| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vascular and Interventional Radiology |
1 From the Department of Radiology (T.A.M., A.W.S., C.M.J.) and Division of Vascular Surgery (G.S.O., J.M.P.), Mayo Clinic and Foundation, 200 First St SW, Rochester, MN 55905; and Department of Radiology, Queen Elizabeth Hospital, Woodville, SA, Australia (M.L.T.). Received December 9, 2002; revision requested February 27, 2003; final revision received August 1; accepted August 22. Address correspondence to T.A.M. (e-mail: macedo.thanila@mayo.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Review of records of patients with surgical and/or microbiologic proof of infected aortic aneurysm obtained over a 25-year period revealed 31 aneurysms in 29 patients. This study included 21 men and eight women (mean age, 70 years). One radiologist reviewed 28 computed tomographic (CT) studies (22 patients underwent CT once and three patients underwent CT twice), 12 arteriograms (12 patients underwent arteriography once), eight nuclear medicine studies (six patients underwent nuclear medicine imaging once and one patient underwent nuclear medicine imaging twice), and three magnetic resonance (MR) studies (three patients underwent MR imaging once). Features evaluated included aneurysm size, shape, and location; branch involvement; aortic wall calcification; gas; radiotracer uptake on nuclear medicine studies; and periaortic and associated findings. The location of infected aortic aneurysms was compared with that of arteriosclerotic aneurysms.
RESULTS: Aneurysms were located in the ascending aorta (n = 2, 6%), descending thoracic aorta (n = 7, 23%), thoracoabdominal aorta (n = 6, 19%), paravisceral aorta (n = 2, 6%), juxtarenal aorta (n = 3, 10%), infrarenal aorta (n = 10, 32%), and renal artery (n = 1, 3%). Two patients had two infected aortic aneurysms. CT revealed 25 saccular (93%) and two fusiform (7%) aneurysms with a mean diameter at initial discovery of 5.4 cm (range, 1–11 cm). Paraaortic soft-tissue mass, stranding, and/or fluid was present in 13 (48%) of 27 aneurysms, and early periaortic edema with rapid aneurysm progression and development was present in three (100%) patients with sequential studies. Other findings included adjacent vertebral body destruction with psoas muscle abscess (n = 1, 4%), kidney infarct (n = 1, 4%), absence of calcification in the aortic wall (n = 2, 7%), and periaortic gas (n = 2, 7%). Angiography showed 13 saccular aneurysms with lobulated contour in 10 (77%). Nuclear medicine imaging showed increased activity consistent with infection in six (86%) of seven aneurysms. MR imaging showed three saccular aneurysms. Adjacent abnormal vertebral body marrow signal intensity was seen in one (33%) of three patients.
CONCLUSION: Saccular aneurysms (especially those with lobulated contour) with rapid expansion or development and adjacent mass, stranding, and/or fluid in an unusual location are highly suspicious for an infected aneurysm.
© RSNA, 2004
Index terms: Aneurysm, aortic, 94.733, 981.733 • Aneurysm, CT, 94.12916, 981.12916 • Aneurysm, MR, 94.12942, 981.12942
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Diagnostic imaging findings of infected aneurysms have been described mostly in case reports. The largest series come from the surgery literature, but these lack detailed information about radiologic findings. The purpose of this study was to determine the imaging characteristics of infected aortic aneurysms.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
The mean interval between the imaging examination and confirmation of the diagnosis at surgery was 5 days (range, 1–19 days). Infected aortic aneurysm was diagnosed with gross intraoperative findings (inflammation and purulence), clinical evidence of infection (fever, leukocytosis, or both), and aneurysm wall cultures that were positive for infection. Aneurysms in which the culture was negative for infection were analyzed individually. Patients had to have convincing surgical findings, a fever of unknown origin, and had to be using antibiotics before culture samples were collected. We excluded patients with infected aneurysms caused by local complication of prosthetic grafts and patients with infected aneurysms isolated in aortic branches or peripheral locations. Institutional review board approval was obtained, and informed consent was not required for this retrospective study.
Record and Image Review
Demographics, clinical characteristics, and surgical findings were collected from the patient records by either of two authors (G.S.O., T.A.M.). Available imaging studies were reviewed by a vascular radiologist (A.W.S.) with 29 years of experience. Though the reviewer was aware that all the images were of infected aneurysms, he was blinded to the clinical and surgical findings.
The following 51 imaging studies were available for review: 28 computed tomographic (CT) studies (22 patients underwent CT once, three patients underwent CT twice, and four patients did not undergo CT), 12 arteriographic studies (12 patients underwent arteriography once and 17 patients did not undergo arteriography), three magnetic resonance (MR) angiographic studies (three patients underwent MR angiography once and 26 patients did not undergo MR angiography), and eight nuclear medicine studies (six patients underwent nuclear medicine imaging once, one patient underwent nuclear medicine imaging twice, and 22 patients did not undergo nuclear medicine imaging). Multiple studies in a given patient were reviewed at the same time.
Three patients underwent multiple CT examinations within 24–50 days. CT examinations were performed both before and after administration of intravenous contrast material in seven patients, after administration of contrast material in 16 patients, and without contrast material in five patients. The section thickness used for CT was 6–10 mm. Aneurysms were measured on hard-copy film images. Angiograms obtained before 1992 were available in conventional film format, while angiograms obtained after 1992 were obtained with digital subtraction technology. All procedures were performed in at least two views. Nuclear medicine studies included six indium 111 (111In)-tagged white blood cell examinations, one hexamethylpropyleneamine oxime (HMPAO)-tagged white blood cell examination, and one gallium 67 (67Ga) examination. MR imaging and MR angiography were performed with a 1.5-T magnet. Three-dimensional time-of-flight MR angiograms were obtained with gadolinium enhancement in two of three patients, one of whom also underwent transverse T1-weighted MR imaging. One of these three patients underwent transverse T1-weighted MR imaging only.
Features evaluated both on CT and MR studies included aneurysm location, size, and shape (saccular or fusiform); branch involvement; adjacent soft-tissue mass, stranding, and/or fluid; and any additional findings. Calcification in the adjacent or involved aortic wall and presence of gas were evaluated on CT scans only. When consecutive CT studies were available for a given patient, the progression of findings was evaluated, and early subtle edema was defined as stranding in the surrounding fat present on the initial scan.
The location of infected aneurysms was compared with the location of arteriosclerotic aneurysms described in the literature. On conventional arteriograms, the shape of the aneurysm, contour lobulation, branch involvement, measurements, and any additional findings were described. Aneurysm measurements were given as a ratio relative to adjacent aortic diameter. Nuclear medicine scans were evaluated for presence or absence of increased radiotracer uptake. Microbiology of infected aneurysms was correlated with increased radiotracer uptake on nuclear medicine scans, the presence of gas on CT scans, vertebral body changes, multiplicity of aneurysms, and periaortic soft-tissue mass, stranding, and/or fluid.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Conventional Angiography
Twelve patients underwent arteriography. In one patient (8%), an arteriogram was negative for aneurysm and a CT study—which was obtained 2 weeks later—was positive for aneurysm. In two (17%) additional patients, two infected aneurysms were found, yielding a total of 13 aneurysms. Aneurysm shape was saccular in 13 (100%) of 13 aneurysms. Contour lobulation was appreciated in 10 (77%) of 13 aneurysms (Fig 1). Vertebral body erosion was noted in one (8%) of 13 aneurysms (Fig 2). On average, the ratio of the aneurysm diameter was increased by a factor of 2.2, and the length was 1.9 times the diameter of the healthy adjacent aorta. In 13 patients, aneurysms were located in the thoracoabdominal aorta (n = 4, 31%), the descending thoracic aorta (n = 3, 23%), the paravisceral aorta (adjacent to the origin of the celiac, superior mesenteric, and renal arteries) (n = 2, 15%), the infrarenal aorta (n = 2, 15%), the juxtarenal aorta (n = 1, 8%), and the renal artery (n = 1, 8%). Branch involvement was present in five (42%) of 12 patients and included the celiac artery (n = 4, 33%), the superior mesenteric artery (n = 3, 25%), and the renal artery (n = 3, 25%) (the sum of percentages is higher than the total because of the combination of branches involved).
|
|
|
|
|
|
|
|
|
|
|
MR Imaging
Three aneurysms were demonstrated with MR imaging, and saccular shape was present in all three. Two aneurysms (66%) were located in the paravisceral aorta and one (33%) was located in the infrarenal aorta. In one aneurysm, associated adjacent vertebral body abnormal bone marrow signal intensity consistent with that of osteomyelitis was noted (Fig 2). The average aneurysm diameter was 7.5 cm (range, 5–10 cm) and the average length was 8.0 cm (range, 6–10 cm). In one aneurysm, the celiac, superior mesenteric, and renal arteries were involved. Adjacent soft-tissue mass, stranding, and/or fluid were not appreciated in two aneurysms on T1-weighted images.
Nuclear Medicine
Increased radiotracer uptake was present in six (86%) of the seven patients examined with nuclear medicine. Aneurysms were located in the ascending aorta (n = 1, 14%), thoracoabdominal aorta (n = 2, 29%), juxtarenal aorta (n = 1, 14%), and infrarenal aorta (n = 3, 43%). The area of increased radiotracer uptake corresponded to the area of concern on the CT scan (Figs 1, 4).
Summary of Imaging Findings
Figure 6 illustrates the distribution of infected aortic aneurysms and compares with the distribution of arteriosclerotic aortic aneurysms reported in the literature (7,8). Aneurysm shape was saccular in 29 (94%) of 31 aneurysms and fusiform in two (6%). Vertebral body changes at CT, MR imaging, or radiography were noted in association with two (6%) of 31 aneurysms. Perianeurysmal gas was depicted on CT studies in two (7%) of 27 aneurysms, and periaortic soft-tissue mass, stranding, and/or fluid was seen on CT studies in 13 (48%) aneurysms. The average aneurysm diameter was 5.4 cm (range, 1–11 cm). The correlation of microbiologic results and imaging findings is shown in the Table.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
The pathophysiology of infected aortic aneurysm has been reported previously by several authors (9,10). Aortic infection develops when a pathogen offends a vulnerable vessel wall. The focus of infection can develop in a normal caliber vessel or an existing aneurysm. The infectious agent can either travel through the blood stream and harbor in the vasa vasorum of the arterial wall or implant on damaged intima, ulcerated arteriosclerotic plaques, or mural thrombus. Alternatively, the infectious agent can reach the artery either by contiguous spread of adjacent infectious process or by traumatic and/or iatrogenic inoculation. An infected aneurysm can occur anywhere in the vascular system.
Symptoms are nonspecific, laboratory data may be inconclusive, and a diagnosis of infected aortic aneurysm is often not suspected during clinical evaluation. Leukocytosis and an elevated erythrocyte sedimentation rate can be helpful in indicating this diagnosis but are not always present. Blood cultures can be negative for infection in as many as 47% of patients (5). Treatment consists of antibiotics and urgent surgery. The overall mortality rate is 16%–40% (3,11) but has been reported to be as high as 67% (12).
We present a series of imaging findings in 29 patients with infected aortic aneurysms. Imaging findings of infected aortic aneurysms are often suggestive of the diagnosis. Several characteristic features were commonly encountered in our series. Saccular shape, especially with a lobulated contour, was noted in all but two fusiform aneurysms. There was extensive mural thrombus in one of the fusiform aneurysms, and the surgical report described the infection confined to the mural thrombus.
The distribution of infected aortic aneurysms is different from the more commonly seen arteriosclerotic aneurysms. While the infrarenal aorta is by far the most common location (85%–87%) for arteriosclerotic aneurysms (7,8,13), in our series only 10 of 31 infected aneurysms were found in this location. The thoracic and abdominal aorta at or above the renal arteries was involved in almost 70% of infected aneurysms. A saccular aneurysm with lobulated contour was consistently seen with infected aneurysms on arteriography. Multiple aneurysms were uncommon and were seen in only two patients.
Cross-sectional imaging is valuable to demonstrate characteristic adjacent findings such as periaortic soft-tissue mass, stranding, and/or fluid. Helpful findings such as gas and vertebral body abnormalities were seen less commonly. Relatively rapid progression compared with the natural history of arteriosclerotic aneurysm was a feature that was present when sequential examinations were obtained. In the early stage, subtle periaortic inflammatory changes were detected; however, if these are not searched for, they may be overlooked. Nuclear medicine studies helped confirm the diagnosis in correlation with the cross-sectional findings; however, a study that does not reveal increased radiotracer uptake does not necessarily exclude this diagnosis. Previous reports of imaging findings in infected aortic aneurysm are mostly concordant with our findings. The CT appearance of infected aneurysms has been previously described (4,14–23).
In our report, we chose to use the descriptor terms periaortic soft-tissue mass, stranding, and/or fluid. The following terms have been used to describe similar findings that are commonly associated with infected aneurysms: periaortic inflammation and/or soft tissue (4,17), retroperitoneal mass (15,16,24), encasing or contiguous mass (19), and paraaortic mass (23). Other findings that have been described previously in association with infected aneurysms are presence of gas (4,14,15,21) and rapid expansion or development (14–16,20).
The association with vertebral body abnormalities has been recognized (4,14,18,19,22), including a review of the literature by McHenry et al that reported 70 cases of osteomyelitis and aortic infection (20). Lack of calcium in the involved aorta has been described as suggestive of an infected aneurysm (4,15,22); however, in our experience calcium was frequently found in association with infected aneurysms and was not a helpful differentiating feature from aneurysms of arteriosclerotic origin, which also have a high likelihood of calcification. The importance of this finding in the pathogenesis is unknown to us, since we did not have pathologic correlation. Renal infarct has been reported in a patient with an infected aneurysm; however, as previously hypothesized by Gonda et al (23), it appears that the adjacent location rather than the infected origin would be the cause. 111In white blood cell studies have been described as helpful in the diagnosis of infected aneurysms (16,25). Nuclear medicine examinations performed with white blood cell tagging (111In or HMPAO) or 67Ga are physiologic procedures that reproduce the migration of defense cells to the disturbed site. It cannot precisely aid in localization of the site of increased radiotracer uptake, but correlation with CT findings allows organ localization and facilitates the diagnosis of infected aneurysms.
Few reports have described MR findings in infected aortic aneurysms (16,26). MR angiography can demonstrate luminal caliber changes; however, T1-weighted sequences are necessary to demonstrate surrounding abnormalities that are typical of infected aneurysms. Ultrasonographic findings of infected aortic aneurysms have been previously described (27); however, disadvantages inherent to the technical limitations and aspects of this modality—such as patient body habitus, overlying bowel gas, and operator-dependent examination quality—limit the utility of this modality in the diagnosis of infected aortic aneurysms.
Awareness and recognition of imaging findings associated with infected aneurysms are critical for early diagnosis and institution of adequate therapy. Arteriography was formerly used (11,12) but has been largely replaced. At the Mayo Clinic, CT is the preferred imaging modality because it is widely available, fast, and able to depict associated findings, such as gas and periaortic findings. Furthermore, it is typically the first imaging examination that patients with vague general and abdominal complaints undergo. Newer techniques, such as multi–detector row helical CT and rapid three-dimensional reformatting, have further improved evaluation of the vascular system. CT has the advantage of showing both vessel anatomy and surrounding findings that are crucial in the diagnosis of infected aneurysms. Other cross-sectional imaging procedures, such as MR imaging, have similar diagnostic potential.
There are limitations to our study. First, the reviewer knew the diagnosis before the images were evaluated, which directed attention to the specific involved system and findings. Second, this was a retrospective study of patients from a single-institution referral center. Third, infected aortic aneurysm is an uncommon diagnosis, with a small number of cases. The statistical significance of the findings cannot be predicted with such small numbers and lack of control cases to calculate sensitivities and specificities. Fourth, the pool of imaging studies was heterogeneous, with use of different imaging modalities spanning a considerable period of time, during which the technology was changing. In addition, different modalities were used in all patients.
In summary, saccular aneurysm (especially lobulated), rapid expansion or development, and adjacent soft-tissue mass, stranding, and/or fluid in an unusual location are imaging findings that are highly suspicious for infected aortic aneurysm. Gas and vertebral body changes are helpful features, although they occur infrequently.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  W.-K. Lee, P. J. Mossop, A. F Little, G. J. Fitt, J. I Vrazas, J. K. Hoang, and O. F. Hennessy Infected (Mycotic) Aneurysms: Spectrum of Imaging Appearances and Management RadioGraphics, November 1, 2008; 28(7): 1853 - 1868. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A C PANKHANIA and E N KUMAR Abdominal pain in a young man -- a rare condition to consider Br. J. Radiol., October 1, 2008; 81(970): 837 - 838. [Full Text] [PDF]
|
|
|
|
|
|
C. S. Restrepo, A. Eraso, D. Ocazionez, J. Lemos, S. Martinez, and D. F. Lemos The Diaphragmatic Crura and Retrocrural Space: Normal Imaging Appearance, Variants, and Pathologic Conditions RadioGraphics, September 1, 2008; 28(5): 1289 - 1305. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R.-B. Hsu and F.-Y. Lin Surgery for infected aneurysm of the aortic arch. J. Thorac. Cardiovasc. Surg., November 1, 2007; 134(5): 1157 - 1162. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Rakita, A. Newatia, J. J. Hines, D. N. Siegel, and B. Friedman Spectrum of CT Findings in Rupture and Impending Rupture of Abdominal Aortic Aneurysms RadioGraphics, March 1, 2007; 27(2): 497 - 507. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.-B. Woo, L.-C. Cheng, and W.-C. Wong Mycotic Aortic Aneurysm Following Treatment of Pyogenic Vertebral Osteomyelitis Asian Cardiovasc Thorac Ann, October 1, 2006; 14(5): e102 - e105. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Reyes, J.-L. Navarro, C. Gamallo, and M.-C. delas Cuevas Type A aortic dissection associated with Dietzia maris Interactive CardioVascular and Thoracic Surgery, October 1, 2006; 5(5): 666 - 668. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M R Jones and J H Reid Emergency chest radiology: thoracic aortic disease and pulmonary embolism Imaging, September 1, 2006; 18(3): 122 - 138. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. C. Bower Commentary Perspectives in Vascular Surgery and Endovascular Therapy, June 1, 2006; 18(2): 190 - 191. [Abstract] [PDF]
|
|
|
|
 |
|
 R.-B. Hsu and F.-Y. Lin Risk factors for bacteraemia and endovascular infection due to non-typhoid salmonella: a reappraisal QJM, November 1, 2005; 98(11): 821 - 827. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C.-Y. Yang, K.-L. Liu, C.-W. Lee, Y.-M. Tsang, and S.-J. Chen Mycotic Aortic Aneurysm Presenting Initially as an Aortic Intramural Air Pocket Am. J. Roentgenol., August 1, 2005; 185(2): 463 - 465. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
