Grace J. Wang

Risk factors, outcomes, and clinical manifestations of spinal cord ischemia following thoracic endovascular aortic repair

OBJECTIVE The purpose of this study was to assess the incidence, risk factors, and clinical manifestations of spinal cord ischemia (SCI) after thoracic endovascular aortic repair (TEVAR). METHODS A retrospective review of a prospectively collected database was performed for all patients undergoing TEVAR at a single academic institution between July 2002 and June 2010. Preoperative demographics, procedure-related variables, and clinical details related to SCI were examined. Logistic regression analysis was performed to identify risk factors for the development of SCI. RESULTS Of the 424 patients who underwent TEVAR during the study period, 12 patients (2.8%) developed SCI. Mean age of this cohort with SCI was 69.6 years (range, 44-84 years), and 7 were women. One-half of these patients had prior open or endovascular aortic repair. Indication for surgery was either degenerative aneurysm (n = 8) or dissection (n = 4). Six TEVARs were performed electively, with the remaining done either urgently or emergently due to contained rupture (n = 2), dissection with malperfusion (n = 2), or severe back pain (n = 2). All 12 patients underwent extent C endovascular coverage. Multivariate regression analysis demonstrated chronic renal insufficiency to be independently associated with SCI (odds ratio [OR], 4.39; 95% confidence interval [CI], 1.2-16.6; P = .029). Onset of SCI occurred at a median of 10.6 hours (range, 0-229 hours) postprocedure and was delayed in 83% (n = 10) of patients. Clinical manifestations of SCI included lower extremity paraparesis in 9 patients and paraplegia in 3 patients. At SCI onset, average mean arterial pressure (MAP) and lumbar cerebrospinal fluid (CSF) pressure was 77 mm Hg and 10 mm Hg, respectively. Therapeutic interventions increased blood pressure to a significantly higher average MAP of 99 mm Hg (P = .001) and decreased lumbar CSF pressure to a mean of 7 mm Hg (P = .30) at the time of neurologic recovery. Thirty-day mortality was 8% (1 of 12 patients). The single patient who expired, never recovered any lower extremity neurologic function. All patients surviving to discharge experienced either complete (n = 9) or incomplete (n = 2) neurologic recovery. At mean follow-up of 49 months, 7 of 9 patients currently alive continued to exhibit complete, sustained neurologic recovery. CONCLUSION Spinal cord ischemia after TEVAR is an uncommon, but important complication. Preoperative renal insufficiency was identified as a risk factor for the development of SCI. Early detection and treatment of SCI with blood pressure augmentation alone or in combination with CSF drainage was effective in most patients, with the majority achieving complete, long-term neurologic recovery.

Presentation, complications, and natural history of penetrating atherosclerotic ulcer disease

OBJECTIVES Increased utilization of computed tomography angiography (CTA) has increased the radiologic diagnosis of penetrating atherosclerotic ulcers (PAUs), which are defined as the ulceration of atherosclerotic plaque through the internal elastic lamina into the aortic media. However, the presentation, treatment indications, and natural history of this disease process remain unclear. METHODS The radiology database at a single university hospital was searched retrospectively for the CTA diagnosis of PAU from January 2003 to June 2009. All scans were interpreted by a cardiovascular radiologist. Information on PAU characteristics and need for surgical repair due to PAU disease was collected. PAU stability or progression was assessed by follow-up CTA, if available. Only PAUs in the aortic arch, descending thoracic aorta, and abdominal aorta were included. RESULTS Three hundred eighty-eight PAUs were diagnosed by CTA interpretation. PAU location was in the aortic arch in 27 (6.8%) cases, the descending thoracic aorta in 243 (61.2%) cases, and the abdominal aorta in 118 (29.7%) cases. Two hundred twenty-four (57.7%) PAUs were isolated (without saccular aneurysm or intramural hematoma); 108 (27.8%) PAUs had associated saccular aneurysms; and 56 (14.4%) PAUs had associated intramural hematoma. Rupture was present in 16 (4.1%) cases. Fifty (12.9%) PAUs underwent repair with thoracic endovascular aortic repair (TEVAR) (n = 30), endovascular aneurysm repair (EVAR) (n = 10), or open surgery (n = 10); primary indications for repair were saccular aneurysm (n = 26), rupture (n = 16), and persistent or recurrent symptoms (n = 8). Even if initially treated conservatively with resolution of pain, symptomatic PAU disease was more likely to require repair than asymptomatic PAU disease (36.2% vs 7.8%, P < .001). Follow-up CTA was available for 87 PAUs, 20 (23.0%) of which demonstrated radiographic disease progression at a mean follow-up of 8.4 ± 10.3 months. Symptomatic PAU disease was more likely to progress than asymptomatic disease (42.9% vs 16.7%, P = .029). CONCLUSIONS For PAUs diagnosed on CTA at a single institution, 4.1% were ruptured and 12.9% underwent repair. Close follow-up imaging appears to be indicated for PAUs, particularly in the case of symptomatic disease, which is more likely to require repair and to undergo radiographic progression.

Limb Ischemia During Femoral Cannulation for Cardiopulmonary Support

OBJECTIVES Extracorporeal membrane oxygenation and extracorporeal cardiopulmonary support (ECMO/CPS) are potentially life-saving techniques for patients with cardiopulmonary collapse. Complications include lower extremity ischemia from femoral artery cannulation. We examined the outcomes of patients placed on ECMO/CPS, including the rate of limb ischemia. METHODS All instances of ECMO/CPS over a 3-year period (2006-2009) at a single university hospital were examined retrospectively for cannulation strategy, perfusion strategy, mortality, and limb ischemia. Potential predictors of limb ischemia with femoral artery cannulation were age, gender, body surface area (BSA), body mass index (BMI), and arterial cannula size. RESULTS Fifty-eight patients were placed on ECMO/CPS. Of these, 43 patients (74%) had femoral arterial cannulation. In 10 patients, the superficial femoral artery (SFA) was cannulated prophylactically (without antecedent limb ischemia) and perfused in the antegrade direction from a branch of the ECMO/CPS circuit. In 7 of the remaining 33 patients (21%), limb ischemia developed requiring decannulation with fasciotomy (n = 4) or additional cannulation of the SFA with branching of the ECMO/CPS circuit (n = 3). One patient with ipsilateral leg ischemia required eventual amputation. Patients with limb ischemia were significantly younger than those who did not develop limb ischemia (P = .001). BSA, BMI, and cannula size did not predict limb ischemia. Overall 30-day mortality following the initiation of ECMO/CPS was 79%. There was no correlation between limb ischemia and mortality. CONCLUSIONS Younger patients may be at increased risk for lower extremity arterial insufficiency with femoral cannulation for ECMO/CPS. Prophylactic or expectant SFA cannulation are reasonable approaches.

Low rehospitalization rate for vascular surgery patients

OBJECTIVES Reducing rehospitalization rates has been proposed to improve care, reduce costs, and as a pay-for-performance criterion. Recent review of Medicare claims data indicates that vascular surgery patients have among the highest rates of 30-day rehospitalization at 23.9%. METHODS We retrospectively examined all live patient discharges (n = 799) from the vascular surgery service at a single university hospital over 12 months. Planned and unplanned 30-day rehospitalizations were distinguished, and predictors of unplanned 30-day rehospitalization were determined. To identify whether patients were readmitted to other hospitals, a prospective study of patient discharges (n = 66) over 1 month was also performed. RESULTS Ninety-five (11.9%) of the 799 patient discharges from the vascular surgery service were rehospitalized within 30 days. Of these, 71 were unplanned; therefore, the unplanned rehospitalization rate was 8.9%. The most common causes of unplanned 30-day rehospitalization were related to wound complications. Diabetes (P = .039) predicted unplanned 30-day rehospitalization by multivariate analysis. Patients with the diagnosis of critical limb ischemia (14.9%) and patients undergoing open lower extremity revascularization (14.6%) had the highest rates of unplanned 30-day rehospitalization. In the prospective portion of this study, no patient was readmitted to any other hospital. CONCLUSIONS Relatively low 30-day rehospitalization was accomplished in vascular surgery patients at a single university hospital. Moreover, planned rehospitalizations accounted for approximately 25% of readmissions in vascular surgery patients. Strategies designed to reduce rehospitalization in diabetics may be warranted.

The Powerlink system for endovascular abdominal aortic aneurysm repair: six-year results.

OBJECTIVE We compared the results of endovascular repair using the Powerlink endovascular graft with conventional open abdominal aortic aneurysm repair through a 6-year follow-up period. METHODS Two hundred fifty-eight patients with abdominal aortic aneurysms were prospectively enrolled in a multicenter trial and underwent endovascular repair (N = 192) or conventional open surgery (N = 66). All endovascular repairs were approached through a surgically exposed femoral artery and a percutaneously accessed femoral artery. Study endpoints included all-cause mortality and morbidity. Follow-up imaging consisted of contrast-enhanced CT scans and plain abdominal x-rays at 1, 6, 12 months, and annually postoperatively. RESULTS Technical success was achieved in 97.9% of test patients, with four failed insertions (three early conversions because of deployment issues, one access failure). Mean follow-up was 4.1 +/- 1.7 years (test group) and 3.1 +/- 1.9 years (control group). Perioperative morbidity and mortality were significantly reduced in the test group compared with the control group (P < .05). At 6 years, all-cause mortality and morbidity was no different in the Powerlink group compared with the open repair group. There were no reported stent fractures, graft disruptions, or aneurysm ruptures. Core laboratory-reported endoleaks included proximal or distal type I (n = 1) and type I/II (n = 3), with no type III or type IV endoleaks. One explant (0.5%) was undertaken to resolve a refractory type I endoleak. A total of 37 secondary procedures were performed in 26 patients to treat site-reported endoleak (n = 26; 7 for type I and 19 for type II), graft limb occlusion (n = 7), native artery occlusion (n = 3), or endograft migration (n = 1). A reduction in mean aneurysm sac diameters and volumes has been noted at every follow-up interval. CONCLUSION Consistent with other reports, perioperative morbidity and mortality were significantly reduced in the endovascular group compared with the open repair group. Six-year follow-up of patients treated with the Powerlink system demonstrates the continued safety and efficacy of its treatment of abdominal aortic aneurysm.

Late open conversion and explantation of abdominal aortic stent grafts

OBJECTIVES To evaluate indications for, operative strategy during, and outcomes following late open surgical conversion following endovascular aneurysm repair (EVAR). METHODS Between 2002 and 2009, patients undergoing open abdominal aortic aneurysm repair at a university hospital were entered prospectively into a database which was examined to identify patients undergoing open conversion >30 days after EVAR. RESULTS Over 7 years, 21 patients required late open conversion of EVAR. The average patient age was 75 years (range, 59-88), and there were 16 male (76%) patients. The mean interval to conversion was 33.4 months (range, 2-73). Eight patients (38%) presented with proximal type I endoleak; 4 patients (19%) presented with type II endoleak and aneurysm expansion; 5 patients (24%) presented with graft migration and aneurysm expansion; and 5 patients (24%) presented with de novo visceral aneurysms. Rupture (1) and infection (1) were also observed. There were five (24%) emergent cases. Most patients (12/21, 57%) had more than one reason for conversion. There were no perioperative deaths; three patients (14%) had major complications. Grafts requiring conversion were AneuRx (6; Medtronic AVE, Santa Rosa, Calif), Zenith (6; Cook Inc, Bloomington, Ind), Talent (3; Medtronic), Excluder (2; W. L. Gore, Flagstaff, Ariz), Anaconda (1; TERUMO Corp, Ann Arbor, Mich), Ancure (1; Guidant, Menlo Park, Calif), Quantum LP (1; Cordis Corp, Miami Lakes, Fla), and Powerlink (1; Endologix, Irvine, Calif). The surgical approach was retroperitoneal in 16 (76%) and transperitoneal in four (19%) patients. Initial proximal aortic control was supraceliac (9/21), suprarenal (7/21), or infrarenal (5/21), with stepwise distal clamping to reduce ischemic time. Complete endograft removal was performed in 17/21 patients; in 4/21 the distal anastomosis was performed to the endograft after proximal segment explantation. Reconstruction was completed with tube (19/21) or aortoiliac (2/21) grafts; in one case, homograft was used. Mean intraoperative blood loss was 1.9 L (range, 0.4-6.5 L), mean intensive care unit (ICU) stay was 3 days (range, 2-6), and the mean hospital stay was 10 days (range, 4-39). CONCLUSIONS While technically challenging, delayed open conversion of EVAR can be accomplished with low morbidity and mortality in both the elective and emergent settings. These results reinforce the justification for long-term surveillance of endografts following EVAR.

The management of endograft infections following endovascular thoracic and abdominal aneurysm repair

OBJECTIVE The management of infected aortic endografts is a challenging endeavor. Treatment of this problem has not been well defined as it is fairly uncommon. However, the incidence is increasing. This study examines the results of treatment at a single center for this morbid process. METHODS A retrospective review was performed of patients treated for infected abdominal or thoracic endograft infection following previous abdominal or thoracic endovascular aneurysm repair. Data was reviewed for patient demographics, details of initial endograft implantation, presentation and timeline of subsequent infection, management of infected grafts, and outcomes during follow-up. RESULTS Overall, 18 patients were treated for infected endografts (thoracic: six, abdominal:12). Three patients were treated between 2000 and 2006, corresponding to a 0.6% institutional incidence of endograft infection (3/473). There were no transfers for infected endografts from outside institutions. From 2006 to 2011, 15 patients underwent treatment. Six were institutional cases of infections (6/945, 0.6% infection rate), however, there was an increase in transfers (n = 9). Median time to presentation with infection from endograft implant was 90 days, with over one-half (61%) presenting within the first 3 months. Tissue and/or blood cultures were positive in 12/16 growing Escherichia coli (n = 1), group A streptococcus (n = 3), methicillin-resistant Staphylococcus aureus (n = 3), or polymicrobial infections (n = 7). The other four patients were culture negative with computed tomography evidence of gas surrounding the endograft and clinical sepsis. Ten patients (abdominal: eight, thoracic: two) were treated with endograft explantation. The remaining eight patients were considered too high-risk for explant or refused open surgery and were therefore managed conservatively without explant (abdominal: four, thoracic: four). At a mean follow-up of 24.7 months, aneurysm-related mortality was 38.9% (n = 7) and was higher for patients presenting with aortoenteric or aortobronchial fistulas (n = 6/10, 60%) (P = .04) and for thoracic stent infections (n = 5/6; 83%) (P = .03). The only survivor of a thoracic infection was managed surgically. Overall survival for patients with abdominal endografts (n = 12) was similar between the eight patients managed surgically (n = 6/8; 75%) and the four selected for medical management (n = 4/4; 100%) (P = .39). All survivors remain on long-term suppressive antibiotics. Two additional patients died of unrelated causes during follow-up. CONCLUSIONS Endograft infection is a rare but increasing complication after abdominal or thoracic endovascular aneurysm repair, which carries significant associated morbidity and mortality. Most endograft infections occurred in proximity to other types of infection, suggesting that bacterial seeding of the endograft was the source. Aortoenteric and aortobronchial fistulas are common presentations, which portend a significantly worse prognosis. Thoracic endograft infections, which have the highest rate of fistulization, have the worst outcomes. Surgical excision continues to be standard of care but conservative management with intravenous antibiotics may be of benefit in certain patients with abdominal endograft infections.

Vascular distribution of stroke and its relationship to perioperative mortality and neurologic outcome after thoracic endovascular aortic repair

OBJECTIVE This study assessed the vascular distribution of stroke after thoracic endovascular aortic repair (TEVAR) and its relationship to perioperative death and neurologic outcome. METHODS A retrospective review was performed for patients undergoing TEVAR between 2001 and 2010. Aortic arch hybrid and abdominal debranching cases were excluded. Demographics, operative variables, and neurologic complications were examined. Stroke was defined as any new focal or global neurologic deficit lasting>24 hours with radiographic confirmation of acute intracranial pathology. RESULTS Perioperative stroke occurred in 20 of 530 patients (3.8%) undergoing TEVAR. The cohort was 55% male and a mean age of 75.2±8.9 years (range, 57-90 years). Among patients with perioperative strokes, the indication for surgery was degenerative aneurysm in 14 (mean diameter, 6.8 cm), acute type B dissection in four, penetrating atherosclerotic aneurysm in one, and aortic transection in one. Cases were performed urgently or as an emergency in 60%. The proximal landing zone was zone 2 in 11 or zone 3 in nine. All strokes were embolic. The vascular distribution of stroke involved the anterior cerebral (AC) circulation in eight (zone 2, n=5) and the posterior cerebral (PC) circulation in 12 (zone 2, n=6). Laterality of cerebral infarction included five right-sided, eight left-sided, and seven bilateral strokes. Nine strokes were diagnosed<24 hours after operation. There was no difference in baseline demographics, aortic pathology, acuity, zone coverage, preoperative left subclavian artery revascularization, number of stents, or estimated blood loss between stroke groups based on vascular distribution. Independent risk factors for any perioperative stroke were chronic renal insufficiency (odds ratios [OR], 4.65; 95% confidence interval [CI], 1.22-17.7; P=.02) and history of prior stroke (OR, 4.92; 95% CI, 1.69-14.4; P=.004); the risk factor for AC stroke was prior stroke (OR, 7.67; 95% CI, 1.25-46.9; P=.03) and the risk factors for PC stroke were age (OR, 1.11; 95% CI, 1.00-1.23; P=.04), prior stroke (OR, 7.53; 95% CI, 1.78-31.8; P=.006), zone 2 coverage (OR, 6.11; 95% CI, 1.15-32.3; P=.03), and penetrating atherosclerotic ulcer (OR, 32.7; 95% CI, 1.33-807.2; P=.03). Overall in-hospital mortality was 20% (n=4), with those sustaining PC strokes observed to trend toward increased mortality (33% vs 0%; P=.12). Patients with AC strokes were more likely than those with PC strokes to achieve complete recovery of neurologic deficits before discharge (75% vs 17%; P=.02). CONCLUSIONS Perioperative stroke after TEVAR is primarily an embolic event. Although infrequent, stroke was associated with significant morbidity and death, particularly among those with strokes involving the PC circulation.

Prophylactic muscle flaps in vascular surgery

BACKGROUND Vascular surgery-related groin complications can lead to catastrophic outcomes and pose a significant healthcare burden. We have taken steps to reduce potential complications at the time of initial surgery by performing prophylactic muscle flaps. The purpose of this study is to evaluate the efficacy and benefit of prophylactic flaps in high-risk patients. METHODS A retrospective cohort study was performed on patients undergoing open vascular surgery involving the femoral vessels through a groin incision between 2005 and 2010. Patients receiving prophylactic muscle flaps at their initial surgery were compared with those patients not receiving a flap (control). RESULTS Sixty-eight prophylactic flaps in 53 patients were compared with 195 open vascular procedures without flaps in 178 patients. The most frequent indication was reoperative bypass surgery with prosthetic reconstruction (63%). The prophylactic patient group exhibited significantly higher rates of comorbidities, including chronic obstructive pulmonary disease (25.0% vs 12.6%; P = .018) and hyperlipidemia (80.9% vs 59.1%; P = .002). Patients receiving prophylactic flaps had lower rates of overall complications (16.2% vs 50.3%; P < .001), infections (1.5% vs 38.5%; P < .001), seroma (0% vs 7.2%; P = .023), and lymphocele (1.5% vs 15.4%; P = .002). Multivariate regression demonstrated that obesity (odds ratio [OR], 2.1 [1.001-4.49]; P = .05), smoking (OR, 2.7 [1.37-5.16]; P = .004), reoperation (OR, 3.5 [1.41-8.63]; P = .007), and prosthetic graft reconstruction (OR, 2.0 [1.03-3.78]; P = .04) were associated with postoperative complications. Additionally, in analyzing all groin complications in all patients, we found that patients who received a prophylactic flap experienced significantly less groin wound complications (OR, 0.17; P < .001). CONCLUSIONS Complications following open groin surgery are common, lead to significant morbidity, and are very costly. Performing prophylactic muscle flaps at the initial surgery to cover the femoral vessels and reduce dead space can significantly reduce complications in select high-risk patients. Prophylactic flaps are safe, effective, and should be considered in patients with multiple comorbidities undergoing high-risk groin surgery, such as reoperative prosthetic bypass surgery.

Predictors of decreased short- and long-term survival following open abdominal aortic aneurysm repair

OBJECTIVES The purpose of this study was to identify predictors of decreased survival after open abdominal aortic aneurysm (AAA) repair at a single university hospital. METHODS Patients undergoing open AAA repair from June 2003 to June 2009 were identified. Primary outcomes were 30-day and 5-year survival. Preoperative, intraoperative, and postoperative variables were assessed for their influence on outcomes using univariate and multivariate analysis, as appropriate. One- and 5-year survival were determined by Kaplan-Meier analysis. RESULTS Four hundred eight patients (289 men; 70.8%) with a mean age of 72.4 ± 8.3 years underwent open AAA repair. Sixty-seven patients (16.4%) underwent nonelective repair. The clamp site was infrarenal in 137 patients (33.6%), suprarenal in 97 patients (23.8%), and supraceliac in 174 patients (42.6%). Thirty-day survival was 95.6%. One- and 5-year survival were 90.0% ± 1.5% and 65.1% ± 3.0%, respectively. Seventy-nine patients (19.4%) had decreased renal function postoperatively compared to preoperatively, 71 patients (17.4%) sustained cardiac complications, and 45 patients (11.0%) sustained pulmonary complications. Patients with chronic obstructive pulmonary disease (91.9% vs 97.2%; P = .004) and chronic renal insufficiency (92.0% vs 98.3%; P = .009) had decreased 30-day survival. Patients with chronic obstructive pulmonary disease (55.8% ± 5.8% vs 67.3% ± 3.6%; P = .013), chronic renal insufficiency (51.2% ± 5.2% vs 72.8% ± 3.7%; P = .043), and cerebrovascular disease (46.8% ± 7.4% vs 67.4% ± 3.4%; P = .003) had decreased 5-year survival. Patients who had decreased postoperative renal function (41.0% ± 7.4% vs 72.2% ± 3.4%; P = .004), and patients who sustained pulmonary complications (45.6% ± 8.8% vs 66.3% ± 3.3%; P = .042) had worse 5-year survival. CONCLUSIONS Open AAA repair can be done with low morbidity and mortality in the era of endovascular aneurysm repair. Careful consideration should be given to preoperative optimization and perioperative care in patients with chronic obstructive pulmonary disease, chronic renal insufficiency, and cerebrovascular disease. Postoperative decrease in renal function and pulmonary complication portend decreased 5-year survival; strategies to ameliorate these factors should be sought.