Case Report
Perigraft Seroma of Open Aortic Reconstruction
Audio Available | 
Share
 |
Perigraft seroma is a persistent enlarging sterile fluid collection confined in a nonsecretory fibrous pseudomembrane surrounding an intact prosthetic vascular graft [1, 2]. When a vascular graft in a native vessel is involved, such as in open abdominal aortic aneurysm repair, the accumulating collection expands the vessel diameter, but no demonstrable graft leak is seen on imaging. This clinicopathologic entity is not well described in the radiology literature. However, as more aortic reconstructions are performed and the consequent imaging surveillance increases, it is postulated aortic reconstruction perigraft seromas, although they rerely occur, may be more frequently recognized. The mechanism of development of this entity may have relevance to endoluminal aortic reconstructions and the increasingly recognized complication of continued abdominal aortic aneurysm growth in the absence of a demonstrable endoleak. We report a case of perigraft seroma complicating an open aortoiliac graft in one patient and similar imaging findings and clinical course in another.
| Case Reports |  |
|---|
A 60-year-old man presented with abdominal pain 5 years after undergoing uneventful elective repair of a 7.5-cm infrarenal nonleaking abdominal aortic aneurysm (Fig. 1A) with a polytetrafluoroethylene aortoiliac graft. He was afebrile with a normal WBC and erythrocyte sedimentation rate. Contrast-enhanced helical CT revealed an intact undilated aortoiliac graft. However, the aneurysm had expanded. It measured 11.5 cm in the maximal transverse diameter and had a wide-based focal wall protrusion anteriorly, presumably a form of vascular hernia at the site of previous aortic wall surgery (Fig. 1B). Sonography revealed the heterogeneous nature of the perigraft collection, which had anechoic pockets of fluid in solid components and septa (Figs. 1C and 1D). Conventional angiography confirmed aortoiliac graft patency, and no vascular leak was seen.
 View larger version (195K) | Fig. 1A. —60-year-old man with periaortoiliac graft seroma seen 5 years after elective open reconstruction of abdominal aortic aneurysm. Contrast-enhanced CT scan obtained before elective surgery shows infrarenal fusiform abdominal aortic aneurysm with maximal transverse diameter of 7.5 cm. Intraluminal thrombus and wall calcification may be seen. Aortic wall is thickened anterolaterally, thought to be result of previously contained leak or inflammatory aneurysm that caused wall inflammation and fibrosis. |
 View larger version (208K) | Fig. 1B. —60-year-old man with periaortoiliac graft seroma seen 5 years after elective open reconstruction of abdominal aortic aneurysm. Contrast-enhanced helical CT scan obtained 5 years after A shows intact nondilated aortoiliac graft. Aneurysm wall has expanded to 11.5 cm in transverse diameter. Wider-based focal protrusion may be seen anteriorly, presumably at site of previous aortic wall surgery. |
 View larger version (175K) | Fig. 1C. —60-year-old man with periaortoiliac graft seroma seen 5 years after elective open reconstruction of abdominal aortic aneurysm. Sonogram shows heterogeneous nature of perigraft collection. Note anechoic pockets of fluid in solid components and septa. |
At laparotomy, the aortic sac wall appeared thickened and inflamed; the duodenum and proximal jejunum adherent to the wall required resection of the small bowel. When the aortic wall was opened, tense gelatinous material was revealed surrounding the intact arterial graft. The proximal and distal anastomoses were intact. The sac contents were evacuated, and the sac was closed tightly around the graft.
Histologic examination of the aortic sac wall showed atheromatous change and a fibrinous exudate over the luminal surface; a patchy chronic inflammatory infiltrate was found only in the deeper layers near the adventitia. No evidence was found of acute infection in the sac wall or in the gelatinous contents. The gelatinous material contained no cells and was sterile. The patient's symptoms improved. Findings on routine follow-up CT performed 3 months after surgery revealed that the sac had returned to its preoperative size.
Similar imaging findings were noted in a 73-year-old man 5 years 6 months after he underwent open abdominal aortic aneurysm repair with a GORE-TEX (W. L. Gore, Flagstaff, AZ) graft for a 14-cm nonleaking infrarenal aneurysm. Imaging performed after surgery revealed a progressively enlarging abdominal aortic aneurysm (Fig. 2A,2B,2C,2D). No evidence of a graft leak was seen on multislice CT angiography. The patient has remained symptomatic from the mass effect of the aneurysm, experiencing back pain and constipation but no further sequelae. He is considered a poor surgical candidate.
 View larger version (175K) | Fig. 2A. —73-year-old man with periaortoiliac graft seroma monitored for 5 years 6 months after elective infrarenal abdominal aortic aneurysm repair. Contrast-enhanced CT scan obtained before surgery shows huge nonleaking infrarenal abdominal aortic aneurysm. Transverse diameter measures 14 cm, with intraluminal thrombus with minor laminated calcification. We found no pelvic extension. |
 View larger version (181K) | Fig. 2B. —73-year-old man with periaortoiliac graft seroma monitored for 5 years 6 months after elective infrarenal abdominal aortic aneurysm repair. Contrast-enhanced helical CT scans obtained 5 years 6 months after A show further enlargement of aneurysm sac wall to 16 cm in transverse diameter, but now with considerable inferior extension of seroma into pelvis. |
 View larger version (140K) | Fig. 2C. —73-year-old man with periaortoiliac graft seroma monitored for 5 years 6 months after elective infrarenal abdominal aortic aneurysm repair. Contrast-enhanced helical CT scans obtained 5 years 6 months after A show further enlargement of aneurysm sac wall to 16 cm in transverse diameter, but now with considerable inferior extension of seroma into pelvis. |
 View larger version (180K) | Fig. 2D. —73-year-old man with periaortoiliac graft seroma monitored for 5 years 6 months after elective infrarenal abdominal aortic aneurysm repair. Sonogram shows loculated nature of collection with solid and fluid components. |
| Discussion | |
|---|
Fluid between the open aortic graft and the sac wall is a normal finding on imaging in the period immediately after surgery. However, after 3 months, any perigraft hematoma or fluid should have resolved, and the vessel wall should approximate the graft. Uncomplicated graft incorporation involves periprosthetic fibrosis [1].
In the early postoperative phase, perigraft fluid cannot be differentiated from infection on the basis of imaging [3]. In the absence of a demonstrable leak, the differential diagnoses of the perigraft fluid collection are graft infection or a sterile fluid collection—lymphatic or serous. Most graft infections are clinically apparent, and imaging techniques are used primarily for diagnostic confirmation and surgical planning.
Lymphatic fluid collections are well recognized after vascular surgery when surgical disruption of lymphatic channels has occurred in regions of high interstitial fluid (ischemic legs and femoral—popliteal grafts) where lymph flow is high, or after renal transplantations when external iliac lymphatic chains has been disrupted. Lymphatic fluid collections have been documented, using electrophoresis, as perigraft reactions surrounding Dacron (DuPont, Wilmington, DE) velour axillofemoral or femoral—femoral grafts [4, 5]. In the six cases described [4, 5], an associated intense inflammatory reaction (vascular proliferation, lymphocyte, plasma cells, and foreign body macrophages) was seen. Lymphatic perigraft fluid appears to be a delayed allergic hypersensitivity reaction to the graft by the host [4, 5].
The pathogenesis of perigraft seromas involves both failure of graft incorporation into the native vessel wall and increased graft porosity. Failure of perigraft fibrosis in the surrounding connective tissue or vessel wall [1, 2] is associated with a slowly expanding collection caused by transudation of an ultrafiltrate of serum through the graft wall into the extrinsic milieu [2]. Unlike lymphatic collections, perigraft seromas contain no acute inflammatory cells or foreign body giant cells [2]. Reports have varied in their descriptions of the consistency of this material from clear dilute fluid [2] to a gelatinous coagulum identified histologically as compacted fibrin and proteinaceous material [6] surrounded by a pseudocapsule [2].
Perigraft seroma is a rare, late complication of polytetrafluoroethylene and Dacron grafts [1, 2], rarely involving autologous saphenous vein grafts [2]. These seromas are most often reported in superficial grafts such as axillofemoral reconstructions and in hemodialysis arteriovenous fistulas, because of their easy clinical recognition [1]. Mediastinal perigraft seromas have been described in pediatric polytetrafluoroethylene grafts in modified Blalock-Taussig shunt procedures [2]. The incidence of perigraft seromas in open aortic reconstructions may be underreported, possibly because of lack of imaging surveillance or clinical detection.
Why this entity occurs rarely and sporadically is unknown, but multiple theories exist [1, 2]. Polytetrafluoroethylene grafts are chemically inert, highly electronegative and hydrophobic, and impermeable to whole blood. However, excess wetting of the graft by organic solvents, such as alcohol, or excess manipulation or forced irrigation through the walls of the graft before graft insertion are thought to create a condition conducive to plasma leakage [6]. Increased permeability of the graft may also be due to previous infection [2]. Increased levels of humoral fibroblast inhibition, antigenically induced by the graft acting as a foreign body, may have a role in the failure of graft incorporation by preventing maturation and proliferation of perigraft fibroblasts [1]. Populations susceptible to humoral fibroblast inhibition production have not been defined.
Perigraft seromas in any vascular reconstruction are resistant to treatment and prone to recurrence [1], thereby posing further treatment dilemmas. Reported, usually unsuccessful, treatments include aspiration, pseudocapsule resection, graft replacement, and closed drainage [1]. Injection of a sclerosing agent is not recommended because of the risk of graft thombosis [2]. Nevertheless, case reports of inserting microfibrillar collagen (the end product of mature fibroblasts) into the space surrounding an axillobifemoral graft have documented successful graft incorporation into the surrounding tissue with no fluid reaccumulation [1]. We know of only one case report that has described a surgical communication between the seroma and the peritoneal cavity via a polytetrafluoroethylene membrane that allows the fluid to drain; no recurrence had occurred in that patient at 10 months after surgery [7].
Although perigraft seromas have been documented only in open vascular arterial grafts, we postulate that this finding may have relevance to endoluminal stent-grafts. Increase in the size of the aneurysmal sac after endoluminal exclusion of the aortic sac without apparent endoleak has been documented [6, 8]. In the context of endoluminal repair, ruptures have occurred in this situation [8], and they are thought to be caused by persistent intrasac pressure close to systemic arterial pressure. This phenomenon is referred to as endotension—that is, continued sac pressurization as documented via direct pressure lines in open surgery [6]. Mechanisms leading to increased sac volume in conjunction with endotension have not been fully elucidated in the literature [6]. We propose that increased graft permeability, failure of graft incorporation into the native vessel wall, and failure of perigraft fibrosis in the surrounding connective tissue may be contributing factors, and we think that further research may be indicated. White et al. [6] noted a case of expanding aortic aneurysm despite the absence of demonstrable endoleak of an endovascular abdominal aortic aneurysm repair. The endovascular graft was replaced with an open graft. Endotension was confirmed at surgery using a direct pressure line, and a gellike semiliquid thrombus, similar macroscopically to a perigraft seroma, was noted around the graft. Therefore, understanding the underlying pathogenesis of a perigraft seroma may aid in finding a solution to this new treatment dilemma.
Our two patients who underwent open repair (Figs. 1A,1B,1C and 2A,2B,2C,2D) had minimal complications and mild symptoms, despite their having aneurysms that were found on imaging to have increased in size. These outcomes are in contrast to reports in the literature of endovascular graft rupture caused by endotension [6].
In the absence of a clinical suspicion of graft infection, perigraft seroma should be considered in the differential diagnosis of an aneurysmal sac that increases in diameter after an open aortoiliac graft procedure. The expansion may be noted incidentally on imaging, or it may be a cause of symptoms reported by the patient. Its significance lies in its recognition, the potential complications, and in the treatment dilemma. The underlying pathogenesis may have relevance to the understanding of complications of endoluminal aortic aneurysm repair. The apparent contrasting natural history of aneurysm sac expansion in otherwise macroscopically successful open and endoluminal aortic repairs is noted.
Address correspondence to E. Kat.
| References | |
|---|
