- Email: [email protected]
Autogenous Tissue Reconstruction in the Management of Aortoiliofemoral Graft Infection
Autogenous Tissue Reconstruction in the Management of Aortoiliofemoral Graft Infection William J. Quifiones-Baldrich, MD, Hugh A. Gelabert, MD, Los Angeles, California.
Management of infected prosthetic arterial grafts remains a challenging clinical problem. Revascularization following removal of the infected arterial prosthesis is frequently necessary and usually achieved using remote extraanatomic bypasses. We present three cases where complete removal of the infected prosthesis was followed by autogenous reconstruction using reversed saphenous vein, endarterectomized segments of superficial femoral artery, or endarterectomy of the native occluded vessels. Eradication of the infection was seen in all three patients. The reconstructions have remained patent in the three patients with follow-up from four months to three years. Percutaneous balloon dilation was successful in the management of a distal limb stenosis in one patients. Autogenous vascular reconstruction following removal of infected prosthetic vascular grafts remains a viable alternative in the treatment of difficult, selected cases of infected arterial prosthesis. It is probably an underutilized alternative, when one considers that the surgical techniques employed are familiar to vascular surgeons. Cure of the infection and prepreservation of critical vascular beds can both be achieved with autogenous tissue, either in situ or from remote areas. Available percutaneous techniques may play an important role in maintaining durability of these reconstructions. (Ann Vasc Surg 1990;4:223-228). KEY WORDS: Grafts, arterial; autogenous tissue; reconstruction; infection; extraanatomic bypass.
Management of infected prosthetic arterial grafts remains challenging and carries a high morbidity and mortality. In a thorough review of the collective experience, Liekweg and Greenfield [1] reported an average mortality of 36% and an amputation rate From the Section of Vascular Surgeo', UCLA Center for the Health Sciences, Los Angeles, California. Presented at the Annual Meeting of the Peripheral Vascular Surgeo' Socieo', New York, New York, June 17. 1989. Reprint requests: William J. Qui~ones-Baldrich, MD, Department of Surgery, UCLA Center for the Health Sciences, 10833 LeConte Avenue, Los Angeles, California 90024-6904.
ranging from 23-36%, depending on location of the graft. Current standard management dictates removal of all infected prosthetic material with remote extraanatomic revascularization, if necessary. Revascularization following removal of an infected aortic prosthesis is frequently necessary. A large number of these patients were originally operated on for significant occlusive disease and others were operated on for aneurysmal disease and lack of adequate collateral circulation to sustain viability of the affected vascular bed. Extraanatomic reconstructions are usually possible and remain the backbone of reconstruction after prosthetic graft removal. Not infrequently, the extraanatomic approach requires unusual expo223
G R A F T I N F E C T I O N A N D TISSUE R E C O N S T R U C T I O N
224
a
ANNALS OF VASCULAR SURGERY
b
Fig. 1. (a) Aortogram showing failing aortobifemoral prosthesis, which on exploration proved to be infected. Note layered thrombus in right limb of graft and absence of flow into internal lilac bilaterally. (b) Aortogram two months following bilateral common internal and external thromboendarterectomy and autogenous aortobiiliac bypass using autogenous saphenous vein. Note restoration of flow ,o both internal and external lilac arteries. sures, imaginative planning, and long-term results which often depend on multiple reoperations [2]. This has led some authors to recommend in situ replacement with a new prosthetic graft. Although this is against accepted principles, successful outcomes with intermediate follow-up have been reported [3]. The resistance of autogenous tissue to infection has been previously demonstrated by others [4]. We report our experience with three patients in whom complete autogenous graft reconstruction was carded out following removal of infected aortic or iliopopliteal grafts. This approach has the potential advantage of eradicating the infection, preserving continuity of flow through the infrarenal aorta, and maintaining adequate perfusion to distal critical beds. PATIENTS AND METHODS
Patient No. 1. A 30-year-old white man was referred to the UCLA Medical Center with a two-week history of ischemic rest
pain of his right foot. Four months earlier the patient had undergone an aortobifemoral bypass after he came to an emergency room with acute bilateral lower extremity ischemia. The cause of the ischemia was undetermined, but presumed to be secondary to a traumatic aortic injury. During his initial postoperative course following aortobifemoral bypass, 13 thrombectomies were required to maintain patency of the right limb to his graft. Eventually, the patient was discharged and came to our vascular clinic with severe ischemia of his fight lower extremity. Aortography identified a failing aortofemoral graft with layered thrombus of the fight limb, complete occlusion of the right superficial femoral artery with reconstitution of a very irregular posterior tibial vessel in the calf, and occlusion of the left superficial femoral artery. No flow was seen into either internal lilac system (Fig. 1). The patient was taken to the operating room in an attempt to revise his aortofemoral bypass. On exploration, there was purulent material lining the body of the aortofemoral graft, which eventually grew Candida albicans. At the initial operation, the surgeon had stapled both common lilac arteries at their origin. After removal of the graft from the aorta, it was noted that the entire
VOLUME4 No 3 - 1990
GRAFT INFECTION AND TISSUE RECONSTRUCTION
225
a b Fig. 2. (a) Preoperative arteriogram in patient with bilateral iliopopliteal graft infection showing completely occluded left common and external lilac artery and patent ilio-right femoral graft with severe stenosis at its origin. (b) Bifurcated autogenous saphenous vein graft constructed by joining two-thirds of beveled end of proximal saphenous vein. Valve lysis using valve cutter allowed for antegrade flow.
aorta was filled with thrombotic material with minimal atherosclerotic changes. A thromboendarterectomy of the infrarenat aorta was performed, both iliac arteries were opened, and chronic thrombectomies of both external and internal iliac systems were done. A bifurcated graft using autogenous saphenous vein was constructed and placed in the terminal aorta with distal end-to-end anastomosis to both common iliac arteries. After operation the patient had palpable femoral pulses. Six days later, the patient was operated on again, and a right transpopliteal thrombectomy and in situ poplitealto-posterior tibial bypass was performed. There was difficulty in maintaining patency of this very tenuous tibial graft because of limited outflow, and thus heparin was continuously administered during and after surgery. Four days later, the patient developed intraabdominal bleeding, requiring exploration. Suture repair of an anastomotic leak on the left side was required. Heparin was discontinued and five days later, thrombosis of the right popliteal-to-tibial bypass occurred. He eventually required an above-knee amputation on the right side. Two months after his autogenous reconstruction, it was noted that his left femoral pulse had decreased. Arteriography revealed a stenosis, probably at the distal anastomosis on
the left, which was successfully managed with balloon dilatation (Fig. 1B). The patient eventually underwent a left femoropopliteal saphenous vein bypass and has been ambulating without a prosthesis. Three years following autogenous reconstruction the patient maintains a left ankle/arm index of 0.92.
Patient No, 2. A 79-year-old white woman initially came to our vascular service in 1967 because of severe bilateral lower extremity claudication. She underwent bilateral common iliac-to-popliteal bypasses. Six years later, she was diagnosed with an anastomotic pseudoaneurysm which was repaired, but recurred in 1984 and 1986. Subsequently, both iliopopliteal grafts thrombosed and she developed bilateral lower extremity ischemia with early gangrene in her left foot. She also developed draining sinuses at the popliteal incision of both legs, with cultures positive for Staphylococcus epidermidis. Her arteriogram is seen in Figure 2. The patient was taken to the operating room, where both iliopopliteal grafts and an additional patent graft that
226
G R A F T I N F E C T I O N A N D TISSUE R E C O N S T R U C T I O N
ANNALS OF VASCULAR SURGERY
the heel of his left foot, and draining osteomyelitis (Staphylococcus aureus) in the first toe of the left foot. Ankle/
C Fig. 2. (c) Aortogram showing patent bilateral aortofemoral autogenous saphenous vein bypass two months following removal of infected bilateral iliopopliteal grafts.
arm indices at that time were 0.5 and 0.3, right and left, respectively. Arteriography revealed the left iliac artery and both superficial femoral arteries to be occluded and the right iliac artery to be diseased. An aortobifemoral bypass with a distal anastomosis to both profundas was performed. His postoperative ankle/arm indices were 0.76 and 0.72, respectively. Two months later, he underwent a toe amputation without ditficulty and all wounds, including the heel ulceration, healed. Four months after his aortobifemoral bypass, he came to the emergency room with a chief complaint of drainage from the left groin. He denied fever, chills, or other systemic symptoms. On examination, he had an abscess measuring approximately 2 cm in the left groin. This had drained spontaneously, yielding a purulent drainage which eventually grew Staphylococcus aureus. A computed tomographic (CT) scan of the abdomen and pelvis, contrast fistulogram, and indium-labeled white cell scan showed aortobifemoral graft infection with involvement of both limbs of this graft. Arteriogram is shown in Figure 3. He was taken to the operating room where, through transabdominal and bilateral femoral incisions, the entire aortobifemorai prosthesis was removed. An open aortic endarterectomy and closed iliofemoral endarterectomy, using the Hall oscillating device, was performed (Fig. 3). Angioscopy of both iliac systems was used to assess the completeness of the endarterectomy. Arterial patch closure of the femoral arteries was accomplished using endarterectomized segments of superficial femoral arteries. The aorta was closed primarily. His postoperative course was uneventful, with ankle/arm indices of 0.7 bilaterally restored. He has been continued on a three month course of antibiotics and it has been four months since his surgery. His postoperative arteriogram is shown in Figure 3.
DISCUSSION had been placed from the right iliac to the femoral artery during the treatment of one of her pseudoaneurysms were removed. Her reconstruction was done by creating a bifurcated autogenous vein graft, with the proximal anastomosis in the distal aorta and distal anastomosis to both profunda femoris arteries (Fig. 2). She did well until two months later when she developed breakdown of her left heel. Repeat arteriography revealed a patent bilateral aortofemoral autogenous bypass with a superficial femoral artery occlusion and popliteal reconstitution with significant tibial disease (Fig. 2). She underwent a left distal aortofemoral limb to popliteal artery bypass using reversed cephalic vein. Her postoperative ankle/arm indices improved; however, three months later she developed thrombosis of the distal reconstruction, requiring below-knee amputation. Her aortobifemoral bypass remains patent with a follow-up of nine months and a right ankle/arm index of 0.43.
Patient No. 3. A 55-year-old white man had symptoms of severe bilateral lower extremity claudication, an ulceration over
There are several potential advantages to an autogenous reconstruction following removal of an infected vascular prosthesis. Avoidance of the use of prosthetic material, even when it was located in a remote area, is advantageous in certain clinical situations. Patients who have systemic symptoms from the graft infection are at risk of infecting the extraanatomic bypasses. In our overall experience at U C L A , 20% of patients have recurrent infection following removal of an aortic infected prosthesis [5]. Thus a combination of proper surgical drainage and debridement, adequate antibiotic therapy, and avoidance of prosthetic material may yield the best cure rate. The three patients we discuss would have been difficult candidates for extraanatomic reconstruction. The above-knee popliteal arteries were not available in either one or both lower extremities and thus were not suitable for axiUopopliteal bypasses. The long-term patency o f axillopopliteal (aboveknee) reconstructions is estimated to be around 40% at three years [5]. The long-term results o f
VOLUME 4 NO 3 - 1990
G R A F T I N F E C T I O N A N D TISSUE R E C O N S T R U C T I O N
a
227
b
Fig. 3. (a) Arteriogram prior to removal of infected aortobifemoral bypass graft. Note bilateral distal stenosis. (b) infected aortobifemoral bypass graft following removal (right) and aortobiilial femoral endarterectomy specimen obtained using the Hall arterial oscillator (closed endarterectomy).
axillo below knee-popliteal bypasses is unknown, since this procedure is rarely required. Using autogenous techniques, we were able anatomically to revascularize these patients. Although two amputations eventually resulted, these were the result of severe occlusive distal disease rather than failure of the autogenous reconstruction. Maintaining continuity of the infrarenal aorta following aortic graft removal may be of significant importance. It will avoid retrograde thrombosis with eventual renal failure and potential bowel ischemia. Restoration of pelvic circulation can be achieved with anatomic reconstruction. These are important collaterals and their preservation desirable [6]. In addition, it may decrease the incidence of aortic stump blowout. The latter is a significant cause of mortality following removal of infected aortic prosthesis [1,7]. One of the disadvantages of autogenous graft reconstruction is the inherent risk of recurrent stenosis following regional endarterectomy, use of endarterectomized segments, and saphenous vein bypasses. With
the availability of percutaneous techniques, this problem is only of relative concern. It does suggest, however, that very careful follow-up with both meticulous physical examination and noninvasive laboratory testing is of paramount importance in recognizing this problem prior to thrombosis. The largest series of autogenous tissue reconstruction in the management of infected prosthetic grafts has been reported by Ehrenfeld and colleagues [4]. In their experience with 16 patients in whom aortic grafts were removed, three patients died. Only two amputations resulted in this group, none of them secondary to inherent failure of the autogenous reconstruction. Follow-up in most patients averaged less than one year. Reporting their late results following removal of an infected prosthetic vascular graft, Reiily and associates noted a lower operative mortality and long-term survival free of related complications in 56 patients undergoing autogenous reconstruction, compared with 29 patients revascularized using remotely placed prosthetic grafts [2]. They attributed this difference to
228
GRAFT INFECTION AND TISSUE RECONSTRUCTION
ANNALS OF VASCULAR SURGERY
Although conservative treatment of prosthetic graft infection has been advocated by some [9,10] it has failed to gain popularity mainly because of its unpredictable success rate, prolonged hospitalization, and the risk of rupture and hemorrhage of an anastomosis. We firmly believe that graft removal, when possible, remains the treatment of choice. CONCLUSION Autogenous vascular reconstruction following removal of an infected prosthetic vascular graft remains a viable alternative in the treatment of difficult selected cases. It is perhaps an underutilized alternative when one considers that the surgical techniques employed are familiar to vascular surgeons. In properly selected cases, it will result in both cure of the infection and preservation of critical vascular beds. Autogenous tissue, either in situ or from remote areas, appears resistant to infection, provided that all infected tissue is adequately debrided. Careful follow-up of these patients is mandatory, given the propensity of this type of reconstruction to gradually fail over the course of time. Available percutaneous techniques may play a significant role in maintaining the durability of autogenous reconstruction in the management of infected vascular prosthesis.
C Fig. 3. (c) Aortogram two months following removal of infected aortobifemoral graft, and autogenous revascularization by aortobiilial femoral thromboendarterectomy, and bilateral femoral patch profundaplasties.
the ease of eradicating the infection in the absence of prosthetic material. Aortobifemoral saphenous vein bypass was reported by Lorentzen and Neilsen from Denmark in 1986 [8]. Early thrombosis of one of the vein grafts occurred, resulting in an amputation. Two patients died, one eight days following implantation, from rupture of the infected aorta without disruption of the autogenous reconstruction, and a second following amputation 96 days postoperatively. The remaining three patients were followed from two to ten months without evidence of infection or ischemia. This small series points out the potential of aortic disruption even after autogenous reconstruction, underscoring the importance of adequate surgical debridement, irrigation, and antibiotic therapy. In addition, we hesitate to proceed with autogenous reconstruction in the presence of pseudomonas as a causative organism, given the propensity of this bacteria to cause necrotizing changes in affected tissues. I l l
REFERENCES 1. LIEKWEG WG, GREENFIELD LJ. Vascular prosthetic infections: collected experience and results of treatment. Surgery 1977;81:335-342. 2. REILLY LM, ALTMAN H, LUSBY RJ, et al. Late results following surgical management of vascular graft infection. J Vasc Surg 1984;1:36--44. 3. WALKER WE, COOLEY DA, DUNCAN JM, et al. The management of aorto-duodenal fistula by in situ replacement of the infected abdominal aortic graft. Ann Surg 1987; 205:727-731. 4. EHRENFELD WK, WILBUR BG, OLCOTF CN, STONEY RJ. Autogenous tissue reconstruction in the management of infected prosthetic grafts. Surgery 1979;85:82-92. 5. HERNANDEZ J J, QUIlqONES-BALDRICH WJ, MOORE WS. Long-term outcome following removal of infected aortic prosthesis. Data on file; manuscript in preparation, 1989. 6. ILIOPOULUS Jl, HERMRECK AS, THOMAS JH, PIERCE GE. Hemodynamics of the hypogastric arterial circulation. J Vasc Surg 1989;9:637--642. 7. CONN JH, HARDY JD, CHAVEZ CM, FAIN WR. Infected arterial grafts: experience in 22 cases with emphasis on unusual bacteria and techniques. Ann Surg 1970;17I: 7(14-714. 8. LORENTZEN JE, NIELSEN OM. Aortobifemoral bypass with autogenous saphenous vein in treatment of pan-infected aortic bifurcation graft. J Vasc Surg 1986;3:666--668. 9. YASHAR J J, WEYMAN AK, BURNARD RJ, YASHAR J. Survival and limb salvage in patients with infected arterial prosthesis. Am J Surg 1978;135:499-504. 10. CASALI RE, TUCKER WE, THOMPSON BW, READ RC. Infected prosthetic grafts. Arch Surg 1980;115:577-580.
Management of infected prosthetic arterial grafts remains a challenging clinical problem. Revascularization following removal of the infected arterial prosthesis is frequently necessary and usually achieved using remote extraanatomic bypasses. We present three cases where complete removal of the infected prosthesis was followed by autogenous reconstruction using reversed saphenous vein, endarterectomized segments of superficial femoral artery, or endarterectomy of the native occluded vessels. Eradication of the infection was seen in all three patients. The reconstructions have remained patent in the three patients with follow-up from four months to three years. Percutaneous balloon dilation was successful in the management of a distal limb stenosis in one patients. Autogenous vascular reconstruction following removal of infected prosthetic vascular grafts remains a viable alternative in the treatment of difficult, selected cases of infected arterial prosthesis. It is probably an underutilized alternative, when one considers that the surgical techniques employed are familiar to vascular surgeons. Cure of the infection and prepreservation of critical vascular beds can both be achieved with autogenous tissue, either in situ or from remote areas. Available percutaneous techniques may play an important role in maintaining durability of these reconstructions. (Ann Vasc Surg 1990;4:223-228). KEY WORDS: Grafts, arterial; autogenous tissue; reconstruction; infection; extraanatomic bypass.
Management of infected prosthetic arterial grafts remains challenging and carries a high morbidity and mortality. In a thorough review of the collective experience, Liekweg and Greenfield [1] reported an average mortality of 36% and an amputation rate From the Section of Vascular Surgeo', UCLA Center for the Health Sciences, Los Angeles, California. Presented at the Annual Meeting of the Peripheral Vascular Surgeo' Socieo', New York, New York, June 17. 1989. Reprint requests: William J. Qui~ones-Baldrich, MD, Department of Surgery, UCLA Center for the Health Sciences, 10833 LeConte Avenue, Los Angeles, California 90024-6904.
ranging from 23-36%, depending on location of the graft. Current standard management dictates removal of all infected prosthetic material with remote extraanatomic revascularization, if necessary. Revascularization following removal of an infected aortic prosthesis is frequently necessary. A large number of these patients were originally operated on for significant occlusive disease and others were operated on for aneurysmal disease and lack of adequate collateral circulation to sustain viability of the affected vascular bed. Extraanatomic reconstructions are usually possible and remain the backbone of reconstruction after prosthetic graft removal. Not infrequently, the extraanatomic approach requires unusual expo223
G R A F T I N F E C T I O N A N D TISSUE R E C O N S T R U C T I O N
224
a
ANNALS OF VASCULAR SURGERY
b
Fig. 1. (a) Aortogram showing failing aortobifemoral prosthesis, which on exploration proved to be infected. Note layered thrombus in right limb of graft and absence of flow into internal lilac bilaterally. (b) Aortogram two months following bilateral common internal and external thromboendarterectomy and autogenous aortobiiliac bypass using autogenous saphenous vein. Note restoration of flow ,o both internal and external lilac arteries. sures, imaginative planning, and long-term results which often depend on multiple reoperations [2]. This has led some authors to recommend in situ replacement with a new prosthetic graft. Although this is against accepted principles, successful outcomes with intermediate follow-up have been reported [3]. The resistance of autogenous tissue to infection has been previously demonstrated by others [4]. We report our experience with three patients in whom complete autogenous graft reconstruction was carded out following removal of infected aortic or iliopopliteal grafts. This approach has the potential advantage of eradicating the infection, preserving continuity of flow through the infrarenal aorta, and maintaining adequate perfusion to distal critical beds. PATIENTS AND METHODS
Patient No. 1. A 30-year-old white man was referred to the UCLA Medical Center with a two-week history of ischemic rest
pain of his right foot. Four months earlier the patient had undergone an aortobifemoral bypass after he came to an emergency room with acute bilateral lower extremity ischemia. The cause of the ischemia was undetermined, but presumed to be secondary to a traumatic aortic injury. During his initial postoperative course following aortobifemoral bypass, 13 thrombectomies were required to maintain patency of the right limb to his graft. Eventually, the patient was discharged and came to our vascular clinic with severe ischemia of his fight lower extremity. Aortography identified a failing aortofemoral graft with layered thrombus of the fight limb, complete occlusion of the right superficial femoral artery with reconstitution of a very irregular posterior tibial vessel in the calf, and occlusion of the left superficial femoral artery. No flow was seen into either internal lilac system (Fig. 1). The patient was taken to the operating room in an attempt to revise his aortofemoral bypass. On exploration, there was purulent material lining the body of the aortofemoral graft, which eventually grew Candida albicans. At the initial operation, the surgeon had stapled both common lilac arteries at their origin. After removal of the graft from the aorta, it was noted that the entire
VOLUME4 No 3 - 1990
GRAFT INFECTION AND TISSUE RECONSTRUCTION
225
a b Fig. 2. (a) Preoperative arteriogram in patient with bilateral iliopopliteal graft infection showing completely occluded left common and external lilac artery and patent ilio-right femoral graft with severe stenosis at its origin. (b) Bifurcated autogenous saphenous vein graft constructed by joining two-thirds of beveled end of proximal saphenous vein. Valve lysis using valve cutter allowed for antegrade flow.
aorta was filled with thrombotic material with minimal atherosclerotic changes. A thromboendarterectomy of the infrarenat aorta was performed, both iliac arteries were opened, and chronic thrombectomies of both external and internal iliac systems were done. A bifurcated graft using autogenous saphenous vein was constructed and placed in the terminal aorta with distal end-to-end anastomosis to both common iliac arteries. After operation the patient had palpable femoral pulses. Six days later, the patient was operated on again, and a right transpopliteal thrombectomy and in situ poplitealto-posterior tibial bypass was performed. There was difficulty in maintaining patency of this very tenuous tibial graft because of limited outflow, and thus heparin was continuously administered during and after surgery. Four days later, the patient developed intraabdominal bleeding, requiring exploration. Suture repair of an anastomotic leak on the left side was required. Heparin was discontinued and five days later, thrombosis of the right popliteal-to-tibial bypass occurred. He eventually required an above-knee amputation on the right side. Two months after his autogenous reconstruction, it was noted that his left femoral pulse had decreased. Arteriography revealed a stenosis, probably at the distal anastomosis on
the left, which was successfully managed with balloon dilatation (Fig. 1B). The patient eventually underwent a left femoropopliteal saphenous vein bypass and has been ambulating without a prosthesis. Three years following autogenous reconstruction the patient maintains a left ankle/arm index of 0.92.
Patient No, 2. A 79-year-old white woman initially came to our vascular service in 1967 because of severe bilateral lower extremity claudication. She underwent bilateral common iliac-to-popliteal bypasses. Six years later, she was diagnosed with an anastomotic pseudoaneurysm which was repaired, but recurred in 1984 and 1986. Subsequently, both iliopopliteal grafts thrombosed and she developed bilateral lower extremity ischemia with early gangrene in her left foot. She also developed draining sinuses at the popliteal incision of both legs, with cultures positive for Staphylococcus epidermidis. Her arteriogram is seen in Figure 2. The patient was taken to the operating room, where both iliopopliteal grafts and an additional patent graft that
226
G R A F T I N F E C T I O N A N D TISSUE R E C O N S T R U C T I O N
ANNALS OF VASCULAR SURGERY
the heel of his left foot, and draining osteomyelitis (Staphylococcus aureus) in the first toe of the left foot. Ankle/
C Fig. 2. (c) Aortogram showing patent bilateral aortofemoral autogenous saphenous vein bypass two months following removal of infected bilateral iliopopliteal grafts.
arm indices at that time were 0.5 and 0.3, right and left, respectively. Arteriography revealed the left iliac artery and both superficial femoral arteries to be occluded and the right iliac artery to be diseased. An aortobifemoral bypass with a distal anastomosis to both profundas was performed. His postoperative ankle/arm indices were 0.76 and 0.72, respectively. Two months later, he underwent a toe amputation without ditficulty and all wounds, including the heel ulceration, healed. Four months after his aortobifemoral bypass, he came to the emergency room with a chief complaint of drainage from the left groin. He denied fever, chills, or other systemic symptoms. On examination, he had an abscess measuring approximately 2 cm in the left groin. This had drained spontaneously, yielding a purulent drainage which eventually grew Staphylococcus aureus. A computed tomographic (CT) scan of the abdomen and pelvis, contrast fistulogram, and indium-labeled white cell scan showed aortobifemoral graft infection with involvement of both limbs of this graft. Arteriogram is shown in Figure 3. He was taken to the operating room where, through transabdominal and bilateral femoral incisions, the entire aortobifemorai prosthesis was removed. An open aortic endarterectomy and closed iliofemoral endarterectomy, using the Hall oscillating device, was performed (Fig. 3). Angioscopy of both iliac systems was used to assess the completeness of the endarterectomy. Arterial patch closure of the femoral arteries was accomplished using endarterectomized segments of superficial femoral arteries. The aorta was closed primarily. His postoperative course was uneventful, with ankle/arm indices of 0.7 bilaterally restored. He has been continued on a three month course of antibiotics and it has been four months since his surgery. His postoperative arteriogram is shown in Figure 3.
DISCUSSION had been placed from the right iliac to the femoral artery during the treatment of one of her pseudoaneurysms were removed. Her reconstruction was done by creating a bifurcated autogenous vein graft, with the proximal anastomosis in the distal aorta and distal anastomosis to both profunda femoris arteries (Fig. 2). She did well until two months later when she developed breakdown of her left heel. Repeat arteriography revealed a patent bilateral aortofemoral autogenous bypass with a superficial femoral artery occlusion and popliteal reconstitution with significant tibial disease (Fig. 2). She underwent a left distal aortofemoral limb to popliteal artery bypass using reversed cephalic vein. Her postoperative ankle/arm indices improved; however, three months later she developed thrombosis of the distal reconstruction, requiring below-knee amputation. Her aortobifemoral bypass remains patent with a follow-up of nine months and a right ankle/arm index of 0.43.
Patient No. 3. A 55-year-old white man had symptoms of severe bilateral lower extremity claudication, an ulceration over
There are several potential advantages to an autogenous reconstruction following removal of an infected vascular prosthesis. Avoidance of the use of prosthetic material, even when it was located in a remote area, is advantageous in certain clinical situations. Patients who have systemic symptoms from the graft infection are at risk of infecting the extraanatomic bypasses. In our overall experience at U C L A , 20% of patients have recurrent infection following removal of an aortic infected prosthesis [5]. Thus a combination of proper surgical drainage and debridement, adequate antibiotic therapy, and avoidance of prosthetic material may yield the best cure rate. The three patients we discuss would have been difficult candidates for extraanatomic reconstruction. The above-knee popliteal arteries were not available in either one or both lower extremities and thus were not suitable for axiUopopliteal bypasses. The long-term patency o f axillopopliteal (aboveknee) reconstructions is estimated to be around 40% at three years [5]. The long-term results o f
VOLUME 4 NO 3 - 1990
G R A F T I N F E C T I O N A N D TISSUE R E C O N S T R U C T I O N
a
227
b
Fig. 3. (a) Arteriogram prior to removal of infected aortobifemoral bypass graft. Note bilateral distal stenosis. (b) infected aortobifemoral bypass graft following removal (right) and aortobiilial femoral endarterectomy specimen obtained using the Hall arterial oscillator (closed endarterectomy).
axillo below knee-popliteal bypasses is unknown, since this procedure is rarely required. Using autogenous techniques, we were able anatomically to revascularize these patients. Although two amputations eventually resulted, these were the result of severe occlusive distal disease rather than failure of the autogenous reconstruction. Maintaining continuity of the infrarenal aorta following aortic graft removal may be of significant importance. It will avoid retrograde thrombosis with eventual renal failure and potential bowel ischemia. Restoration of pelvic circulation can be achieved with anatomic reconstruction. These are important collaterals and their preservation desirable [6]. In addition, it may decrease the incidence of aortic stump blowout. The latter is a significant cause of mortality following removal of infected aortic prosthesis [1,7]. One of the disadvantages of autogenous graft reconstruction is the inherent risk of recurrent stenosis following regional endarterectomy, use of endarterectomized segments, and saphenous vein bypasses. With
the availability of percutaneous techniques, this problem is only of relative concern. It does suggest, however, that very careful follow-up with both meticulous physical examination and noninvasive laboratory testing is of paramount importance in recognizing this problem prior to thrombosis. The largest series of autogenous tissue reconstruction in the management of infected prosthetic grafts has been reported by Ehrenfeld and colleagues [4]. In their experience with 16 patients in whom aortic grafts were removed, three patients died. Only two amputations resulted in this group, none of them secondary to inherent failure of the autogenous reconstruction. Follow-up in most patients averaged less than one year. Reporting their late results following removal of an infected prosthetic vascular graft, Reiily and associates noted a lower operative mortality and long-term survival free of related complications in 56 patients undergoing autogenous reconstruction, compared with 29 patients revascularized using remotely placed prosthetic grafts [2]. They attributed this difference to
228
GRAFT INFECTION AND TISSUE RECONSTRUCTION
ANNALS OF VASCULAR SURGERY
Although conservative treatment of prosthetic graft infection has been advocated by some [9,10] it has failed to gain popularity mainly because of its unpredictable success rate, prolonged hospitalization, and the risk of rupture and hemorrhage of an anastomosis. We firmly believe that graft removal, when possible, remains the treatment of choice. CONCLUSION Autogenous vascular reconstruction following removal of an infected prosthetic vascular graft remains a viable alternative in the treatment of difficult selected cases. It is perhaps an underutilized alternative when one considers that the surgical techniques employed are familiar to vascular surgeons. In properly selected cases, it will result in both cure of the infection and preservation of critical vascular beds. Autogenous tissue, either in situ or from remote areas, appears resistant to infection, provided that all infected tissue is adequately debrided. Careful follow-up of these patients is mandatory, given the propensity of this type of reconstruction to gradually fail over the course of time. Available percutaneous techniques may play a significant role in maintaining the durability of autogenous reconstruction in the management of infected vascular prosthesis.
C Fig. 3. (c) Aortogram two months following removal of infected aortobifemoral graft, and autogenous revascularization by aortobiilial femoral thromboendarterectomy, and bilateral femoral patch profundaplasties.
the ease of eradicating the infection in the absence of prosthetic material. Aortobifemoral saphenous vein bypass was reported by Lorentzen and Neilsen from Denmark in 1986 [8]. Early thrombosis of one of the vein grafts occurred, resulting in an amputation. Two patients died, one eight days following implantation, from rupture of the infected aorta without disruption of the autogenous reconstruction, and a second following amputation 96 days postoperatively. The remaining three patients were followed from two to ten months without evidence of infection or ischemia. This small series points out the potential of aortic disruption even after autogenous reconstruction, underscoring the importance of adequate surgical debridement, irrigation, and antibiotic therapy. In addition, we hesitate to proceed with autogenous reconstruction in the presence of pseudomonas as a causative organism, given the propensity of this bacteria to cause necrotizing changes in affected tissues. I l l
REFERENCES 1. LIEKWEG WG, GREENFIELD LJ. Vascular prosthetic infections: collected experience and results of treatment. Surgery 1977;81:335-342. 2. REILLY LM, ALTMAN H, LUSBY RJ, et al. Late results following surgical management of vascular graft infection. J Vasc Surg 1984;1:36--44. 3. WALKER WE, COOLEY DA, DUNCAN JM, et al. The management of aorto-duodenal fistula by in situ replacement of the infected abdominal aortic graft. Ann Surg 1987; 205:727-731. 4. EHRENFELD WK, WILBUR BG, OLCOTF CN, STONEY RJ. Autogenous tissue reconstruction in the management of infected prosthetic grafts. Surgery 1979;85:82-92. 5. HERNANDEZ J J, QUIlqONES-BALDRICH WJ, MOORE WS. Long-term outcome following removal of infected aortic prosthesis. Data on file; manuscript in preparation, 1989. 6. ILIOPOULUS Jl, HERMRECK AS, THOMAS JH, PIERCE GE. Hemodynamics of the hypogastric arterial circulation. J Vasc Surg 1989;9:637--642. 7. CONN JH, HARDY JD, CHAVEZ CM, FAIN WR. Infected arterial grafts: experience in 22 cases with emphasis on unusual bacteria and techniques. Ann Surg 1970;17I: 7(14-714. 8. LORENTZEN JE, NIELSEN OM. Aortobifemoral bypass with autogenous saphenous vein in treatment of pan-infected aortic bifurcation graft. J Vasc Surg 1986;3:666--668. 9. YASHAR J J, WEYMAN AK, BURNARD RJ, YASHAR J. Survival and limb salvage in patients with infected arterial prosthesis. Am J Surg 1978;135:499-504. 10. CASALI RE, TUCKER WE, THOMPSON BW, READ RC. Infected prosthetic grafts. Arch Surg 1980;115:577-580.