Primary Aortic Infections Research Articles

Primary Aortic Endograft Infections Have Worse Outcomes Compared to Infections Involving Aortic Surgical Grafts or Primary Mycotic Aortic Infections

Aortic infection is associated with high morbidity and mortality, but the outcomes of prosthetic (surgical or endovascular) grafts compared to primary aortic infections are poorly defined. This large single-center retrospective study aims to compare the outcomes of primary and prosthetic aortic graft infections. Patients diagnosed with primary or infected aortic grafts between January 2000 and December 2022 were included. Patients were group based on the infection type (primary, surgical graft and endograft). Baseline demographics, symptoms, imaging, interventions, and outcomes were evaluated. Primary outcomes included overall and reintervention-free survival. Kaplan-Meier survival and Cox proportional hazards analysis were performed. A total of 145 patients presented during the study period with infections of the native aorta (34.5%) and aortic graft (65.5%). Baseline characteristics and treatments are summarized in the Table. In-hospital mortality (36.7% vs 15.4% vs 10%; P = .01) and 30-day complication rates (67% vs 40% vs 38%; P = .03) were highest among endografts, followed by surgical grafts, and lowest among primary infections respectively. Primary aortic infections had the highest 30-day, 1-year, and 5-year survival on Kaplan-Meier analysis compared to surgical and endograft infections (Figure). Reintervention-free survival was also highest in primary infections at 1 and 5 years (log-rank P < .001), while no difference was found at 30 days. On multivariate analysis, infected surgical grafts and endografts were associated with a significantly higher 30-day mortality (hazard ratio [HR], 9.18; P = .015 and HR 7.58; P = .02, respectively). Visceral artery involvement or isolated groin infections did not influence mortality at any timepoint nor did treatment type (in situ repair vs extra-anatomical vs endovascular), while gram-positive tissue cultures were associated with higher 30-day and 1-year mortality (HR, 17.6; P = .01 and HR 4.14; P = .004, respectively) when compared to patients who had no growth on tissue cultures. There was no difference in the duration of intravenous antibiotic treatment between groups (42 vs 45 vs 44 days; P = .84). Infected aortic endografts have lower short and long-term survival independent of treatment type as well as a lower rate of long-term intervention-free survival when compared to surgical graft and primary aortic infections. Gram positive intraoperative cultures were associated with increased mortality. Predictors of endograft infection need to be determined to minimize complications and careful consideration is necessary before employing endovascular interventions if there is concern an underlying mycotic process.TableBaseline demographics, presenting symptoms, and treatmentsPrimary infections, n = 50 (34.5%)Surgical grafts, n = 65 (44.8%)Endografts, n = 30 (20.7%)P valuePrior aortic pathology Aneurysmal disease26 (40%)27 (90%)<.001 Occlusive disease8 (12%)1 (3%) Other (dissection)5 (8%)1 (3%) Unknown26 (40%)1 (3%)Visceral involvement8 (16.3%)5 (7.8%)3 (9.7).35Infected graft location Groin4 (6.3%)0 (0%).01 Abdomen38 (59.4%)28 (93.4%) Abdomen + groin15 (23.4%)1 (3.3%) Chest7 (10.9%)1 (3.3%)Age, mean (SD)64.9 (9.1)67.8 (10.9)73.1 (10.5).003Gender Male34 (68%)42 (65%)26 (87%).078 Female16 (32%)23 (35%)4 (13%)Race Non-Hispanic White44 (88%)61 (94%)24 (80%).059 Black5 (10%)1 (2%)2 (7%) Other1 (2%)3 (5%)4 (13%)History of peripheral artery disease2 (4%)31 (48%)9 (30%)<.001Diabetes15 (30%)9 (14%)3 (10%).046End-stage renal disease4 (8%)1 (2%)3 (10%).13Hypertension36 (72%)48 (74%)19 (63%).56Coronary artery disease15 (30%)19 (29%)16 (53%).060Recent infection (3 weeks)24 (48%)15 (23%)6 (20%).007Recent surgery (3 weeks)1 (2%)8 (12%)3 (10%).11Antibiotics on presentation2 (4%)17 (26%)9 (30%).001Abdominal pain30 (60%)21 (32%)11 (37%).010Back pain26 (52%)6 (9%)10 (33%)<.001Groin symptoms0 (0%)15 (23%)2 (7%)<.001Fever2 (4%)8 (12%)10 (33%).002Rupture on computed tomography26 (53%)3 (6%)3 (11%)<.001Surgical/tissue culture results Fungal0 (0%)2 (5%)1 (4%).015 Gram negative4 (13%)9 (21%)1 (4%) Gram positive10 (32%)7 (17%)9 (36%) Mixed0 (0%)7 (17%)6 (24%) No growth17 (55%)17 (40%)8 (32%)Blood culture results Fungal0 (0%)1 (2%)0 (0%).017 Gram negative4 (9%)2 (4%)2 (7%) Gram positive15 (34%)5 (9%)6 (22%) No growth25 (57%)47 (85%)19 (70%)Antibiotics given21 (42%)64 (98%)30 (100%)<.001Antibiotic duration, median (IQR), days42 (9, 96)45 (30, 61)44 (29, 52).84Intraoperative purulence noted0 (0%)27 (55%)25 (100%)<.00130-day complications19 (38%)26 (40%)20 (67%).02830-Day readmission12 (27%)11 (19%)3 (11%).29Treatment Open in situ repair18 (36%)13 (20%)7 (23%).007 Extra-anatomical bypass15 (30%)31 (48%)16 (53%) Endovascular16 (32%)8 (13%)1 (3%) Medical0 (0%)5 (8%)3 (10%) Palliative0 (0%)2 (3%)1 (3%) Other (debridement)1 (2%)5 (8%)2 (8%)Recurrent infections3 (33.3%)6 (27.3%)1 (14.3%).6830-Day survival44 (89.8%)52 (89%)21 (66.7%).041-Year survival35 (73.3%)41 (69.1%)10 (36.7%)<.0015-Year survival20 (58.8%)22 (42.8%)10 (36.7%).0230-Day reintervention-free survival41 (83.7%)45 (61.4%)19 (62.7%).121-Year reintervention-free survival31 (65.1%)31 (52.3%)5 (20.7%)<.0015-Year reintervention-free survival16 (48.3%)15 (27.1%)2 (16.5%)<.001 Open table in a new tab

Management and Outcome of Non-Aneurysmal Primary Aortic Infection.

Aortic infection without prior intervention or aneurysm is exceedingly rare. We report the presentation, diagnosis, management, and outcome of patients with this unusual entity. Retrospective chart and imaging review of patients with primary aortic infection. 5 patients (3 male, mean age 71.2years) presented between 2014 and 2022. All had abdominal, back, or flank pain. Four had constitutional symptoms. All were evaluated with a complete blood count; 3 had leukocytosis. Both serum C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR) were elevated in the 4 patients evaluated with these tests. All were studied with peripheral blood culture on the first hospital day prior to any antibiotic administration. Blood culture was positive in only 1 patient. Computed tomography (CT) scan showed periaortic inflammation without aneurysm in all. Fluorodeoxyglucose positron emission tomography (PET) was obtained in 3 and a radiolabeled leukocyte single-photon emission CT (SPECT) scan was performed in 2. All demonstrated periaortic concentration of the radioisotope consistent with inflammation or infection. Intraoperative cultures were positive in 3. One patient who had a negative intraoperative culture was examined with broad range polymerase chain reaction (PCR) and DNA sequencing which identified a causative bacterium. The other patient with a negative intraoperative culture had periaortic abscess but was on antibiotics preoperatively, potentially confounding the culture. All patients underwent in-situ repair with rifampin impregnated polyester (N = 2), cryopreserved aortic allograft (N = 2), or autogenous femoral vein (N = 1). No patient developed recurrent infection or aortic related complications following surgery with an average follow up of 31.8months (range 8-88months). Patients with primary aortic infection present similarly with the triad of abdominal or back pain, laboratory markers of infection, and imaging demonstrating periaortic inflammation. Patients were treated successfully with in-situ repair. Preoperative identification of a causative organism was difficult, and PCR may be useful to help identify an organism.

Outcome in Patients after Autologous Femoral Vein Reconstruction for Primary Aortic Infection and Aortic Graft Infection: A Case Series

For surgical treatment of primary aortic infection and aortic graft infection, in situ reconstruction with autologous vein(s) has the lowest rates of re-infection and of graft thrombosis. In this study, we have assessed the outcome after autologous femoral vein reconstruction in patients with aortic (graft) infection and we provide insights into the specific technical surgical considerations of the procedure. In this retrospective single-center study, all patients who underwent autologous femoral vein reconstruction because of primary aortic infection or aortic graft infection between January 2012 and January 2020 were included. The primary outcome parameter was 30-day mortality. Twenty-nine patients with autologous femoral vein reconstruction for a primary aortic infection (n=3) or aortic graft infection (n=26) were included. An aorto-enteral fistula was detected in 13 patients (49%). Venous reconstruction of the aorta was performed with a single femoral vein in 17 patients (59%), and two femoral veins in 12 patients (41%). Thirty-day mortality was 17%. Relapse of infection occurred in two patients (7%) and no amputations were needed. One year after surgery, only three patients (10%) still needed stockings and after 2 years none of the patients used stockings. Central aortic reconstruction with femoral veins is a durable solution for primary aortic and aortoiliac graft infections with a low incidence of reinfections, amputations, and venous hypertension.

Femoral Vein Reconstruction for Aortic Infections.

While rare, abdominal aortic infections remain one of the most technically and emotionally challenging cases that a vascular surgeon may face. Secondary infections of either endovascular, or open aortic reconstructions range from 0.2% to 8%. Primary aortic infections are much more rare. Diagnosis can be elusive, depending upon the virulence of the causative microbes, and extent of the infection. Patients are often brittle, with immunocompromise and malnutrition prevalent in this patient population. The gold standard diagnostic test remains a computed tomographic angiogram. The mainstay of management requires vascular control, and wide debridement of all infected materials and revascularization. Multiple methods exist to reconstruct the vascular supply. The neo-aortoiliac system (NAIS) is attractive as it utilizes the patient’s own femoral veins to reconstruct the vascular supply after the infection has been extirpated. The procedure is demanding upon the patient and surgeons alike. Also, the rarity of aortic infections limit experiences the literature to centers of excellence. However, the NAIS resists infection well, leaving the patient without any remaining foreign bodies. No further costs for conduit are incurred. Moreover, multiple experiences show excellent durability. While comparative effectiveness literature remains sparse, we believe the NAIS to be the optimal method of revascularization for select patients. In this article, we will review the use of NAIS for primary and secondary aortic infections. In particular, we will emphasize procedural details to help enable the reader to apply this procedure most effectively to their own patients.

Rifampin Soaking Dacron-Based Endografts for Implantation in Infected Aortic Aneurysms—New Application of a Time-Tested Principle

Infections involving the aorta are associated with high rates of morbidity and mortality, and their management is complex. Saturating Dacron grafts in rifampin (60 mg/mL) inhibits the growth of organisms commonly found to be involved in both primary aortic infections and aortoenteric fistulas. Open repair and replacement of the aorta with rifampin-soaked Dacron grafts is frequently used in clinical practice and is considered a viable option for open repair with a low recurrence of infection; however, the morbidity and mortality of the procedure is significant. More recently, patients who are high risk for open surgery have been managed with endografts to treat infected aortas and aortoenteric fistulas with limited success, a high recurrence rate, and elevated mortality. We describe a technique to expose Dacron endografts with rifampin delivered via injection port or into the sheath before deployment in selected patients with aortic infections. We used this novel technique in 2 patients who were high risk for open repair: 1 with a bleeding aortoenteric fistula and 1 with mycotic abdominal aortic aneurysm. The first patient tolerated 1.5 years without surgical correction of the duodenal defect after placement of a rifampin-treated endograft. This allowed her to recover and ultimately undergo definitive repair under elective circumstances. Our second patient remains without evidence of recurrence 1 year after implantation for a mycotic abdominal aortic aneurysm. Following the principles of rifampin use in open vascular repairs, treating Dacron endografts with rifampin may add similar antimicrobial resistance when used to treat selected aortic infections.

Use of Cryopreserved Aortoiliac Allograft (CAA) for Aortic Reconstruction in the United States

Aortic infections, even with treatment, have a high mortality and risk of recurrent infection and limb loss. CAA has been proposed for in-line reconstruction to improve outcomes in this high-risk population. A multi-center study using a standardized database was performed at 14 of the highest volume institutions who used CAA for aortic infection. Two hundred and twenty patients (mean age = 65; M:F = 1.6/1) were treated since 2000 with 283 CAAs for prosthetic graft infection (59%), primary aortic infection (17%), enteric fistula/erosion (16%), mycotic aneurysm (4%), and other (4%). Intra-op cultures indicated infection in 66%, most frequently polymicrobial. Distal anastomosis was to the femoral artery, iliac, then distal aorta. 30-day mortality was 9% and procedure related major complications occurred in 24%, including persistent sepsis (n = 17), graft thrombosis (n = 9), graft/stump rupture (n = 8), recurrent CAA/aortic infection (n = 8), pseudoaneurysm (n = 6), recurrence of AE fistula (n = 4), and compartment syndrome (n = 1). Hospital LOS was 24 days. Ten (5%) required allograft explant; 2 developed CAA aneurysm requiring resection at 23 and 40 mo. Primary graft patency and freedom from limb loss were 93% and 97%, respectively, at 5 yr. Patient survival was 75% at 1 yr and 51% at 5 yr. This largest study indicates that CAA allows in-line reconstruction of aortic infection with lower early and long-term morbidity and mortality than other previously reported treatment options. Repair with CAA is associated with low rates of aneurysm formation, recurrent infection, and limb loss.

Contained Ruptured Paravisceral Aortic Aneurysm Related to Immunoglobulin G4 Aortitis

To describe a case of autoimmune inflammatory abdominal aortic aneurysm (AAA) associated with rupture. A 63-year-old woman presented with 5 days of abdominal pain, malaise, fever, and chills after 6 months of debilitating back pain with a 3-kg weight loss. On examination, she was shown to have a tender palpable pulsatile abdominal mass. Computed tomographic angiography revealed a multilobulated paravisceral AAA (5.5 cm in maximal diameter) and bilateral popliteal aneurysms. The appearance of the aneurysms was indicative of primary aortic infection. Laboratory examinations demonstrated a white blood cell (WBC) count of 12.3×10(9)/L, erythrocyte sedimentation rate of 131 mm/hr, normal antinuclear antibody level, and C-reactive protein level of 211 mg/L. Nuclear WBC scan showed no uptake of tracer around the aorta. Blood and urine cultures were negative. Because of the AAA size and symptoms, open repair was expedited. The operation was performed through a transabdominal midline incision with a mediovisceral rotation. Extensive retroperitoneal inflammation extending into the paravisceral aorta was encountered. Supraceliac clamping was possible. The aorta was replaced from the level of the superior mesenteric artery to the aortic bifurcation with a 16-mm rifampin-soaked graft with reimplantation of the left renal artery. Cultures and biopsies were done. Histology demonstrated vessel wall rupture, adventitial fibrosis and inflammatory cell infiltration, obliterative phlebitis, lymphoid follicles, perineural inflammation, and immunoglobulin G4 (IgG4) plasma cell infiltration, consistent with a contained ruptured aneurysm associated with IgG4 periaortitis. The patient had a long postoperative course with prolonged intubation and renal failure requiring hemodialysis, which resolved 8 weeks postoperatively. Immunosuppression was paramount for her improvement. IgG4-related inflammatory AAAs are rare; this is the first report of one with a contained rupture. The patient's symptoms, the unusual appearance on computed tomography, the presence of popliteal aneurysms in a woman, and the normal WBC scan were indicative of an inflammatory etiology. Tissue biopsy was critical to obtain histological diagnosis and direct treatment.

Primary aortic infections and infected aneurysms.

Primary infections of the aorta and infected aortic aneurysms are rare and are life threatening. Most of them are due to bacterial infection occurring in an atheromatous plaque or a pre existing aneurysm during bacteremia. Rarely spread from a contiguous septic process may be the cause. The reported hospital mortality ranges from 16-44%. Gram positive bacteria are still the most common causative organisms. More recently, Gram negative bacilli are seen increasingly responsible. The mortality rate is higher for the Gram negative infection since they most often cause supra renal aneurysms and are more prone for rupture. Best results are achieved by appropriate antibiotics and aggressive surgical treatment. Excision of the infected aneurysm sac as well as surrounding tissue and in situ reconstruction of aorta is the preferred treatment. Pedicled omental cover also helps to reduce infection. Long term antibiotic is needed to prevent reinfection. Mortality is high for those who undergo emergency operation, with advanced age and for nonsalmonella infection.

Primary Aortic Infections and Infected Aneurysms

Primary Aortic Infections and Infected Aneurysms

Successful Treatment of Ruptured Mycotic Aneurysm of the Iliac Artery in a Patient with Salmonella Sepsis

The term ‘mycotic aneurysm’ was first used by Sir William Osler in 1885 when he described multiple bead-like aneurysms resembling a fungal growth, which were associated with bacterial endocarditis. Currently, however, the term ‘mycotic’ is applied to any aneurysm from which an organism can be cultured. In 25% of patients over 50 years of age with Salmonella sepsis, arteritis due to Salmonella has been reported. Infection of pre-existing atherosclerotic lesions has become more prevalent and this mechanism now accounts for most mycotic aneurysms. Different species of Salmonella were the causative organism in 33–36% of reported cases of primary aortic infection prior to 1965, but thereafter only 10%. The majority of mycotic aneurysms due to Salmonella species involve the abdominal aorta or the iliac arteries. The reported overall mortality rate for this entity is reported as over 50%, however, the mortality and morbidity are much higher in patients with rupture. The first successfully treated case of Salmonella aortitis was reported in 1962. Since then, sporadic instances have been reported in the literature with a consistent pattern of delayed diagnosis and treatment. The complete excision of the aneurysmal sac is generally accepted, however, in situ versus extraanatomic reconstruction remains controversial. Case Report

In situ revascularization with silver-coated polyester grafts to treat aortic infection: early and midterm results

In this prospective study we analyzed the immediate and midterm outcome in patients with abdominal aorta infection (mycotic aneurysm, prosthetic graft infection) managed by excision of the aneurysm or the infected vascular prosthesis and in situ replacement with a silver-coated polyester prosthesis. From January 2000 to December 2001, 27 consecutive patients (25 men, 2 women; mean age, 69 years) with an abdominal aortic infection were entered in the study at seven participating centers. Infection was managed with either total (n = 18) or partial (n = 6) excision of the infected aorta and in situ reconstruction with an InterGard Silver (IGS) collagen and silver acetate-coated polyester graft. Assessment of outcome was based on survival, limb salvage, persistent or recurrent infection, and prosthetic graft patency. Twenty-four patients had prosthetic graft infections, graft-duodenal fistula in 12 and graft-colonic fistula in 1; and the remaining 3 patients had primary aortic infections. Most organisms cultured were of low virulence. The IGS prosthesis was placed emergently in 11 patients (41%). Mean follow-up was 16.5 months (range, 3-30 months). Perioperative mortality was 15%; all four patients who died had a prosthetic graft infection. Actuarial survival at 24 months was 85%. No major amputations were noted in this series. Recurrent infection developed in only one patient (3.7%). Postoperative antibiotic therapy did not exceed 3 months, except in one patient. No incidence of prosthetic graft thrombosis was noted during follow-up. Preliminary results in this small series demonstrate favorable outcome with IGS grafts used to treat infection in abdominal aortic grafts and aneurysms caused by organisms with low virulence. Larger series and longer follow-up will be required to compare the role of IGS grafts with other treatment options in infected fields.

Long-term results of cryopreserved arterial allograft reconstruction in infected prosthetic grafts and mycotic aneurysms of the abdominal aorta

Purpose: This prospective, observational study determined the long-term outcome in patients with abdominal aortic infection (primary or prosthetic graft) who were treated with simultaneous aortic/graft excision and cryopreserved arterial allograft reconstruction. Methods: From April 1992 to March 2000, patients with abdominal aortic infection underwent complete or partial excision of the infected aorta/prosthetic graft and cryopreserved arterial allograft reconstruction. Arterial allografts were harvested from multiple organ donors and cryopreserved at –80°C without rate-controlled freezing. The patients were observed for survival, limb salvage, persistence and/or recurrence of infection, and allograft patency. The results were calculated with life-table methods. Results: During the 8-year study period, 28 consecutive patients (27 men, 1 woman; mean age, 64 years) underwent treatment for abdominal aortic infection (23 graft infections, including 7 graft-enteric fistulas and 5 primary aortic infections). Allograft reconstruction was performed as an emergency procedure in 13 patients (46%). The mean follow-up period was 35.4 months (range, 6-101 months). The overall treatment-related mortality rate was 17.8% (17% for graft infection, 20% for primary aortic infection). The overall 3-year survival was 67%. There was no early or late amputation. There was no persistent or recurrent infection, and none of the patients received long-term (> 3 months) antibiotic therapy. Reoperation for allograft revision, excision, or replacement was necessary in four patients (17%) who were available for examination, with no reoperative perioperative death. The 3-year primary and secondary allograft patency rates were 81% and 96%, respectively. Conclusion: Our experience with cryopreserved arterial allograft in the management of abdominal aortic infection suggests that this technique seems to be a useful option for treating one of the most dreaded vascular complications. (J Vasc Surg 2001;34:616-22.)

Use of Rifampin-Soaked Gelatin-Sealed Polyester Grafts for in Situ Treatment of Primary Aortic and Vascular Prosthetic Infections

Background. In situ treatment of artery/graft infection has distinct advantages compared to vessel excision and extra-anatomic bypass procedures. Based on animal studies of a rifampin-soaked, gelatin-impregnated polyester graft that demonstrated prolonged in vivo antibacterial activity, this antibiotic-bonded graft was used selectively in patients for in situ treatment of low-grade Gram-positive prosthetic graft infections or primary aortic infections not amenable to excision and ex situ bypass.Methods. In a 5-year period (1995–1999), 27 patients with prosthetic graft infection (aortofemoral, n = 18, femorofemoral, n = 3; axillofemoral, n = 1) or primary aortic infection (mycotic aneurysm, n = 3; infected AAA, n = 2) underwent excision of the infected vessel and in situ replacement with a rifampin soaked (45–60 mg/ml for 15 min) gelatin-impregnated polyester graft. All prosthetic graft infections were low grade in nature, caused Gram-positive bacteria (Staphylococcus epidermidis, 16; Staphylococcus aureus, 5; Streptococcus, 1), and were treated electively. Patients with mycotic aortic aneurysm presented with sepsis and underwent urgent or emergent surgery.Results. Two (8%) patients died—1 as a result of a ruptured Salmonella mycotic aortic aneurysm and the other from methicillin-resistant S. aureus infection following deep vein replacement of an in situ replaced femorofemoral graft. No amputations or late deaths as the result of vascular infection occurred in the 25 surviving patients. Two patients developed recurrent infection caused by a rifampin-resistant S. epidermidis in a replaced aortofemoral graft limb and were successfully treated with graft excision and in situ autogenous vein replacement. Eighteen patients remain alive and clinically free of infection after a mean follow-up interval of 17 months.Conclusions. In situ replacement treatment using a rifampin-bonded prosthetic graft for low-grade staphylococcal arterial infection was safe, durable, and associated with eradication of clinical signs of infection. Failure of this therapy was the result of virulent and antibiotic-resistant bacterial strains.

Improved results with conventional management of infrarenal aortic infection

Purpose: Interest in alternative methods, such as autogenous vein grafts and aortic allografts, for the management of infrarenal aortic infection (IRAI) has been stimulated by the historically disappointing results with conventional surgical management. Recently, there have been dramatic improvements in the results of axillofemoral bypass grafting (AXFB) followed by excision of the IRAI that have gone relatively unrecognized. The purpose of this report is the presentation of modern-day results in the treatment of IRAI with conventional surgical methods. Methods: From January 1, 1983, through June 30, 1998, patients with IRAI underwent treatment with AXFB and complete excision of the IRAI. The patients were followed for survival, limb salvage, and AXFB graft patency. The results were tabulated with life-table methods. Results: During the 15-year study period, 60 patients (51 men, nine women; mean age, 68 years) underwent treatment for IRAI (50 graft infections, including 16 graft-enteric fistulae, and 10 primary aortic infections). The mean follow-up period was 41 months. The perioperative mortality rate was 13% (12% for graft infection, and 20% for primary infection). The overall 2-year and 5-year survival rates were 67% and 47%, respectively. The limb salvage rates at 2 and 5 years were 93% and 82%, respectively. The 5-year primary AXFB graft patency rate was 73%. Conclusion: These results show an improvement with the conventional management of IRAI equal or superior to those results reported with alternative methods, including femoral vein grafts or aortic allografts. These results should be regarded as the modern standard with which alternative therapies can be compared. (J Vasc Surg 1999;30:76-83.)