Ted R. Kohler

Outcomes Following Endovascular vs Open Repair of Abdominal Aortic Aneurysm: A Randomized Trial

CONTEXT Limited data are available to assess whether endovascular repair of abdominal aortic aneurysm (AAA) improves short-term outcomes compared with traditional open repair. OBJECTIVE To compare postoperative outcomes up to 2 years after endovascular or open repair of AAA in a planned interim report of a 9-year trial. DESIGN, SETTING, AND PATIENTS A randomized, multicenter clinical trial of 881 veterans (aged > or = 49 years) from 42 Veterans Affairs Medical Centers with eligible AAA who were candidates for both elective endovascular repair and open repair of AAA. The trial is ongoing and this report describes the period between October 15, 2002, and October 15, 2008. INTERVENTION Elective endovascular (n = 444) or open (n = 437) repair of AAA. MAIN OUTCOME MEASURES Procedure failure, secondary therapeutic procedures, length of stay, quality of life, erectile dysfunction, major morbidity, and mortality. RESULTS Mean follow-up was 1.8 years. Perioperative mortality (30 days or inpatient) was lower for endovascular repair (0.5% vs 3.0%; P = .004), but there was no significant difference in mortality at 2 years (7.0% vs 9.8%, P = .13). Patients in the endovascular repair group had reduced median procedure time (2.9 vs 3.7 hours), blood loss (200 vs 1000 mL), transfusion requirement (0 vs 1.0 units), duration of mechanical ventilation (3.6 vs 5.0 hours), hospital stay (3 vs 7 days), and intensive care unit stay (1 vs 4 days), but required substantial exposure to fluoroscopy and contrast. There were no differences between the 2 groups in major morbidity, procedure failure, secondary therapeutic procedures, aneurysm-related hospitalizations, health-related quality of life, or erectile function. CONCLUSIONS In this report of short-term outcomes after elective AAA repair, perioperative mortality was low for both procedures and lower for endovascular than open repair. The early advantage of endovascular repair was not offset by increased morbidity or mortality in the first 2 years after repair. Longer-term outcome data are needed to fully assess the relative merits of the 2 procedures. TRIAL REGISTRATION clinicaltrials.gov Identifier: NCT00094575.

Long-Term Comparison of Endovascular and Open Repair of Abdominal Aortic Aneurysm

BACKGROUND Whether elective endovascular repair of abdominal aortic aneurysm reduces long-term morbidity and mortality, as compared with traditional open repair, remains uncertain. METHODS We randomly assigned 881 patients with asymptomatic abdominal aortic aneurysms who were candidates for both procedures to either endovascular repair (444) or open repair (437) and followed them for up to 9 years (mean, 5.2). Patients were selected from 42 Veterans Affairs medical centers and were 49 years of age or older at the time of registration. RESULTS More than 95% of the patients underwent the assigned repair. For the primary outcome of all-cause mortality, 146 deaths occurred in each group (hazard ratio with endovascular repair versus open repair, 0.97; 95% confidence interval [CI], 0.77 to 1.22; P=0.81). The previously reported reduction in perioperative mortality with endovascular repair was sustained at 2 years (hazard ratio, 0.63; 95% CI, 0.40 to 0.98; P=0.04) and at 3 years (hazard ratio, 0.72; 95% CI, 0.51 to 1.00; P=0.05) but not thereafter. There were 10 aneurysm-related deaths in the endovascular-repair group (2.3%) versus 16 in the open-repair group (3.7%) (P=0.22). Six aneurysm ruptures were confirmed in the endovascular-repair group versus none in the open-repair group (P=0.03). A significant interaction was observed between age and type of treatment (P=0.006); survival was increased among patients under 70 years of age in the endovascular-repair group but tended to be better among those 70 years of age or older in the open-repair group. CONCLUSIONS Endovascular repair and open repair resulted in similar long-term survival. The perioperative survival advantage with endovascular repair was sustained for several years, but rupture after repair remained a concern. Endovascular repair led to increased long-term survival among younger patients but not among older patients, for whom a greater benefit from the endovascular approach had been expected. (Funded by the Department of Veterans Affairs Office of Research and Development; OVER ClinicalTrials.gov number, NCT00094575.).

Increased blood flow inhibits neointimal hyperplasia in endothelialized vascular grafts.

Intimal hyperplasia is a primary cause of failure after vascular reconstruction and may be affected by blood flow. We have studied the effects of increased blood flow on intimal hyperplasia in porous polytetrafluoroethylene grafts implanted in baboons. These grafts develop an endothelial lining by 2 weeks and neointimal thickening due to proliferation of underlying smooth muscle cells by 1 month. Creation of a distal arteriovenous fistula increased flow (from 230 +/- 35 to 785 +/- 101 ml/min, p less than 0.001) and mean shear (from 26 +/- 4 to 78 +/- 10 dynes/cm2, p less than 0.001) without causing a drop in pressure across the grafts. Fistula flow did not alter the pattern of endothelial coverage but did cause a marked reduction in the cross-sectional area of the neointima (from 2.60 +/- 0.52 to 0.42 +/- 0.07 mm2 at 3 months, p less than 0.01). Detailed morphometric analysis revealed an equivalent percentage decrease in smooth muscle cells and matrix content, suggesting that the primary effect of increased flow was to reduce smooth muscle cell number without affecting the amount of matrix produced by individual cells. The neointima remained sensitive to changes in flow at late times; ligation of the fistula after 2 months resulted in a rapid increase in neointimal thickness (from 0.60 +/- 0.03 mm2 after 2 months of fistula flow to 3.88 +/- 0.55 mm2 1 month after ligation of fistula, p less than 0.01). These results support the hypothesis that changes in blood flow affect the structure of diseased as well as normal vessels.

Duplex ultrasound scanning in the diagnosis of renal artery stenosis: A prospective evaluation

Since ultrasonic energy can be used to interrogate vessels at great depth, it is only natural that it should be applied to deeply placed arteries in the abdomen. Early studies suggested that high-grade stenoses of the renal artery could be detected by this approach as long as the peak systolic velocity in the renal artery was normalized by that measured in the abdominal aorta. A retrospective study comparing the peak velocity in the renal artery to that from the adjacent abdominal aorta (the renal aortic ratio) showed that if this value exceeded 3.5, it is likely to be associated with a greater than 60% diameter-reducing stenosis. To test this hypothesis, we used duplex scanning to prospectively evaluate 58 renal arteries in 29 patients in whom arteriograms were available. There were 39 renal arteries with 0% to 59% stenosis, 14 with 60% to 99% stenosis, and five occlusions by angiography. Renal duplex scanning accurately diagnosed 38 of 39, 11 of 14, and four of five of these, respectively, giving a sensitivity of 84%, a specificity of 97%, and a positive predictive value of 94% for the detection of a greater than 60% diameter-reducing stenosis. The overall agreement with angiography was 93%. These data show that renal duplex scanning can be used to diagnose renal artery stenosis in patients with hypertension or renal dysfunction, thus providing a rational basis for the selection of patients for angiography.

Noninvasive diagnosis of renal artery stenosis by ultrasonic duplex scanning.

We retrospectively studied the results of duplex scanning for evaluation of renal artery disease in 158 patients. Satisfactory examinations were achieved in 144 patients (90%). Arteriograms were available for 43 renal arteries. We used the ratio of the peak velocities in the renal artery and the aorta (RAR) to separate nonstenotic arteries (less than 60% diameter reduction) from stenotic arteries (greater than 60% diameter reduction). With an RAR of greater than 3.5 to indicate stenotic lesions, duplex scanning had a sensitivity of 91% (20 of 22 diseased arteries correctly identified) and specificity of 95% (20 of 21 normal or insignificantly diseased arteries correctly identified). One of four occluded arteries was incorrectly interpreted as patent because of misidentification of a collateral vessel. Prospective studies will be necessary to validate this test and establish other criteria for a more detailed classification of renal artery stenosis. The ratio of the end-diastolic to peak systolic velocities in the renal artery (EDR) tended to decrease with increasing serum creatinine levels, presumably because renal vascular resistance increases with end-stage parenchymal disease. EDR may prove useful in the detection of advanced parenchymal disease before renal artery revascularization is attempted.

Increased Blood Flow Induces Regression of Intimal Hyperplasia

We have previously shown that high shear stress inhibits growth of developing neointima in a primate model of polytetrafluoroethylene (PTFE) graft healing. We used this model to test the hypothesis that increased shear stress can cause atrophy of an established neointima. High porosity PTFE grafts were inserted into the aorto-iliac circulation bilaterally in baboons. These grafts develop neointimal hyperplasia comprising smooth muscle cells and a luminal surface of confluent endothelium. Neointima was allowed to develop for 2 months. At that time 8 animals were sacrificed. In eight other animals blood flow in one of two grafts was increased by construction of a femoral arterio-venous fistula. These animals were sacrificed 2 months later (4 months after graft placement). At four months, intimal cross sectional area was smaller on the high shear stress side compared to the contralateral, normal shear stress side (2.53 +/- 0.75 versus 6.83 +/- 0.65 mm2, P < .05). Neointima from grafts exposed to 2 months normal shear stress followed by 2 months of high shear stress had regressed when compared to normal-shear stress grafts studied at 2 months (2.53 +/- 0.75 versus 4.56 +/- 0.68 mm2, P < .05). Morphometric analysis using transmission electron microscopy revealed that the decrease in intimal cross sectional area was attributable to a loss of both smooth muscle cells and matrix. Endothelial nitric oxide synthase was induced in high-flow graft intima. These observations support the conclusion that elevated shear stress can cause vessel wall atrophy. This process might be mediated by nitric oxide.

Intraabdominal paraanastomotic aneurysms after aortic bypass grafting

Although the reported incidence of intraabdominal paraanastomotic aneurysms after abdominal aortic bypass grafting ranges from 1% to 15%, the true incidence is unknown because few studies have used routine, serial radiographic or sonographic imaging studies. Since July 1, 1988, we have used yearly abdominal sonography examinations to monitor our patients with aortic grafts. In the first 33 months we studied 138 patients. Medical records of 111 of these were available for review and form the basis of this report. Eleven patients (10%) were found to have intraabdominal paraanastomotic aneurysms ranging in overall size from 4.1 to 6.2 cm (mean, 5.0 +/- 0.7 cm). The mean time between operation and detection of an aneurysm was 144 +/- 101 months (range, 8 to 336 months). Three paraanastomotic aneurysms occurred within 3 years of operation, and the remaining eight occurred late (7 to 28 years). By life-table analysis, the incidence of paraanastomotic aneurysms was 27% at 15 years. Paraanastomotic aneurysms were classified as either pseudoaneurysms (presumed disruption of the anastomotic suture line, n = 7) or as true aneurysms (widening of the adjacent aorta, n = 4). True aneurysms occurred only after repair of an abdominal aortic aneurysm, whereas pseudoaneurysms were more frequent after bypass for occlusive disease. The finding of paraanastomotic aneurysms in 10% of our patients supports the use of yearly sonography for routine follow-up after aortic grafting.

Flow affects development of intimal hyperplasia after arterial injury in rats.

This study examined the effects of blood flow on intimal hyperplasia after balloon catheter injury of the rat common carotid artery. Flow was altered by ligation of the opposite common carotid artery (increased flow) or of the ipsilateral internal carotid artery (decreased flow). Blood flow decreased by 35% in the low-flow group and increased by 29% in the high-flow group. Similar changes in mean velocity were observed. Cross-sectional intimal area was significantly greater in the low- than the high-flow group at 2 weeks (0.11 +/- 0.01 versus 0.06 +/- 0.01 mm2, p less than 0.01) and 4 weeks (0.17 +/- 0.02 versus 0.12 +/- 0.01 mm2, p = 0.01) but not at 1 or 8 weeks. Smooth muscle cell proliferation rates (thymidine labeling indexes) were not different in high- and low-flow groups at 2 days and at 1 and 4 weeks. Matrix accumulation at 2 and 4 weeks was the same in both groups. Mature neointima did not respond to changes in flow; when vessel ligation was delayed until 2 months after injury, there was no effect on neointimal area. These data indicate that early neointimal hyperplasia is increased when flow is reduced, possibly because of alteration of smooth muscle cell migration.

Time course of flow-induced smooth muscle cell proliferation and intimal thickening in endothelialized baboon vascular grafts.

Polytetrafluoroethylene (PTFE) grafts placed into the arterial circulation of baboons for 8 weeks under high blood flow (HF) conditions develop a thin intima composed of smooth muscle cells (SMCs) and extracellular matrix beneath an endothelial monolayer. When these grafts are returned abruptly to normal flow (NF), they develop marked intimal thickening within 1 month. The mechanisms underlying this thickening are unclear. We studied the SMC response to altered flow by placing bilateral aortoiliac PTFE grafts into baboons with bilateral femoral arteriovenous fistulas. After 8 weeks, one fistula was closed, returning the graft flow on that side to NF. The opposite graft remained under HF conditions. Flow differences were monitored with duplex ultrasound (for all grafts: NF, 135 +/- 21 [mean +/- SEM] mL/min; HF, 507 +/- 35 mL/min; P < .001). Grafts were removed 2, 4, 7, 14, or 28 days later (five animals per group). Endothelial coverage, as assessed by scanning electron microscopy, was intact in each graft. Intimal area and SMC number increased progressively in NF grafts through 28 days (for area: NF, 3.0 +/- 0.3 mm2; HF, 0.6 +/- 0.2 mm2; P < .001; and for SMCs per cross section: NF, 11.8 +/- 1.1 x 10(3); HF, 2.6 +/- 1.0 x 10(3); P < .002). Intimal SMC proliferation (thymidine labeling) was increased significantly in NF grafts at 4 and 7 days (at 4 days: NF, 5.9 +/- 1.5%; HF, 1.4 +/- 0.6%; P < .05). Extracellular matrix accounted for an equal proportion of intimal mass in NF and HF grafts (percent matrix at 28 days: NF, 62.9 +/- 1.6%; HF, 63.7 +/- 4.7%; P = NS). We conclude that intimal thickening in this model of flow-induced vascular remodeling is due to increased SMC proliferation and accumulation of SMCs with a proportionate amount of extracellular matrix.

Shear stress regulates smooth muscle proliferation and neointimal thickening in porous polytetrafluoroethylene grafts.

High shear stress appears to decrease wall thickening in diseased arteries and vascular grafts. To determine if increased shear stress diminishes smooth muscle (SMC) proliferation, we studied the effect of increased blood flow on neointimal thickening in porous polytetrafluoroethylene grafts implanted in baboons. An aorto-aortic 5-mm graft was placed in tandem with a pair of aorto-iliac 5-mm grafts, so that the proximal graft supplied all flow to both distal grafts. At 12 weeks, calculated luminal shear stress in proximal grafts was twice that in distal grafts (24 +/- 8 versus 11 +/- 5 dynes/cm2; p less than 0.05). All grafts were completely endothelialized. The neointimal cross-sectional area in proximal grafts was about half as large as in distal grafts (3.36 +/- 1.61 versus 5.93 +/- 0.61 mm2; p less than 0.05). Proximal grafts also had significantly less SMC proliferation (0.14 +/- 0.05% versus 0.24 +/- 0.10%; p less than 0.05) and SMC volume (6.1 +/- 4.0 versus 12.4 +/- 2.6 mm3/cm graft; p less than 0.01) when compared with distal grafts. We conclude that the elevation in shear stress in the proximal graft, which remained within the physiological range, inhibits SMC proliferation and neointimal thickening in these grafts.