Eur J Vasc Surg6, 282-287 (1992)
The Late Complications of Aorto-ilio-femoral Dacron Prostheses: Dilatation and Anastomotic Aneurysm Formation P. T. den Hoed and H. F. Veen Department of Surgery, Ikazia Hospital, Rotterdam, The Netherlands Dilated prosthetic grafts and anastomotic aneurysms are recognised as a long-term complication of aorto-ilio-femoral reconstruction. In the literature an incidence of 1-24% for anastomotic aneurysms and a mean dilatation of 85.7% for Dacron aortic prostheses is reported. In our own clinic the frequency of these complications was not known, and we started a follow-up study in order to establish: the mean dilatation of the prosthetic grafts, the actual incidence of anastomotic aneurysms and the relationship between these two. All patients (n = 61) who underwent aorto-ilio-femoral reconstruction between 1980 and 1985 were retrospectively studied. During the mean 8 years follow-up period 16 patients died (26%). Complete data were availablefrom 36 patients, who underwent physical examination and ultrasonography. In the whole series the degree of dilatation varied from 0 to 44 % with an average of 5.9 % at the aortic level and 8.9 % at the distal level. The diameters of 25 grafts remained unchanged, and the maximum dilatation observed was 44%. The incidence of anastomotic aneurysms in these patients was 13.8%. The anastomotic aneurysms were located at thefollowing anatomical sites: aortic anastomoses 5.9 %, iliac anastomoses 0 and 30.4 % for thefemoral anastomoses. There was no relationship between the dilated grafts and anastomotic aneurysms. In our clinic the degree of dilatation and the incidence of anastomotic aneurysms seems to be low, especially if we compare these results with the results reported by other investigators. None of the patients who underwent aorto-ilio-femoral reconstruction between 1980 and 1985 died or had to undergo emergency surgery because of graft failure or a ruptured anastomotic aneurysm. Key Words: Aorto-ilio-femoral reconstruction; Dacron prostheses; Dilatation of the prostheses; Anastomotic aneurysm.
Introduction
Dilatation of a prosthetic graft and anastomotic aneurysms are potentially serious late complications after aorto-ilio-femoral reconstructions. These complications do not necessarily cause symptoms and can remain unnoticed for years. Ultimately the anastomotic aneurysm or the dilated graft may rupture and the outcome of such a complication requiring repair under emergency circumstances is less favourable than in an elective situation. Despite improving technical skills and new prosthetic grafts, the incidence of anastomotic aneurysms and dilated grafts has not decreased in the last few years. In our clinic we have been using the Meadox-Cooley Double Velour prostheses (Meadox Medicals, Oakland, NJ, U.S.A) for many years. Numerous publications and 30 years of clinical Please address all correspondence to: H. F. Veen, Department of Surgery, Ikazia Hospital, Montesorriweg1, 3083 AN Rotterdam, The Netherlands. 0950-821X/92/030282+06 $03.00/0© 1992Grune & StrattonLtd.
experience attest to the generally satisfactory performance of Dacron as an aortic substitute. Despite general acceptance, however, Dacron prostheses are not ideal because of their susceptibility to dilatation and anastomotic aneurysm formation. Nunn 1 found a mean dilatation of 17.6% and concluded that Dacron grafts as a whole tend to dilate early after implantation, and that the dilatation increases slowly with time. His study has also shown that ultrasound is a reliable, convenient and relatively inexpensive method for evaluating postoperative changes in Dacron graft diameter. May and Stephen 2 stated that the aetiology of the dilatation is not always clear, but multiple defects in construction seemed the most likely cause of multiple aneurysms in Dacron Velour grafts in most cases. Cooley et al. 3 on the other hand found no instance of aneurysm or dilatation of Dacron fabric prostheses after reviewing all their patients who had received a Meadox-Cooley Double Velour graft. The aim of this study was to determine in a systematic clinical investigation the incidence of anasto-
Complications of Aorto-ilio-femoral Dacron Prostheses
motic aneurysms and dilatation of prosthetic grafts in patients after aorto-itio-femoral reconstruction during long-term follow-up.
Patients and Methods
All patients who underwent aorto-ilio-femoral reconstruction between 1980 and 1985 at Ikazia Hospital, Rotterdam, were studied retrospectively. The indications for surgery were aneurysmal or arterial occlusive disease or a combination of both. Patients received either a tube or a bifurcated graft. Data from 61 patients with a follow-up period of 6-10 years were collected. As expected the available data were found to be incomplete in many cases. During a visit, which these patients were requested to make to the outpatient department, informed consent was obtained and physical examination undertaken. This was followed by ultrasonography of the aorto-ilio-femoral tract and if necessary a computed tomography (CT) scan. From the primary group of 61 patients, 16 (26%) died from non-related causes and nine (15%) were not included in the study for other reasons leaving 36 patients who participated in the investigation. This study group was not significantly different from the primary group and a comparison of the patient characteristics in both groups is presented in Table 1. Table 1. Patient characteristics of the primary group (n = 61) and the study group (n = 36)
Primary group
Study group
Age distribution(years)
50-86
50-85
Mean age (years)
68
65
Sex distribution Men
49 (80%)
28 (78%)
12 (20%)
8 (22%)
Indicationsfor surgery Occlusivedisease
32 (52%)
21 (58%)
Aneurysmaldisease
23 (38%)
11 (31%)
6 (10%)
4 (11%)
Women
Occlusive/aneurysmal
In all patients the prosthetic material used was a Meadox-Cooley Double Velour Dacron prosthesis. There was a total of 103 anastomoses. For tube grafts, both the proximal and distal anastomoses were endto-end; for the bifurcated grafts the proximal anastomoses were end to end in most cases (24; 77%) as
283
opposed to 7 (23%) that were end to side. The distal anastomoses were end to side in 87% of patients who underwent aorto-ilio-femoral reconstruction and end to end in 13%. The diagnosis of an anastomotic aneurysm was based on only one criterion: the diameter of the anastomosis had to be equal to or greater than 1.5 times the diameter of the prostheses. 4 Calculation of the percentage change in diameter of the grafts was based on the difference between the original graft size, as designated by the manufacturer, and the postoperative size as determined by ultrasonography. A generalised graft dilatation of 15-20% greater than the manufacturer's stated diameter usually occurs after aortic insertion. 1,5 The diagnosis of graft dilatation was based on the criterion that the dilatation of the prosthetic graft had to be more than 20% of its original diameter. The images were evaluated by five experienced radiologists, blind to the results of the physical examination. If there was any doubt about the ultrasound result a CT-scan was performed.
Results
Complete documentation was available in 36 patients. By careful physical examination, two graft dilatations were diagnosed and a third patient, who was suspected of having a dilatation, was found to be normal on ultrasound.
Incidence of anastomotic aneurysms Of the total of 103 anastomoses at risk, nine aneurysms were found, representing an incidence of 8.7%. The incidence for the different anatomical locations was 5.9% (two cases) for aortic anastomoses, zero for iliac anastomoses and 30.4% (seven cases) for femoral artery anastomoses (Table 2). The nine anastomotic aneurysms were found in five different patients (13.8%), with four patients having two and one a single anastomotic aneurysm. Of four patients with two aneurysms, one had a tube graft, which was affected at both the proximal and distal ends. The other three had a bifurcated graft and in these, both the anastomotic aneurysms were distal (Table 3). There was no connection between the anastomotic aneurysms and the original diameter of the prosthetic graft. Eur J VascSurg Vol 6, May 1992
284
P.T. den Hoed and H. F. Veen
Extent of dilatation
Table 2. Incidence of anastomotic aneurysms per anatomical site Number of anastomoses Site
at risk
Number of anastomotic aneurysms
Percentage
Aorta End to end
34
2
5.9
End to side
7
0
0
Iliac End to end
9
0
0
End to side
29
0
0
Femoral End to end
1
0
0
End to side
23
7
30.4
103
9
8.7
Total
Over the whole series, the extent of graft dilatation varied from 0 to 44%, with an average at the aortic level of 5.9% and at the distal level of 8.9% (Fig. 1). The diameter of 25 grafts remained unchanged, and the maximum dilatation observed was 8turn (44%). At the proximal aortic level three dilatations were found: two tube grafts of 18mm, which had dilated by 22 and 44% respectively and one bifurcated graft of 18 x 10 mm which had dilated by 33% (Fig. 2). Four bilateral dilatations were found in bifurcated graft limbs, one of 22% and three of 33% (Fig. 3). Thus the seven dilatations were found in five patients. There was no connection between the extent of dilatation and the original diameter of the graft.
Table 3. Anastomotic aneurysms diagnosed by ultrasonography Ultrasonic diameter
Dilatation
Original diameter
Site
(ram)
(ram)
(ram)
Anastomoses
Patient
Proximal
35
+ 15
20
ETE
A
Distal Left iliac
.
Left femoral
.
.
.
.
23
+ 14
16 x 9
ETS
B
23
+14
16 x 9
ETS
C
30
+19
20 x 11
ETS
D
19
+9
18 x 10
ETS
E
20
+11
16 x 9
ETS
C
20
+11
16 x 9
ETS
B
18
+8
18 x 10
ETS
E
48
+28
20
ETE
A
Right iliac Right femoral
Aorta 30 25
._o
20
,o Z
5 HI Normal
+1
+2
+5
+4 +5 Dilefotion (mm)
t
+6
+7
mml +8
Fig. 1. Dilatation of the prosthetic graft at different anatomical levels. NN:aorta; []: left iliac/femoral; [] right iliac/femoral. Eur J Vasc Surg Vol 6, May 1992
Complications of Aorto-ilio-femoral Dacron Prostheses
285
I2
EO
u,
8
=
6
J;3
"5 E
15%
2:
I1%
2 0
19% 17%
33% ~t Normal
I +1
+2
+5
, I +4 +5 Dilatation (mm)
N
+6
I +7
I +8
Fig. 2. Relationship b e t w e e n the diameter of the bifurcated graft and the dilatation at the aortic level. I1:14 x 8; []: 16 x 9; E]: 18 x 10; []: 20 x 11; ~: 22 x 12.
2O
I6
S 12 ~S
Z
2 2 % 20% 0
I Normal
~ +1
+2
H
I +5 Dilal"a lion (rnm)
I +4
I +5
Fig. 3, Relationship b e t w e e n the diameter of the bifurcated grafts and the dilatation at the distal level. I1:14 x 8; []: 16 x 9; [3:18 x 10; I~: 20 x 11; B: 22 × 12.
Discussion
The reported incidences of anastomotic aneurysms vary widely, from 1 to 2 4 % 4,6,7,8 and several factors have been suggested to explain these wide variations. First, there are major differences in the types of arterial reconstruction under study. Endarterectomies and autologous graft procedures may have been included, although it is generally agreed that the occurrence of anastomotic aneurysms is largely associated with synthetic prostheses. The location of the anastomoses is often not taken into account, although femoral anastomoses bear the greatest risk of developing an anastomoti c aneurysm. 6 The use of a corset over the proximal anastomosis is also not taken into account, although it may be preventative. 9
Second, there is the question of which technique should be used for the detection of anastomotic aneurysms. The clinician often detects an abdominal aortic aneurysm first by palpation, but this estimate of size is often inaccurate. The diameter may be more accurately estimated by CT-scan or ultrasonography. Ultrasonography offers a rapid, convenient, sensitive method of screening for abdominal aortic aneurysm. Ellis et al. 1° showed that maximum diameter using ultrasonography was larger than that using CT-scan, the difference being least for anterior-posterior measurements. Finally, there are major differences in the length and completeness of follow-up. Because anastomotic aneurysms can develop at any postoperative interval, their incidence will increase with the length of the Eur J Vasc Surg Vol 6, May 1992
286
P.T. den Hoed and H. F. Veen
postoperative observation period, and a number of anastomotic aneurysms may easily remain undetected when patients are lost to follow-up. The above factors mean that a reliable comparison of various series is extremely difficult. The incidence of anastomotic aneurysms per anatomical site in our study is 8.7%, which seems low if we compare it with the incidence reported by others and take into account the duration of our follow-up. However in our opinion the incidence of anastomotic aneurysms in our clinic is still too high. The efficiency of end-to-end as opposed to endto-side anastomoses in providing the more haemodynamically suitable flow for the prevention of anastomotic aneurysms has already been demonstrated.11 According to the laws of fluid flow relating to arterial grafting the larger the angle between the graft and the recipient artery, the smaller the rate of flow through the anastomosis and the greater the turbulence and stress on the suture line. Therefore most surgeons try to fashion their femoral anastomoses with a small angle and a smooth, elliptical orifice. From the nine anastomotic aneurysms in our study seven were end to side and two were end to end. It may be possible to reduce the occurrence of anastomotic aneurysms by doing as many end-to-end anastomoses as possible. Aneurysmal changes in knitted Dacron grafts have been reported by N u n n et al., 1 May and Stephen, 2 and Cook et al. 1"12-17 Cooley et al. 3 on the other hand found no dilatation or aneurysm formation in a review of their patients and our results are closest theirs. In our series a mean proximal dilatation of 5.9% and distal dilatation of 8.9% was found compared with 17.6% found by N u n n et al. 1 The diameter of 25 grafts remained unchanged and the maximum dilatation was 44% in one graft. This study tells us nothing about the causes of graft dilatation. In the literature intraoperative damage to the graft of biodegradation have been suggested as causes of graft failure. 3 •4 ' 16 - 19 None of these causes were found in our study, and no relationship was observed between anastomotic aneurysms and graft dilatation. Patients with a dilated graft did not have anastomotic aneurysms and vice versa. Only one patient with an anastomotic aneurysm, confirmed by ultrasonography, underwent reoperation. The other patients with anastomotic aneurysms or dilated grafts are simply being observed. Several of the patients are of advanced age or in poor mental or physical health and would be unlikely to tolerate another operation, and there are a few borderline cases in which we agreed upon a policy of observation. Eur J Vasc Surg Vol 6, May 1992
Conclusion
Crimped knitted Dacron grafts are commonly employed as arterial substitutes and in most instances have proven to be highly effective and relatively durable. Graft failures and ruptured anastomotic aneurysms have occurred and have usually been associated with serious haemorrhage. In our clinic the incidence of anastomotic aneurysms and the degree of dilatation are very acceptable, especially if we compare our results with those reported by other investigators. None of the patients, who underwent aorto-ilio-femoral reconstruction between 1980 and 1985, died or had to undergo acute surgery because of graft dilatation or a ruptured anastomotic aneurysm.
Acknowledgement The images were made by the Department of Radiology, Ikazia Hospital, Rotterdam. Their excellent assistance and pleasant cooperation are gratefully appreciated. We also like to thank Mr Aubrey L. Waddy for his remarks and contribution in writing this article.
References 1 NUNN DB, FREEMANMH, HUDGINSPC. Postoperative alterations in size of Dacron aortic grafts. Ann Surg 1979; 189: 741-744. 2 MAY J, STEPHEN M. Multiple aneurysms in Dacron velour grafts. Arch Surg 1978; 113: 320-321. 3 COOLEY DA, SUBRAM A, HOUCHIN DP. Clinical experience in 1040 patients with double-velour knitted Dacron vascular prostheses. Bull Texas Heart Inst 1981; 8: 320-332. 4 SIESWERDA C, SROTNICKI SH, BARENTSZJO, et al. Anastomotic a n e u r y e m s - - a n underdiagnosed complication after aorto-iliac reconstructions. Eur J Vasc Surg 1989; 3: 233-238. 5 BERGERK, SAUVAGELR. Late fiber deterioration in Dacron arterial grafts. Ann Surg 1980; 193: 477-491. 6 VAN DEN AKKER PJ, BRAND R. False aneurysms after prosthetic reconstructions for aorto-iliac obstructive disease. Ann Surg 1989; 210: 658-666. 7 BRIGGSRM, BRUCESJ, COLLINS GJ. Anastomotic aneurysms. Am J Surg 1983; 146: 770-773. 8 KALSBEEKHL. False aneurysms in patients with an aortic.bifurcation prosthesis. Ned Tijolschr Geneeskd 1989; 133: 681-683. 9 VAN HOUTEN H. Over de operatieve behandeling van het aneurysma aortae abdominalis atheroscleroticum. Diss 1962. 10 ELLISM, POWELLJT, GREENHALGHRIV[. Limitations of ultrasonography in surveillance of small abdominal aortic aneurysms. Br J Surg 1991;.78: 614-616. 11 SZILAGYIDE, WHITCOMUjG, SCHENKERW, et al. The laws of fluid flow and arterial grafting. Surgery 1960; 47: 55. 12 COOKE PA, NOBIS PA, STONEY RJ. Dacron aortic graft failure. Arch Surg 1974; 108: 101-103. 13 LORD RS, NASH PA, ILaJ BT, et al. Prospective randomized trial of polytetrafluoroethylene and Dacron aortic prosthesis. Ann Vasc Surg 1988; 3: 248-254.
Complications of Aorto-ilio-femoral Dacron Prostheses
14 MILLILIJJ, LANESJS, NEMIR P. A study of anastomotic aneurysms following aortofemoral prosthetic bypass. Ann Surg 1980; 192: 69-73. 15 OTTINGERLW, DARLINGC, WIRTHLINLS, et al. Failure of ultralightweight knitted Dacron grafts in arterial reconstruction. Arch Surg 1976; 111: 146-149. 16 PERRYMO. Early failure of Dacron prosthetic grafts. ] Cardiovasc Surg 1975; 16: 318-321. 17 RAZSO, LUNDSTRUMB, ANGQUISTKA, et al. Bilateral aneurysm of
287
Dacron graft following aorto-femoral graft operation. Acta Chir Scand 1976; 142: 479-482. 18 RAMSAYC, WILLIAM A. Aneurysms of a double velour aortic graft. Arch Surg 1984; 119: 1183-1184. 19 YASHARJJ, RICHMANMH, DYCKMANJ, et al. Failure of Dacron prostheses caused by structural defect. Surge~ 1978; 84: 659663. Accepted 16 November 1991
Eur J Vasc Surg Vol 6, May 1992
The Late Complications of Aorto-ilio-femoral Dacron Prostheses: Dilatation and Anastomotic Aneurysm Formation P. T. den Hoed and H. F. Veen Department of Surgery, Ikazia Hospital, Rotterdam, The Netherlands Dilated prosthetic grafts and anastomotic aneurysms are recognised as a long-term complication of aorto-ilio-femoral reconstruction. In the literature an incidence of 1-24% for anastomotic aneurysms and a mean dilatation of 85.7% for Dacron aortic prostheses is reported. In our own clinic the frequency of these complications was not known, and we started a follow-up study in order to establish: the mean dilatation of the prosthetic grafts, the actual incidence of anastomotic aneurysms and the relationship between these two. All patients (n = 61) who underwent aorto-ilio-femoral reconstruction between 1980 and 1985 were retrospectively studied. During the mean 8 years follow-up period 16 patients died (26%). Complete data were availablefrom 36 patients, who underwent physical examination and ultrasonography. In the whole series the degree of dilatation varied from 0 to 44 % with an average of 5.9 % at the aortic level and 8.9 % at the distal level. The diameters of 25 grafts remained unchanged, and the maximum dilatation observed was 44%. The incidence of anastomotic aneurysms in these patients was 13.8%. The anastomotic aneurysms were located at thefollowing anatomical sites: aortic anastomoses 5.9 %, iliac anastomoses 0 and 30.4 % for thefemoral anastomoses. There was no relationship between the dilated grafts and anastomotic aneurysms. In our clinic the degree of dilatation and the incidence of anastomotic aneurysms seems to be low, especially if we compare these results with the results reported by other investigators. None of the patients who underwent aorto-ilio-femoral reconstruction between 1980 and 1985 died or had to undergo emergency surgery because of graft failure or a ruptured anastomotic aneurysm. Key Words: Aorto-ilio-femoral reconstruction; Dacron prostheses; Dilatation of the prostheses; Anastomotic aneurysm.
Introduction
Dilatation of a prosthetic graft and anastomotic aneurysms are potentially serious late complications after aorto-ilio-femoral reconstructions. These complications do not necessarily cause symptoms and can remain unnoticed for years. Ultimately the anastomotic aneurysm or the dilated graft may rupture and the outcome of such a complication requiring repair under emergency circumstances is less favourable than in an elective situation. Despite improving technical skills and new prosthetic grafts, the incidence of anastomotic aneurysms and dilated grafts has not decreased in the last few years. In our clinic we have been using the Meadox-Cooley Double Velour prostheses (Meadox Medicals, Oakland, NJ, U.S.A) for many years. Numerous publications and 30 years of clinical Please address all correspondence to: H. F. Veen, Department of Surgery, Ikazia Hospital, Montesorriweg1, 3083 AN Rotterdam, The Netherlands. 0950-821X/92/030282+06 $03.00/0© 1992Grune & StrattonLtd.
experience attest to the generally satisfactory performance of Dacron as an aortic substitute. Despite general acceptance, however, Dacron prostheses are not ideal because of their susceptibility to dilatation and anastomotic aneurysm formation. Nunn 1 found a mean dilatation of 17.6% and concluded that Dacron grafts as a whole tend to dilate early after implantation, and that the dilatation increases slowly with time. His study has also shown that ultrasound is a reliable, convenient and relatively inexpensive method for evaluating postoperative changes in Dacron graft diameter. May and Stephen 2 stated that the aetiology of the dilatation is not always clear, but multiple defects in construction seemed the most likely cause of multiple aneurysms in Dacron Velour grafts in most cases. Cooley et al. 3 on the other hand found no instance of aneurysm or dilatation of Dacron fabric prostheses after reviewing all their patients who had received a Meadox-Cooley Double Velour graft. The aim of this study was to determine in a systematic clinical investigation the incidence of anasto-
Complications of Aorto-ilio-femoral Dacron Prostheses
motic aneurysms and dilatation of prosthetic grafts in patients after aorto-itio-femoral reconstruction during long-term follow-up.
Patients and Methods
All patients who underwent aorto-ilio-femoral reconstruction between 1980 and 1985 at Ikazia Hospital, Rotterdam, were studied retrospectively. The indications for surgery were aneurysmal or arterial occlusive disease or a combination of both. Patients received either a tube or a bifurcated graft. Data from 61 patients with a follow-up period of 6-10 years were collected. As expected the available data were found to be incomplete in many cases. During a visit, which these patients were requested to make to the outpatient department, informed consent was obtained and physical examination undertaken. This was followed by ultrasonography of the aorto-ilio-femoral tract and if necessary a computed tomography (CT) scan. From the primary group of 61 patients, 16 (26%) died from non-related causes and nine (15%) were not included in the study for other reasons leaving 36 patients who participated in the investigation. This study group was not significantly different from the primary group and a comparison of the patient characteristics in both groups is presented in Table 1. Table 1. Patient characteristics of the primary group (n = 61) and the study group (n = 36)
Primary group
Study group
Age distribution(years)
50-86
50-85
Mean age (years)
68
65
Sex distribution Men
49 (80%)
28 (78%)
12 (20%)
8 (22%)
Indicationsfor surgery Occlusivedisease
32 (52%)
21 (58%)
Aneurysmaldisease
23 (38%)
11 (31%)
6 (10%)
4 (11%)
Women
Occlusive/aneurysmal
In all patients the prosthetic material used was a Meadox-Cooley Double Velour Dacron prosthesis. There was a total of 103 anastomoses. For tube grafts, both the proximal and distal anastomoses were endto-end; for the bifurcated grafts the proximal anastomoses were end to end in most cases (24; 77%) as
283
opposed to 7 (23%) that were end to side. The distal anastomoses were end to side in 87% of patients who underwent aorto-ilio-femoral reconstruction and end to end in 13%. The diagnosis of an anastomotic aneurysm was based on only one criterion: the diameter of the anastomosis had to be equal to or greater than 1.5 times the diameter of the prostheses. 4 Calculation of the percentage change in diameter of the grafts was based on the difference between the original graft size, as designated by the manufacturer, and the postoperative size as determined by ultrasonography. A generalised graft dilatation of 15-20% greater than the manufacturer's stated diameter usually occurs after aortic insertion. 1,5 The diagnosis of graft dilatation was based on the criterion that the dilatation of the prosthetic graft had to be more than 20% of its original diameter. The images were evaluated by five experienced radiologists, blind to the results of the physical examination. If there was any doubt about the ultrasound result a CT-scan was performed.
Results
Complete documentation was available in 36 patients. By careful physical examination, two graft dilatations were diagnosed and a third patient, who was suspected of having a dilatation, was found to be normal on ultrasound.
Incidence of anastomotic aneurysms Of the total of 103 anastomoses at risk, nine aneurysms were found, representing an incidence of 8.7%. The incidence for the different anatomical locations was 5.9% (two cases) for aortic anastomoses, zero for iliac anastomoses and 30.4% (seven cases) for femoral artery anastomoses (Table 2). The nine anastomotic aneurysms were found in five different patients (13.8%), with four patients having two and one a single anastomotic aneurysm. Of four patients with two aneurysms, one had a tube graft, which was affected at both the proximal and distal ends. The other three had a bifurcated graft and in these, both the anastomotic aneurysms were distal (Table 3). There was no connection between the anastomotic aneurysms and the original diameter of the prosthetic graft. Eur J VascSurg Vol 6, May 1992
284
P.T. den Hoed and H. F. Veen
Extent of dilatation
Table 2. Incidence of anastomotic aneurysms per anatomical site Number of anastomoses Site
at risk
Number of anastomotic aneurysms
Percentage
Aorta End to end
34
2
5.9
End to side
7
0
0
Iliac End to end
9
0
0
End to side
29
0
0
Femoral End to end
1
0
0
End to side
23
7
30.4
103
9
8.7
Total
Over the whole series, the extent of graft dilatation varied from 0 to 44%, with an average at the aortic level of 5.9% and at the distal level of 8.9% (Fig. 1). The diameter of 25 grafts remained unchanged, and the maximum dilatation observed was 8turn (44%). At the proximal aortic level three dilatations were found: two tube grafts of 18mm, which had dilated by 22 and 44% respectively and one bifurcated graft of 18 x 10 mm which had dilated by 33% (Fig. 2). Four bilateral dilatations were found in bifurcated graft limbs, one of 22% and three of 33% (Fig. 3). Thus the seven dilatations were found in five patients. There was no connection between the extent of dilatation and the original diameter of the graft.
Table 3. Anastomotic aneurysms diagnosed by ultrasonography Ultrasonic diameter
Dilatation
Original diameter
Site
(ram)
(ram)
(ram)
Anastomoses
Patient
Proximal
35
+ 15
20
ETE
A
Distal Left iliac
.
Left femoral
.
.
.
.
23
+ 14
16 x 9
ETS
B
23
+14
16 x 9
ETS
C
30
+19
20 x 11
ETS
D
19
+9
18 x 10
ETS
E
20
+11
16 x 9
ETS
C
20
+11
16 x 9
ETS
B
18
+8
18 x 10
ETS
E
48
+28
20
ETE
A
Right iliac Right femoral
Aorta 30 25
._o
20
,o Z
5 HI Normal
+1
+2
+5
+4 +5 Dilefotion (mm)
t
+6
+7
mml +8
Fig. 1. Dilatation of the prosthetic graft at different anatomical levels. NN:aorta; []: left iliac/femoral; [] right iliac/femoral. Eur J Vasc Surg Vol 6, May 1992
Complications of Aorto-ilio-femoral Dacron Prostheses
285
I2
EO
u,
8
=
6
J;3
"5 E
15%
2:
I1%
2 0
19% 17%
33% ~t Normal
I +1
+2
+5
, I +4 +5 Dilatation (mm)
N
+6
I +7
I +8
Fig. 2. Relationship b e t w e e n the diameter of the bifurcated graft and the dilatation at the aortic level. I1:14 x 8; []: 16 x 9; E]: 18 x 10; []: 20 x 11; ~: 22 x 12.
2O
I6
S 12 ~S
Z
2 2 % 20% 0
I Normal
~ +1
+2
H
I +5 Dilal"a lion (rnm)
I +4
I +5
Fig. 3, Relationship b e t w e e n the diameter of the bifurcated grafts and the dilatation at the distal level. I1:14 x 8; []: 16 x 9; [3:18 x 10; I~: 20 x 11; B: 22 × 12.
Discussion
The reported incidences of anastomotic aneurysms vary widely, from 1 to 2 4 % 4,6,7,8 and several factors have been suggested to explain these wide variations. First, there are major differences in the types of arterial reconstruction under study. Endarterectomies and autologous graft procedures may have been included, although it is generally agreed that the occurrence of anastomotic aneurysms is largely associated with synthetic prostheses. The location of the anastomoses is often not taken into account, although femoral anastomoses bear the greatest risk of developing an anastomoti c aneurysm. 6 The use of a corset over the proximal anastomosis is also not taken into account, although it may be preventative. 9
Second, there is the question of which technique should be used for the detection of anastomotic aneurysms. The clinician often detects an abdominal aortic aneurysm first by palpation, but this estimate of size is often inaccurate. The diameter may be more accurately estimated by CT-scan or ultrasonography. Ultrasonography offers a rapid, convenient, sensitive method of screening for abdominal aortic aneurysm. Ellis et al. 1° showed that maximum diameter using ultrasonography was larger than that using CT-scan, the difference being least for anterior-posterior measurements. Finally, there are major differences in the length and completeness of follow-up. Because anastomotic aneurysms can develop at any postoperative interval, their incidence will increase with the length of the Eur J Vasc Surg Vol 6, May 1992
286
P.T. den Hoed and H. F. Veen
postoperative observation period, and a number of anastomotic aneurysms may easily remain undetected when patients are lost to follow-up. The above factors mean that a reliable comparison of various series is extremely difficult. The incidence of anastomotic aneurysms per anatomical site in our study is 8.7%, which seems low if we compare it with the incidence reported by others and take into account the duration of our follow-up. However in our opinion the incidence of anastomotic aneurysms in our clinic is still too high. The efficiency of end-to-end as opposed to endto-side anastomoses in providing the more haemodynamically suitable flow for the prevention of anastomotic aneurysms has already been demonstrated.11 According to the laws of fluid flow relating to arterial grafting the larger the angle between the graft and the recipient artery, the smaller the rate of flow through the anastomosis and the greater the turbulence and stress on the suture line. Therefore most surgeons try to fashion their femoral anastomoses with a small angle and a smooth, elliptical orifice. From the nine anastomotic aneurysms in our study seven were end to side and two were end to end. It may be possible to reduce the occurrence of anastomotic aneurysms by doing as many end-to-end anastomoses as possible. Aneurysmal changes in knitted Dacron grafts have been reported by N u n n et al., 1 May and Stephen, 2 and Cook et al. 1"12-17 Cooley et al. 3 on the other hand found no dilatation or aneurysm formation in a review of their patients and our results are closest theirs. In our series a mean proximal dilatation of 5.9% and distal dilatation of 8.9% was found compared with 17.6% found by N u n n et al. 1 The diameter of 25 grafts remained unchanged and the maximum dilatation was 44% in one graft. This study tells us nothing about the causes of graft dilatation. In the literature intraoperative damage to the graft of biodegradation have been suggested as causes of graft failure. 3 •4 ' 16 - 19 None of these causes were found in our study, and no relationship was observed between anastomotic aneurysms and graft dilatation. Patients with a dilated graft did not have anastomotic aneurysms and vice versa. Only one patient with an anastomotic aneurysm, confirmed by ultrasonography, underwent reoperation. The other patients with anastomotic aneurysms or dilated grafts are simply being observed. Several of the patients are of advanced age or in poor mental or physical health and would be unlikely to tolerate another operation, and there are a few borderline cases in which we agreed upon a policy of observation. Eur J Vasc Surg Vol 6, May 1992
Conclusion
Crimped knitted Dacron grafts are commonly employed as arterial substitutes and in most instances have proven to be highly effective and relatively durable. Graft failures and ruptured anastomotic aneurysms have occurred and have usually been associated with serious haemorrhage. In our clinic the incidence of anastomotic aneurysms and the degree of dilatation are very acceptable, especially if we compare our results with those reported by other investigators. None of the patients, who underwent aorto-ilio-femoral reconstruction between 1980 and 1985, died or had to undergo acute surgery because of graft dilatation or a ruptured anastomotic aneurysm.
Acknowledgement The images were made by the Department of Radiology, Ikazia Hospital, Rotterdam. Their excellent assistance and pleasant cooperation are gratefully appreciated. We also like to thank Mr Aubrey L. Waddy for his remarks and contribution in writing this article.
References 1 NUNN DB, FREEMANMH, HUDGINSPC. Postoperative alterations in size of Dacron aortic grafts. Ann Surg 1979; 189: 741-744. 2 MAY J, STEPHEN M. Multiple aneurysms in Dacron velour grafts. Arch Surg 1978; 113: 320-321. 3 COOLEY DA, SUBRAM A, HOUCHIN DP. Clinical experience in 1040 patients with double-velour knitted Dacron vascular prostheses. Bull Texas Heart Inst 1981; 8: 320-332. 4 SIESWERDA C, SROTNICKI SH, BARENTSZJO, et al. Anastomotic a n e u r y e m s - - a n underdiagnosed complication after aorto-iliac reconstructions. Eur J Vasc Surg 1989; 3: 233-238. 5 BERGERK, SAUVAGELR. Late fiber deterioration in Dacron arterial grafts. Ann Surg 1980; 193: 477-491. 6 VAN DEN AKKER PJ, BRAND R. False aneurysms after prosthetic reconstructions for aorto-iliac obstructive disease. Ann Surg 1989; 210: 658-666. 7 BRIGGSRM, BRUCESJ, COLLINS GJ. Anastomotic aneurysms. Am J Surg 1983; 146: 770-773. 8 KALSBEEKHL. False aneurysms in patients with an aortic.bifurcation prosthesis. Ned Tijolschr Geneeskd 1989; 133: 681-683. 9 VAN HOUTEN H. Over de operatieve behandeling van het aneurysma aortae abdominalis atheroscleroticum. Diss 1962. 10 ELLISM, POWELLJT, GREENHALGHRIV[. Limitations of ultrasonography in surveillance of small abdominal aortic aneurysms. Br J Surg 1991;.78: 614-616. 11 SZILAGYIDE, WHITCOMUjG, SCHENKERW, et al. The laws of fluid flow and arterial grafting. Surgery 1960; 47: 55. 12 COOKE PA, NOBIS PA, STONEY RJ. Dacron aortic graft failure. Arch Surg 1974; 108: 101-103. 13 LORD RS, NASH PA, ILaJ BT, et al. Prospective randomized trial of polytetrafluoroethylene and Dacron aortic prosthesis. Ann Vasc Surg 1988; 3: 248-254.
Complications of Aorto-ilio-femoral Dacron Prostheses
14 MILLILIJJ, LANESJS, NEMIR P. A study of anastomotic aneurysms following aortofemoral prosthetic bypass. Ann Surg 1980; 192: 69-73. 15 OTTINGERLW, DARLINGC, WIRTHLINLS, et al. Failure of ultralightweight knitted Dacron grafts in arterial reconstruction. Arch Surg 1976; 111: 146-149. 16 PERRYMO. Early failure of Dacron prosthetic grafts. ] Cardiovasc Surg 1975; 16: 318-321. 17 RAZSO, LUNDSTRUMB, ANGQUISTKA, et al. Bilateral aneurysm of
287
Dacron graft following aorto-femoral graft operation. Acta Chir Scand 1976; 142: 479-482. 18 RAMSAYC, WILLIAM A. Aneurysms of a double velour aortic graft. Arch Surg 1984; 119: 1183-1184. 19 YASHARJJ, RICHMANMH, DYCKMANJ, et al. Failure of Dacron prostheses caused by structural defect. Surge~ 1978; 84: 659663. Accepted 16 November 1991
Eur J Vasc Surg Vol 6, May 1992
