(Radiology. 1999;212:573-577.)
© RSNA, 1999
Composite Graft Replacement of the Ascending Aorta: Leakage Detection with Gadolinium-enhanced MR Imaging1
Rossella Fattori, MD,
Benedetta Descovich, MD,
Paola Bertaccini, MD,
Francesca Celletti, MD,
Ilaria Caldarera, MD,
Angelo Pierangeli, MD and
Giampaolo Gavelli, MD
1 From the Institute of Radiology and Cardiovascular Surgery, Ospedale S Orsola, Via Massarenti 9, 40100 Bologna, Italy. Received March 10, 1998; revision requested May 12; revision received November 30; accepted March 16, 1999. Address reprint requests to R.F. (e-mail: ross@med.unibo.it).
 |
Abstract
|
|---|
PURPOSE: To assess the value of magnetic resonance (MR) imaging in the detection of postoperative complications after composite valve graft replacement.
MATERIALS AND METHODS: Spin-echo and gradient-echo MR imaging was performed in 52 patients 
to 200 months after composite graft replacement of the ascending aorta (22 for dissection, 30 for aneurysm). The prosthetic aortic segment, distal and proximal anastomoses, general morphologic characteristics, and diameter of the reimplanted coronary arteries were evaluated. In patients with abnormal perigraft thickening, additional spin-echo imaging was performed after injection of gadopentetate dimeglumine.
RESULTS: Normal postoperative perigraft thickening (
10 mm) was observed in 42 patients. Ten patients had abnormal periprosthetic thickening of 15–52 mm. Gadolinium-enhanced MR imaging demonstrated leakage in five of those 10 patients. The lack of enhancement excluded the presence of bleeding in the remaining five patients (three with chronic hematomas, one with infection, and one with granulation tissue). These findings were confirmed at surgery or with subsequent follow-up MR examinations.
CONCLUSION: MR imaging was an optimal imaging modality for evaluating the morphologic characteristics of composite grafts and reimplanted coronary arteries. Gadolinium-enhanced MR imaging is a simple, accurate, and noninvasive method for detecting a leak, which necessitates urgent repeat surgery.
Index terms: Aorta, grafts and prostheses, 56.1269, 941.1269 • Aorta, MR, 56.121411, 56.121412, 56.12143, 941.129411, 941.129412, 941.12942
|
Introduction
|
|---|
Advances in surgical techniques have substantially improved the survival rate in patients undergoing repair of the ascending aorta. Composite graft replacement of the ascending aorta and aortic valve, first described by Bentall and De Bono (1) (who used a Dacron tube with a prosthetic aortic valve sewn into one end), or their modification has become the procedure of choice in patients with annuloaortic ectasia or aortic dissection involving the aortic sinuses. Although this technique reduces the risk of recurrent aortic aneurysm, a relatively high frequency of early and late postoperative complications has been reported, justifying the need for close follow-up (2–4). Surgical replacement of the ascending aorta includes the risk of paravalvular leakage and false aneurysm formation as well as leakage or chronic partial dehiscence due to tension at the anastomosis of the coronary arteries. Postoperative assessment has been performed with angiography, echocardiography, and computed tomography, but magnetic resonance (MR) imaging has been shown to be the most effective procedure, providing multiplanar evaluation of the entire aorta and its surrounding structures (5–7).
The purpose of this study was to demonstrate the value of MR imaging in the depiction of postoperative anatomic details and complications after composite valve graft insertion.
|
MATERIALS AND METHODS
|
|---|
The study included 52 patients, 22 of whom underwent surgery for Stanford type A aortic dissection (14 in the acute phase and eight in the chronic phase) and 30 of whom underwent surgery for aneurysm of the ascending aorta, annulus, or both between January 1982 and October 1996. Of the 22 patients with dissection, 20 were men and two were women; the mean age was 48.7 years ± 14 (SD), and the follow-up time ranged from to 200 months after surgery. Of the 30 patients with aneurysm, 26 were men and four were women; the mean age was 43.2 years ± 16, and the follow-up time ranged from to 120 months after surgery. The diagnosis of Marfan syndrome was established with clinical criteria (8) in 24 patients, six with aortic dissection and 18 with aortic aneurysm.
All patients underwent composite graft replacement of the ascending aorta. Among the 22 patients with aortic dissection, the Bentall technique (1) (direct suture of coronary arteries to the conduit and wrapping of the native aortic wall around the prosthetic conduit) (Fig 1) was used in nine (among them five without preclotting grafts), the button technique (9) (reimplantation of coronary arteries with a small button of surrounding tissue to the conduit, without wrapping) (Fig 2) was used in 10, and the Cabrol technique (10) (interposition of Dacron T tube between coronary arteries and the main conduit) (Fig 3) was used in three.
View larger version (124K):
[in this window]
[in a new window]
|
Figure 1. Sagittal oblique spin-echo MR image (20/650) of ascending aorta replaced with the Bentall technique (direct suture of coronary arteries [arrows] to the conduit).
|
|
View larger version (125K):
[in this window]
[in a new window]
|
Figure 2. Sagittal oblique spin-echo MR image (20/830) of ascending aorta replaced with the button technique (reimplantation of coronary arteries [arrows] with a small button of surrounding tissue attached to the conduit).
|
|
View larger version (135K):
[in this window]
[in a new window]
|
Figure 3. Sagittal oblique spin-echo MR image (20/600) of ascending aorta replaced with the Cabrol technique (interposition of Dacron T tubes [arrows] between coronary arteries and the main conduit [*]).
|
|
Among the 30 patients with aortic aneurysm disease, the Bentall technique associated with wrapping of the aneurysm wall around the aortic graft was used in 15. In two of the 15 patients, the Bentall operation was combined with coronary artery bypass graft replacement (left internal mammary artery on the descending anterior coronary artery and saphenous venous graft on the circumflex coronary artery) and replacement of the aortic arch. The remaining 15 patients underwent the button operation with an open technique (no wrapping).
Eighteen patients underwent surgery before January 1991, and they underwent a single MR examination at a variable time after surgery (range, 6–200 months). Thirty-four patients underwent surgery between January 1991 and December 1997. These patients were included in a follow-up program in which Institutional Review Board approval and informed consent were required and in which MR imaging was performed 15 days after surgery and, subsequently, if no major pathologic findings were present, at 6 months and once a year thereafter. The number of MR studies available for each patient ranged from two to nine.
In all patients, electrocardiographically gated multisection spin-echo imaging of the entire thoracic aorta was performed in the axial, sagittal oblique, and coronal planes. The repetition time was dependent on the R-R interval on the electrocardiograph. The echo time was 20 msec. The section thickness varied from 5 to 7 mm, and the matrix size ranged from 128 x 160 to 256 x 256. In addition, gradient-echo imaging (25/17 [repetition time msec/echo time msec]; flip angle, 35°) was performed in the sagittal oblique plane.
In each patient, the prosthetic aortic segment, along with surrounding structures, distal and proximal anastomoses, general morphologic characteristics, and diameter of reimplanted coronary arteries were evaluated by two authors (R.F., B.D.) together, who rendered a consensus opinion. The image quality was graded as good, fair, or inadequate. The diameter of reimplanted coronary arteries was measured with the spin-echo sequences, which are not influenced by the metallic artifacts of the prosthetic valve.
Special attention was directed to the identification of periprosthetic collections, and, according to findings of previous studies (11), a perigraft thickening greater than 10 mm was considered abnormal. In patients with perigraft thickening greater than 10 mm, to identify the presence of bleeding and the site of its origin, a multisection spin-echo sequence was performed in the coronal plane, transaxial plane, or both after manual intravenous bolus injection of gadopentetate dimeglumine (Magnevist; Schering, Berlin, Germany; 0.2 mmol per kilogram of body weight) via a catheter placed in a vein in the antecubital fossa.
|
RESULTS
|
|---|
In all patients, a fair to good image quality was achieved (43 good; nine, fair). The diameter of composite grafts ranged from 28 to 37 mm (mean, 32.2 mm ± 2.8 [SD]). No abnormality of the distal anastomosis was observed. The proximal segment of both coronary vessels was visualized in 48 patients; only one coronary artery was visualized in four patients. In 47 patients, the diameter of the coronary artery was normal (range, 6–10 mm). Slight to moderate dilatation of reimplanted coronary ostia (range, 11–16 mm) was observed with the classic Bentall technique in five patients with Marfan syndrome (Fig 4).
View larger version (129K):
[in this window]
[in a new window]
|
Figure 4. Coronal spin-echo MR image (20/720) in a patient with Marfan syndrome shows moderate dilatation (arrow) of the reimplanted left coronary artery ostium.
|
|
Forty-two patients had perigraft thickening of less than 10 mm. The area of perigraft thickening was isointense to skeletal muscle. As described in previous articles (6,11), perigraft thickening up to 10 mm, attributable to fibrosis, normal granulation tissue, or both, is the most common postoperative MR finding (Fig 5). Ten patients had perigraft thickening of more than 10 mm (range, 15–52 mm). On spin-echo images, the signal intensity of the thickening was homogeneously intermediate in eight patients, and the thickening was isointense but not homogeneous, with some areas of flow void, in the other two. After injection of gadopentetate dimeglumine, five of the 10 patients with abnormal perigraft thickening showed a marked increase in signal intensity. Enhancement was concentric around the graft in two patients. In the remaining three, enhancement appeared to be localized to the site of the reimplanted coronary arteries (Fig 6), which is indicative of bleeding that originates from the coronary ostial sutures. The lack of enhancement excluded the presence of bleeding in the remaining five patients (three with chronic hematomas, one with infection, and one with granulation tissue). Gradient-echo sequences did not conclusively demonstrate flow inside any of the collections owing to the artifact of the prosthetic valve.
View larger version (120K):
[in this window]
[in a new window]
|
Figure 5. Axial spin-echo MR image (20/850) of ascending aorta and descending aorta. Normal perigraft thickening (black arrows), due to fibrosis and surrounding the prosthetic conduit of the ascending aorta, is visible. The perigraft thickening is less than 10 mm, and the signal intensity is intermediate compared to that of skeletal muscle. Residual dissection (white arrow) is present in the descending aorta.
|
|
View larger version (146K):
[in this window]
[in a new window]
|
Figure 6a. (a) Nonenhanced and (b) contrast-enhanced axial spin-echo MR images (20/700) of ascending aorta and descending aorta at the anastomoses of the coronary arteries. a shows a wide perigraft collection (arrows) characterized by homogeneously low signal intensity. b shows high signal intensity at the site (arrows) of the left coronary arterial anastomosis, which helps identify the presence and origin of bleeding.
|
|
View larger version (153K):
[in this window]
[in a new window]
|
Figure 6b. (a) Nonenhanced and (b) contrast-enhanced axial spin-echo MR images (20/700) of ascending aorta and descending aorta at the anastomoses of the coronary arteries. a shows a wide perigraft collection (arrows) characterized by homogeneously low signal intensity. b shows high signal intensity at the site (arrows) of the left coronary arterial anastomosis, which helps identify the presence and origin of bleeding.
|
|
Among patients who showed enhancement with gadopentetate dimeglumine, four underwent successful repeat surgery. One patient was unable to undergo surgery because his general clinical condition was poor. At surgery, the site of suture detachment (reimplanted coronary arteries in two patients and paravalvular suture lines causing a false aneurysm in the other two) was confirmed.
|
DISCUSSION
|
|---|
The underlying disease processes in patients referred for replacement of the ascending aorta represent a wide spectrum of conditions and acuity requiring different surgical strategies (10,12). Despite advances in surgical techniques that have substantially improved the survival rate over the years, follow-up data have documented the possible development of early and late complications (9). Most of the problems seen during long-term follow-up are due to the underlying disease process of the residual distal aorta, whereas early complications such as infection, suture dehiscence, or false aneurysm are mainly related to the prosthetic graft. Repeat surgery for bleeding at the site of repair has been reported in 8% of patients after 30 days and 4% of patients after 1 year.
The higher incidence of bleeding has been reported at the site of coronary arteries reimplanted according to classic Bentall technique (range, 4.2%–13.6%) (13–15). Moreover, aneurysmatic dilatation of the reimplanted coronary arteries may develop during long-term follow-up. For these reasons, this surgical procedure was modified (button technique) during the 1980s, particularly with regard to coronary arterial anastomosis to the prosthetic conduit, to avoid traction at the suture sites as a potential cause of bleeding.
There has also been controversy about the need to wrap the native aorta around the graft (2,9,16). Even though the wrapped wall may ensure better hemostasis and help avoid late sepsis, several reports describe a high incidence of false aneurysm formation. Our findings are in agreement with these data: False aneurysms were seen in two patients treated with the classic Bentall technique and none of the patients treated with button and Cabrol techniques (Fig 7). The largest surgical series (9) we know of that compares early and long-term survival among the most common methods used for composite valve graft replacement shows no substantial differences in perioperative bleeding, thus encouraging elimination of the wrapping procedure.
View larger version (151K):
[in this window]
[in a new window]
|
(a) Axial spin-echo MR image (20/800) of ascending aorta replaced with the classic Bentall technique and wrapping of the native aortic wall. A false aneurysm (*) is seen surrounding the prosthetic ascending aorta. (b) Sagittal oblique spin-echo MR image (20/800) of the same case shows extension of the false aneurysm (*) to the distal anastomosis.
|
|
View larger version (136K):
[in this window]
[in a new window]
|
(a) Axial spin-echo MR image (20/800) of ascending aorta replaced with the classic Bentall technique and wrapping of the native aortic wall. A false aneurysm (*) is seen surrounding the prosthetic ascending aorta. (b) Sagittal oblique spin-echo MR image (20/800) of the same case shows extension of the false aneurysm (*) to the distal anastomosis.
|
|
In our series, two patients treated with the button technique had a hematoma. In both cases, the suture detachment was not related to the surgical technique but was due to an inflammatory process (one patient had a Behçet aortitis and the other had a long-standing mediastinitis). The three cases of chronic hematoma were observed in patients treated with the Bentall technique before the introduction of the preclotting technique (the graft is impregnated with collagen or human serum albumin to control transgraft blood loss, therefore eliminating the need for inclusion techniques and their associated complications) (Fig 8) (17). This type of hematoma was reported in the first series of patients who underwent the Bentall operation but is not clinically relevant because it is not an evolving lesion.
View larger version (160K):
[in this window]
[in a new window]
|
Figure 8. Sagittal spin-echo MR image (20/600) shows asymmetric perigraft thickening (arrows) with heterogeneous signal intensity due to chronic hematoma.
|
|
Results of several studies have documented the superior performance of MR imaging in the evaluation of the thoracic aorta (5–7). This technique is able to provide multiplanar depiction of the aortic lumen and wall along with mediastinal structures; moreover, its reproducibility and noninvasive nature are optimal characteristics during the follow-up. In our series, spin-echo MR images enabled excellent visualization of the proximal portion of the reimplanted coronary arteries.
Dacron graft obstruction and thrombosis may be responsible for late sudden death observed during long-term follow-up after the Cabrol operation (9,18). In our three patients treated with the Cabrol technique, we did not detect any alteration in the anastomotic tube or in the proximal part of coronary arteries, whereas moderate dilatation of both reimplanted coronary arteries was present in five patients treated with the Bentall technique. Four of these five patients underwent three subsequent MR examinations at 1, 2, and 3 years: No substantial modification in diameters was visualized. Dilatation of the reimplanted coronary ostia after the classic Bentall operation has been reported (14), particularly in patients with Marfan syndrome, as a result of abnormal tension forces applied to a diseased wall. Even though this alteration does not seem to be clinically relevant, to our knowledge, results of imaging follow-up have not been reported. When we consider the young age of most of the patients undergoing composite valve graft replacement and the actual better survival rate, noninvasive monitoring of the reimplanted coronary arteries might be recommended.
Repeat surgery in patients with a composite valve graft has a high risk, and precise details, such as the identification of bleeding or the exact site of suture dehiscence, are necessary for a correct surgical approach. Nevertheless, the presence of a collection does not necessarily indicate bleeding. Perigraft thickening of less than 10 mm appears to be a normal postoperative appearance. Gaubert et al (11) analyzed the MR imaging aspect of 36 surgically treated aortic dissections and found perigraft thickening of less than 10 mm in 23 cases and perigraft thickening of up to 10 mm in 13 cases (nine false aneurysms). On MR images, normal granulation or fibrous tissue is hypo- to isointense compared with skeletal muscle. The distribution is usually uniform and concentric around the ascending aortic graft (11), whereas pathologic collections are often asymmetric (Fig 9).
View larger version (185K):
[in this window]
[in a new window]
|
Figure 9a. (a) Axial nonenhanced spin-echo MR image (20/730) of the ascending aorta replaced with the button technique in a patient with a prosthetic collection (arrows), which is 18 mm thick. (b) Contrast-enhanced image demonstrates a minimal, insubstantial increase in signal intensity (arrows).
|
|
View larger version (179K):
[in this window]
[in a new window]
|
Figure 9b. (a) Axial nonenhanced spin-echo MR image (20/730) of the ascending aorta replaced with the button technique in a patient with a prosthetic collection (arrows), which is 18 mm thick. (b) Contrast-enhanced image demonstrates a minimal, insubstantial increase in signal intensity (arrows).
|
|
Because of the possible different causes of perigraft thickening, it is important to differentiate this entity from other pathologic collections. On the basis of the signal intensity on spin-echo images, the differential diagnosis can be difficult because of the interaction of the different degradation compounds of hemoglobin that may result in variable signal intensity. Furthermore, in a periprosthetic hemorrhage collection, different aged foci can easily coexist.
Perigraft flow, outside the graft but contained within the wrapped aneurysmal sac, as in a false aneurysm, shows perigraft areas of signal void on spin-echo images or marked changes in signal intensity throughout the cardiac cycle on cine gradient-echo images. Gradient-echo sequences can enable visualization of moving blood between two different compartments, similar to angiography or Doppler echocardiography. Nevertheless, if the suture dehiscence is small, the flow could be too limited to be depicted with such techniques, as happened in three of our patients. Even though the bleeding was small in the five patients with bleeding in our study, a large, growing hematoma contained in the periaortic space was present and possibly evolved into graft compression and rupture. In our experience, gadolinium-enhanced MR imaging enabled us to diagnose suture detachment, which necessitated urgent repeat surgery. It clearly depicted the site of bleeding as an area of high signal intensity within the organized hematoma. Furthermore, it enabled us to define the bleeding site, clearly depicting the relationship with the reimplanted coronary artery.
In conclusion, routine follow-up studies with MR imaging should be recommended in all patients with composite graft replacement of the ascending aorta. MR imaging helped define the morphologic characteristics of the prosthetic graft and the anastomotic sites. The possibility of achieving a noninvasive evaluation of the proximal segment of reimplanted coronary arteries may be clinically relevant in long-term survival. Moreover, gadolinium-enhanced MR imaging can clearly help identify hemorrhagic complications, which represent a potential asymptomatic cause of death.
|
Acknowledgments
|
|---|
The authors are indebted to Albert de Roos, MD, for providing a critical review of the manuscript.
|
Footnotes
|
|---|
Author contributions: Guarantor of integrity of entire study, R.F.; study concepts, P.B.; study design, F.C.; definition of intellectual content, B.D.; clinical studies, I.C.; manuscript preparation, B.D.; manuscript editing, A.P.; manuscript review, G.G.
|
References
|
|---|
-
Bentall H, De Bono A. A technique for complete replacement of ascending aorta. Thorax 1968; 23:338-339.[Abstract/Free Full Text]
-
Kouchoukos NT, Marshall WG, Wedige-Stecher TA. Eleven-year experience with composite graft replacement of the ascending aorta and aortic valve. Thorac Cardiovasc Surg 1986; 92:691-705.
-
Midulla PS, Ergin A, Galla J, et al. Three faces of the Bentall procedure. J Cardiovasc Surg 1994; 9:466-481.
-
Svensson LG. Approach to the insertion of composite valve graft. Ann Thorac Surg 1992; 54:376-378.[Abstract]
-
Nienaber CA, von Kodolitsch Y, Nicolas V, et al. The diagnosis of thoracic aortic dissection by noninvasive imaging procedures. N Engl J Med 1993; 328:1-9.[Abstract/Free Full Text]
-
Lepore V, Lamm C, Bugge M, Larsson S. Magnetic resonance imaging in the follow-up of patients after aortic root reconstruction. Thorac Cardiovasc Surg 1996; 44:188-192.[Medline]
-
Fattori R, Celletti F, Bertaccini P, et al. Delayed surgery of traumatic aortic rupture: role of magnetic resonance imaging. Circulation 1996; 94:2865-2870.[Abstract/Free Full Text]
-
De Paepe A, Devereux RB, Dietz HC, Hannekam M, Pyeritz RE. Revised diagnostic criteria for the Marfan syndrome. Am J Med Genet 1996; 62:417-426.[Medline]
-
Svensson LG, Crawford ES, Hess KR, et al. Composite graft replacement of the proximal aorta: comparison of techniques in 348 patients. Ann Thorac Surg 1992; 54:427-439.[Abstract]
-
Cabrol C, Pavie A, Mesnildrey P, et al. Long-term result with total replacement of the ascending aorta and reimplantation of the coronary arteries. J Thorac Cardiovasc Surg 1986; 91:17-25.[Abstract]
-
Gaubert JY, Moulin G, Mesana T, et al. Type A dissection of the thoracic aorta: use of MR imaging for long-term follow-up. Radiology 1995; 196:363-369.[Abstract/Free Full Text]
-
Bachet J, Termignon JL, Goudot B, et al. Aortic root replacement with a composite graft: factors influencing immediate and long-term results. Eur J Cardiothorac Surg 1996; 10:207-213.[Abstract]
-
Taniguchi K, Nakano S, Matsuda H, et al. Long-term survival and complication after composite graft replacement for ascending aortic aneurysm associated with aortic regurgitation. Circulation 1991; 84(suppl 3):31-39.
-
Svensson LG, Crawford ES, Coselli JS, Safi HJ, Hess KR. The impact of cardiovascular operation on survival in the Marfan patient. Circulation 1989; 80(suppl 1):233-242.
-
Coselli JS, Crawford ES. Composite valve graft replacement of aortic root using separate Dacron tube for coronary reattachment. Ann Thorac Surg 1989; 47:558-565.[Abstract]
-
Crawford ES, Svensson LG, Coselli JS, Safi HJ, Hess KR. Surgical treatment of aneurysm and/or dissection of the ascending aorta, transverse aortic arch, and ascending aorta and transverse aortic arch. J Thorac Cardiovasc Surg 1989; 98:659-674.[Abstract]
-
Savunen T, Inberg M, Niinikoski J, Rantakokko V, Vanttinen E. Composite graft in annulo-aortic ectasia: nineteen years' experience without graft inclusion. Eur J Cardiothorac Surg 1996; 10:428-432.[Abstract]
-
Sekine S, Tadaaki A, Keiji S, Shibata Y, Yamagishi I, Kamada M. Dacron coronary graft obstruction after composite graft replacement of aortic root. Ann Thorac Surg 1995; 60:1123-1126.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
A. D. Milano, S. Pratali, G. Mecozzi, P. Boraschi, G. Braccini, E. Magagnini, and U. Bortolotti
Fate of coronary ostial anastomoses after the modified Bentall procedure
Ann. Thorac. Surg.,
June 1, 2003;
75(6):
1797 - 1801.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
Abstract
Introduction
