J. Gregory Modrall

Trends and outcomes of concurrent carotid revascularization and coronary bypass

BACKGROUND The management of concurrent carotid and coronary artery disease is controversial. Although single-center observational studies have revealed acceptable outcomes of combined carotid endarterectomy (CEA) and coronary artery bypass grafting (CABG), community-based outcomes have been substantially inferior. Recently, carotid artery stenting (CAS) has been introduced for the management of high-risk patients with carotid stenosis, including those with severe coronary artery disease. This study was undertaken to evaluate the nationwide trends and outcomes of CAS before CABG vs combined CEA and CABG and to assess the risk for adverse events. METHODS The Nationwide Inpatient Sample (NIS) was used to identify patients discharged after concurrent carotid and coronary revascularization procedures. All patients that underwent CAS before CABG and combined CEA-CABG during the years 2000 to 2004 were included. The type of revascularization and major adverse events (ie, in-hospital stroke and death rates) were determined by cross-tabulating discharge diagnostic and procedural codes. Risk stratification was performed using the Charlson Comorbidity Index. Weighted exact Cochrane-Armitage trend test and multivariate logistic regression were used to assess the association between types of revascularization, comorbidities, complications, and risk-adjusted mortality. RESULTS During the 5-year period, 27,084 concurrent carotid revascularizations and CABG were done. Of these, 96.7% underwent CEA-CABG, whereas only 3.3% (887 patients) had CAS-CABG. From 2000 to 2004, the proportion of patients undergoing CAS-CABG vs CEA-CABG did not significantly changed (P = .27). Patients undergoing CAS-CABG had fewer major adverse events than those undergoing CEA-CABG. CAS-CABG patients had a lower incidence of postoperative stroke (2.4% vs 3.9%), and combined stroke and death (6.9% v. 8.6%) than the combined CEA-CABG group (P < .001), although in-hospital death rates were similar (5.2% vs 5.4%). After risk-stratification, CEA-CABG patients had a 62% increased risk of postoperative stroke compared with patients undergoing CAS before CABG (odds ratio [OR], 1.62; 95% confidence interval [CI], 1.1-2.5; P = .02). However, no differences in the risk of combined stroke and death were observed (OR, 1.26; 95% CI, 0.9-1.6; P = NS). CONCLUSION Although CAS may currently be performed for high-risk patients, it is still infrequently used in patients who require concurrent carotid and coronary interventions. In the United States, patients who undergo CAS-CABG have significantly decreased in-hospital stroke rates compared with patients undergoing CEA-CABG but similar in-hospital mortality. CAS may provide a safer carotid revascularization option for patients who require CABG.

Long-term results of the treatment of aortic graft infection by in situ replacement with femoral popliteal vein grafts

OBJECTIVE Graft excision and neo-aortoiliac system (NAIS) reconstruction with large caliber, femoral popliteal vein (FPV) grafts have been reported as successful treatment of aortic graft infection (AGI) in several small series with limited follow-up. The goal of this study was to evaluate long-term outcomes in large cohort of consecutive patients treated with NAIS for AGI. METHODS From 1990 to 2006, 187 patients (age: 63 +/- 10 years) with AGI were treated with in situ reconstructions using 336 FPV grafts. Data from a prospectively maintained data base were analyzed. RESULTS NAIS reconstruction was performed for 144 infected aortofemoral bypasses, 21 infected aortic-iliac grafts, and 22 infected axillofemoral bypasses that had been placed to treat AGI. Polymicrobial cultures were present in 37% while 17% showed no growth. There were 55% gram positive, 32% gram negative, 13% anaerobic, and 18% fungal infections. The mean Society for Vascular Surgery run-off resistance score was 4.5 +/- 2.3. Concomitant infrainguinal bypass was necessary in 27 (14%) patients (32 limbs). Major amputations were performed in 14 (7.4%) patients. Out of 14 amputations, five patients had irreversible ischemia and in four, there was no conduit available. Graft disruption from reinfection occurred in 10 patients (5%). While 30-day mortality was 10%, procedure-related mortality was 14%. Independent risk factors for perioperative death on multivariate analysis were: preoperative sepsis (odds ratio [OR] 3.5) ASA class 4 (OR 2.9), Candida species (OR 3.4), Candida glabrata (OR 7.6), Klebsiella pneumoniae (OR 3.5), and Bacteroides fragilis (OR 4.1). Perioperative factors included use of platelets (OR 2.4), blood loss >3.0 liters (OR 9.5). Cumulative primary patency at 72 months was 81%; secondary/assisted primary patency was 91%. Limb salvage at 72 months was 89%. Five-year survival was 52%. CONCLUSIONS These results compare favorably with other methods of treating AGI, especially in patients with multilevel occlusive disease. Principle advantages include acceptable perioperative mortality, low amputation rate, superior durability with excellent long-term patency, and freedom from secondary interventions and recurrent infections.

Advances in the treatment of phlegmasia cerulea dolens

Phlegmasia cerulea dolens (PCD) is an uncommon, severe form of lower extremity deep venous thrombosis characterized by extremity swelling, cyanosis, and pain. Progression of the thrombotic process may result in extremity gangrene, amputation, and death. The relative value of specific therapeutic regimens in the treatment of this disease remains uncertain. Twelve patients, 9 females and 3 males, with PCD were treated during a 10-year period. Eighteen lower extremities were involved. Pre-existing conditions included malignancy (eight), postoperative state (four), diabetes (three), previous deep venous thrombosis (three), and hypercoagulation (two). Venous gangrene was present in four patients. All patients were treated initially with bedrest, fluid resuscitation, extremity elevation, and systemic high-dose heparin therapy. Five patients had complete resolution with this regimen alone. One patient required cessation of heparin therapy due to heparin-induced thrombocytopenia and developed gangrenous toes. Two patients whose condition failed to respond to heparin therapy underwent catheter-based delivery of urokinase with marked clinical improvement. Four patients, two with venous gangrene, died, three of whom had disseminated malignant disease. A significant percentage of patients with PCD will respond to extremity elevation, fluid resuscitation, and aggressive systemic anticoagulation therapy. Thrombolytic therapy selectively administered is beneficial in patients whose disease fails to respond promptly. Venous thrombectomy should be reserved for patients with contraindications to thrombolysis.

Operative mortality for renal artery bypass in the United States: Results from the National Inpatient Sample

BACKGROUND The mortality rate for renal artery bypass grafting (RABG) is reported to be 0% to 4% for patients with renovascular hypertension and 4% to 7% for patients with ischemic nephropathy. However, these data come from high-volume referral centers known for their expertise in treating these conditions. Because of the relative infrequency of these operations in most vascular surgery practices, the nationwide outcomes for RABG are not known. The purpose of this study was to define the operative mortality rate for RABG in the United States and to identify risk factors for perioperative mortality. METHODS The National Inpatient Sample was analyzed to identify patients undergoing RABG for the years 2000 to 2004. Categoric data were analyzed using chi(2) and the Cochran-Armitage trend tests. Multivariate logistic regression analyses were performed to identify risk factors for perioperative mortality after RABG. RESULTS During the study period, 6608 patients underwent RABG, representing a frequency of 3.51 operations per 100,000 discharges. More than two-thirds were performed at teaching hospitals (4564 vs 2,044; P < .0001). The frequency of RABG decreased by 30.7% between 2000 and 2004 (4.28 vs 2.96 RABGs per 100,000 discharges; P for trend < .0001). The in-hospital mortality for RABG was 10.0%. On univariate analysis, in-hospital mortality after RABG varied with increasing age, race, region of the country, and a preoperative history of chronic renal failure, congestive heart failure, or chronic lung disease. Logistic regression models identified advanced age (odds ratio [OR] 1.57; 95% confidence interval [CI], 1.44-1.72], female gender (OR, 1.20; 95% CI, 1.02-1.41), and a history of chronic renal failure (OR, 2.21; 95% CI, 1.75-2.78), congestive heart failure (OR, 1.94; 95% CI, 1.44-2.62), or chronic lung disease (OR, 1.40; 95% CI, 1.18-1.67) as independent markers of risk-adjusted, in-hospital mortality (P < .0001 for each of these five variables). CONCLUSIONS Nationwide in-hospital mortality after RABG is higher than predicted by prior reports from high-volume referral centers. Advanced age, female gender, and a history of chronic renal failure, congestive heart failure, or chronic lung disease were predictive of perioperative death. For the typical vascular practice, these data may provide a rationale for lower risk alternatives, such as renal artery stenting or referral to high-volume referral centers for RABG.

Randomized clinical trial of open-cell vs closed-cell stents for carotid stenting and effects of stent design on cerebral embolization

OBJECTIVE The effect of stent design on cerebral embolization has not been established. The purpose of this trial was to contrast the incidence of subclinical cerebral embolization in high-risk patients undergoing carotid artery stenting (CAS) with open-cell vs closed-cell stents. METHODS During an 18-month period, 40 patients were randomized (1:1) to undergo CAS with open-cell (Acculink, n = 20) or closed-cell stents (Xact, n = 20). A single filter device for embolic protection (Accunet filter) was used. Transcranial Doppler (TCD)-detected microembolic signals (MES) during CAS and preprocedural and 24-hour postprocedural diffusion-weighted magnetic resonance imaging (DW-MRI) were used to determine cerebral embolization. Univariate and nonparametric analyses were used to assess associations between stent design and cerebral embolization. RESULTS CAS was performed in 17 symptomatic patients (43%) and 23 asymptomatic patients (57%) with a similar number of open-cell and closed-cell stents (9/8 and 11/12, respectively). The total and poststenting median ipsilateral MES counts detected by TCD were 264 (interquartile range [IQR], 222-343) and 48 (IQR, 41-66) for open-cell stents and 339 (IQR, 163-408) and 53 (IQR, 23-88) for closed-cell stents, respectively (P > .56). New acute cerebral emboli detected with DW-MRI occurred in 53% and 47% of patients undergoing CAS with open-cell and closed-cell stents, respectively (P = 1.0). The total and ipsilateral median numbers of DW-MRI lesions between groups were not statistically significantly different (ie, 2 [IQR, 0-4] and 1 [IQR, 0-3] for open-cell stents and 1 [IQR, 0-3] and 1 [IQR, 0-2] for closed cell-stents, respectively; P > .4). One asymptomatic patient undergoing CAS with an open-cell stent sustained a minor stroke; the 30-day stroke-death rate in this series was 2.5%. CONCLUSION Cerebral embolization, as detected by TCD and DW-MRI, occurs with similar frequency after CAS with open-cell and closed-cell stents. This randomized trial does not support the superiority of any stent design with respect to cerebral embolization.

Distinct roles for angiotensin‐converting enzyme 2 and carboxypeptidase A in the processing of angiotensins within the murine heart

Angiotensin‐converting enzyme 2 (ACE2), a homologue of angiotensin‐converting enzyme (ACE), converts angiotensin (Ang) I to Ang(1–9) and Ang II to Ang(1–7), but does not directly process Ang I to Ang II. Cardiac function is compromised in ACE2 null mice; however, the importance of ACE2 in the processing of angiotensin peptides within the murine heart is not known. We determined the metabolism of angiotensins in wild‐type (WT), ACE (ACE−/−) and ACE2 null mice (ACE2−/−). Angiotensin II was converted almost exclusively to Ang(1–7) in the cardiac membranes of WT and ACE−/− strains, although generation of Ang(1–7) was greater in the ACE−/− mice (27.4 ± 4.1 versus 17.5 ± 3.2 nmol−1 mg h−1 for WT). The ACE2 inhibitor MLN4760 significantly attenuated Ang II metabolism and the subsequent formation of Ang(1–7) in both strains. In the ACE2−/− hearts, Ang II metabolism and the generation of Ang(1–7) were significantly attenuated; however, the ACE2 inhibitor reduced the residual Ang(1–7)‐forming activity in this strain. Angiotensin I was primarily converted to Ang(1–9) (WT, 28.9 ± 3.1 nmol−1 mg h−1; ACE−/−, 49.8 ± 5.3 nmol−1 mg h−1; and ACE2−/−, 35.9 ± 5.4 nmol−1 mg h−1) and to smaller quantities of Ang(1–7) and Ang II. Although the ACE2 inhibitor had no effect on Ang(1–9) formation, the carboxypeptidase A inhibitor benzylsuccinate essentially abolished the formation of Ang(1–9) and increased the levels of Ang I in cardiac membranes. In conclusion, our studies in the murine heart suggest that ACE2 is the primary pathway for the metabolism of Ang II and the subsequent formation of Ang(1–7), a peptide that, in contrast to Ang II, exhibits both antifibrotic and antiproliferative actions.

Volumetric analysis of type B aortic dissections treated with thoracic endovascular aortic repair

BACKGROUND Type B aortic dissections are being successfully treated by thoracic endovascular aortic repair (TEVAR). Postoperative false lumen patency has been associated with aneurysmal dilatation and rupture of the thoracic aorta, necessitating further intervention. This is the first volumetric analysis of type B aortic dissections comparing patients with and without false lumen thrombosis (FLT) after TEVAR. We hypothesized that a greater increase in postoperative true lumen volume will lead to FLT, and without this change, false lumen patency will result. METHODS Preoperative and postoperative computed tomography angiography (CTA) imaging was analyzed using three-dimensional reconstruction to measure the short- and long-axis diameter and cross-sectional area of the true lumen, false lumen, and total aorta. Measurements were taken at 5-cm intervals from the left subclavian artery to the aortic bifurcation. Pre- and postoperative volumetric data were calculated and compared in patients with and without postoperative FLT. RESULTS Between 2006 and 2010, 132 patients underwent thoracic aortic stent grafting. Of these, 31 (23%) had thoracic endografting for type B aortic dissection. Pre- and postoperative CTA images were available for analysis in 23 patients with a mean age of 59 ± 14 years treated for acute, complicated (n = 8, 35%), and chronic (n = 15, 65%) indications. Mean follow-up imaging was 9 months (range, 1-39 months). Thirteen patients (56%) had postoperative FLT and 10 (43%) had persistent false lumen patency. The dissections involved the left subclavian artery (n = 12), visceral arteries (n = 14), renal arteries (n = 16), and iliac arteries (n = 15). The left subclavian artery was intentionally covered in 15 patients (65%). There were no significant differences in age, acute vs chronic dissection, branch vessel involvement, coverage of the left subclavian artery, or distal extent of the endograft between patients with and without postoperative FLT. Patients with postoperative FLT had a significantly smaller preoperative maximum thoracic aortic diameter (5.05 ± 1.0 vs 6.30 ± 1.4 cm; P = .02). Volumetric analysis demonstrated significantly smaller preoperative true lumen volume (141.3 ± 68 vs 230.5 ± 92 cm(3); P = .01) in patients with FLT, but no difference in preoperative false lumen volume. Patients with FLT had a significant increase in the volume percentage of the true lumen from 42.7% to 61.7% (P = .02) after stent graft repair, compared with an increase from 46.7% to 47.7% (P = .75) in patients with persistent false lumen patency. CONCLUSIONS This volumetric study of type B aortic dissection treated with TEVAR suggests that the ability of the endograft to significantly increase the true lumen volume as a percent of the total aorta most accurately predicts postoperative FLT. This is best demonstrated in a nonaneurysmal dissection regardless of timing since dissection.

Open-cell versus closed-cell stent design differences in blood flow velocities after carotid stenting

OBJECTIVE The differential effect of open-cell vs closed-cell stent design configuration on carotid velocities detected by duplex ultrasound (DUS) imaging has not been established. To identify possible stent design differences in carotid velocities, we analyzed DUS studies obtained before and immediately after carotid artery stenting (CAS). METHODS In a series of 141 CAS procedures performed during a 3-year period, data from the first postinterventional DUS images and carotid angiograms were evaluated for each patient. Peak systolic velocities (PSV), end-diastolic velocities (EDV), and internal carotid artery/common carotid artery (ICA/CCA) PSV ratios were compared according to stent design. Differences in carotid velocities were analyzed using nonparametric statistical tests. RESULTS Completion angiograms revealed successful revascularization and <30% residual stenosis in each case. The 30-day stroke-death rate in this series was 1.6% and was unrelated to stent type. Postintervention DUS images were obtained a median of 5 days (interquartile range [IQR], 1-25 days) after CAS. Closed-cell stents were used in 41 procedures (29%) and open-cell stents in 100 (71%). The median PSV was 95.9 cm/s (IQR, 77-123 cm/s) for open-cell stents and 122 cm/s (IQR, 89-143 cm/s) for closed-cell stents, which was significantly higher (P = .007). Closed-cell stents also had significantly higher median EDVs (36 vs 29 cm/s; P =.006) and ICA/CCA PSV ratios (1.6 vs 1.1; P =.017). By DUS criteria, the carotid velocities in 45% of closed-cell stents exceeded the threshold of 50% stenosis for a nonstented artery compared with 26% of open-cell stents (P =.04). Closed-cell stents had a 2.2-fold increased risk of yielding abnormally elevated carotid velocities after CAS compared with open-cell stents (odds ratio, 2.2; 95% confidence interval, 1.02-4.9). CONCLUSIONS Carotid velocities are disproportionately elevated after CAS with closed-cell stents compared with open-cell stents. This suggests that the velocity criteria for quantifying stenosis may require modification according to stent design. The importance of these differences in carotid velocities related to stent design and the potential relationship with recurrent stenosis remains to be established.

Emerging role of endovascular grafts in complex aortoiliac occlusive disease

BACKGROUND Aortobifemoral bypass is the standard therapy for complex aortoiliac occlusive disease. The purpose of this study was to examine the use of endovascular grafts as an alternative to aortobifemoral bypass in patients with advanced aortoiliac occlusive disease at high risk. METHODS Endovascular grafts were placed in 23 limbs in 22 patients with TransAtlantic Inter-Society Consensus document (TASC) type C and D lesions. All procedures were performed in the operating room, and images were obtained with portable digital fluoroscopy. Surgical exposure of the ipsilateral common femoral artery was performed to enable safe closure of 9F to 12F sheath sites and to facilitate ipsilateral interventions in the distal external iliac artery. Concomitant infrainguinal outflow procedures were performed in 6 patients. RESULTS Twenty of 22 patients were men; mean patient age was 63.2 +/- 3.2 years. Indications for intervention were rest pain in 12 of 23 limbs and tissue loss in 9 of 22 limbs. Risk factors included hostile abdomen or pelvis in 8 patients, coronary artery disease in 11 patients, end-stage renal disease in 3 patients, and severe chronic obstructive pulmonary disease in 3 patients. Each patient received a mean of 1.6 grafts. Initial technical success was 95.2%, with one technical failure. There was no 30-day mortality. All patients experienced at least one grade improvement per Society for Vascular Surgery reporting standards. Primary patency at 24 months was 84.2% +/- 8.0%, with a limb salvage rate of 95.3% +/- 5.0%. Mean (+/- SD) ankle brachial index improved from 0.49 +/- 0.22 to 0.87 +/- 0.26 (P <.001). CONCLUSION Endovascular grafting to treat advanced aortoiliac occlusive disease can be accomplished with good clinical outcome and acceptable short-term patency. This endovascular technique can be a viable alternative to conventional surgical revascularization in patients with advanced aortoiliac occlusive disease at high risk.

The progressive nature of peripheral arterial disease in young adults: A prospective analysis of white men referred to a vascular surgery service

OBJECTIVE The onset of symptomatic peripheral arterial disease at a young age (premature PAD) has been associated with rapid progression, bypass graft failure, and amputation. This study was performed to document the incidence of these complications and to determine the risk factors for poor outcome in patients with premature PAD. METHODS This study was designed as a prospective longitudinal analysis, with patients who were ambulatory or hospitalized at a single vascular referral institution. The subjects were 51 white men with onset of PAD symptoms before the age of 45 years (mean age of onset, 41 +/- 0.5 years) and represented consecutive patients who were seen at the vascular surgery service during a 4-year period. Thirty of the study subjects (58%) were recruited during the first 2 years. The main outcome measures were number and type of lower extremity revascularization procedures or amputations that were necessitated during the follow-up period. RESULTS During a mean follow-up period of 73 +/- 6 months, 15 patients (29%) had PAD that remained stable without interventions and 15 (29%) had PAD that remained stable for a mean of 76 +/- 13 months after a single intervention. Twenty-one patients (41%) required multiple operations or major amputations. In a comparison of the 30 PAD patients whose conditions were stable with or without a single intervention with the 21 PAD patients who required multiple interventions (REDO), there were no differences in smoking, hypertension, diabetes, or dyslipidemias. The REDO group had a younger mean age at the onset of symptoms (39 +/- 1 years vs 43 +/- 2 years; P <.001). At entry, the REDO patients had a higher prevalence of infrainguinal or multilevel disease (57% vs 20%; P =.03), a lower mean ankle brachial index (0. 44 +/- 0.04 vs 0.56 +/- 0.03; P =.02), and more frequent tissue loss (24% vs 0; P =.005). The REDO patients had a higher mean lipoprotein (a) level than did the patients with stable conditions (51 +/- 11 mg/dL vs 27 +/- 5 mg/dL; P =.03), but there were no significant differences in the mean plasma homocysteine levels (19 +/- 2 micromol/L vs 16 +/- 1 micromol/L) or in the proportion of patients with hypercoagulable states (33% vs 30%). The only predictive variables that were selected with stepwise logistic regression analysis were age at onset (P <.002; odds ratio, 1.4; 95% confidence interval, 1.11 to 1.81) and ankle brachial index of less than 0.5 (P <.008; odds ratio, 6.4; 95% confidence interval, 1.5 to 27.3). CONCLUSION Although 60% of the white men with premature PAD who were referred to a vascular surgery service had conditions that appeared to remain stable, these data show that approximately 40% of the patients will require multiple interventions because of disease progression or bypass graft failure. Clinical indicators, not serum markers, are predictors of poor outcome in patients with premature PAD. The results of this study suggest that patients with onset of PAD before the age of 43 years who have objective evidence of advanced disease are predisposed to multiple interventions.