Matthew J. Eagleton

A comparison of endovascular revascularization with traditional therapy for the treatment of acute mesenteric ischemia

OBJECTIVES Few centers have adopted endovascular therapy for the treatment of acute mesenteric ischemia (AMI). We sought to evaluate the effect of endovascular therapy on outcomes for the treatment of AMI. METHODS A single-center, retrospective cohort review was performed on all consecutive patients with thrombotic or embolic AMI presenting between 1999 and 2008. Patients with mesenteric venous thrombosis, nonocclusive mesenteric ischemia, and ischemia associated with aortic dissection were excluded. Demographic factors, preoperative metabolic status, and etiology were compared. Primary clinical outcomes included endovascular technical success, operative complications, and in-hospital mortality. RESULTS Seventy consecutive patients were identified with AMI (mean age, 64 ± 13 years). Etiology of mesenteric ischemia was 65% thrombotic and 35% embolic occlusions. Endovascular revascularization was the preferred treatment (81%) vs operative therapy (19%). Successful endovascular treatment was achieved in 87%. Endovascular therapy required laparotomy in 69% vs traditional therapy in 100% (P < .05), with a median 52-cm necrotic bowel resected (interquartile range [IQR], 11-140 cm) vs 160 cm (IQR, 90-250 cm; P < .05), respectively. Acute renal failure and pulmonary failure occurred less frequently with endovascular therapy (27% vs 50%; P < .05 and 27% vs 64%; P < .05). Successful endovascular treatment resulted in a mortality rate of 36% compared with 50% (P < .05) with traditional therapy, whereas the mortality rate for endovascular failures was 50%. Endovascular therapy was associated with improved mortality in thrombotic AMI (odds ratio, 0.10; 95% confidence interval, 0.10-0.76; P < .05). CONCLUSIONS Endovascular therapy has altered the management of AMI, and there are measurable advantages to this approach. Using endovascular therapy as the primary modality for AMI reduces complications and improves outcomes.

Durability of branches in branched and fenestrated endografts

OBJECTIVE Branched and fenestrated repair has been shown to be effective for treatment of complex aortic aneurysms. However, the long-term durability of branches is not well reported. METHODS Prospective data collected for all patients enrolled in a physician-sponsored investigational device exemption trial for branched and fenestrated endografts were analyzed. Retrospective review of imaging studies and electronic records was used to supplement the dataset. Incidences of branch stent secondary intervention, stent fracture, migration, branch-related rupture, and death were calculated. A time-to-event analysis was performed for secondary intervention for any branch. Univariable and multivariable analyses were performed to identify related variables. Branch instability, a composite outcome of any branch event, was reported as a function of exponential decay to capture the loss of freedom from complications over time. RESULTS Between the years 2001 and 2010, 650 patients underwent endovascular aortic repair with branched or fenestrated devices. Over 9 years of follow-up (mean [standard deviation], 3 [2.3] years), secondary procedures were performed for 0.6% of celiac, 4% of superior mesenteric artery (SMA), 6% of right renal artery, and 5% of left renal artery stents. Mean time to reintervention was 237 (354) days. The 30-day, 1-year, and 5-year freedom from branch intervention was 98% (95% confidence interval [CI], 96%-99%), 94% (95% CI, 92%-96%), and 84% (95% CI, 78%-90%), respectively. Death from branch stent complications occurred in three patients, two related to SMA thrombosis and one due to an unstented SMA scallop. Multivariable analysis revealed no factors as independent predictors of need for branch reintervention. CONCLUSIONS Branches, after branched or fenestrated aortic repair, appear to be durable and are rarely the cause of patient death. The absence of long-term data on branch patency in open repair precludes comparison, yet the lower morbidity and mortality risk coupled with longer-term durability data will further alter the balance of repair options.

Late conversion of aortic stent grafts

OBJECTIVES The frequency of late removal of endovascular abdominal aortic repair (EVAR) parallels the rise of endovascular aortic repair. Evaluation of outcomes for EVAR explants may identify risks for complications and alter clinical management. METHODS A patient database was used to identify EVAR patients requiring explant >1 month after implant. A retrospective analysis was conducted of the type of graft, duration of implant, reason for removal, operative technique, death, and length of stay. RESULTS During 1999 through 2007, 1606 EVARs were performed, and 25 patients required explantation, with an additional 16 referred from other institutions (N = 41). The average age was 73 years (range, 50-87 years); 90% were men. Grafts were excised after a median of 33.3 months (range, 3-93 months). Explanted grafts included 16 AneuRx (40%), 7 Ancure (17%), 6 Excluder (15%), 4 Zenith (10%), 4 Talent (10%), 1 Cook Aortomonoiliac rupture graft, 1 Endologix, 1 Quantum LP, and 1 homemade tube graft. Overall hospital mortality was 19% and occurred after conversion for rupture in 4, and in infected graft, aortoenteric fistula, repair of new aneurysm of the visceral segment, and claudication due to graft stenosis in one patient each. Elective EVAR-related mortality was 3.3%. Mortality was higher in patients with rupture compared with nonrupture (4 of 6 vs 3 of 35, P <or= .01). Thirty patients (73%) had one or more endoleaks (type I, 16; II, 9; III, 9; endotension, 5). Migration (n = 10), rupture (n = 6), aortoenteric fistula (n = 3), infection (n = 1), limb thrombosis (n = 3), and claudication (n =1) were also factors. Proximal aortic control was above the endograft (supravisceral, 23; suprarenal, 12; infrarenal, 6). Reconstruction was an aortoiliac repair in 63% and tube graft in 25%. Grafts with suprarenal fixation required longer proximal aortic clamp time of 43 minutes vs 28 minutes for infrarenal fixation. Complete graft removal was achieved in 85%. Proximal or distal portions of the endograft were incorporated into the repair in the remaining six. CONCLUSION Elective EVAR conversion, although technically challenging may be done with mortality similar to primary open repair. Mortality for conversion for infected grafts and ruptured aneurysms remains high. EVAR is associated with continued risk of conversion, and surveillance may identify late complications that require removal, justifying lifelong monitoring. Aggressive management of late complications and elective conversion may minimize the mortality associated with this procedure.

Endovascular repair of complicated chronic distal aortic dissections: Intermediate outcomes and complications

INTRODUCTION Patients with chronic distal aortic dissection (CDAD) remain at high risk for late aorta-related events and reinterventions, and the ideal management strategy remains undefined. Open surgical procedures carry morbidity, but scant data for thoracic endovascular aortic repair (TEVAR) of CDAD exist. This study reports our intermediate-term results with TEVAR for complicated CDAD. METHODS All cases of TEVAR for complicated (aortic growth, malperfusion, intractable pain) CDAD at our institution between 2000 and 2007 were retrospectively reviewed. Demographic information, indications for repair, complications, and aortic morphologic changes were collected from medical records and imaging studies. Aortic morphology (aneurysm size, false lumen thrombosis) was assessed at multiple levels with 3-dimensional image analysis techniques. Kaplan-Meier analysis was used to estimate survival, freedom from reintervention, and likelihood of false lumen thrombosis, with log-rank tests used to discriminate between Kaplan-Meier curves. RESULTS In total, 144 stent-grafts were implanted in 76 consecutive patients (49 male) with complicated CDAD. Early (<30 postoperative days) mortality was 5%. There was no paraplegia, and 1 patient died of stroke. At mean follow-up of 34 months, 12 patients had died (1 aorta-related death). Seventeen patients (22%) underwent 19 secondary aortic reinterventions, mainly for enlargement of the untreated aorta remote to stent-graft repair. Three secondary procedures treated retrograde proximal dissections. Estimated survivals were 86%, 82%, and 80% at 12, 24, and 36 months, respectively, and freedoms from both death and reintervention were 72%, 64%, and 59% at similar time points. Of 67 patients (88%) with complete imaging follow-up, TEVAR resulted in significantly decreased aortic diameter through the stent-grafted segment but not untreated segments. Complete thrombosis of the entire false lumen was uncommon in patients with extensive dissections (13% vs 78% P < .001). CONCLUSIONS Management of complicated CDAD remains challenging for clinicians. TEVAR is a reasonable treatment modality for dissections limited to the thoracic aorta and for prevention of focal aortic growth in extensive dissections. Late complications and the need for secondary interventions emphasize the complexity of this patient population and the need for long-term follow-up.

Twelve-year results of fenestrated endografts for juxtarenal and group IV thoracoabdominal aneurysms

OBJECTIVE The practice of using fenestrated endografts to treat juxtarenal and group IV thoracoabdominal aortic aneurysms (TAAAs) has become more accepted, but long-term outcomes are still unknown. We report long-term survival, complications, and branch-related outcomes from a single-center experience. METHODS The study included consecutive patients enrolled prospectively into a physician-sponsored investigational device exemption classified as undergoing group IV TAAA or juxtarenal aneurysm repair by the treating surgeon using fenestrated endografts. Device morphology was used to subclassify this group of patients. Long-term survival and a composite outcome of secondary intervention, branch occlusion, stent migration, endoleak, aneurysm growth, or spinal cord injury were calculated. Descriptive analysis of branch-related outcomes and need for any reintervention was performed. Univariate and multivariate analysis of mortality and the composite outcome was performed to determine associative risks. RESULTS Long-term survival for patients with juxtarenal and group IV TAAA aneurysms treated with fenestrated stent grafts was 20% at 8 years. Multivariate analysis showed long-term survival for this patient population was negatively associated with increasing age, congestive heart failure, cancer, and previous aneurysm repair. The risk of spinal cord ischemia (SCI) in this group was 1.2% and of aortic-related mortality was 2%. The risk of a spinal event increased with coverage above the celiac artery (52 mm of coverage above the celiac artery in patients with SCI vs 33 mm without SCI; P = .099). More complex device configurations were more likely to require an increased rate of reinterventions, and patients with celiac fenestrations were more likely to experience celiac occlusion over time (3.5% vs 0.5%; P = .019). However, less complex designs were complicated by an increased risk of type I endoleak over time (10.4% for renal fenestrations only vs 1.9% for others; P < .01). As experience evolved, there was a trend to increase the number of fenestrations in devices treating the same anatomy. CONCLUSIONS The use of fenestrated devices to treat juxtarenal and group IV TAAA is safe and effective in long-term follow-up. Mortality in this patient population is largely not aortic-related. Devices designed for fenestrated repair of juxtarenal and group IV thoracoabdominal aneurysms within a physician sponsored investigational device exemption have changed over time. Further research is needed to determine the best configuration to treat aneurysms requiring coverage proximal to the celiac artery.

Hypogastric and subclavian artery patency affects onset and recovery of spinal cord ischemia associated with aortic endografting.

OBJECTIVE Spinal cord ischemia (SCI) is a devastating complication associated with aortic aneurysm repair. The aim of the current study was to evaluate factors affecting outcomes from SCI associated with endovascular aortic aneurysm repair. METHODS A total of 1251 patients underwent endovascular repair of aortic aneurysm as part of a device trial between 1998 and 2010 utilizing endovascular abdominal aortic aneurysm repair (n = 351), thoracic endovascular aortic aneurysm repair (n = 201), fenestrated endovascular aortic aneurysm repair (n = 227), and visceral branched endografts (n = 472). Records and imaging studies were reviewed to supplement prospective outcome data. Demographics, type of repair, collateral bed (hypogastric/subclavian) patency, clinical presentation, and outcomes were evaluated on patients with SCI. Survival was calculated using life-table analysis. RESULTS SCI occurred in 2.8% (n = 36) of patients: abdominal aortic aneurysm, 0.3%, juxtarenal, 0.4%, thoracic aortic aneurysm, 4.6%, and thoracoabdominal aortic aneurysm, 4.8%). Four (11%) required carotid-subclavian bypass prior to endografting, and two underwent coverage of the left subclavian artery. Unilateral hypogastric artery occlusion was present in 11 (31%) patients prior to endograft placement, and three had bilateral occlusions. An additional seven patients had occlusion of at least one hypogastric artery during surgery. SCI was apparent immediately in 15 (42%) patients. Immediate onset of symptoms was observed in 73% of patients with at least one occluded collateral bed but in only 24% of those with patent collateral beds (P = .021). Of those presenting in a delayed fashion, nine (43%) had a clear precipitating event prior to onset of SCI (hypotension, n = 6, and segmental artery drain removal, n = 3). Recovery occurred in 24 (67%) patients, most within 7 days. Immediate presentation was a negative predictor of recovery (P = .025), as was occlusion of at least one collateral bed (P = .035). Mean follow-up was 22 ± 4 months with 30-day and 1-year survival of 92 ± 4.6% and 56 ± 8.3%. Survival was only 36% at 3 months in those with permanent SCI compared with 92% (P < .001) in those with temporary symptoms. CONCLUSIONS SCI continues to complicate aortic surgery despite the advent of endovascular therapy. Occlusion of a single collateral bed is associated with an increased risk for immediate onset of SCI and lack of recovery. These factors are harbingers of poor outcomes and increased short-term mortality. This may be prevented by preserving collateral bed patency in patients undergoing extensive endovascular procedures.

Zenith p-branch standard fenestrated endovascular graft for juxtarenal abdominal aortic aneurysms

OBJECTIVE This article reports the early clinical outcomes and experiences of Zenith pivot branch device (p-branch) standard fenestrated endovascular graft (Cook, Bloomington, Ind) for treating juxtarenal abdominal aortic aneurysm (AAA) originating below the superior mesenteric artery (SMA). METHODS A physician-sponsored investigational device exemption study was used to assess enrolled elective and emergency patients from August 2011 to September 2012 for treatment with an off-the-shelf Zenith p-branch device. Patients were included provided an anatomic seal could be established 4 mm below the SMA and the renal geometry fit the protocol based on reconstructed computed tomography data. The celiac artery was addressed with a scallop and the SMA with an 8-mm fenestration. The renal fenestrations were constructed as a modified design to allow a range of renal locations (7.5 mm radially from the center of the fenestration) to be acceptably treated with a single configuration. Two anatomic configurations were created for renal arteries (origins at the same level, or disparate renal arteries with left lower than right). Outcomes are reported in coherence with endovascular reporting standards documents. RESULTS The study enrolled 16 patients (94% men; median age, 75 years [range, 59-87 years]) with a mean aneurysm size of 61 mm (range, 52-82 mm). Two were treated for aneurysm rupture. Technical success was achieved in all patients. The median fluoroscopy time was 62 minutes (range, 38-105 minutes), and the amount of contrast media was 69 mL (range, 31-121 mL). There were no aortic-related deaths, aneurysm ruptures, open surgical conversion, or type I/III endoleaks. One right renal artery occluded during follow-up in the setting of a conically shaped visceral aortic segment and was successfully treated with endovascular recanalization. CONCLUSIONS The use of the p-branch device for aneurysms originating infra-SMA is associated with a high rate of technical success and minimal problems during the short follow-up duration. The off-the-shelf design allows for the treatment of ruptures and other urgent situations. Additional patients and more follow-up will be required to further define the risks and benefits of such a treatment strategy.

Fenestrated and branched endovascular aortic repair for chronic type B aortic dissection with thoracoabdominal aneurysms

OBJECTIVE The treatment of patients with arch and thoracoabdominal aortic aneurysms (TAAAs) and chronic dissections is challenging. We report the results of fenestrated and branched endovascular aortic repair (FEVAR) of such aneurysms. METHODS A single-center prospective FEVAR trial enrolled 356 patients (2006 to 2011), of whom 30 had chronic dissections with arch aneurysm or TAAAs, or both. Patients were divided into group A, 15 patients (mean age, 58 years) with extensive dissections extending from the arch through the visceral segment, and group B, 15 patients (mean age, 74 years old) with focal dissections and no extension into the thoracic aorta. Inclusion criterion was aneurysm size >5.5 cm in diameter. Customized grafts were implanted into the true lumen, and branches were extended into the true lumen of the supra-aortic trunk (arch branch devices) and visceral vessels. Patients were monitored annually with clinical, imaging, and laboratory studies. Outcome analyses included survival, rupture, spinal cord ischemia, endoleak, morbidity (cardiac, renal or pulmonary), reinterventions, dissection, and aneurysm growth. RESULTS The mean time from the onset of dissection to the FEVAR performed in group A was 10.4 years. The mean maximum aneurysm diameter was 60 mm. Follow-up averaged 1.7 years. There were no perioperative deaths. One aortic-related death occurred at 87 days due to progression of a pre-existing untreated arch dissection. No ruptures, cardiac, renal, pulmonary, or spinal cord ischemia complications occurred. Despite the initially narrow true lumen dimensions, stent grafts expanded to their nominal diameters after implantation without any blood flow disturbance of branched visceral vessels and distal aorta. No graft compression occurred. Post-FEVAR growth was noted in two patients, related to type II endoleaks. Sac regression was similar (-6.8 vs -11.4 mm; P = .43), but early endovascular reinterventions were more common in group A (8 patients). Patients with extensive dissection were younger, and the dissection more likely to be associated with a defined connective tissue disease (Marfan syndrome or Loeys-Dietz mutations, 40% vs 0%; P = .006). CONCLUSIONS FEVAR is feasible for patients with chronic dissections and TAAA. Concerns regarding visceral vessel access and graft compression resulting from narrow true lumen diameters were not relevant in our experience. Favorable sac and lumen morphologic changes, coupled with a low mortality and complication risk, makes this an attractive means of handling this clinical problem.

Staged endovascular repair of thoracoabdominal aortic aneurysms limits incidence and severity of spinal cord ischemia

OBJECTIVE Neurologic dysfunction remains a persistent complication of extensive aortic repair owing to disruption of the spinal collateral network. We hypothesized that staged repair might mitigate the incidence and severity of this spinal cord ischemia (SCI). METHODS We conducted a retrospective cohort study of patients undergoing a Crawford type II repair of a thoracoabdominal aortic aneurysm between January 2008 and July 2013. Baseline demographics, incidence of prior aortic surgery, comorbidities, and outcomes were prospectively recorded. Staged repair was defined as intentional completion of the endovascular repair as two temporally separate procedures, referred to as a two-stage repair. Extent of aortic cover was calculated by three-dimensional imaging and reported as the proportion of the aorta covered between the left subclavian artery and the aortic bifurcation. Primary outcome measures were incidence and severity of SCI and mortality. RESULTS The study included 87 patients, divided into the following subgroups: single-stage repair (n = 32; repair in a single procedure, without prior aortic surgery), two-stage repair (n = 27; repair in two separate procedures, without prior aortic surgery), and unintentionally staged repair (n = 28; those with prior aortic surgery, without an intention to stage). Median time between stages was 5 months (range, 1-60 months). All groups were equivalent in terms of demographics and risk factors; however, the staged group had significantly greater proximal aortic cover (P = .001). The overall rates of SCI in the nonstaged and staged groups were 37.5% (12 of 32) and 11.1% (3 of 27), respectively (P = .03). Furthermore, all neurologic injuries in the staged group were temporary. The 30-day survival in the single-stage, two-stage, and unintentionally staged repairs was 18.8%, 0%, and 10.7%, respectively (P = .52). CONCLUSIONS Staged repair appears both to protect against SCI and to enhance overall survival in extensive aortic repair.

Angiography Underestimates Peripheral Atherosclerosis: Lumenography Revisited

Purpose: To compare angiograms, considered the gold standard for diagnostic imaging of peripheral arterial disease (PAD), to the corresponding histological sections of popliteal and tibial vessels obtained after amputation to determine if angiography fails to define atheroma burden in “normal appearing” arteries in patients with PAD. Methods: Between 2004 and 2006, 69 patients underwent amputation of a lower extremity for severe tissue loss, gangrene, or pedal sepsis precluding limb salvage. Popliteal and tibial vessels were harvested, perfusion-fixed, and analyzed histologically. Thirty-four of these patients had pre-amputation angiography during attempted salvage procedures. Angiograms with patent or minimally diseased vessel segments (n=19) were assessed for stenoses, diameter, and calcification by 3 vascular surgeons (n=72 evaluations). These results were compared to corresponding cross-sectional histological slides (n=66) in a blinded manner. Results: Angiograms performed prior to above-knee (n=9) or below-knee (n=10) amputation revealed 24 stenoses with a mean (±SD) diameter-reducing stenosis of 19.5%±15.2%. Corresponding histological cross sections revealed greater linear stenoses measured via boundaries of the internal elastic lamina (IEL stenosis, 28.9%±20.2%, p=0.003 versus angiography) or via boundaries of the external elastic membrane (vessel stenosis, 43.1%±15.2%, p<0.0001). Stenosis calculated by area methods (IEL area) were greater and measured 39.2%±24.2% (p<0.0001) and 60.9%±15.2% (vessel area, p<0.0001). Popliteal arteries had greater discrepancy in stenosis measurement than tibial arteries (18.5%±14.6% versus 34.9%±21.0%, p=0.0005). However, evaluations of tibial arteries for concentricity of plaque (44% versus 69%, p=0.08) and calcification grade (1.6 versus 2.2, p=0.002) by angiography were discordant with histological analyses. Measurement of arterial diameter by histology for popliteal arteries (6.2±0.9 mm) and tibial arteries (3.1±0.7 mm) was greater than angiographic diameter determination (p<0.001). Conclusion: Angiography provides information on luminal characteristics of peripheral arteries but severely underestimates the extent of atherosclerosis in patients with PAD even in “normal appearing” vessels.