Mourad Boufi

Endovascular Management of Chronic Disabling Ilio-caval Obstructive Lesions: Long-Term Results

OBJECTIVE To report the long-term results of stenting for chronic ilio-caval obstructive lesions. MATERIAL AND METHODS From January 1996 to January 2008, 89 patients (72 women, 17 men; median age 43 years) were admitted for endovascular treatment of chronic disabling non-malignant obstructive ilio-caval lesions. Patients were classified as C2 in 15 cases, C3 in 59, C4 in seven, C5 in two and C6 in six. Median preoperative venous disability score (VDS) and venous clinical severity score (VCSS) were 2 and 9, respectively. Aetiology was primary in 52 patients, secondary in 35 and congenital in two. Lesions were bilateral in seven cases, eight patients had inferior vena cava (IVC) involvement and 18 had common femoral vein (CFV) obstructive lesions. Complete occlusion was found in 30 cases. RESULTS Technical success was achieved in 98%. The median hospital stay was 2 days. During a median follow-up of 38 months (range: 1-144 months), one patient died and five cases of thromboses occurred. Iterative stenting was performed for restenosis in six cases. Primary, assisted-primary and secondary patency rates, in terms of intention to treat, were 83%, 89% and 93%, respectively, at 3 and 10 years, with a median VDS of 1. Univariate analysis found that significant factors affecting patency were CFV involvement for primary patency and history of deep venous thrombosis (DVT) and CFV involvement for secondary patency. The last 46 patients had statistically more severe lesions than the first 43 (higher VDS, more secondary lesions, more occlusions, more stented segments, higher length of stented vein), and in spite of which patency rates are not different. CONCLUSION Endovenous angioplasty, combined with stenting, is a sure, safe, effective and very minimally invasive technique which provides good long-term patency rates. Currently, it is recognised as the technique of choice for the treatment of ilio-caval obstructive lesions. Surgery should be proposed only in case of failure.

Late results of surgical venous thrombectomy with iliocaval stenting.

PURPOSE Iliac vein occlusive disease leads to 73% of rethrombosis that occurs after venous thrombectomy when left untreated. The goal of this study is to present our long-term results of stenting of iliocaval occlusive lesions persisting after surgical venous thrombectomy. METHODS From November 1995 to April 2007, 29 patients (19 women), with a median age of 38 years, had surgical venous thrombectomy with creation of an arteriovenous fistula and angioplasty and stenting. All were admitted for acute (<10 days) deep venous thrombosis (DVT) involving the iliocaval segment, of which eight had concomitant acute pulmonary embolism. Six patients had a history of DVT (2 with previous venous thrombectomy), two were pregnant, and three had postpartum DVT. No patients had short- or mid-term life-threatening factors. The underlying lesion was left iliocaval compression (May-Thurner syndrome) in 22 patients, chronic left common iliac vein occlusion in 3, residual clot in 3, and compression of the left external iliac vein by the left internal iliac artery in 1. RESULTS Neither perioperative death nor pulmonary embolism occurred. Four early complications occurred after stenting (13.8%). Median hospital length of stay was 8 days (range, 5-22 days). Median follow-up was 63 months (range, 2-137 months). Three late complications occurred (10.3 %): one rethrombosis due to stent crushing during pregnancy and two restenosis, which were treated by iterative stenting. At the end of the follow-up, the median venous clinical severity score was 3 (range 1-12) and the venous disability score was 1 (range 0-2). Primary, assisted primary and secondary patency rates were, respectively, 79%, 86%, and 86% at 12, 60, and 120 months. Patients with patent iliocaval segments had significantly fewer infrainguinal obstructive lesions (4% vs 50%) and a higher rate of valvular competence (76% vs 0%) than those who experienced rethrombosis. Venous scores were also worse in patients with rethrombosis. CONCLUSION Stenting is a safe, efficient, and durable technique to treat occlusive iliocaval disease after venous thrombectomy. Its use can prevent most of the rethrombosis that occurs after venous thrombectomy without major adverse effects.

Elective Stent-graft Treatment for the Management of Thoracic Aorta Mural Thrombus

BACKGROUND Optimal management of aorta mural thrombus (AMT) continues to be controversial. The aim of this study was to describe the management of AMT in the thoracic aorta with either conservative or stent-graft treatment and to analyze the role of morphological characteristics of thrombus in the selection of suitable candidates for intervention. METHODS A retrospective review was conducted of all patients admitted for thoracic AMT. Clinical data, treatment used, and outcomes were recorded. Patients were divided in two groups according to the treatment used: either conservative or stent-graft. Morphological features of thrombus, including size, sessile or pedunculated aspect and mobility, were compared between the two groups. RESULTS From January 2006 to March 2013, 13 consecutive patients (nine male, mean age 53, range 37-76) were admitted for symptomatic (n = 8) or asymptomatic AMT (n = 5). All patients received unfractionated heparin. Management of primary aortic thrombus required stent-graft in seven patients, aortic thrombectomy in one, and anticoagulation therapy alone in five. Indications for intervention were recurrent embolism (n = 4), occurrence of embolism under heparin (n = 1), or persistent thrombus (n = 2). Endovascular exclusion of AMT was successful in all cases, with no complications or deaths at 30 days and no recurrence at midterm. Analysis of the morphological features of the thrombus identified solely the high degree mobility as associated with adverse outcome (p = .048). CONCLUSION In our experience, stent-graft exclusion of AMT is an effective approach. Systematic evaluation of thrombus mobility by a real-time imaging study can be helpful to better define the indications for radical treatment of the aortic lesion.

Examination of factors in type I endoleak development after thoracic endovascular repair

OBJECTIVE The objective of this study was to assess the effects of operative indication, anatomy, and stent graft on type I endoleak occurrence after thoracic endovascular aortic repair. METHODS A retrospective review was conducted of patients admitted for thoracic endovascular aortic repair between 2007 and 2013. All computed tomography angiography imaging was analyzed for the presence of endoleak and measurement of diameters and lengths. Variables studied included underlying disease, emergency, achieved aortic neck length, difference between proximal and distal neck diameters, landing zone 2, and stent graft characteristics (diameter, number, type of device, oversizing degree, and covered aorta length). RESULTS The study population involved 84 patients (mean age, 56 years; range, 17-94 years) who were treated for thoracic aortic aneurysm (TAA) (n = 29; 34.5%), traumatic aortic rupture (n = 27; 32%), type B aortic dissection (n = 19; 22.5%), intramural hematoma (n = 2; 2%), penetrating aortic ulcer (n = 5; 6%), and aortoesophageal fistula (n = 2; 2%). Of these, 60 patients (71.5%) were treated emergently and 24 (28.5%) electively. Primary type I endoleak was noted in eight patients (9.5%), of which two resolved spontaneously. After a mean follow-up of 32 months (range, 3-76 months), secondary type I endoleak was detected in four patients (4.5%). All of them occurred after emergent TAA treatment. Comparison between emergent and elective groups revealed no significant differences in neck length (19.5 mm vs 26.5 mm; P = .197), oversizing degree (11.1% vs 10.9%; P = .811), or endoleak rates (13.3% vs 8.3%; P = .518). Hemorrhagic shock was not predictive of endoleak (P = .483). Cox regression analysis of the different anatomic and stent graft-related factors revealed short proximal landing zone as the unique independent predictor of type I endoleak (hazard ratio, 0.89; 95% confidence interval, 0.81-0.99; P = .032). CONCLUSIONS Endoleak risk seems not to be increased by an emergency setting. However, the relatively high rate of late endoleak observed after emergent TAA repair advocates for close follow-up, contrary to traumatic aortic rupture. Furthermore, regardless of the pathologic process, a longer proximal landing zone is likely to guarantee early and late success.

Laparoscopic transposition of the left renal vein into the inferior vena cava for nutcracker syndrome

Reimplantation of the left renal vein into the infrarenal inferior vena cava is the standard surgical procedure for nutcracker syndrome. A 40-year-old woman with a solitary left kidney suffered from left lumbar pain and hematuria. Imaging techniques found a large kidney with nutcracker syndrome. A totally laparoscopic transposition of the left renal vein was performed. Twelve months later, the patient is improved and has no more hematuria. Duplex scan showed no residual stenosis. Laparoscopic transposition of the left renal vein into the inferior vena cava is feasible with short length of stay and good short-term result.

A comparison of the standard bolia technique versus subintimal recanalization plus Viabahn stent graft in the management of femoro-popliteal occlusions

OBJECTIVE To assess the potential benefit of the addition of a covered stent to a subintimal recanalized artery in patients with femoro-popliteal occlusions. METHODS From September 2003 to October 2005, we retrospectively analyzed all patients admitted for severe claudication or critical limb ischemia related to long femoro-popliteal occlusions and treated with subintimal recanalization. Patients were divided into two groups depending on whether they received a stent or not. All patients in the group treated with stent received a stent graft, and the entire length of the recanalized artery was covered in each case. Demographic data, indications, procedure, and outcomes were examined using survival analysis statistical techniques. RESULTS Fifty-three patients (54 limbs) were treated consecutively for severe claudication (n=19) or critical limb ischemia (n=34). Thirty-four (64%) had a stent placed, while 19 (35.8%) did not. The mean length of the lesions treated was 20.11 cm (range, 5-35 cm). Statistically, there was no significant difference in lesion length, Rutherford stage of peripheral-artery disease, Transatlantic Inter-Society Consensus classification, and distal run-off between the two groups. The technical success rate was 94.5%, and two out of the three failures were treated with surgical bypass in one case and major amputation in the other. The third patient received only medical treatment. Combined procedures were required in the treatment of 68.2% of limbs in the no-stent group and 55.8% in the stent group. Mean follow up was 16.9 months (range, 1-35 months). At 1 year, primary, primary-assisted, and secondary patency for the stent vs no-stent groups was, respectively, 61.8% vs 78.9% (P=.49), 70.6% vs 78.9% (P=.78), and 88.2% vs 78.9% (P=.22). The 1-year limb salvage rate for the stent vs no-stent group was 94.1% vs 100% (P=.7). CONCLUSION Combining subintimal angioplasty with a stent graft in femoro-popliteal lesions does not improve patency. The limb salvage rate remains high after addition of a stent graft. Rigorous monitoring is recommended to diagnose and treat restenosis early in order to improve patency.

Emergency stent graft implantation for ruptured visceral artery pseudoaneurysm

BACKGROUND Literature series that include visceral artery pseudoaneurysms rarely separate them from true aneurysms, although they address different issues. Guidelines for optimal management of these lesions are lacking. We report our experience of stent graft treatment of these lesions with midterm results. METHODS We retrospectively reviewed all patients with a visceral pseudoaneurysm who were treated with a stent graft in our institution. Patient history, clinical characteristics, procedure details, and outcome were recorded and analyzed. RESULTS From March 2004 to June 2009, 10 consecutive patients (9 men), who were a mean age of 59 years, were treated for symptomatic visceral artery pseudoaneurysm, with hemorrhagic shock in 8 patients (80%), after pancreaticoduodenectomy in 8, gastrectomy in 1, and abdominal trauma in 1. A mean of 24 days (range, 7-60 days) passed between the initial surgery or trauma and pseudoaneurysm diagnosis. Septic complications were associated in six patients (60%). The pseudoaneurysm was in the hepatic artery in 8 patients, the splenic artery in 1, and the superior mesenteric artery in 1. Technical and clinical success was achieved in 80% of patients. Two failures of catheterization were followed by redo surgery and death (20%). No patients died postoperatively, and no complications among the patients who were treated successfully. Mean follow-up was 37 months (range, 10-63 months). All stent grafts were patent, with no signs of infection. Two patients died secondary to neoplasm. No rebleeding or recurrent aneurysms were noted. CONCLUSION Stent graft exclusion of visceral artery pseudoaneurysm seems to be a valid therapeutic approach regardless of the patients septic or hemodynamic status.

Management of pregnancy in women with previous left ilio-caval stenting

BACKGROUND Ilio-caval stenting now represents the first line treatment for disabling obstructive ilio-caval lesions. Most patients are young women of child-bearing age. We herein report our experience of pregnancy in women who have a history of ilio-caval stenting. MATERIALS AND METHODS From November 1995 to April 2008, 119 patients had ilio-caval stenting for obstructive venous disease in our department. Of these, 62 women were able to become pregnant. When pregnancy occurred, they received preventive treatment with low molecular weight heparin (LMWH) from the 3rd month of pregnancy to 1 month after delivery and had to wear elastic stockings. Patients also had to sleep on their right side if possible. They were followed during the pregnancy by duplex scanning at 3, 6, and 8 months, and then 1 month after delivery. RESULTS Eight pregnancies occurred in 6 patients (mean age 26.5 years) who had a patent self-expanding stent (1 patient had 3 pregnancies). They had stenting for May-Thurner disease in 3 patients, for post-deep venous thrombosis (DVT) left common iliac vein occlusion in 1 patient, and during venous thrombectomy in 2 patients. All stents were self-expanding metallic stents located on the left common iliac vein. One patient had unrelated spontaneous abortion after 2 months of pregnancy. No DVT or symptomatic pulmonary embolism occurred during pregnancy, delivery, or during the postpartum period. Four patients needed cesarean delivery and none had hemorrhagic complications. None of the patients had adverse effects from the treatment. Duplex scan showed compression of the stent(s) at 8 months in 4 patients with inflow obstruction in 3 patients. Postpartum duplex-scan showed no remaining stenosis in all patients. No stents had structural damage. CONCLUSION Ilio-caval stent compression can occur during pregnancy but does not lead to structural damage to the self-expanding stents. Despite this, no cases of DVT occurred with preventive LMWH treatment.

Risk Factor Analysis of Bird Beak Occurrence after Thoracic Endovascular Aortic Repair

OBJECTIVES The aim was to analyze the role played by anatomy and stent graft in the incidence of incomplete apposition to aortic arch. METHODS Between 2007 and 2014 data including available and suitable computed tomographic angiography (CTA) imaging of patients who had undergone thoracic endovascular aortic repair were reviewed. The study included 80 patients (65 men, 54 ± 21 years) treated for traumatic aortic rupture (n = 27), thoracic aortic aneurysm (n = 15), type B aortic dissection (n = 24), penetrating aortic ulcer (n = 5), intramural hematoma (n = 2), aorto-oesophageal fistula (n = 2), and aortic mural thrombus (n = 5). Pre- and post-operative CTA images were analyzed to characterize bird beak in terms of length and angle, and to calculate aortic angulation within a 30 mm range at the proximal deployment zone. RESULTS Bird beak configuration was detected in 46 patients (57%): mean stent protrusion length was 16 mm (range: 8-29 mm) and mean bird beak angle was 20° (range: 7-40°). The bird beak effect was significantly more frequent after traumatic aortic rupture treatment (p = .05) and in landing zone 2 (p = .01). No influence of either stent graft type or generation, or degree of oversizing was observed (p = .29, p = .28, p = .81 respectively). However, the mean aortic angle of patients with bird beak was higher in the Pro-form group than that in the Zenith TX2 group (62° vs. 48°, p = .13). Multivariate analysis identified the aortic angle of the deployment zone as the unique independent risk factor of malapposition (HR = 1.05, 95% CI 1-1.10, p = .005). The cutoff value of 51° was found to be predictive of bird beak occurrence with a sensitivity of 58% and a specificity of 85%. CONCLUSIONS Assessment of proximal landing zone morphology to avoid deployment zones generating an aortic angle of over 50° can be recommended to improve aortic curvature apposition with the current available devices.

Experimental study of a novel mechanical connector for sutureless open arterial anastomosis

OBJECTIVES This study assessed the feasibility and efficacy of a new sutureless connector for end-to-end arterial anastomosis. METHODS The anastomotic device is a connector consisting of a bare-metal stent with spikes covering its outer surface, which is introduced through the prosthesis. The seal of the anastomosis is obtained by inflating a balloon anchoring the stent with the spikes in the prosthesis and in the receiving artery. This experiment was conducted in three phases: (1) A feasibility study was done on four cadaveric femoral arteries using a polytetrafluoroethylene prosthesis, with measurement of the penetration of the spikes into the layers of the arterial wall. (2) Bench tests were conducted in seven automated and in seven sutured anastomoses. Anastomosis sealing was tested using a pump system (≤250 mm Hg) in a water-filled closed circuit. (3) The infrarenal aorta of seven pigs was replaced with a 6- to 8-mm diameter prosthesis, using this automated device for the proximal anastomosis. The distal anastomosis was handsewn on the aortic bifurcation. After completion angiography, animals were euthanized for macroscopic and histologic studies of the aorta, connector, and prosthesis. Explantations were done immediately (2 pigs), and at 15 (2 pigs), at 30 (2 pigs), and 42 days (1 pig). Study end points were the automated anastomosis time compared with manual suturing, leakage, mechanical strength, and patency of the anastomosis together with the histologic changes of the aortic wall in contact with the spiked stent. RESULTS Tests performed on four cadaveric arteries showed complete penetration of the spikes into the arterial wall layers without metal fracture. Tests of traction showed that the median force needed to rupture the automated anastomosis was 18.3 N (interquartile range [IQR], 17.7-19.9 N), with no significant difference from the handmade anastomosis (19.5 N; IQR, 17.9-20.2 N, P = .33). No leakage was demonstrated in vitro with a pulsatile flow and a pressure up to 250 mm Hg. The median automated anastomosis time on pig aorta was 2.4 minutes (IQR, 1.4-3.3 min) vs 17.0 minutes (IQR, 15.1-17.2 minutes) for the handmade aortic anastomosis performed on the same animals (P = .002). There was no anastomotic leak. Histology showed incorporation of the connector spikes in the aortic wall, without intimal hyperplasia or false aneurysm. CONCLUSIONS This preliminary study confirms the feasibility of this sutureless system, suggesting its usefulness for a faster and simpler anastomosis in hybrid arterial surgery.