Audra A. Noel

Open surgical and endovascular treatment of superior vena cava syndrome caused by nonmalignant disease

OBJECTIVES The purpose of this study was to evaluate the role of endovascular and open surgical reconstructions in patients with superior vena cava (SVC) syndrome caused by nonmalignant disease. METHODS Clinical data from 32 consecutive patients who underwent endovascular or open surgical reconstruction of central veins because of symptomatic benign SVC syndrome between November 1983 and June 2001 were retrospectively reviewed. RESULTS The study included 17 male and 15 female patients (mean age, 38 years; range, 5-69 years). Presenting symptoms were head fullness (n = 26), dyspnea or orthopnea (n = 23), headache (n = 17), or dizziness (n = 11); physical signs were head swelling (n = 31), chest wall collateral vessels (n = 29), facial cyanosis (n = 18), or arm swelling (n = 17). Etiologic factors included mediastinal fibrosis (n = 19), indwelling catheter (n = 8), idiopathic thrombosis (n = 4), or post-surgery (n = 1). Two patients were heterozygous for factor V Leiden; 1 patient had antithrombin III deficiency. Twenty-nine patients underwent surgical reconstruction with 31 bypass grafts: spiral saphenous vein (n = 20), superficial femoral vein (n = 4), human allograft (n = 1), or expanded polytetrafluoroethylene (ePTFE, n = 6). Eleven patients underwent percutaneous transluminal angioplasty or stenting; 3 primary and 8 secondary endovascular procedures were performed to treat graft stenosis (n = 7) or occlusion (n = 1). There were no early deaths. Five early graft failures in 3 ePTFE grafts and 2 bifurcated vein grafts (thrombosis, n = 4; stenosis, n = 1) were successfully treated with open surgical revision. Over a mean follow-up of 5.6 years (range, 0.4-16.6 years) in surgical patients, 17 additional secondary interventions were performed in 8 patients, 14 endovascular and 3 surgical. Primary, assisted primary, and secondary patency rates of surgical bypass grafts were 63%, 79%, and 85%, respectively, at 1 year, and 53%, 68%, and 80%, respectively, at 5 years. Graft patency was significantly higher in vein grafts compared with ePTFE grafts (P =.02). Mean follow-up after percutaneous transluminal angioplasty or stenting was 3.1 years (range, 1 day-11.7 years). Twelve secondary endovascular interventions were performed in 6 patients (primary group, 3 of 3; secondary group, 3 of 9 grafts in 8 patients) to maintain patency in 11 of 12 reconstructions. Mean follow-up in the entire patient cohort was 5.3 years (range, 0.4-16.6 years). In 79% of patients symptoms had resolved or were significantly improved at last follow-up. CONCLUSIONS Surgical treatment of benign SVC syndrome is effective over the long term, with secondary endovascular interventions to maintain graft patency. Straight spiral saphenous vein graft remains the conduit of choice for surgical reconstruction, with results superior to those with bifurcated vein and ePTFE. Endovascular treatment is effective over the short term, with frequent need for repeat interventions. It does not adversely affect future open surgical reconstruction and may prove to be a reasonable primary intervention in selected patients. Patients who are not suitable for or who fail endovascular intervention merit open surgical reconstruction.

Most patients with abdominal aortic aneurysm are not suitable for endovascular repair using currently approved bifurcated stent-grafts.

Strict morphologic criteria must be used for patient selection to achieve durable success with endovascular aortic aneurysm repair (EVAR). The goal of this study was to assess morphologic suitability (MS) of abdominal aortic aneurysms (AAAs) for 2 currently approved bifurcated stent grafts and identify reasons for exclusion from EVAR. The authors reviewed the electronic charts of 1,795 consecutive patients who were diagnosed as having AAA between January 1999 and July 2001 at their institution. Three hundred and twenty patients had an AAA with a diameter of =5.0 cm, measured on computed tomography (CT). The records of 301 patients, 254 men, 47 women, with a mean age of 74 years were available for review, and these patients constituted the study cohort. Criteria used for MS included a proximal neck length =15 mm; neck diameter between 18 and 26 mm; neck angulation =60°; common or external iliac artery (CIA or EIA) diameters of 7–16 mm and 8–13 mm, respectively, for AneuRx (Medtronic Ave, Santa Rosa, CA) and Ancure (Guidant Cardiac and Vascular Division, Menlo Park, CA) bifurcated grafts. AAAs were suitable for AneuRx device in 14% of patients (43 of 301; 95% CI = 11–19%) and for Ancure in 5% (16 of 301; 95% CI = 3.1–9%). The main reason for exclusion was an inadequate proximal aortic neck (73%). The neck was too short in 49.5%, too wide in 64% and badly angulated in 12% of the patients. Iliac artery morphology precluded EVAR with AneuRx and Ancure devices in 52% and 80%. Both CIAs were too wide for EVAR in 43% and 77%, respectively. When iliac artery diameter =20 mm was accepted, iliac suitability for AneuRx increased from 49% to 70% and overall suitability increased from 14% to 20%. When more permissive criteria were used for MS (neck length =10 mm, neck diameter =30 mm, CIA =20), 39% of patients became candidates for EVAR. More than three fourths of the patients with an AAA =5.0 cm in size, seen in a tertiary referral center, are morphologically not suitable for EVAR using 2 currently approved bifurcated endografts. The main reasons for exclusion are a short or wide proximal aortic neck. Considerable changes in size of the devices and in proximal attachment techniques have to occur before most AAAs will be suitable for EVAR.

Outcome of complex venous reconstructions in patients with trauma.

PURPOSE The role of complex venous reconstructions (CVRs) in patients with major trauma remains a controversial topic. This study evaluates the patency and clinical outcome of CVRs in a major urban trauma center. METHODS Between 1979 and 1994 the records of 92 patients with 100 injuries to the iliac, femoral, and popliteal venous system were reviewed. The incidence of edema, pulmonary embolism, and limb loss was documented in 75 men and 17 women (mean age of 27 years, range 14 to 59 years). The 30-day patencies were assessed in all patients with either impedance plethysmography (n = 16), venography (n = 40), or duplex scan (n = 36). Long-term patencies were assessed in 14 patients monitored for 0.5 to 9 years (mean 3.2 years). RESULTS Mechanisms of injury consisted of 58 gunshot wounds, 23 stab wounds, 6 shotgun wounds, and 5 blunt injuries. There were 112 associated injuries, 41 of which were concomitant arterial injuries. Forty-five of the 100 venous injuries were repaired with CVRs and included 6 (13%) spiral vein grafts, 8 (18%) panel vein grafts, 8 (18%) reversed saphenous vein interposition grafts, 8 (18%) end-to-end repairs, and 15 (33%) vein patch repairs. Thirty-day patency rates for these repairs were 50%, 50%, 75%, 88%, and 87%, respectively, and an overall patency rate of 73% was observed. The remaining 55 injuries were treated with ligation (n = 27) or lateral venorrhaphy (n = 28). The cumulative 30-day patency rate for all venous repairs was 81% (59 of 73). Fourteen patients, nine of whom had CVRs, were available for long-term follow-up. In this group CVRs demonstrated a 100% patency. One patient with a spiral vein graft repair of the common femoral vein had severe reflux causing intermittent edema and mild lipodermatosclerosis. No pulmonary emboli, limb loss, or deaths were identified in patients undergoing CVRs. CONCLUSION Patients with CVRs had a 30-day patency rate of 73%. Of this group panel and spiral vein grafts were less successful, exhibiting only a 50% 30-day patency rate, whereas end-to-end and vein patch repairs were successful in 88% and 87% of cases, respectively. Our overall evaluation suggests that use of CVRs results in successful venous repair; however, the postoperative patency of interposition panel and spiral grafts suggests selective use of these techniques.

Endovascular repair of abdominal aortic aneurysms: initial experience with 100 consecutive patients.

OBJECTIVE To review early results of endovascular repair of abdominal aortic aneurysms (AAAs). PATIENTS AND METHODS The first 100 patients who underwent endovascular repair of AAA (EVAR) between June 26,1996, and October 31, 2001, at the Mayo Clinic in Rochester, Minn, were studied retrospectively to evaluate technical success, freedom from reinterventions, and early clinical outcome. RESULTS A total of 89 men and 11 women (mean +/- SD age, 76 +/- 7 years; range, 47-92 years) underwent EVAR. The procedure was successful in 97 patients. There was no early death. Major complications occurred in 25 patients. The 30-day technical success rate was 86% (95% confidence interval [CI], 77%-92%). The median intensive care unit stay was 1 day (range, 1-15 days), and the median hospital stay was 3 days (range, 1-35 days). Median follow-up was 7 months (range, 1-60 months). Endoleak (incomplete seal of the endovascular graft) at discharge was observed in 14 patients; 13 developed endoleak during follow-up. There were 23 reinterventions, 65% of which were percutaneous procedures. One aneurysm ruptured at 5 months, but the patient was successfully treated by open repair. Primary and secondary graft patency rates at 1 year were 83% (95% CI, 74%-93%) and 94% (95% CI, 87 %-99%), respectively. The freedom from reintervention rate at 1 year was 71% (95% CI, 59%-84%), with an overall success rate from EVAR of 92% (95% CI, 84%-100%). There were no differences in early patency, reinterventions, and success rates between unibody and modular devices. CONCLUSION EVAR can be performed with high technical success and low mortality rates; however, nonfatal complications and catheter-based reinterventions are frequent, and EVAR may not prevent aneurysm rupture. Although stent graft repair for high-risk patients is appealing, current data are insufficient to support EVAR as the preferred treatment of AAAs.

In Vivo Gene Transfer of Endothelial Nitric Oxide Synthase to Carotid Arteries From Hypercholesterolemic Rabbits Enhances Endothelium-Dependent Relaxations

Background and Purpose —Hypercholesterolemia is associated with abnormal endothelium-dependent vasorelaxation due to decreased nitric oxide bioavailability. Our aim was to examine the effect of adenovirus-mediated gene transfer of endothelial nitric oxide synthase (eNOS) to the hypercholesterolemic rabbit carotid artery in vivo. In addition, we examined whether adenovirus-mediated gene transfer was associated with vascular dysfunction. Methods —Rabbits were fed a 1% cholesterol diet for 4 weeks followed by a 0.5% cholesterol diet for 6 weeks. Vascular reactivity was assessed in nontransduced carotid arteries from chow- and cholesterol-fed animals. In addition, carotid arteries were surgically isolated, and 2 separate doses of adenoviral vectors encoding eNOS or β-galactosidase (AdβGal) on the contralateral side were delivered to the lumen (1×1010 and 5×1010 pfu/mL). Results —Abnormal acetylcholine-mediated endothelium-dependent vasorelaxation was detected in the carotid artery from cholesterol-fed animals, whereas responses to calcium ionophore A23187 and diethylamine NONOate were normal. Vascular reactivity was similar in nontransduced and AdβGal-transduced hypercholesterolemic vessels. In vessels transduced with eNOS, transgene expression was demonstrated by immunostaining in both the endothelium and the adventitia and by Western blot analysis. High-dose but not low-dose eNOS gene transfer enhanced endothelium-dependent relaxation in vessels from cholesterol-fed rabbits. Conclusions —Adenovirus-mediated gene transfer of eNOS to carotid arteries of cholesterol-fed animals improves endothelium-dependent relaxation when an optimal viral titer is administered.

Evaluation of a Xenogeneic Acellular Collagen Matrix as a Small-Diameter Vascular Graft in Dogs—Preliminary Observations

Autogenous veins are the materials of choice for arterial reconstruction. In the absence of autogenous material, prosthetic materials are used. However, vascular prostheses of less than 0.4 cm in diameter have low long-term patency. This study was designed to determine if cells would infiltrate an engineered xenogeneic biomaterial used as a smalldiameter arterial graft in dogs and, if so, to determine the phenotype of the infiltrating cells. Nine acellular xenogeneic grafts (0.4 cm in diameter, 5 cm long), composed of porcine collagen derived from the submucosa of the small intestine and type I bovine collagen, were implanted as end-to-end interposition grafts in femoral arteries of five male mongrel dogs (total of nine grafts). All dogs received daily aspirin (325 mg). Patency of implanted grafts was monitored weekly by Duplex ultrasonography. After 9 weeks, or earlier in case of blood flow reduction by at least 75%, grafts were explanted and prepared for light or electron microscopy to evaluate cellularization. Eight of nine grafts remained patent up to 9 weeks. At explant, diameters were 0.31 - 0.02 cm at the midgraft, and 0.14 - 0.01 and 0.19 - 0.01 cm at the proximal and distal anastomoses. At explant, cells of mesenchymal origin (endothelial cells, smooth muscle cells, myofibroblasts) were embedded in the extracellular matrix of the graft scaffold. Minimal evidence of cellular inflammatory reaction and no aneurysmal dilatation or thrombus formation was detected. Variable degrees of hyperplasia were present at proximal and distal anastomoses. This preliminary study demonstrates that a collagen-based xenogeneic biomaterial provides a scaffold for cellularization when used for arterial reconstruction in dogs.

Intraoperative duplex ultrasound of visceral revascularizations: optimizing technical success and outcome

PURPOSE The purpose of this study was to evaluate the use of intraoperative duplex ultrasound scanning (IOUS) during visceral revascularizations and correlate its results with clinical outcome. METHODS We studied 68 patients (15 men and 53 women, mean age 66.5 years, range 27-86 years) who underwent visceral revascularization with concomitant IOUS examination of 120 visceral arteries (52 celiac, 60 superior mesenteric, and 8 inferior mesenteric arteries) from 1992 to 2002. Patients were divided into two groups on the basis of ultrasound findings: normal and abnormal IOUS. The incidence of early and late graft-related complications (thrombosis, restenosis, recurrent symptoms, reintervention) and graft-related death was compared in both groups. RESULTS One-hundred and two (85%) arteries had normal IOUS. Eight (6.6%) arteries had minor defects, including small kinks (4), mild residual stenoses (3), and small intimal flap (1). Ten (8.4%) arteries had major defects, consisting of hemodynamically significant residual stenoses (4), thrombus (2), kinks (2), bidirectional flow (1), and intimal flap (1). Major defects were successfully revised in all except three cases: two persistent mild stenoses and one bidirectional flow. Patients with abnormal IOUS at the end of the operation had increased incidence of graft-related complications and/or death (55.5% vs 7.8%; P =.004), early graft thrombosis (14.2% vs 1.0; P =.04), reintervention (21.4% vs 3.2%; P =.03), and graft-related death (33.3% vs 1.9%; P =.02), compared with patients with normal IOUS. CONCLUSION This study supports the routine use of IOUS during visceral revascularizations to optimize technical success and outcome. Persistent ultrasound scanning abnormalities are associated with risk of early graft failure, reintervention, and death. Patients with normal ultrasound scans can expect excellent results.

Hand-assisted laparoscopic aorto-aortic bypass: Initial experience in a porcine model

Videoscopic surgical techniques have been developed to reduce morbidity of open aortic reconstructions. The advantage of hand-assisted laparoscopic surgery (HALS) technique is the introduction of the surgeons hand into the peritoneal cavity. The aims of this study were to assess the feasibility and to examine the learning curve, limitations, and pitfalls of the HALS technique to perform aortic reconstruction in a porcine model for training purposes. HALS aorto-aortic 8 mm polytetrafluoroethylene (PTFE) interposition grafts were placed in 12 pigs. Proficiency was judged by measuring operative time points, satisfactory completion of the operation, and the need to convert to open procedure. The strength of the relationship between order number in which a procedure was performed and the various surgical time point measures was described with the Spearman rank correlation. HALS aortic grafting was successful in the last 8 pigs. The first 2 pigs required conversion to open repair, and the graft of the third and fourth animals occluded early. Median operative time was 115 minutes (range: 75 to 205), median intestinal retraction time was 28 minutes (range: 10 to 40), median aortoiliac dissection time was 30 minutes (range: 20 to 60), and median aortic cross-clamp time was 48 minutes (range: 35 to 82). The Spearman rank correlations and p values between the order of the procedure and the intestinal retraction time, aortoiliac dissection time, clamping time, and total operative time were –0.62 (0.06), –0.47 (0.17), –0.69 (0.03), and –0.83 (0.03), respectively. HALS facilitates intestinal retraction and completion of laparoscopic aortoiliac dissection. It offers adequate exposure in pigs for aortic grafting and allows open sutured aortic anastomosis. The learning curve for HALS aortic surgery in a porcine model is short and within reach of surgeons with standard laparoscopic surgery skills, since no laparoscopic suturing is required. Training on this porcine model may be an efficient and safe way to introduce surgeons to HALS for aortoiliac reconstruction.

Optimal Management of Abdominal Aortic Aneurysms and Urologic Malignancies: Benefits of Simultaneous Surgical Treatment

The coexistence of urologic malignancy (UM) and abdominal aortic aneurysm (AAA) is rare. Simultaneous treatment may increase morbidity, whereas staged operations delay necessary treatment. We reviewed our experience to develop guidelines for evaluation and management. Clinical data of all patients diagnosed between 1980 and 1998 with both AAA and UM, who also had staged or simultaneous surgical treatment were reviewed. Four thousand forty-seven patients underwent AAA repair during the study period at our institution. Of these, 18 (0.44%) patients, 16 men and two women (mean age: 74 years, range: 61 to 92) had UM. UM was discovered incidentally in nine patients with AAA with computed tomography (CT) scan (7) and ultrasonography (2). In one patient with AAA, hematuria raised the suspicion of UM. Eight patients were diagnosed with AAA during evaluation of UM by CT scan or ultrasound (7) or during laparotomy (1). Initial signs and symptoms included abdominal pain (33%), gross hematuria (33%), and urinary tract infections (17%), although 33% were asymptomatic. The AAA was symptomatic at first examination in three (17%) patients, 14 (78%) patients had hypertension, and two (11%) had renal insufficiency (overall mean serum creatinine: 1.6, range: 0.8 to 3.7). Mean AAA diameter was 5.8 cm (range: 3.8 to 8 cm). Aortography documented significant contralateral renal artery disease in two patients. Twelve patients underwent simultaneous AAA repair and resection of UM; the operations were staged in six (AAA repair first in 2, nephrectomy first in 4). One patient required emergent repair of a ruptured AAA three days after nephrectomy. UM was treated with nephrectomy in 17 patients, and with bilateral ureterectomy and cystectomy in one. A straight aortic graft was implanted in seven patients, a bifurcated graft in ten, and one had extra-anatomic reconstruction. Two patients required reconstruction of the contralateral renal artery. The 30-day mortality rate was 0% after staged and 6% after simultaneous repair (p= >0.1). Major perioperative complications occurred more frequently after simultaneous (42%) than after staged repair (33%), although the difference was not significant (p= >0.1). Two patients developed postoperative renal failure. Seventeen patients were followed for an average of 3.9 years (14 days to 10.8 years). Three (17%) patients died of recurrent cancer at 1, 10.7, and 10.8 years after surgery. Patients with resectable UM have satisfactory long-term survival, which justifies aggressive treatment of concomitant AAA. Aortography in these patients is suggested to exclude contralateral renal artery disease. Complications were frequent, but simultaneous repair did not increase morbidity or mortality significantly. The risk of AAA rupture after nephrectomy, need to correct contralateral renal artery disease at the time of nephrectomy, and disadvantage of delaying treatment of UM are compelling reasons to favor simultaneous treatment of AAA and UM.

Experience with Greenfield Filters in Pregnant Women for Deep Venous Thrombosis and Pulmonary Embolism Case Reports

The indications for inferior vena caval filter placement in the gravid female are ill defined. During pregnancy, however, pulmonary embolism (PE) secondary to venous thrombosis is the most common cause of maternal mortality. Pregnant women are at risk for deep venous thrombosis (DVT) due to hypercoagulability caused by increased levels of coagulation factors and decreased fibrinolytic activity. In addition, decreased venous tone and velocity of blood flow in the lower extremities lead to venous stasis. Although heparin is the treatment of choice for DVT associated with pregnancy, propagation of thrombus or development of bleeding diathesis mandates discontinuation of anticoagulant therapy and consideration for caval interruption. In this review, two patients are presented who required vena caval filters during pregnancy, and indications for their usage in this patient population are defined. Filter placement is recommended during pregnancy in the presence of extensive iliofemoral thrombus, free-floating thrombus, bleeding complications, or pulmonary embolism despite adequate anticoagulation.