Background
Vascular graft prosthesis infection is a serious complication of vascular surgery. Its diagnosis is often elusive. [
18 F] fluoro-2-deoxy-D-glucose positron emission tomography/computed tomography (
18 F-FDG PET/CT) is an established imaging tool for tracing suspected inflammation or infection [
1,
2]. The fusion of
18 F-FDG-PET and computed tomography scans (
18 F-FDG-PET/CT) acquired in a single session enables precise localization of abnormal
18 F-FDG uptake [
2]. These properties make it a promising modality for the noninvasive diagnosis of graft infection [
1,
3‐
6].
We report two clinical cases where 18 F- FDG-PET/CT was very useful to arrive at a definitive diagnosis and to determine proper treatment.
Discussion
Graft infection following prosthetic vascular reconstruction is an uncommon but severe complication. The clinical presentation is sometimes subtle and non-specific and may occur long after surgery. Early diagnosis and treatment are essential to prevent further complications and to avoid unnecessary repeat surgery and removal of uninfected grafts [
1,
3]. Infection of aortic grafts is generally more serious than that of peripheral arteries, because repeated surgery is so invasive. If prosthetic graft infection is suspected, a patient should be hospitalized, and cultures from both the surgical wound and blood should be taken followed by the initiation of broad-spectrum antibiotics. At the same time, definitive diagnostic imaging is essential. CT scan is the gold standard for assessment of vascular graft infection, though ultrasound scanning and MRI are widely used noninvasive diagnostic modalities [
3]. Generally, intact grafts are covered by normal or sometimes fibrotic tissue and are well incorporated into the surrounding tissue. In contrast, infected grafts often appear unattached to the surrounding tissues, and are sometimes surrounded by turbid fluid, air bubbles, or purulent material. It is important to remember that: 1) peri-graft air is present in only about half of all cases, 2) a normal air collections can be seen on CT up to 4–7 weeks following reconstructive surgery [
1,
7], 3) the presence of lymphocele and non-infected hematoma in the vicinity of the implant hampers diagnosis, 4) low-grade infection also may increase the incidence of false-negative CT studies, 5) the accuracy of CT scans decreases in the chronic phase (acute vascular graft infection is diagnosed with a sensitivity and specificity reaching 100%, but that decreases to 55% in cases of chronic infection) [
3]. To increase diagnostic accuracy, repeat examinations over time are often required.
Whereas CT scans afford visualization of the structural changes secondary to infection, nuclear medicine techniques facilitate diagnosis on the basis of molecular biological changes.
67Gallium scintigraphy was acknowledged to be a useful examination for inflammation. Unfortunately, it has been shown to be of limited value in the assessment of vascular graft infection, with a relatively low reported sensitivity [
1]. While
99mTc- or
111In-labeled leukocyte scintigraphy is highly accurate for the detection of infected grafts, other infectious processes localized in the vicinity of the graft can account for false-positive results [
1,
3].
As opposed to conventional scintigraphy,
18 F-FDG-PET/CT has become an established imaging tool.
18 F-FDG is a marker of increased intracellular glucose metabolism and is taken up by malignant as well as infectious and inflammatory processes [
8]. The diagnostic sensitivity of
18 F-FDG/PET-CT is extremely high compared with other diagnostic modalities. In a prospective study including 39 patients with suspected vascular graft infections,
18 F-FDG-PET/CT had a sensitivity, specificity, positive predictive value, and negative predictive value of 93%, 91%, 88%, and 96%, respectively [
5]. According to Tokuda et al. [
9], an SUVmax value of greater than 8 in perigraft tissue is suspicious for graft infection. This cut-off value must be further assessed, as its experience was limited to only nine cases(thoracic graft infections; 4 positive
vs 5 negative cases). Actually in our Case 2, a definite graft infection, SUVmax around the graft was lower than 8.
The whole-body imaging capabilities of
18 F-FDG-PET/CT allow the detection of infectious foci distant from the graft. Faster examination time than conventional scintigraphy required is an additional clinical advantage of
18 F-FDG-PET/CT, with preliminary results available within 2 hours [
10].
One of the drawbacks of
18 F-FDG-PET/CT is its relatively low specificity in infection/inflammation imaging [
1,
4,
10,
11]. Fukuchi et al., using PET alone, showed that
18 F-FDG-PET had a sensitivity of 91% but a specificity of only 64%, as compared with CT, which had a lower sensitivity of 64% but a specificity of 86% [
4]. The early phase of wound healing is physiologically accompanied by an increased accumulation of
18 F-FDG [
5,
12]. Extensive linear FDG uptake of mild-to-moderate intensity along vascular grafts that have no evidence of infection has been previously described [
1,
4]. Venous thrombosis, sterile inflammation, vasculitis, chronic polyarthritis, or retroperitoneal fibrosis may all lead to an abnormal accumulation of
18 F-FDG.
18 F-FDG-PET/CT enabled us to find an infectious nidus remote from the graft in Case 1 and to diagnose aortic graft infection in Case 2. Graft infection could have preceded and, in fact, caused the lumbar pyogenic spondylitis in Case 1, as recurrent bacteremia with the same microorganism suggested an intravascular focus. But we speculate that it was an extraaortic source due to the lack of change in the appearance of the graft between the first and second 18 F-FDG-PET/CT studies. Regardless, 18 F-FDG-PET/CT enabled the diagnosis of an infectious focus, which could then be addressed.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
ET wrote the draft of the manuscript. HH and UT obtained the data and written consent. KT, TH and YT performed the literature review and participated in the manuscript writing of the paper. All authors have read and approved the final manuscript.