Figure 88-3. CT scan demonstrating perigraft fluid surrounding a prosthetic aortic graft as a sign of an aortic graft infection.

Figure 88-4. CT scan demonstrating perigraft air (A) surrounding an endograft as a sign of graft infection associated with a

possible aortoenteric fistula.

Leukocyte Scintigraphy

Leukocyte scintigraphy may be performed using either 111Indium- or 99mTechnetium-labeled leukocytes

to localize infection within the arterial wall or graft. A theoretical advantage of leukocyte scintigraphy

is the diagnosis of indolent infections with low-virulence organisms for which the symptoms are most

ambiguous. Unfortunately there are numerous pitfalls to this modality. Leukocyte scintigraphy is timeconsuming, labor-intensive, and prone to false-positive results due to undesired cross-labeling of

platelets that may bind to grafts.54,55 Uptake of the radionuclide-labeled white blood cells requires

intact chemotaxis and a minimum total white blood cell count of 2,000/μL.55 The net effect of these

issues is that the sensitivity and specificity for tagged white blood cell scanning is highly variable

between institutions, with sensitivity and specificity both ranging between 50% and 100%.51,54,55

Single Photon Emission Tomography (SPECT) Scanning

SPECT scanning, when combined with CT scanning, may enhance the specificity of leukocyte

scintigraphy in the detection of vascular infections.56,57 While there are limited data available at this

time, the early results appear promising. Erba recently evaluated SPECT–CT in 55 subjects with

suspected vascular graft infections. SPECT–CT reduced the number of false positives by 37%, compared

to CT alone. The sensitivity and specificity of SPECT–CT was 100%, which was superior to SPECT or CT

alone.57

Fluorodeoxyglucose Positron Emitting Tomography (FDG-PET)

FDG-PET utilizes glucose labeled with 18fluoride, which is utilized by cells proportional to their

metabolic activity. Initially, this technique was utilized to detect occult metastases in oncology patients.

Recently the protocols have been adapted for the detection of occult arterial infections.58,59 FDG-PET

can be advantageous in diagnosing endograft or peripheral stent infections since the images are less

prone to metal artifact. Compared with leukocyte scintigraphy, FDG-PET offers the advantages of

shorter examination times, less radiation exposure, higher resolution, and improved interobserver

agreement.60,61 Unfortunately there is a high false-positive rate for diagnosing arterial graft infections

2529

because of the inflammatory response associated with prosthetic grafts.58 Poor uptake of the FDGglucose can also yield nonspecific results, which is a limitation of using this test in diagnosing lowvirulence infections. While the sensitivity and specificity of FDG-PET alone are inferior to CT scan

alone, combining FDG-PET with CT scanning markedly improves the localization of infection and the

sensitivity and specificity of the examination. Using the approach, Spacek reported a diagnostic accuracy

exceeding 95% in a series of 95 grafts.59 For FDG-PET-CT, Bruggink reported a sensitivity, specificity,

positive predictive value, and negative predictive value of 93%, 70%, 82%, and 88%, respectively,

which was superior to FDG-PET alone. 60 Further study is required to define the threshold for a positive

FDG scan in the setting of cavitary arterial infections.61–63 Drawbacks of the fusion technology include

longer study times and increased radiation exposure. Furthermore, expertise with FDG-PET-CT

techniques is not widely available. Despite these limitations, some authorities foresee the combined

modality becoming the study of choice in the future.61

Adjunctive Diagnostic Modalities

Arteriography, endoscopy, and image-guided biopsy should be viewed as adjunctive diagnostic

modalities that have a limited role in diagnosing graft infections. Arteriography may be useful in

planning reconstruction. Rarely, an occult AEF can be discovered with angiography by filling of the

enteric structure.64 Esophagogastroduodenoscopy (EGD) is typically performed to exclude other sources

of upper gastrointestinal hemorrhage in a patient with a suspected AEF. EGD was successful in

diagnosing AEF in approximately 25% cases in one small series of primary AEFs.40 In spite of the low

diagnostic yield, the authors recommended endoscopy in hemodynamically stable patients with a

suspected AEF because of the devastating consequences of a missed diagnosis. Image-guided biopsy of

the fluid or rind surrounding the affected vascular structure has been advocated when the diagnosis of

graft infection is equivocal.65,66 This approach could potentially distinguish between patients with

benign early postoperative perigraft gas versus an early graft infection. However, sufficient sampling is

difficult to achieve in all cases, as the tissue is often fibrotic and in close proximity to other major

vascular structures and bowel.66 Further study will be required to clarify the role of this modality in the

evaluation of arterial infections.

CONSERVATIVE MANAGEMENT

Primary Arterial Infections

Hsu et al. provided insight on the outcome of conservative management of subjects with primary aortic

infections.67 Hsu examined 22 subjects with infected aortic aneurysms with perceived contraindications

to surgery. The predominant organism in the cohort was Salmonella, with two-thirds of subjects having

infections of the abdominal aorta. All subjects received culture-specific intravenous antibiotics for at

least 6 to 8 weeks, or until normalization of white blood cell count, daily temperatures, and C-reactive

protein levels. After the acute phase treatment, oral antibiotics were continued indefinitely. In Hsu’s

series, the inhospital mortality was 50% due to rupture or sepsis. The 1-year survival rate was 32%.

Patients with Salmonella infections had a lower aneurysm-related mortality rate than patients with nonSalmonella infections.

Arterial Graft Infections

Calligaro advocated for complete or partial prosthetic bypass preservation in subjects with severe

comorbidities precluding surgical repair.68 Over a 20-year period, 51 extracavitary grafts (aortofemoral

limbs or lower extremity bypasses) were treated with graft preservation and aggressive wound

management without surgical graft resection. When wounds were present, aggressive debridement of

all necrotic tissue and exudate surrounding the artery was performed. Nine cases utilized adjunctive

muscle flaps to aid healing. Povidone–iodine 1% or antibiotic-soaked dressing changes were performed

three times a day. With this regimen, 2 of 51 patients (4%) experienced major hemorrhage from the

infected graft at 3 and 12 months after presentation. Eleven of the patients (22%) failed the treatment

regimen, with nonhealing wounds or recurrent infections requiring total graft excision. Thirty-two

subjects (63%) were ultimately treated successfully. Calligaro concluded that conservative management

of extracavitary grafts is an acceptable option in the absence of sepsis or hemorrhage, especially when

graft excision and revascularization is not feasible due to patient comorbidities. Pseudomonas infections,

however, were particularly prone to failure with this regimen.

2530

Building on the early reports of graft preservation therapy, subsequent reports have emerged utilizing

vacuum-assisted negative pressure therapy with antibiotics and wound debridement to preserve

extracavitary grafts with Szilagyi III wounds.69,70 Wound healing with these regimens have ranged

between 82% and 91% with excellent graft preservation rates.69,70 Episodes of hemorrhage remain an

issue with this approach,69 and the failures have been associated with Pseudomonas.69,70 Based on these

reports, it appears that carefully selected patients with early infections involving extracavitary grafts

may be treated without surgical graft excision with acceptable results.

Calligaro et al. used a nonoperative approach in six patients with intracavitary graft infections with a

hostile abdomen or severe medical comorbidities.71 All subjects underwent either open or percutaneous

placement of drains along the graft with instillation of 10 mL of 1% povidone–iodine in 100 mL of

normal saline or antibiotics (neomycin or a cephalosporin) three times a day. All subjects were treated

concurrently with culture-directed intravenous antibiotic therapy for 6 weeks, followed by 6 months of

oral antibiotic therapy. Of these 6 subjects, only one died due to sepsis. One patient ultimately required

total graft excision and survived. The remaining patients (66%) were alive at last follow-up. These

authors showed that in highly selected patients, conservative management can have reasonable

outcomes. In the absence of larger series, this approach should be reserved for patients who are not

reasonable candidates for graft excision.

SURGICAL MANAGEMENT OF AORTIC ARTERIAL AND GRAFT

INFECTIONS

Surgical treatment is based upon the central tenets of appropriate resuscitation, preoperative planning,

resection of all infected material and tissue, and arterial revascularization. The optimal method of

arterial reconstruction remains unresolved, as few direct comparisons between approaches have been

published. Based on our extensive institutional experience with a variety of techniques, we have

concluded that no single approach is suitable for all patients. We advocate tailoring the treatment

algorithm based on each patient’s clinical presentation and the local surgical expertise.

Aortic Infections Presenting with Hemorrhage or Sepsis

A minority of patients present with septic or hemorrhagic shock. For these patients, rapid intravenous

access and resuscitation are essential prior to emergent surgery. In the authors’ experience, the

complexity of the operations required to manage these patients mandates a rapid CTA to obtain

essential anatomic information. It is recognized that some of these patients will have either acute

kidney injury or chronic kidney disease, but managing the life-threatening complications of the aortic

graft infection outweighs the risk of contrast-induced nephropathy. Ideally, blood and wound cultures

are obtained prior to the administration of broad-spectrum gram-positive and gram-negative antibiotic

therapy. The surgical approach to managing patients presenting with hemorrhage or sepsis should be

individualized based on the patient’s anatomy and the expertise of the surgeons.

Endografting as Bridge Therapy

Endografts are being used with increasing frequency as a method to temporize hemorrhage in patients

with aortoenteric or aortobronchial fistulas.38,39,72,73 When the anatomy permits, endografting may

temporarily prevent exsanguination and allow time for resuscitation, medical stabilization, and further

investigations to plan definitive surgery.72 Endografting should not be viewed as definitive therapy

since recurrent hemorrhage occurs in majority of cases without further surgical treatment.39,73

Definitive surgery should be planned within 2 weeks of endografting, since most episodes of recurrent

bleeding occurred greater than 2 weeks after initial endografting.73 The exception to this principle is an

aortobronchial fistulae for which endografting may be viewed as definitive therapy.38

Definitive Surgery for AEF

Definitive surgery for an AEF involves resection of the infected graft, repair of the enteric defect, and

revascularization of the lower extremities. The most appropriate surgical approach to achieve these

goals has been the focus of considerable debate. For decades, the most common approach to AEFs was

total graft excision with ligation of the infrarenal aortic stump and extraanatomic bypass.74–78 In recent

years, a variety of groups have touted the merits of in situ reconstruction of the aorta after graft

excision, using a variety of conduits. The advantages, disadvantages, and technical details of these

approaches are discussed later in the chapter (see “Aortic Infections in the Stable Patient”).

2531