- Email: [email protected]
Descending thoracic aorta-to-femoral artery bypass grafts
Descending
Thoracic Aorta-tocFemora1 Bypass Grafts
Artery
Paolo Sapienza, MD, Andrea Mingoli, MD, Richard J. Feldhaus, MD, Omaha, Nebraska, Filippo Napoli, MD, Andre Marsan, MD, Marco Franceschini, MD, Monaco, Luca di Marzo, MD, Omaha, Nebraska, Antonino Cavallaro, MD, Rome, Italy
BACKGROUND: Descending thoracic aorta-to-femoral artery (DTAFA) bypass graft is an alternative procedure to revascularize lower limbs. METHODS: Between 1976 and lg96,41 patients underwent DTAFA bypass grafts. Operative indications consisted of previous abdominal graft thrombosis (22 cases, group l), abdominal operations (6, group 2), initial vascular operation in the presence of difficult aortas (6, group 3), and infection of aortic grafts (5, group 4). RESULTS:Perioperative mortality was 5%. Cumulative lo-year primary patency, limb salvage, and survival rates were 64%, 7g%, and 55%, respectively. Breaking down the result on the basis of the four groups, DTAFA bypass grafts performed for infection of previous aottic grafts had a significantly lower primary patency rate (25% at 24 months; P < 0.004) with dismal limb salvage (31% at 24 months; P < 0.001) and survival rates (0% at 24 months; P < 0.005). CONCLUSIONS: DTAFA bypass grafts can be safely and durably used in patients who had thrombosis of previous abdominal grafts or had a difficult abdomen or as the initial vascular operation in the presence of difficult aortas. Conversely, dismal results are obtained in the treatment of aortic graft infection. Am J Surg. lgg7;174:662-666. 0 1997 by Excerpta Medica, Inc.
T
he first descending thoracic aorta-to-femoral artery (DTAFA) bypass graft was performed by Stevenson et al’ in June 1956 and was reported in the international literature in September 1961. The DTAFA bypass graft was inserted in a patient who presented with disabling intermittent claudication and a thrombosed aortoiliac graft. A transabdominal approach with a median sternotomy extending into the left fourth intercostal space was used to reach the thoracic aorta. A preserved aortic homograft placed intraperitoneally was sutured end-to-side between From the Department of Surgery (PS, AM, RJF, LDM), Creighton University, Omaha, Nebraska; Princess Grace Hospital (FN, AM, MF), Monaco, Principality of Monaco; and University of Rome “La Sapienza” (AC), Rome, Italy. Requests for reprints should be addressed to Andrea Mingoli, MD, 1”’ Department of Surgery, University of Rome “La Sapienza,” Policlinico Umberto I’, Viale del Policlinico, 155, 00161 Rome, Italy. Presented at the 49th Annual Meeting of the Southwestern Surgical Congress, Ranch0 Mirage, California, April 13-l 6, 1997.
662
0 1997 by Excerpta All rights reserved.
Medica,
Inc.
the thoracic aorta and the common femoral arteries bilaterally. The patient returned to a normal life, and the graft remained patent for 20 months. The first extraperitoneal DTAFA bypass graft was performed in October 1961 by Blaisdell et al’ in a patient requiring the removal of an infected aortoiliac Teflon bifurcated graft. Through an anterolateral approach in the eighth intercostal space, the descending thoracic aorta was reached, and a Dacron graft measuring 14 mm was tunnellized retroperitoneally and anastomosed end-to-side to the left femoral artery. A sovrapubic branch graft anastomosed to the right femoral artery was used to revascularize the limb. Since these original descriptions and despite its conceptual simplicity, DTAFA bypass grafts were rarely used, and their operative indications are still under debate.3-5 The aim of the present study was to investigate a series of 41 patients who underwent a DTAFA at two institutions to assess operative indications, morbidity and mortality rates, and long-term outcome.
MATERIAL
AND
METHODS
Between 1976 and 1996,41 patients underwent DTAFA bypass grafts for a total of 71 limbs revascularized. Demographics and preoperative data were retrospectively collected from hospital charts. Postoperative information was obtained from discharge notes. Long-term follow-up data were obtained from patient reexamination and noninvasive vascular laboratory investigation performed every 6 months. Surgical Technique The operative technique is similar to that described by Blaisdell et al.* Briefly, all operations were performed with end-to-side anastomoses to the lower descending thoracic aorta and the left femoral artery. Access to the thoracic aorta was gained through an anterolateral approach (left seventh interspace). The descending thoracic aorta was isolated for 5 to 6 cm. A partial occluding clamp was placed on the lower descending thoracic aorta, and a synthetic graft was anastomosed end-to-side to the aorta. A circular opening was made posteriorly in the diaphragm at its attachment to the twelfth rib. Through a small anterolateral abdominal incision, a tunnel was dissected bluntly in the left retroperitoneum. This additional incision permits an easier passage of the graft to the left femoral artery. An endto-side anastomosis was then performed at the femoral level between the common and the profunda femoris arteries. A femorofemoral bypass was used to revascularize the right limb when indicated. Right femoral anastomosis was always performed between the common and the profunda femoris 0002-9610/97/$17.00 PII SOOO2-9610(97)00184-O
1
1 DESCENDING
arteries. In 8 same inguinal flow dynamics, cases the right fascia.
THORACIC
cases to avoid a double anastomosis at the site, and on the basis of theoretically better an inverted Y graft was constructed. In these limb was passed beneath the posterior rectal
Statistical Analysis Data were entered into a computer spreadsheet (Microsoft Excel for Windows 95 version 7.0, 1985-95) and statistically analyzed with the software packages SPSS for Windows 95 (SPSS Inc., Chicago, Illinois, basic and advanced program version 6.1.3, 1995) running on a Zenith, Data Systems, Z-Station ES/P computer. Statistical comparisons were obtained either by Fisher’s exact test or chi-square test with Yates’ correction. All values are expressed as means f standard deviation. Primary and secondary patency rates and limb salvage rates were assessed by the Kaplan-Meier method. Patency and limb salvage rates were calculated on the basis of the number of limbs at risk and survival rates from the number of patients. Comparisons between curves were performed by log-rank test. P to.05 was the criterion for significance.
RESULTS There were 31 (76%) men and 10 (24%) women. Age ranged between 50 and 77 years, and mean age was 63 f 7 years (median 61). Risk factors included tobacco use in all patients, cardiac diseases (prior myocardial infarction, stable or unstable angina, ST-segment alteration on electrocardiogram) in 14 (34%), hypertension (diastolic pressure greater than 90 mm Hg) in 12 (29%), diabetes (insulin dependent or independent) in 1 (2%), pulmonary diseases (PO2 less than 60 mm Hg, PCOz greater than 50 mm Hg, pulmonary function tests less than 80% of predicted) in 1 (2%), and renal diseases (blood urea nitrogen greater than 50 mg/dL, creatinine level greater than 3 mg/dL, creatinine clearance less than 50 mL/min) in 5 (12%). Associated diseases were represented by obesity (body weight exceeding 20% of ideal) in 12 (29%) and alteration in coagulation assay in 1 (2%). Preoperative symptoms tabulated according to the parameters proposed by the Ad Hoc Committee on Reporting Standard6 were severe claudication in 22 (31%) limbs, rest pain in 36 (51%), minor tissue loss in 11 (15%), and major tissue loss in 2 (3%). Indications for operation consisted of occlusion of previous aortic reconstruction in 22 patients (group 1; mean operations per patients were 2.3), multiple nonvascular abdominal operations in 8 patients (group 2; mean operations per patients were 1.9), initial vascular operation in 6 patients with difficult aortas (group 3), and infection of previous aortic grafts in 5 patients (group 4). In the latter group of patients an axillodistal bypass graft was initially performed after infected aortic graft removal. Then, the infection of this extraanatomic reconstruction required its removal and the placement of a DTAFA. The waiting period before placement of the new graft material at a previously contaminated abdomen or femoral groin ranged between 9 and 24 months (mean 14 2 9). Risk factors, associated diseases, and symptoms are described in Tables I and 11 for each group of patients according to the indications for operation. THE
AMERICAN
AORTA-TO-FEMORAL
ARTERY
BYPASS/SAPIENZA
ET Al
All patients were preoperatively evaluated with selective angiography and duplex ultrasound scan or Doppler continuous wave examination. Seven (17%) procedures were performed from 1976 to 1982, 18 (44%) from 1983 to 1989, and 16 (39%) from 1989 to 1996. The reconstructions were carried out to both femoral arteries in 30 patients and to a single femoral artery in 11, making a total of 7 1 limbs revascularized. Dacron or expanded polytetrafluoroethylene prosthetic materials were used in 21 and 20 cases, respectively. An additional outflow procedure was associated in 10 (24%) patients (6 femoropopliteal and 4 femorotibial bypass grafts). Thirty-day mortality included 2 (5%) postoperative deaths caused by myocardial infarction and congestive heart failure. Postoperative morbidity consisted of 1 (2%) myocardial infarction, 1 (2%) renal failure requiring urgent dialysis, 1 (2%) permanent minor stroke, and 1 graft occlusion (2 limbs, 5%). Graft failure was managed with thrombectomy, and in 1 limb also by improving the outflow by means of a femoropopliteal bypass graft (the patient died on the fifth postoperative day of myocardial infarction). These features represent a cumulative 16% mortality and morbidity rate. No major limb amputations were recorded during the postoperative period. Two (5%) patients were submitted to a minor amputation. Mean intraoperative blood loss was 850 2 210 mL (range 530 mL to 1,230 mL, median 800 mL). Mean postoperative hospital stay averaged 13 -+ 4 days (range 10 to 18, median 14) and intensive care stay was 4 + 2 days (range 2 to 7, median 4). Complete follow-up information (mean 55 + 5 months, median 51, range 1 to 144) were available for 35 (90%) patients (62 limbs, 87%). Four patients (6 limbs) were in fact lost to follow-up at 29, 42, 57, and 65 months with a patent graft. Eight (13%) patients had a thrombosis of the right graft limb (10,24,46,48,58,64,67, and 72 months). One (2%) patient had a left graft limb thrombosis (16 months), 1 (2%) patient developed a pseudoaneurysm at the left graft limb (65 months) and 3 (5%) patients had a DTAFA bypass graft infection after 6, 16, and 21 months. Graft thromboses were always managed with thrombectomy. In 2 cases the outflow was also improved with a femoropoliteal and a femoroposterior tibia1 bypass graft. The pseudoaneurysm was resected, and the distal anastomosis was distalized to the profunda femoris. Infected grafts were explanted, and an axillodistal bypass graft was performed in 2 cases. In the other patient, DTAFA bypass graft was partially explanted, and a redo graft with a different intraabdominal route was inserted. During follow-up we performed 11 additional peripheral vascular procedures to improve the outflow (7 femoropopliteal and 4 femorotibial bypass grafts). Eight (13%) patients underwent a major amputation at 6, 16, 16, 21, 24, 46, 58 and 72 months. Cumulative 5- and lo-year primary (Figure 1) and secondary patency rates were 80% and 64% and 86% and 76%, respectively. Limb salvage rates were 86% at 5 years and 79% at 10 years. Six (17%) patients died of unrelated causes at 35,46, 54, 62, 72 and 91 months. Specifically, 4 patients died of cardiac diseases and 2 of cancer. Three (9%) patients operated on for previous aortic graft infection died of related causes JOURNAL
OF SURGERY@
VOLUME
174
DECEMBER
1997
663
DESCENDING
THORACIC
AORTA-TO-FEMORAL
TABLE
ARTERY
BYPASWSAPIENZA
ET AL 1
I Demographic
Age (mean ? years) Median Gender Male Female Tobaccouse Cardiac diseases Hypertension Diabetes Pulmonary diseases Renal diseases Obesity Altered coagulation
TABLE
Information
Associated
Disease
Group 2 (n = 8)
Group 3 (n = 8)
Group 4 (n = 5)
64 k 6 65
62 2 4 61
59 2 3 60
64 + 5 63
17 5 (100%) (36%) (36%) (4%) (4%) (8%) (32%) (4%)
8 3 2 0 0 1 2 0
6 2 (100%) (37%) (25%) (0%) (0%) (12%) (25%) (0%)
5 1 (100%) (33%) (17%) (0%) (0%) (17%) (33%) (0%)
5 1 1 0 0 1 1 0
3 2 (100%) (20%) (20%) (0%) (0%) (20%) (20%) (0%)
Preoperative
Symptoms
22 8 8 1 1 2 7 1
assay
and
Group 1 (n = 22)
6 2 1 0 0 1 2 0
II
Category Category Category Category
3 4 5 6
(severe claudication) (rest pain) (minor tissue loss) (major tissue loss)
Group 1 (n = 38)
Group 2 (n = 18)
Group 3 (n = 11)
Group 4 (n = 8)
12 17 6 1
5 8 2 1
3 7 1 0
2 4 2 0
(33%) (47%) (17%) (3%)
(31%) (50%) (12%) (6%)
(27%) (63%) (10%) (0%)
(25%) (50%) (25%) (0%)
r 80
80
I s 5
80
0
12
Figure 1. Primary Ian-Meier method.
24
36
patency
46
rates
60
12
84
tabulated
98
108 120 132 144
according
to the Kap-
because of DTAFA bypass graft infection at 13, 16, and 24 months. Cumulative 5 and lo-year survival rates were 75% and 55%, respectively (Figure 2). Comparative Results Demographics, risk factors, associated diseases and preoperative symptoms were represented similarly among the groups (P = NS). One patient in group 1 and 1 patient in group 4 died in the early postoperative period (P = NS). Patients in group 4 had a higher incidence of postoperative complications (1 permanent minor stoke and 2 graft failures) as compared with the other groups (P <0.009). No statistical differences were observed among the groups with regard to intraoperative blood loss, intensive care and hospital stay (P = NS). Long-term primary patency rates were significantly better in group 1 (65% at 144 months), group 2 (70% at 94 months), and group 3 (86% at 128 months) as compared
664
THE
AMERICAN
JOURNAL
OF SURGERY@
VOLUME
174
I
80
40 20 0 0
12
24
36
48
*cum”,atlw *awp
Figure 2. Survival method.
rates
tabulated
60
72
84
96
108 120 132 IU
MONTHS 3 eGmp2 *molp 3 --Group 1
according
to the Kaplan-Meier
with group 4 (25% at 24 months) (I’
1997
DESCENDING
THORACIC
13, 46, and 91 months), 2 patients of group 2 (35 and 62 months), 2 of group 3 (54 and 72 months), and 4 patients of group 4 (1, 13, 16, and 24 months) died.
COMMENTS Infrarenal aortobifemoral bypass graft is the gold standard procedure for aortoiliofemoral reconstruction in atherosclerotic occlusive disease. The cumulative patency rates in large historical series range from 83% to 92% at 5 years, from 71% to 83% at 10 years, and from 61% to 74% at 15 years.‘-” Late bypass graft failures occur in 6% to 13% of patients because of progressive atherosclerotic involvement of the outflow and/or inflow and infection.7’11”2 Reoperative abdominal aortic surgery is, therefore, not infrequently required to manage graft failure and its sequelae. During the past decades axillofemoral bypass graft was widely used as the treatment of choice in patients with aortic graft infection or difficult abdomen after multiple vascular procedures. Despite historical axillofemoral bypass graft, 5-year patency rates were discouraging and ranged from 9% and 30%.13,14 Re cent series have reported more optimistic patency rates (67% to 80%)‘5,1” because of the introduction of externally supported prosthetic material and more accurate patient selection. The DTAFA bypass graft represents another alternative in the management of aortic graft failures. It has an optimal inflow source, a shorter and more direct course even if the distal anastomosis is placed on the distal profunda femoris or the supragenicular popliteal artery, and it is shielded from extrinsic compression by its retroperitoneal course. These valuable characteristics are confirmed by the good short- and long-term results obtained with this reconstruction.3-5x17 Low perioperative mortality and morbidity rates (5% and 1 l%, respectively) and optimal 5-year primary patency (80%) and limb salvage (86%) rates were observed also in out series. These results indicate that DTAFA bypass graft outcome is comparable with that obtainable with aortobifemoral grafts. Moreover, we observed an overall 5-year survival rate of 77%. It was lower (10 to 15 years less) than that which might be anticipated for a normal age- and gender-matched population but similar to those expectable after aortofemoral reconstruction.“s The contraindications to DTAFA bypass grafts, as suggested by McCarthy et al’ and Criado et a1,4 are the presence of severe pulmonary disease and the theoretical blood flow steal from visceral arteries. In our opinion, although the presence of severe pulmonary function test impairment (contraindicating a thoracotomy) mandates the use an axillofemoral bypass graft, the stealing of blood from visceral arteries, in the absence of hemodynamically significant atherosclerotic lesions, is unlikely. In a previous study from our center, no early or late significant variation of serum creatinine and blood urea nitrogen values were reported in patients who underwent DTAFA bypass grafts.‘s Current experiences with DTAFA bypass grafts state that indications for this procedure are reoperation for recurrent obstruction after aortofemoral bypass graft, when a more proximal source of inflow is required, when a direct approach to the abdominal aorta is not feasible or is ill advised, when repeated failures of axillofemoral grafts have occurred, or when the axillary or subclavian artery is unsuitable for inflow source. THE
AMERICAN
AORTA-TO-FEMORAL
ARTERY
BYPASSlSAPlEN2.A
ET Al
-3
In our opinion, DTAFA bypass grafts could be also safely chosen as the initial vascular operation in the presence of extremely calcified infrarenal aortas. We inserted a DTAFA bypass graft in 6 patients with occlusion of the abdominal aorta at the level of the renal arteries. In this subgroup of patients, no postoperative mortality or morbidity was recorded. Five-year patency and limb salvage rates were optimal. However, although these data are extremely encouraging, we do not currently advocate the DTAFA bypass graft as a primary operation for routine aortic occlusive disease because of the simplicity and familiarity of infrarenal aortic exposure and the small number of patients we have treated. Another indication for DTAFA bypass grafts that has been proposed but is still under debate, is the treatment of previous infected abdominal reconstruction. McCarthy et al2 have recently demonstrated that DTAFA bypass graft is an excellent reconstruction for survivors of an aortic graft infection. The authors reported 1 infection-related mortality out of 12 DTAFA bypass grafts inserted because of previous vascular graft infection. Conversely, despite the complete healing of the abdominal cavity and groins, in our series DTAFA bypass graft infections and sepsis-related mortality occurred in a high percentage of patients. Our results and the extremely demanding management of an infected DTAFA bypass graft suggest the use of an axillofemoral graft or a conservative treatment. In conclusion, the DTAFA bypass graft is a safe and durable procedure for the treatment of aortoiliac disease when opening the abdomen is contraindicated or ill advised. The indications for this type of operation should be limited to patients who have had multiple failures of previous abdominal grafts, had difficult abdomen or as initial vascular operation in the presence of difficult aortas. The choice of DTAFA bypass graft for the treatment of multiple graft infections should be supported by larger successful experiences.
REFERENCES 1. Stevenson JK, Sauvage LR, Harkins HN. A bypass homograft from thoracic aorta to femoral arteries for occlusive vascular disease. Ann Surg. 1961;27:632-637. 2. Blaisdell FW, DeMattei G, Gauder PJ. Extraperitoneal thoracic aorta-to-femoral bypass graft as replacement for an infected aortic bifurcation prosthesis. Am J Surg. 1961;102:583-585. 3. McCarthy W], Mesh CL, McMillan WD, et al. Descending thoracic aorta-to-femoral artery bypass: ten years’ experience with a durable procedure. J Vast Surg. 1993;17:336-348. 4. Criado E, Johnson G Jr, Burnham SJ, et al. Descending thoracic aorta-to-femoral artery bypass as an alternative to aortoiliac reconstruction. .I Vnsc Surg. 1992;15:550-557. 5. Branchereau A, Espinoza H, Rudondy P, et al. Descending thoracic aorta as an inflow source for late occlusive failures following aortoiliac reconstruction. Ann Vast Surg. 1991;5:8-15. 6. Ad Hoc Committee on Reporting Standards. Suggested standards for reports dealing with lower extremity ischemia. J Vast Surg. 1986;4:80-94. 7. Crawford ES, Bomberger RA, Glaeser DH, et al. Aortoiliac occlusive disease: factors influencing survival and function following reconstructive operation over a twenty-five-year period. Surgery. 1981;90:1055-1067. 8. Szilagyi DE, Elliott JP Jr, Smith RF, et al. A thirty-year survey of the reconstructive surgical treatment of aortoiliac occlusive disease. J Vast Surg. 1986;3:421-436. JOURNAL
OF SURGERY@
VOLUME
174
DECEMBER
1997
I
665
ESCENDING
THORACIC
AORTA-TO-FEMORAL
ARTERY
BYPASWSAPIENZA
9. Perdue GD, Smith RB, Veazey CR, Ansley JD. Revascularization for severe limb ischemia. Arch Surg. 1980;115:168-171. 10. Brewster DC, Darling CR. Optimal methods of aortoiliac reconstruction. Surgery. 1978;84:739-748. 11. Thompson WM, Johnsrud IS, Jackson DC, et al. Late complications of abdominal aortic reconstructive surgery: roentgen evaluation. Ann Surg. 1977;185:326-334. 12. Crawford S, Manning LG, Kelly TF. Redo surgery after operations for aneurysm and occlusion of the abdominal aorta. Surgery. 1977;81:41-52. 13. Moore WS, Hall AD, Blaisdell FW. Late results of axillary-femoral bypass grafting. Am ] Surg. 1971;122: 148-154.
ET AL 1
14. Eugene J, Goldstone J, Moore WS. Fifteen-year experience with subcutaneous bypass grafts for lower extremity ischemia. Ann Surg. 1977;186:177-183. 15. Passman MA, Taylor LM Jr, Moneta GL, et al. Comparison of axillofemoral and aortofemoral bypass for aortoiliac occlusive disease. J Vast Surg. 1996;23:263-271. 16. Burrell MJ, Wheeler JR, Gregory RT, Snyder SO. Axillofemoral bypass: a ten-year review. Ann Surg. 1982;195:796-799. 17. Hussain SA. Descending thoracic aorta to bifemoral by-pass graft without laparotomy. Int Surg. 1988;73:260-263. 18. Schultz RD, Sterpetti AV, Feldhaus RJ. Thoracic aorta as source of inflow in reoperation for occluded aortoiliac reconstruction. Surgery. 1986; 100:635-644.
DISCUSSION Charles H. McCollum, MD (Houston, Texas): I think that all vascular surgeons would agree that the best operation for aorto-iliac occlusive disease is the standard bypass graft from the infrarenal aorta to either the iliac or femoral. Yet, alternate surgical procedures are occasionally required. Four indications are described for performing the DTAFA bypass. These represent a mixed group and perhaps should not all be grouped together. Patients requiring surgery for “redo” abdominal aortic surgery and patients with a hostile abdomen, including prior abdominal irradiation or known extensive peritoneal adhesions, are closely related. Another alternate operation that avoids laparotomy in these difficult situations provides a retroperitoneal approach to the abdominal aorta through a thoraco-abdominal incision. The problem of the infected abdominal aortic graft is a very difficult one and perhaps should be examined separately. Management of these problems is controversial. Many surgeons recommend removing the infected infrarenal abdominal aorta graft and restoring arterial flow to the lower extremities utilizing an axillo-femoral artery by pass. Several of my colleagues have had considerable success utilizing an alternate procedure. The infected graft is removed along with a wide local debridement. A new prosthetic graft is replaced in the in-situ position. This must be accompanied by antibiotic catheter irrigations, drainage, and most importantly, coverage of the prosthetic graft by viable tissue, using either omentum or a muscle flap. Lifelong suppressive antibiotic therapy is mandatory. I have several questions for the authors. Several of their patients developed infection in the axillo-femoral bypass graft prior to performing the DTAFA bypass. Do you have any recommendations to avoid this complication, and what was your protocol for antibiotic therapy? Also, has your
abdominal aorta for patients with redo surgery or in patients with a hostile abdomen? The abstract describes a total of 41 patients who had the DTAFA bypass. Thirty had grafts to both extremities, and 11 had revascularization of a single limb. What were the criteria for using a single versus a bilateral distal anastomosis? Did the right leg ever represent the single limb? Group III was described as “difficult aorta,” or the small calcified aorta. Please clarify and elaborate on this indication and how you made this determination preoperatively. DTAFA bypass is one of several alternative methods in dealing with difficult abdominal aortic problems. Every vascular surgeon should be knowledgeable about this technique.
group
thoracic
aorta.
DECEMBER
1997
666
considered
THE
AMERICAN
using
the
JOURNAL
retroperitoneal
OF SURGERY=’
approach
to the
VOLUME
174
CLOSING Paolo Sapienza, MD: In the present series, only 4 patients were operated on for infection of a previous bypass graft. We usually use other kinds of treatment for these patients. This represents their last chance. At my university hospital we do not use antibiotic irrigation or in situ graft replacement after valve debridement. We take out the graft prosthesis. I know that Dr. Feldhaus does the same for his cases in Omaha. We are not used to the retroperitoneal approach to the abdominal aorta. It is much easier, in my opinion, to go through the chest especially if someone is not used to making this approach to the abdominal aorta. So it is only a subjective preference to go through the chest instead of going through the retroperitoneum. This group of patients came from the experience of Dr. Marsan from Monte Carlo. He preferred to approach the descending thoracic aorta because it is much easier than approaching the calcified aorta into the abdomen. It is much
easier
to
put
in
a lateral
clamp
to
the
descending
Thoracic Aorta-tocFemora1 Bypass Grafts
Artery
Paolo Sapienza, MD, Andrea Mingoli, MD, Richard J. Feldhaus, MD, Omaha, Nebraska, Filippo Napoli, MD, Andre Marsan, MD, Marco Franceschini, MD, Monaco, Luca di Marzo, MD, Omaha, Nebraska, Antonino Cavallaro, MD, Rome, Italy
BACKGROUND: Descending thoracic aorta-to-femoral artery (DTAFA) bypass graft is an alternative procedure to revascularize lower limbs. METHODS: Between 1976 and lg96,41 patients underwent DTAFA bypass grafts. Operative indications consisted of previous abdominal graft thrombosis (22 cases, group l), abdominal operations (6, group 2), initial vascular operation in the presence of difficult aortas (6, group 3), and infection of aortic grafts (5, group 4). RESULTS:Perioperative mortality was 5%. Cumulative lo-year primary patency, limb salvage, and survival rates were 64%, 7g%, and 55%, respectively. Breaking down the result on the basis of the four groups, DTAFA bypass grafts performed for infection of previous aottic grafts had a significantly lower primary patency rate (25% at 24 months; P < 0.004) with dismal limb salvage (31% at 24 months; P < 0.001) and survival rates (0% at 24 months; P < 0.005). CONCLUSIONS: DTAFA bypass grafts can be safely and durably used in patients who had thrombosis of previous abdominal grafts or had a difficult abdomen or as the initial vascular operation in the presence of difficult aortas. Conversely, dismal results are obtained in the treatment of aortic graft infection. Am J Surg. lgg7;174:662-666. 0 1997 by Excerpta Medica, Inc.
T
he first descending thoracic aorta-to-femoral artery (DTAFA) bypass graft was performed by Stevenson et al’ in June 1956 and was reported in the international literature in September 1961. The DTAFA bypass graft was inserted in a patient who presented with disabling intermittent claudication and a thrombosed aortoiliac graft. A transabdominal approach with a median sternotomy extending into the left fourth intercostal space was used to reach the thoracic aorta. A preserved aortic homograft placed intraperitoneally was sutured end-to-side between From the Department of Surgery (PS, AM, RJF, LDM), Creighton University, Omaha, Nebraska; Princess Grace Hospital (FN, AM, MF), Monaco, Principality of Monaco; and University of Rome “La Sapienza” (AC), Rome, Italy. Requests for reprints should be addressed to Andrea Mingoli, MD, 1”’ Department of Surgery, University of Rome “La Sapienza,” Policlinico Umberto I’, Viale del Policlinico, 155, 00161 Rome, Italy. Presented at the 49th Annual Meeting of the Southwestern Surgical Congress, Ranch0 Mirage, California, April 13-l 6, 1997.
662
0 1997 by Excerpta All rights reserved.
Medica,
Inc.
the thoracic aorta and the common femoral arteries bilaterally. The patient returned to a normal life, and the graft remained patent for 20 months. The first extraperitoneal DTAFA bypass graft was performed in October 1961 by Blaisdell et al’ in a patient requiring the removal of an infected aortoiliac Teflon bifurcated graft. Through an anterolateral approach in the eighth intercostal space, the descending thoracic aorta was reached, and a Dacron graft measuring 14 mm was tunnellized retroperitoneally and anastomosed end-to-side to the left femoral artery. A sovrapubic branch graft anastomosed to the right femoral artery was used to revascularize the limb. Since these original descriptions and despite its conceptual simplicity, DTAFA bypass grafts were rarely used, and their operative indications are still under debate.3-5 The aim of the present study was to investigate a series of 41 patients who underwent a DTAFA at two institutions to assess operative indications, morbidity and mortality rates, and long-term outcome.
MATERIAL
AND
METHODS
Between 1976 and 1996,41 patients underwent DTAFA bypass grafts for a total of 71 limbs revascularized. Demographics and preoperative data were retrospectively collected from hospital charts. Postoperative information was obtained from discharge notes. Long-term follow-up data were obtained from patient reexamination and noninvasive vascular laboratory investigation performed every 6 months. Surgical Technique The operative technique is similar to that described by Blaisdell et al.* Briefly, all operations were performed with end-to-side anastomoses to the lower descending thoracic aorta and the left femoral artery. Access to the thoracic aorta was gained through an anterolateral approach (left seventh interspace). The descending thoracic aorta was isolated for 5 to 6 cm. A partial occluding clamp was placed on the lower descending thoracic aorta, and a synthetic graft was anastomosed end-to-side to the aorta. A circular opening was made posteriorly in the diaphragm at its attachment to the twelfth rib. Through a small anterolateral abdominal incision, a tunnel was dissected bluntly in the left retroperitoneum. This additional incision permits an easier passage of the graft to the left femoral artery. An endto-side anastomosis was then performed at the femoral level between the common and the profunda femoris arteries. A femorofemoral bypass was used to revascularize the right limb when indicated. Right femoral anastomosis was always performed between the common and the profunda femoris 0002-9610/97/$17.00 PII SOOO2-9610(97)00184-O
1
1 DESCENDING
arteries. In 8 same inguinal flow dynamics, cases the right fascia.
THORACIC
cases to avoid a double anastomosis at the site, and on the basis of theoretically better an inverted Y graft was constructed. In these limb was passed beneath the posterior rectal
Statistical Analysis Data were entered into a computer spreadsheet (Microsoft Excel for Windows 95 version 7.0, 1985-95) and statistically analyzed with the software packages SPSS for Windows 95 (SPSS Inc., Chicago, Illinois, basic and advanced program version 6.1.3, 1995) running on a Zenith, Data Systems, Z-Station ES/P computer. Statistical comparisons were obtained either by Fisher’s exact test or chi-square test with Yates’ correction. All values are expressed as means f standard deviation. Primary and secondary patency rates and limb salvage rates were assessed by the Kaplan-Meier method. Patency and limb salvage rates were calculated on the basis of the number of limbs at risk and survival rates from the number of patients. Comparisons between curves were performed by log-rank test. P to.05 was the criterion for significance.
RESULTS There were 31 (76%) men and 10 (24%) women. Age ranged between 50 and 77 years, and mean age was 63 f 7 years (median 61). Risk factors included tobacco use in all patients, cardiac diseases (prior myocardial infarction, stable or unstable angina, ST-segment alteration on electrocardiogram) in 14 (34%), hypertension (diastolic pressure greater than 90 mm Hg) in 12 (29%), diabetes (insulin dependent or independent) in 1 (2%), pulmonary diseases (PO2 less than 60 mm Hg, PCOz greater than 50 mm Hg, pulmonary function tests less than 80% of predicted) in 1 (2%), and renal diseases (blood urea nitrogen greater than 50 mg/dL, creatinine level greater than 3 mg/dL, creatinine clearance less than 50 mL/min) in 5 (12%). Associated diseases were represented by obesity (body weight exceeding 20% of ideal) in 12 (29%) and alteration in coagulation assay in 1 (2%). Preoperative symptoms tabulated according to the parameters proposed by the Ad Hoc Committee on Reporting Standard6 were severe claudication in 22 (31%) limbs, rest pain in 36 (51%), minor tissue loss in 11 (15%), and major tissue loss in 2 (3%). Indications for operation consisted of occlusion of previous aortic reconstruction in 22 patients (group 1; mean operations per patients were 2.3), multiple nonvascular abdominal operations in 8 patients (group 2; mean operations per patients were 1.9), initial vascular operation in 6 patients with difficult aortas (group 3), and infection of previous aortic grafts in 5 patients (group 4). In the latter group of patients an axillodistal bypass graft was initially performed after infected aortic graft removal. Then, the infection of this extraanatomic reconstruction required its removal and the placement of a DTAFA. The waiting period before placement of the new graft material at a previously contaminated abdomen or femoral groin ranged between 9 and 24 months (mean 14 2 9). Risk factors, associated diseases, and symptoms are described in Tables I and 11 for each group of patients according to the indications for operation. THE
AMERICAN
AORTA-TO-FEMORAL
ARTERY
BYPASS/SAPIENZA
ET Al
All patients were preoperatively evaluated with selective angiography and duplex ultrasound scan or Doppler continuous wave examination. Seven (17%) procedures were performed from 1976 to 1982, 18 (44%) from 1983 to 1989, and 16 (39%) from 1989 to 1996. The reconstructions were carried out to both femoral arteries in 30 patients and to a single femoral artery in 11, making a total of 7 1 limbs revascularized. Dacron or expanded polytetrafluoroethylene prosthetic materials were used in 21 and 20 cases, respectively. An additional outflow procedure was associated in 10 (24%) patients (6 femoropopliteal and 4 femorotibial bypass grafts). Thirty-day mortality included 2 (5%) postoperative deaths caused by myocardial infarction and congestive heart failure. Postoperative morbidity consisted of 1 (2%) myocardial infarction, 1 (2%) renal failure requiring urgent dialysis, 1 (2%) permanent minor stroke, and 1 graft occlusion (2 limbs, 5%). Graft failure was managed with thrombectomy, and in 1 limb also by improving the outflow by means of a femoropopliteal bypass graft (the patient died on the fifth postoperative day of myocardial infarction). These features represent a cumulative 16% mortality and morbidity rate. No major limb amputations were recorded during the postoperative period. Two (5%) patients were submitted to a minor amputation. Mean intraoperative blood loss was 850 2 210 mL (range 530 mL to 1,230 mL, median 800 mL). Mean postoperative hospital stay averaged 13 -+ 4 days (range 10 to 18, median 14) and intensive care stay was 4 + 2 days (range 2 to 7, median 4). Complete follow-up information (mean 55 + 5 months, median 51, range 1 to 144) were available for 35 (90%) patients (62 limbs, 87%). Four patients (6 limbs) were in fact lost to follow-up at 29, 42, 57, and 65 months with a patent graft. Eight (13%) patients had a thrombosis of the right graft limb (10,24,46,48,58,64,67, and 72 months). One (2%) patient had a left graft limb thrombosis (16 months), 1 (2%) patient developed a pseudoaneurysm at the left graft limb (65 months) and 3 (5%) patients had a DTAFA bypass graft infection after 6, 16, and 21 months. Graft thromboses were always managed with thrombectomy. In 2 cases the outflow was also improved with a femoropoliteal and a femoroposterior tibia1 bypass graft. The pseudoaneurysm was resected, and the distal anastomosis was distalized to the profunda femoris. Infected grafts were explanted, and an axillodistal bypass graft was performed in 2 cases. In the other patient, DTAFA bypass graft was partially explanted, and a redo graft with a different intraabdominal route was inserted. During follow-up we performed 11 additional peripheral vascular procedures to improve the outflow (7 femoropopliteal and 4 femorotibial bypass grafts). Eight (13%) patients underwent a major amputation at 6, 16, 16, 21, 24, 46, 58 and 72 months. Cumulative 5- and lo-year primary (Figure 1) and secondary patency rates were 80% and 64% and 86% and 76%, respectively. Limb salvage rates were 86% at 5 years and 79% at 10 years. Six (17%) patients died of unrelated causes at 35,46, 54, 62, 72 and 91 months. Specifically, 4 patients died of cardiac diseases and 2 of cancer. Three (9%) patients operated on for previous aortic graft infection died of related causes JOURNAL
OF SURGERY@
VOLUME
174
DECEMBER
1997
663
DESCENDING
THORACIC
AORTA-TO-FEMORAL
TABLE
ARTERY
BYPASWSAPIENZA
ET AL 1
I Demographic
Age (mean ? years) Median Gender Male Female Tobaccouse Cardiac diseases Hypertension Diabetes Pulmonary diseases Renal diseases Obesity Altered coagulation
TABLE
Information
Associated
Disease
Group 2 (n = 8)
Group 3 (n = 8)
Group 4 (n = 5)
64 k 6 65
62 2 4 61
59 2 3 60
64 + 5 63
17 5 (100%) (36%) (36%) (4%) (4%) (8%) (32%) (4%)
8 3 2 0 0 1 2 0
6 2 (100%) (37%) (25%) (0%) (0%) (12%) (25%) (0%)
5 1 (100%) (33%) (17%) (0%) (0%) (17%) (33%) (0%)
5 1 1 0 0 1 1 0
3 2 (100%) (20%) (20%) (0%) (0%) (20%) (20%) (0%)
Preoperative
Symptoms
22 8 8 1 1 2 7 1
assay
and
Group 1 (n = 22)
6 2 1 0 0 1 2 0
II
Category Category Category Category
3 4 5 6
(severe claudication) (rest pain) (minor tissue loss) (major tissue loss)
Group 1 (n = 38)
Group 2 (n = 18)
Group 3 (n = 11)
Group 4 (n = 8)
12 17 6 1
5 8 2 1
3 7 1 0
2 4 2 0
(33%) (47%) (17%) (3%)
(31%) (50%) (12%) (6%)
(27%) (63%) (10%) (0%)
(25%) (50%) (25%) (0%)
r 80
80
I s 5
80
0
12
Figure 1. Primary Ian-Meier method.
24
36
patency
46
rates
60
12
84
tabulated
98
108 120 132 144
according
to the Kap-
because of DTAFA bypass graft infection at 13, 16, and 24 months. Cumulative 5 and lo-year survival rates were 75% and 55%, respectively (Figure 2). Comparative Results Demographics, risk factors, associated diseases and preoperative symptoms were represented similarly among the groups (P = NS). One patient in group 1 and 1 patient in group 4 died in the early postoperative period (P = NS). Patients in group 4 had a higher incidence of postoperative complications (1 permanent minor stoke and 2 graft failures) as compared with the other groups (P <0.009). No statistical differences were observed among the groups with regard to intraoperative blood loss, intensive care and hospital stay (P = NS). Long-term primary patency rates were significantly better in group 1 (65% at 144 months), group 2 (70% at 94 months), and group 3 (86% at 128 months) as compared
664
THE
AMERICAN
JOURNAL
OF SURGERY@
VOLUME
174
I
80
40 20 0 0
12
24
36
48
*cum”,atlw *awp
Figure 2. Survival method.
rates
tabulated
60
72
84
96
108 120 132 IU
MONTHS 3 eGmp2 *molp 3 --Group 1
according
to the Kaplan-Meier
with group 4 (25% at 24 months) (I’
1997
DESCENDING
THORACIC
13, 46, and 91 months), 2 patients of group 2 (35 and 62 months), 2 of group 3 (54 and 72 months), and 4 patients of group 4 (1, 13, 16, and 24 months) died.
COMMENTS Infrarenal aortobifemoral bypass graft is the gold standard procedure for aortoiliofemoral reconstruction in atherosclerotic occlusive disease. The cumulative patency rates in large historical series range from 83% to 92% at 5 years, from 71% to 83% at 10 years, and from 61% to 74% at 15 years.‘-” Late bypass graft failures occur in 6% to 13% of patients because of progressive atherosclerotic involvement of the outflow and/or inflow and infection.7’11”2 Reoperative abdominal aortic surgery is, therefore, not infrequently required to manage graft failure and its sequelae. During the past decades axillofemoral bypass graft was widely used as the treatment of choice in patients with aortic graft infection or difficult abdomen after multiple vascular procedures. Despite historical axillofemoral bypass graft, 5-year patency rates were discouraging and ranged from 9% and 30%.13,14 Re cent series have reported more optimistic patency rates (67% to 80%)‘5,1” because of the introduction of externally supported prosthetic material and more accurate patient selection. The DTAFA bypass graft represents another alternative in the management of aortic graft failures. It has an optimal inflow source, a shorter and more direct course even if the distal anastomosis is placed on the distal profunda femoris or the supragenicular popliteal artery, and it is shielded from extrinsic compression by its retroperitoneal course. These valuable characteristics are confirmed by the good short- and long-term results obtained with this reconstruction.3-5x17 Low perioperative mortality and morbidity rates (5% and 1 l%, respectively) and optimal 5-year primary patency (80%) and limb salvage (86%) rates were observed also in out series. These results indicate that DTAFA bypass graft outcome is comparable with that obtainable with aortobifemoral grafts. Moreover, we observed an overall 5-year survival rate of 77%. It was lower (10 to 15 years less) than that which might be anticipated for a normal age- and gender-matched population but similar to those expectable after aortofemoral reconstruction.“s The contraindications to DTAFA bypass grafts, as suggested by McCarthy et al’ and Criado et a1,4 are the presence of severe pulmonary disease and the theoretical blood flow steal from visceral arteries. In our opinion, although the presence of severe pulmonary function test impairment (contraindicating a thoracotomy) mandates the use an axillofemoral bypass graft, the stealing of blood from visceral arteries, in the absence of hemodynamically significant atherosclerotic lesions, is unlikely. In a previous study from our center, no early or late significant variation of serum creatinine and blood urea nitrogen values were reported in patients who underwent DTAFA bypass grafts.‘s Current experiences with DTAFA bypass grafts state that indications for this procedure are reoperation for recurrent obstruction after aortofemoral bypass graft, when a more proximal source of inflow is required, when a direct approach to the abdominal aorta is not feasible or is ill advised, when repeated failures of axillofemoral grafts have occurred, or when the axillary or subclavian artery is unsuitable for inflow source. THE
AMERICAN
AORTA-TO-FEMORAL
ARTERY
BYPASSlSAPlEN2.A
ET Al
-3
In our opinion, DTAFA bypass grafts could be also safely chosen as the initial vascular operation in the presence of extremely calcified infrarenal aortas. We inserted a DTAFA bypass graft in 6 patients with occlusion of the abdominal aorta at the level of the renal arteries. In this subgroup of patients, no postoperative mortality or morbidity was recorded. Five-year patency and limb salvage rates were optimal. However, although these data are extremely encouraging, we do not currently advocate the DTAFA bypass graft as a primary operation for routine aortic occlusive disease because of the simplicity and familiarity of infrarenal aortic exposure and the small number of patients we have treated. Another indication for DTAFA bypass grafts that has been proposed but is still under debate, is the treatment of previous infected abdominal reconstruction. McCarthy et al2 have recently demonstrated that DTAFA bypass graft is an excellent reconstruction for survivors of an aortic graft infection. The authors reported 1 infection-related mortality out of 12 DTAFA bypass grafts inserted because of previous vascular graft infection. Conversely, despite the complete healing of the abdominal cavity and groins, in our series DTAFA bypass graft infections and sepsis-related mortality occurred in a high percentage of patients. Our results and the extremely demanding management of an infected DTAFA bypass graft suggest the use of an axillofemoral graft or a conservative treatment. In conclusion, the DTAFA bypass graft is a safe and durable procedure for the treatment of aortoiliac disease when opening the abdomen is contraindicated or ill advised. The indications for this type of operation should be limited to patients who have had multiple failures of previous abdominal grafts, had difficult abdomen or as initial vascular operation in the presence of difficult aortas. The choice of DTAFA bypass graft for the treatment of multiple graft infections should be supported by larger successful experiences.
REFERENCES 1. Stevenson JK, Sauvage LR, Harkins HN. A bypass homograft from thoracic aorta to femoral arteries for occlusive vascular disease. Ann Surg. 1961;27:632-637. 2. Blaisdell FW, DeMattei G, Gauder PJ. Extraperitoneal thoracic aorta-to-femoral bypass graft as replacement for an infected aortic bifurcation prosthesis. Am J Surg. 1961;102:583-585. 3. McCarthy W], Mesh CL, McMillan WD, et al. Descending thoracic aorta-to-femoral artery bypass: ten years’ experience with a durable procedure. J Vast Surg. 1993;17:336-348. 4. Criado E, Johnson G Jr, Burnham SJ, et al. Descending thoracic aorta-to-femoral artery bypass as an alternative to aortoiliac reconstruction. .I Vnsc Surg. 1992;15:550-557. 5. Branchereau A, Espinoza H, Rudondy P, et al. Descending thoracic aorta as an inflow source for late occlusive failures following aortoiliac reconstruction. Ann Vast Surg. 1991;5:8-15. 6. Ad Hoc Committee on Reporting Standards. Suggested standards for reports dealing with lower extremity ischemia. J Vast Surg. 1986;4:80-94. 7. Crawford ES, Bomberger RA, Glaeser DH, et al. Aortoiliac occlusive disease: factors influencing survival and function following reconstructive operation over a twenty-five-year period. Surgery. 1981;90:1055-1067. 8. Szilagyi DE, Elliott JP Jr, Smith RF, et al. A thirty-year survey of the reconstructive surgical treatment of aortoiliac occlusive disease. J Vast Surg. 1986;3:421-436. JOURNAL
OF SURGERY@
VOLUME
174
DECEMBER
1997
I
665
ESCENDING
THORACIC
AORTA-TO-FEMORAL
ARTERY
BYPASWSAPIENZA
9. Perdue GD, Smith RB, Veazey CR, Ansley JD. Revascularization for severe limb ischemia. Arch Surg. 1980;115:168-171. 10. Brewster DC, Darling CR. Optimal methods of aortoiliac reconstruction. Surgery. 1978;84:739-748. 11. Thompson WM, Johnsrud IS, Jackson DC, et al. Late complications of abdominal aortic reconstructive surgery: roentgen evaluation. Ann Surg. 1977;185:326-334. 12. Crawford S, Manning LG, Kelly TF. Redo surgery after operations for aneurysm and occlusion of the abdominal aorta. Surgery. 1977;81:41-52. 13. Moore WS, Hall AD, Blaisdell FW. Late results of axillary-femoral bypass grafting. Am ] Surg. 1971;122: 148-154.
ET AL 1
14. Eugene J, Goldstone J, Moore WS. Fifteen-year experience with subcutaneous bypass grafts for lower extremity ischemia. Ann Surg. 1977;186:177-183. 15. Passman MA, Taylor LM Jr, Moneta GL, et al. Comparison of axillofemoral and aortofemoral bypass for aortoiliac occlusive disease. J Vast Surg. 1996;23:263-271. 16. Burrell MJ, Wheeler JR, Gregory RT, Snyder SO. Axillofemoral bypass: a ten-year review. Ann Surg. 1982;195:796-799. 17. Hussain SA. Descending thoracic aorta to bifemoral by-pass graft without laparotomy. Int Surg. 1988;73:260-263. 18. Schultz RD, Sterpetti AV, Feldhaus RJ. Thoracic aorta as source of inflow in reoperation for occluded aortoiliac reconstruction. Surgery. 1986; 100:635-644.
DISCUSSION Charles H. McCollum, MD (Houston, Texas): I think that all vascular surgeons would agree that the best operation for aorto-iliac occlusive disease is the standard bypass graft from the infrarenal aorta to either the iliac or femoral. Yet, alternate surgical procedures are occasionally required. Four indications are described for performing the DTAFA bypass. These represent a mixed group and perhaps should not all be grouped together. Patients requiring surgery for “redo” abdominal aortic surgery and patients with a hostile abdomen, including prior abdominal irradiation or known extensive peritoneal adhesions, are closely related. Another alternate operation that avoids laparotomy in these difficult situations provides a retroperitoneal approach to the abdominal aorta through a thoraco-abdominal incision. The problem of the infected abdominal aortic graft is a very difficult one and perhaps should be examined separately. Management of these problems is controversial. Many surgeons recommend removing the infected infrarenal abdominal aorta graft and restoring arterial flow to the lower extremities utilizing an axillo-femoral artery by pass. Several of my colleagues have had considerable success utilizing an alternate procedure. The infected graft is removed along with a wide local debridement. A new prosthetic graft is replaced in the in-situ position. This must be accompanied by antibiotic catheter irrigations, drainage, and most importantly, coverage of the prosthetic graft by viable tissue, using either omentum or a muscle flap. Lifelong suppressive antibiotic therapy is mandatory. I have several questions for the authors. Several of their patients developed infection in the axillo-femoral bypass graft prior to performing the DTAFA bypass. Do you have any recommendations to avoid this complication, and what was your protocol for antibiotic therapy? Also, has your
abdominal aorta for patients with redo surgery or in patients with a hostile abdomen? The abstract describes a total of 41 patients who had the DTAFA bypass. Thirty had grafts to both extremities, and 11 had revascularization of a single limb. What were the criteria for using a single versus a bilateral distal anastomosis? Did the right leg ever represent the single limb? Group III was described as “difficult aorta,” or the small calcified aorta. Please clarify and elaborate on this indication and how you made this determination preoperatively. DTAFA bypass is one of several alternative methods in dealing with difficult abdominal aortic problems. Every vascular surgeon should be knowledgeable about this technique.
group
thoracic
aorta.
DECEMBER
1997
666
considered
THE
AMERICAN
using
the
JOURNAL
retroperitoneal
OF SURGERY=’
approach
to the
VOLUME
174
CLOSING Paolo Sapienza, MD: In the present series, only 4 patients were operated on for infection of a previous bypass graft. We usually use other kinds of treatment for these patients. This represents their last chance. At my university hospital we do not use antibiotic irrigation or in situ graft replacement after valve debridement. We take out the graft prosthesis. I know that Dr. Feldhaus does the same for his cases in Omaha. We are not used to the retroperitoneal approach to the abdominal aorta. It is much easier, in my opinion, to go through the chest especially if someone is not used to making this approach to the abdominal aorta. So it is only a subjective preference to go through the chest instead of going through the retroperitoneum. This group of patients came from the experience of Dr. Marsan from Monte Carlo. He preferred to approach the descending thoracic aorta because it is much easier than approaching the calcified aorta into the abdomen. It is much
easier
to
put
in
a lateral
clamp
to
the
descending