosed pneumonia. Surveillance blood cultures after the full course of antibiotic therapy had not revealed growth.
Radiologic examination revealed bony destruction of thoracic vertebrae T4 through T6 (Fig. 1), impingement on the spinal cord and canal by an inflammatory mass (Fig. 1), and an anatomically distinct 2.5-cm, saccular and eccentric mycotic aneurysm of the infrarenal aorta (Fig. 2). A physical examination revealed that the patient had bilateral lower-extremity motor and sensory loss, and she was emaciated and weak after her experience with pneumonia, bacteremia, and osteomyelitis. The aortic aneurysm had not been present on either computed tomography or magnetic resonance imaging that had been performed 2 months earlier; its rapid expansion and potential for rupture were of substantial concern.
The patient's physiological state was such that she was unable to tolerate 2 major open procedures simultaneously. Furthermore, treating only the neurologic problems would have exposed the patient to potential aneurysmal rupture. We therefore decided to exclude the aneurysm percutaneously by use of an AneuRx® 16-mm × 5.5-cm stent-graft (Medtronic, Inc.; Minneapolis, Minn) (Fig. 3). This procedure immediately preceded laminectomy and stabilization of the thoracic spine.
Microbiology specimens obtained during the laminectomy yielded no growth. The patient continued on prophylactic antibiotic therapy for 6 months. Fourteen months postoperatively, her neurologic function continued to improve and was near normal; and neither surveillance blood cultures nor radiologic examinations showed recurrence of infection or aneurysm.
Although usual treatment of a mycotic aneurysm involves resection, débridement, and vascular reconstruction, it may be necessary to consider other measures in patients who face prohibitive risk from surgery because of their debilitated physical condition or difficult anatomy. As with our patient, such risk can be decreased by the use of endovascular techniques in order to achieve either a satisfactory long-term outcome or an acceptable temporary outcome that will enable later intervention.
Although stent-grafts are helpful in certain situations, their use in the treatment of mycotic aneurysms is not without risk. Lee and colleagues5 reported 3 deaths after the use of stent-grafts to treat mycotic aneurysms in various anatomic locations: 2 patients died of sepsis within 30 days, and the other died of aneurysmal rupture at 6 months. Ishida and coworkers6 reported an operative death in a patient with a mycotic aneurysm; 2 days after placement of a thoracic aortic stent-graft, a type 1 endoleak occurred, and an open surgical attempt at salvage was required.
Few data are available regarding long-term outcomes (morbidity, death, and need for further intervention) after such use of stent-grafts. Forbes and Harding7 described the case of a patient who developed bacteremia and aortoenteric fistula 3 years after the stenting of a mycotic aneurysm that had been positive for Salmonella. The patient required enterectomy, removal of a previously placed aortic stent-graft, and axillo–bifemoral bypass. There is no consensus regarding the long-term usefulness of suppressive antibiotics in such cases.
Several authors have reported acceptable initial outcomes when aortic stent-grafts have been used to treat mycotic aneurysms in difficult anatomic locations, including the thoracic aorta,8–13 the abdominal aorta,14–19 or near viscera;16,19 aneurysms that were associated with contained rupture;9,18 or aneurysms that required entry into hostile operative fields after previous surgery.8–10 Acceptable results have also been achieved with aortic stent-grafts that have been used to treat pseudoaneurysms that developed after initial open repair failed.9,10
Multiple factors can influence the decision to accept an immediately good endovascular outcome as a permanent solution or to consider it as a bridge to open surgical intervention. These include the patient's current or evolving physical condition and risk factors for comorbidity, the virulence of the specific infective organism, the persistence of positive blood cultures, the presence of contiguous infection or abscess, the anatomic location of the lesion, and changes in the serial clinical, radiologic, and laboratory evaluations on follow-up. The relative importance of each factor or combination of factors must be weighed for each patient while the experience and technical capabilities of the institution are also considered. Koeppel and colleagues14 successfully treated a debilitated patient (critically ill with a mycotic infrarenal abdominal aortic aneurysm and a contiguous retroperitoneal abscess) by means of endovascular stent-graft exclusion and percutaneous drainage of the abscess. At 1-year follow-up, that patient was alive and well, and magnetic resonance imaging showed no evidence of the abscess.
The differential diagnosis of our patient's lesion included penetrating ulcer of the aorta. However, we believed that this was an unlikely diagnosis, because of the lesion's development during a period of sepsis and the lack of diffuse or locally extensive atherosclerosis that usually is found with penetrating ulcers. Furthermore, the rupture of penetrating ulcers is not uncommon. In a series at Yale, Tittle and colleagues20 found rupture in 33% of 26 patients with untreated penetrating aortic ulcers. Therefore, the same concern regarding potential rupture exists regardless of the diagnosis. Although both Sundt21 and Absi and associates22 have published criteria that should apply in the nonoperative management of penetrating ulcers, the rapid increase in the size of our patient's lesion and the potential hemodynamic fluctuation during emergent spinal surgery under general anesthesia were risk factors for rupture that would have precluded nonoperative management if indeed our patient had had a penetrating ulcer.
In the management of either a mycotic aortic aneurysm or a penetrating ulcer with associated pseudoaneurysm and a separate mycotic process, the standard of care remains open operative reconstruction, with removal and débridement of infected tissue. By contrast, in a salvage situation that involves a critically ill patient who cannot tolerate surgery, the use of a stent-graft can serve as an intermediate measure that enables definitive intervention later.
Although 25% of all mycotic aneurysms are culture-negative,23 little is known about the natural history of patients with mycotic aneurysms—such as our patient —whose cultures are negative after antibiotic treatment for a specific organism. We recognize the continuing possibility of clinical or radiologic deterioration, and we are prepared to intervene immediately if findings during our close, ongoing follow-up warrant surgery.
We conclude that the use of endovascular stent-grafts to exclude culture-negative mycotic aortic aneurysms can reduce the risks of surgery in a severely debilitated patient and enable recovery in the short term. Therapies should be tailored to individual patients by evaluating operative risk, institutional experience, and potential long-term outcome.
Address for reprints: Thomas J. Takach, MD, Department of Cardiothoracic and Vascular Surgery, Carolinas Heart Institute, Carolinas Health Care System, 1000 Blythe Blvd., Charlotte, NC 28203. E-mail: ten.epacsten@hcakatjt