Reoperation for carotid stenosis is as safe as primary carotid endarterectomy

PURPOSE

Patients with recurrent carotid artery stenosis are sometimes referred for carotid angioplasty and stenting because of reports that carotid reoperation has a higher complication rate than primary carotid endarterectomy. The purpose of this study was to determine whether a difference exists between outcomes of primary carotid endarterectomy and reoperative carotid surgery.

METHODS

Medical records were reviewed for all carotid operations performed from September 1993 through March 1998 by vascular surgery faculty at a single academic center. The results of primary carotid endarterectomy and operation for recurrent carotid stenosis were compared.

RESULTS

A total of 390 operations were performed on 352 patients. Indications for primary carotid endarterectomy (n = 350) were asymptomatic high-grade stenosis in 42% of the cases, amaurosis fugax and transient ischemic symptoms in 35%, global symptoms in 14%, and previous stroke in 9%. Indications for reoperative carotid surgery (n = 40) were symptomatic recurrent lesions in 50% of the cases and progressive high-grade asymptomatic stenoses in 50%. The results of primary carotid endarterectomy were no postoperative deaths, an overall stroke rate of 1.1% (three postoperative strokes, one preoperative stroke after angiography), and no permanent cranial nerve deficits. The results of operations for recurrent carotid stenosis were no postoperative deaths, no postoperative strokes, and no permanent cranial nerve deficits. In the primary carotid endarterectomy group, the mean hospital length of stay was 2.6 +/- 1. 1 days and the mean hospital cost was $9700. In the reoperative group, the mean length of stay was 2.6 +/- 1.5 days and the mean cost was $13,700. The higher cost of redo surgery is accounted for by a higher preoperative cerebral angiography rate (90%) in redo cases as compared with primary endarterectomy (40%).

CONCLUSION

In this series of 390 carotid operations, the procedure-related stroke/death rate was 0.8%. There were no differences between the stroke-death rates after primary carotid endarterectomy and operation for recurrent carotid stenosis. Operation for recurrent carotid stenosis is as safe and effective as primary carotid endarterectomy and should continue to be standard treatment.

Effect of exercise on hemodynamic conditions in the abdominal aorta

PURPOSE

The beneficial effect of exercise in the retardation of the progression of cardiovascular disease is hypothesized to be caused, at least in part, by the elimination of adverse hemodynamic conditions, including flow recirculation and low wall shear stress. In vitro and in vivo investigations have provided qualitative and limited quantitative information on flow patterns in the abdominal aorta and on the effect of exercise on the elimination of adverse hemodynamic conditions. We used computational fluid mechanics methods to examine the effects of simulated exercise on hemodynamic conditions in an idealized model of the human abdominal aorta.

METHODS

A three-dimensional computer model of a healthy human abdominal aorta was created to simulate pulsatile aortic blood flow under conditions of rest and graded exercise. Flow velocity patterns and wall shear stress were computed in the lesion-prone infrarenal aorta, and the effects of exercise were determined.

RESULTS

A recirculation zone was observed to form along the posterior wall of the aorta immediately distal to the renal vessels under resting conditions. Low time-averaged wall shear stress was present in this location, along the posterior wall opposite the superior mesenteric artery and along the anterior wall between the superior and inferior mesenteric arteries. Shear stress temporal oscillations, as measured with an oscillatory shear index, were elevated in these regions. Under simulated light exercise conditions, a region of low wall shear stress and high oscillatory shear index remained along the posterior wall immediately distal to the renal arteries. Under simulated moderate exercise conditions, all the regions of low wall shear stress and high oscillatory shear index were eliminated.

CONCLUSION

This numeric investigation provided detailed quantitative data on the effect of exercise on hemodynamic conditions in the abdominal aorta. Our results indicated that moderate levels of lower limb exercise are necessary to eliminate the flow reversal and regions of low wall shear stress in the abdominal aorta that exist under resting conditions. The lack of flow reversal and increased wall shear stress during exercise suggest a mechanism by which exercise may promote arterial health, namely with the elimination of adverse hemodynamic conditions.

AneuRx stent graft versus open surgical repair of abdominal aortic aneurysms: multicenter prospective clinical trial

The results of a prospective, nonrandomized, multicenter clinical trial that compared endovascular stent graft exclusion of abdominal aortic aneurysms with open surgical repair are presented. During an 18-month period, 250 patients with infrarenal aneurysms underwent treatment at 12 study sites-190 patients underwent endovascular repair using the Medtronic AneuRx stent graft (Sunnyvale, Calif), and 60 underwent open surgical repair. There was no significant difference in operative mortality rates between the groups. The patients who underwent stent grafting had significant reductions in blood loss, time to extubation, and days in the intensive care unit and in the hospital, with an earlier return to function. The major morbidity rate was reduced from 23% in the surgery group to 12% (P <. 05) in the stent graft group. There was no difference in the combined morbidity/mortality rates between the two groups. Primary technical success at the time of discharge for the patients with stent grafts was 77%, largely as a result of a 21% endoleak rate. At 1 month, the endoleak rate had decreased to 9%. There was no difference in the primary or secondary procedure success rates at 30 days between the surgery and stent graft groups. The primary graft patency rate at 6 months was 98% in the surgery group and 97% in the stent graft group. The aneurysm exclusion rate at 1 month and 6 months was 100% in patients who underwent surgery and 91% in patients who underwent stent grafting. Stent graft migration occurred in three patients and resulted in late endoleaks; each endoleak was corrected by means of endovascular placement of a stent graft extender cuff. There have been no aneurysm ruptures and no surgical conversions to open repair in the stent graft group. Stent graft repair compares favorably with open surgical repair, with a reduced morbidity rate, shortened hospital stays, and satisfactory short term outcomes.

Predictive medicine: computational techniques in therapeutic decision-making

The current paradigm for surgery planning for the treatment of cardiovascular disease relies exclusively on diagnostic imaging data to define the present state of the patient, empirical data to evaluate the efficacy of prior treatments for similar patients, and the judgement of the surgeon to decide on a preferred treatment. The individual variability and inherent complexity of human biological systems is such that diagnostic imaging and empirical data alone are insufficient to predict the outcome of a given treatment for an individual patient. We propose a new paradigm of predictive medicine in which the physician utilizes computational tools to construct and evaluate a combined anatomic/physiologic model to predict the outcome of alternative treatment plans for an individual patient. The predictive medicine paradigm is implemented in a software system developed for Simulation-Based Medical Planning. This system provides an integrated set of tools to test hypotheses regarding the effect of alternate treatment plans on blood flow in the cardiovascular system of an individual patient. It combines an Internet-based user interface developed using Java and VRML, image segmentation, geometric solid modeling, automatic finite element mesh generation, computational fluid dynamics, and scientific visualization techniques. This system is applied to the evaluation of alternate, patient-specific treatments for a case of lower extremity occlusive cardiovascular disease.

Finite element modeling of three-dimensional pulsatile flow in the abdominal aorta: relevance to atherosclerosis

The infrarenal abdominal aorta is particularly prone to atherosclerotic plaque formation while the thoracic aorta is relatively resistant. Localized differences in hemodynamic conditions, including differences in velocity profiles, wall shear stress, and recirculation zones have been implicated in the differential localization of disease in the infrarenal aorta. A comprehensive computational framework was developed, utilizing a stabilized, time accurate, finite element method, to solve the equations governing blood flow in a model of a normal human abdominal aorta under simulated rest, pulsatile, flow conditions. Flow patterns and wall shear stress were computed. A recirculation zone was observed to form along the posterior wall of the infrarenal aorta. Low time-averaged wall shear stress and high shear stress temporal oscillations, as measured by an oscillatory shear index, were present in this location, along the posterior wall opposite the superior mesenteric artery and along the anterior wall between the superior and inferior mesenteric arteries. These regions were noted to coincide with a high probability-of-occurrence of sudanophilic lesions as reported by Cornhill et al. (Monogr. Atheroscler. 15:13-19, 1990). This numerical investigation provides detailed quantitative data on hemodynamic conditions in the abdominal aorta heretofore lacking in the study of the localization of atherosclerotic disease.

Sequential increases and decreases in blood flow stimulates progressive intimal thickening

OBJECTIVES

To assess the effect of chronic, repetitive increases and decreases in blood flow on an artery.

MATERIALS AND METHODS

Arteriovenous fistulae were created in Japanese male rabbits between the left common carotid artery and the corresponding external jugular vein. Animals were placed into either control groups or one of six cycle groups consisting of flow variations (0.5 cycles, 1.0 cycle, 1.5 cycles, 2.0 cycles, 2.5 cycles and 3.0 cycles). Each complete cycle consisted of 4 weeks of increased flow followed by 6 weeks of normalised flow by fistula ligation.

RESULTS

Arteries exposed to increased flow for 4 weeks (0.5 cycles) had a significant increase in lumen diameter without intimal thickening. After 6 weeks of normalised flow (1.0 cycle), shear stress became subnormal (0.42 +/- 0.17 N/m2), intimal thickening developed. In subsequent cycles, intimal thickening continued to develop with each point of flow normalisation and reduction in shear stress. Histologic and ultrastructural analysis revealed endothelial cells preservation at all time points, with individual strata of smooth muscle cell proliferation in the intima corresponding to the cycle numbers.

CONCLUSION

Progressive intimal thickening occurred in the previously flow-induced remodelled artery when shear stress was reduced to subnormal levels with preserved endothelium, but was inhibited by high flow periods.

Matrix metalloproteinase inhibition limits arterial enlargements in a rodent arteriovenous fistula model

BACKGROUND

We administered a specific, nonselective matrix metalloproteinase (MMP) inhibitor (RS-113,456) to examine the effect of MMP inhibition on flow-mediated arterial enlargement in a rodent arteriovenous fistula (AVF) model.

METHODS

Four groups of male Sprague-Dawley rats were created: sham (sham operated; n = 10), control (2.0 mm left common femoral AVF alone; n = 16), vehicle (AVF plus 0.5 mL vehicle orally twice a day; n = 20), and treatment (AVF plus 25 mg/kg RS-113,456 in 0.5 mL vehicle orally twice a day; n = 16). Heart rate, mean arterial pressure, and body weight were recorded on postoperative days 0, 7, 14, and 21. On day 21, AVF patency was confirmed, the infrarenal aorta and common iliac arteries were exposed, blood flow velocity and external diameter were measured, and wall shear stress (WSS) was calculated. Analysis was performed by paired, two-tailed Student t test, one-way analysis of variance, and the Bonferroni/Dunn procedure for post hoc testing.

RESULTS

Heat rate, mean arterial pressure, and weight did not vary at any time between groups. Aortic and left iliac diameter was larger in the AVF groups than in sham groups (P < .001), and control and vehicle groups were larger than treatment groups (P < .0001). Changes in aortic and left iliac flow were also significant (AVF was more than sham and control, and vehicle was more than treatment). No difference in aortic and left iliac artery velocity and WSS or right iliac diameter, velocity, flow, or WSS was observed between groups.

CONCLUSIONS

MMP inhibition diminishes flow-mediated arterial enlargement in the rat AVF model.

Basic fibroblast growth factor expression precedes flow-induced arterial enlargement

Arteries enlarge in response to increased blood flow, but the molecular signals controlling this enlargement are not well understood. Basic fibroblast growth factor (bFGF) is a potent mitogen for endothelial cells (EC) and smooth muscle cells (SMC) and promotes cellular proliferation and differentiation. In order to determine whether bFGF is expressed in response to increased blood flow in vivo, carotid-jugular arteriovenous fistulas (AVF) were created in Japanese white rabbits. The carotid artery proximal to the fistula was harvested after 1, 3, or 7 days and compared to nonoperated, control carotid arteries. Arterial blood flow increased five- to eightfold in all AVF animals and resulted in a significant increase in wall shear stress. The proximal carotid artery arterial diameter was no different from control after 1 and 3 days (2.3 +/- 0. 1 mm) but enlarged to 2.9 +/- 0.1 mm (P < 0.05) after 7 days. RT-PCR revealed early transcription of bFGF mRNA at 1 and 3 days with increased densitometric band ratio (bFGF/beta-actin) at 7 days. Immunohistochemical analysis revealed bFGF protein localization in EC of control arteries as well as AVF arteries at all time points. SMC and adventitia expression of bFGF was absent in controls, minimal at 1 day, and increased after 3 and 7 days in the experimental groups. Western blotting confirmed the presence of bFGF in samples and transmission immunoelectron microscopy confirmed its nuclear localization. Endothelial cells in vivo express bFGF under both normal and elevated flow conditions. Smooth muscle cells, however, do not express bFGF under normal flow conditions but begin to express bFGF after 1 day of high flow with increased expression after 3 and 7 days. Flow-induced arterial enlargement begins after SMC expression of bFGF. Therefore, bFGF may play a role in arterial enlargement and adaptive remodeling in response to increased flow.

Treatment of aortoiliac aneurysms with use of single-piece tapered stent-grafts

PURPOSE

The authors describe their experience with the use of single-piece, tapered stent-grafts for the treatment of abdominal aortic or aortoiliac aneurysms.

MATERIALS AND METHODS

Single-piece, tapered stent-grafts were placed in 15 patients for the treatment of abdominal aortic aneurysms with small distal necks (n = 13), and aortoiliac aneurysms (n = 2). There were 13 men and two women who ranged in age from 59 to 83 years (mean, 71 years). Usual open surgery was considered high risk in all patients because of comorbid medical conditions. The stent-grafts were made of Z stents covered with polytetrafluoroethylene (PTFE). Additional stent-grafts needed to treat perigraft leaks were made of Z stents covered with woven polyester (n = 5), Wallstents covered with PTFE (n = 2), Z stents covered with PTFE (n = 1), and a PTFE-covered Palmaz stent (n = 1). After stent-graft placement, the contralateral iliac artery was occluded by a blocking device composed of either a PTFE-covered Palmaz (n = 1) or Z stent (n = 13), and a femoral-femoral bypass was created.

RESULTS

After placement of the stent-grafts, immediate perigraft leaks were observed in eight patients (53%). These were at the proximal (n = 5) or the distal end (n = 3). All, except one, were treated successfully with additional stent-grafts. The one failure was in a patient who developed aortic rupture after balloon dilation, requiring open surgical repair. Second procedures were required in four patients (27%), including three leaks treated successfully with coil embolization and/or a back-up stent-graft, and one stent-graft migration and thrombosis treated by thrombolysis and placement of an additional stent-graft. One patient died of respiratory failure 23 days after placement of the stent-graft. The mean follow-up was 12 months (range, 4-26 months). On the last follow-up, the aneurysms in the 13 living patients remained thrombosed.

CONCLUSION

Treatment of aortoiliac aneurysms with use of single-piece, tapered stent-grafts is feasible in selected patients. The morbidity and mortality rates compare favorably with those of the open surgical procedures in a high-risk population. Further improvements in the technique and longer follow-up data are needed before this procedure can be recommended for the treatment of all aortoiliac aneurysms.

Operative repair for aortic aneurysms: the gold standard

Surgical treatment of abdominal aortic aneurysm (AAA) is being challenged by newer, minimally invasive therapies. Such new treatment strategies will need to prove themselves against concurrent results of standard operative AAA repair, within defined medical risk and aneurysm morphological categories. We review the natural history of AAAs, the medical risk levels for elective AAA repair, aneurysm morphology and its impact on operative mortality, the issue of high-risk patient treatment, and the current standard of care for AAAs based on single-center, multicenter, and population-based statistics. In good-risk patients, aneurysms > 5 cm in diameter are best treated by replacement with a prosthetic graft. Operative mortality should be < 5% 1-year survival > 90%. Aortic endograft techniques must meet or exceed these standards if they are to supplant standard surgical repair.

Flow-induced arterial enlargement is inhibited by suppression of nitric oxide synthase activity in vivo

BACKGROUND

Acute flow-induced arterial dilation is mediated by nitric oxide (NO). The role of NO in chronic flow-induced adaptive enlargement is unknown. We assessed the role of NO in arterial adaptation to increased blood flow (BF).

METHODS

Iliac artery BF was increased in adult male rats by creating a left femoral arteriovenous fistula. Left iliac BF and diameter were measured, and wall shear stress was calculated. The effect of the NO synthase inhibitor N omega-nitro-L-arginine-methyl ester (L-NAME) was studied in arteriovenous fistula rats divided into three groups (group 1, vehicle, group 2, 0.5 mg/ml; group 3, 2 mg/ml) in drinking water. Arterial diameter, blood pressure, and medial cell density were assessed after 21 days. Left iliac cyclic guanosine monophosphate content was measured in an additional group of animals.

RESULTS

BF and wall shear stress in the left iliac artery increased fourfold immediately after arteriovenous fistula. Arterial enlargement was evident after 7 days, and wall shear stress normalized after 42 days. Flow-induced arterial enlargement was inhibited by both low- and high-dose L-NAME compared with control (analysis of variance p < 0.05). Blood pressure was elevated only in animals treated with high-dose L-NAME. Left iliac cyclic guanosine monophosphate content was lower in rats treated with L-NAME than in the control group (p < 0.05).

CONCLUSIONS

NO suppression by L-NAME inhibits flow-induced iliac artery enlargement in rats. This finding suggests that NO plays a role in flow-induced arterial remodeling.

Measurements of velocity and wall shear stress inside a PTFE vascular graft model under steady flow conditions

The flow field inside a model of a polytetrafluorethylene (PTFE) canine artery end-to-side bypass graft was studied under steady flow conditions using laser-Doppler anemometry. The anatomically realistic in vitro model was constructed to incorporate the major geometric features of the in vivo canine anastomosis geometry, most notably a larger graft than host artery diameter. The velocity measurements at Reynolds number 208, based on the host artery diameter, show the flow field to be three dimensional in nature. The wall shear stress distribution, computed from the near-wall velocity gradients, reveals a relatively low wall shear stress region on the wall opposite to the graft near the stagnation point approximately one artery diameter in axial length at the midplane. This low wall shear stress region extends to the sidewalls, suture lines, and along the PTFE graft where its axial length at the midplane is more than two artery diameters. The velocity distribution inside the graft model presented here provides a data set well suited for validation of numerical solutions on a model of this type.

What are the characteristics of the ideal endovascular graft for abdominal aortic aneurysm exclusion?

PURPOSE

To review the anatomic factors crucial to successful endoluminal abdominal aortic aneurysm (AAA) repair and propose an ideal endograft design for AAA exclusion.

METHODS AND RESULTS

The anatomic features of critical importance to endovascular AAA exclusion comprise remote arterial access, proximal and distal fixation sites, AAA morphology, and arterial wall pathology. When designing an aortic endograft, the major components to consider are stent selection, graft material, and the delivery system. The ideal endograft design must be sufficiently versatile to treat a broad range of patients. To meet this requirement, the endograft should display a high degree of dimensional adaptability. A modular bifurcated endograft design permits intraoperative customization to tailor the device to each patient's anatomy and pathology.

CONCLUSIONS

The modular stent-graft concept addresses many of the important factors in the evolution toward an ideal aortic endograft. Extensive testing will be needed to determine if the bifurcated stent-graft described here is the optimal design for effective AAA exclusion.

Simultaneous abdominal aortic replacement and thoracic stent-graft placement for multilevel aortic disease

PURPOSE

Patients with aneurysmal disease involving both the descending thoracic and abdominal aorta have historically required simultaneous or sequential conventional operations, but the morbidity rate is high with either approach in these patients, who often exhibit coexisting cardiopulmonary disease. Transluminally placed endovascular grafts have recently been developed for repair of aortic aneurysms, and we have implemented these techniques to eliminate the need for a thoracotomy in patients with multilevel aortic disease.

METHODS

Since January 1994, 18 patients have undergone conventional abdominal aortic replacement with endovascular stent-graft placement into the descending thoracic aorta under fluoroscopic guidance through a 10 mm Dacron side limb off the abdominal graft. Abdominal aortic replacement required a tube graft in eight patients and bifurcated grafts in 10 patients. Thoracic stent-grafts (custom fabricated, woven Dacron covered, self-expandable stents) averaged 12.2 +/- 4.2 cm (mean +/- SD) in length.

RESULTS

One patient died, resulting in a hospital mortality rate of 6%. No patients required further surgical intervention to treat their aortic disease. Seventeen patients (94%) are currently well 14 +/- 8 months after surgery (range, 3 to 29 months) with completely excluded thoracic aortic disease, no stent migration, and no change in stent configuration documented by serial radiologic examinations.

CONCLUSIONS

Simultaneous abdominal aortic replacement and deployment of a thoracic stent-graft can safely exclude multilevel aortic aneurysmal disease and may be a valuable treatment option for these otherwise high-risk patients.

In-vivo measurements of blood flow velocity profiles in canine ilio-femoral anastomotic bypass grafts

In-vivo velocity profiles were recorded with a 20 MHz 80-channel pulsed Doppler ultrasound velocimeter in canine end-to-side ilio-femoral anastomotic grafts. The geometries were obtained from casts of the anastomotic region, and flow rates were measured with electromagnetic flow probes. Three cases reported here include a "standard" geometry, which was similar to previously studied in vitro models, a stenosed geometry, and a case with below average flow rate. Observed flow features include separation at the hood and toe, movement of the floor stagnation point, and skewed profiles in the proximal outflow segment. Out-of-plane curvature and lateral displacement of the anastomosis inlet appear to have a strong effect on the flow fields. In addition, compliance affects the instantaneous flow rates within the proximal and distal branches.

Diminished postprandial hyperemia in patients with aortic and mesenteric arterial occlusive disease. Quantification by magnetic resonance flow imaging

BACKGROUND

Superior mesenteric blood flow in the fasting and postprandial state in humans can be measured accurately by cine phase-contrast (CPC) magnetic resonance (MR) imaging. Postprandial flow changes associated with mesenteric arterial occlusive disease (MAOD) are unknown.

METHODS AND RESULTS

We used CPC MR imaging to measure fasting and postprandial blood flow in the superior mesenteric artery (SMA) and vein (SMV) in 22 patients (mean age, 69 years) with aortic occlusive disease and MAOD and compared the results with similar measurements in 8 younger, asymptomatic volunteers (mean age, 34 years). All 22 patients had stenosis or occlusion of the splanchnic or pelvic arteries demonstrated by contrast aortography; 19 were asymptomatic and 3 had symptoms of chronic mesenteric ischemia. Mean fasting blood flow was higher in patients (4.5 mL.kg-1.min-1) than in volunteers (2.3 mL.kg-1.min-1; P < .01). However, postprandial hyperemia (mean percentage change in SMV blood flow) was less in the asymptomatic (70%; P < .001) and symptomatic patients (29%; P < .01) than in the volunteers. Postprandial SMV flow was similar to SMA flow in the patients but was significantly greater than SMA flow in the volunteers (P < .005).

CONCLUSIONS

Postprandial mesenteric hyperemia is reduced in older patients with MAOD. The role of aging alone has not been determined. Fasting and postprandial flow changes in these patients may predict the onset of chronic mesenteric ischemia.

Endovascular stent-graft repair of thoracic aortic aneurysms

Conventional repair of aneurysms of the descending thoracic aorta entails thoracotomy and graft interposition. For elderly patients and those with previous operations, obesity, respiratory insufficiency, or other comorbidities, such a procedure entails significant mortality and morbidity. Transluminal stent-graft placement offers an alternative approach with potentially less morbidity and quicker recovery; however, the effectiveness and durability of stent-grafts remain uncertain.

METHODS

Since July 1992, thoracic aortic stent-grafts have been placed in 44 patients with a variety of pathologic conditions. Each graft was individually constructed from self- expanding, stainless-steel Z stents covered with a woven Dacron polyester fabric graft. Craft dimensions were determined from spiral computed tomographic scans. All implants were performed in the operating theater under fluoroscopic and transesophageal echocardiographic guidance. Follow-up was by computed tomography and contrast angiography in all cases.

PATIENT DATA

There were 36 men and 8 women. Mean age was 66 years (range 35 to 88 years). Mean aneurysmal diameter was 6.3 cm (range 4.0 to 9.4 cm). Etiologies included 23 degenerative aneurysms, four posttraumatic aneurysms, four pseudoaneurysms, and one chronic aortic dissection.

RESULTS

There were three early deaths (<30 days) and two late deaths. One early death resulted from graft failure. There were two instances of paraparesis or paraplegia, with one associated early death. A single stent was deployed in 27 patients, two stents were required in 14 patients, and three stents were required in three patients. In 23 patients, vascular access was attained through the femoral artery; abdominal aortic access, either native or graft, was necessary in the remaining 21 patients. Twelve grafts were placed in conjunction with open abdominal aortic surgical procedures. Mean follow-up (98% complete) was 12.6 months (range 1 to 34 months). One late death occurred from aneurysmal expansion and rupture in a patient with a persistent midgraft leak. The second late death may have resulted from aneurysmal rupture. Immediate thrombosis was achieved in 36 patients, and late thrombosis was achieved in three others. Failure to achieve complete aneurysmal thrombosis occurred in five patients, however, and one individual (previously noted) died of aneurysmal expansion and rupture; the remaining four are being carefully monitored. Only one patient has required conversion of the stent to an open procedure; a contained rupture of the false lumen of a chronic dissection eventually necessitated total descending thoracic aortic exclusion.

CONCLUSIONS

These early results support the hypothesis that endovascular stent-graft placement may be a safe and durable treatment for selected patients with aneurysmal disease of the descending thoracic aorta. Large introducer size (26F outer diameter) and relatively limited angulation capability, as well as imprecise deployment techniques, currently limit its applicability. Distal embolization and stent migration have not been observed. Failure to achieve complete aneurysmal thrombosis may allow continued aneurysmal expansion and rupture. Further follow-up is clearly necessary to evaluate the true long-term effectiveness of this procedure.

Mechanisms of neurologic deficits and mortality with carotid endarterectomy

OBJECTIVE

To evaluate the incidence and etiology of perioperative complications of carotid endarterectomy.

DESIGN

Retrospective review of carotid endarterectomies performed over 13 years. Risk factors, indications, results of electroencephalographic (EEG) monitoring, and outcomes were evaluated.

SETTING

University medical center.

PATIENTS

Three hundred sixty-seven consecutive primary carotid endarterectomies were performed on 336 patients. Indications for operation included transient ischemic attack (48.5%), asymptomatic stenosis (24%), stroke (17%), nonlateralizing ischemia (9.5%), and stroke-in-evolution (1%).

MAIN OUTCOME MEASURES

Postoperative neurologic deficits (permanent and transient) and deaths were correlated with preoperative symptoms, probable mechanism of the neurologic event, intraoperative EEG changes, and the use of intraoperative shunts.

RESULTS

Four new permanent neurologic deficits (1.1%) and one transient postoperative deficit were noted. Of the five deficits, three were related to undiagnosed intraoperative cerebral ischemia and two were related to perioperative emboli. Three perioperative deaths (0.8%) occurred: two of myocardial infarction and one of an intracerebral hemorrhage from a ruptured arteriovenous malformation. Intraoperative EEG tracings for the most recent consecutive 175 procedures were analyzed. Shunts were used in 45 patients (26%), 38 of whom demonstrated significant EEG changes with carotid clamping.

CONCLUSIONS

Carotid endarterectomy can be performed with a low risk of stroke (1.1%) and death (0.8%). Stroke was due to cerebral ischemia or embolization. With meticulous surgical technique, death is due to myocardial ischemia and not neurologic events.

Hypertension-enhanced monocyte adhesion in experimental atherosclerosis

PURPOSE

Hypertension is a known clinical risk factor for atherosclerosis. In experimental atherosclerosis, monocyte adhesion to the endothelial surface is enhanced and is considered to be an important early stage in plaque formation. We tested the hypothesis that hypertension enhances monocyte adhesion in experimental atherosclerosis.

METHODS

Twenty-two New Zealand White rabbits were fed an atherogenic diet for 3 weeks to induce plaque formation. Aortic coarctation was created in eight rabbits by wrapping a Dacron band around the midportion of the descending thoracic aorta (stenosis group), whereas six rabbits underwent banding without aortic constriction (no stenosis group). Eight rabbits served as nonoperated controls. Monocyte binding to the aortic endothelial surface was counted with epifluorescent microscopy on standard aortic segments proximal and distal to the band. Immunohistochemistry was performed for the following antibodies: VCAM-1, RAM11, CD11b, and factor VIII.

RESULTS

Mean blood pressure was 89 +/- 3 mm Hg in the aorta proximal to the stenosis, compared with 64 +/- 4 mm Hg in the no stenosis group and 74 +/- 3 mm Hg in the control group (p < 0.01). The mean aortic blood pressure gradient across the stenosis was 16 +/- 2 mm Hg in the stenosis group, whereas the aortic blood pressure gradient was 0.2 +/- 0.6 mm Hg in the no stenosis group and -0.3 +/- 0.4 mm Hg in the control group (p < 0.001). Monocyte adhesion to the aortic endothelial surface proximal to the stenosis was increased twofold compared with adhesion to the aorta distal to the stenosis and compared with the proximal aorta in the control group (p < 0.02). The proximal-to-distal aortic ratio of monocyte binding was enhanced in the stenosis group (2.2) compared with the no stenosis (0.76) and control (0.83) groups (p < 0.01). The intima area of the aorta proximal to the stenosis was significantly increased compared with the proximal aortas in the no stenosis and control groups (p < 0.01). RAM11, CD11b, and endothelial VCAM-1 expression were enhanced in the hypertensive region proximal to the stenosis.

CONCLUSIONS

In the hypertensive region in the aorta proximal to the stenosis, monocyte adhesion and endothelial VCAM-1 expression were increased, with intimal thickening and accumulation of macrophages. These findings suggest that hypertension may promote atherosclerotic plaque formation by enhancing monocyte adhesion.

Is completion arteriography mandatory after reversed-vein bypass grafting?

Many surgeons advocate uniform performance of operative completion arteriography after leg bypass surgery to ensure technical success and to optimize short- and intermediate-term graft patency. To determine the impact of this practice on the outcome of reversed-vein bypass surgery and associated patient charges, we reviewed our series of consecutive nonemergent leg bypass procedures. Ninety-three infrainguinal bypass procedures were performed in 80 patients (76 men and 4 women) from September 1991 to August 1994. The patients' average age was 67 years (range, 30 to 92 years). Follow-up (mean, 113.1 months; range, 1 to 36 months) was available on 91 grafts (97%). Indications for surgery included limb salvage in 75 cases, claudication in 12 cases, and popliteal aneurysm exclusion in 6 cases. All patients survived surgery. Primary graft patency rates as determined by life-table analysis were 87%, 81%, 78%, and 78% at 6 months and at 1, 2, and 3 years, respectively. Limb-salvage rates were 95%, 91%, 87% and 87% at the same intervals. Bypass procedures were divided into two groups. The 25 grafts in group 1 were evaluated with inspection, continuous-wave Doppler insonation, and routine completion arteriography. The 68 grafts in group 2 were evaluated by inspection and insonation alone. Fourteen grafts occluded after surgery (average, 5 months; range, 1 to 12 months), five in group 1 and nine in group 2. The likelihood of graft occlusion was similar in both groups (p = 0.42). The optimal method of confirming technical adequacy after bypass surgery in the clinically satisfactory graft remains uncertain. Charges for completion arteriography at our institution average $700, including 15 minutes of additional operative time. In our experience, these charges do not appear to be justified by improved short- or intermediate-term graft patency rates in reversed-vein grafts when completion arteriography is performed.

Carotid endarterectomy: the gold standard

Carotid endarterectomy has been firmly established as the gold standard of therapy for symptomatic and asymptomatic patients with severe carotid stenosis, provided surgical complication rates are within prescribed limits. The procedure-related risk of stroke/death should be < 3% in asymptomatic patients and < 6% in symptomatic patients. New investigational therapies such as balloon angioplasty and stenting for carotid stenosis should be evaluated against the same standard.

Etiology of aortic aneurysm

Inthe past three decades, the prevalence of aortic aneurysms has increased threefold. Incidence of aortic aneurysms increases with age and as the population ages, the prevalence increases. Population-based stud- ies have shown that 10% of men over the age of 70 have abdominal aortic aneurysms. After many years of research, the exact pathogenesis of degenerative aneurysms-the most common form of aneurysm-is still unknown, although a number of factors including genetic, protelytic enzymes, hemodynamics, inflammation, and infection have been implicated.

Phlegmasia complicating prophylactic percutaneous inferior vena caval interruption: a word of caution

PURPOSE

The purpose of this study was to evaluate the incidence of thrombotic complications in patients with deep vein thrombosis (DVT) who were treated with percutaneous inferior vena caval interruption in place of anticoagulation.

METHODS

A retrospective review of all percutaneously placed inferior vena cava filters for 1 year, August 1993 through July 1994, was performed.

RESULTS

Thirty-three percutaneous inferior vena cava filters were placed in 32 patients. The underlying disease was pulmonary embolism in 15 (47%) and DVT in 17 (53%) patients. Of patients with pulmonary embolism, 11 had a documented DVT, and four were not evaluated for DVT. There were 14 men and 18 women, with a mean age of 63.5 years (range 24 to 93 years). Indications for vena caval interruption were recurrent pulmonary embolism with therapeutic anticoagulation (n = 2 [6%]), prophylactic insertion with documented pulmonary embolism and therapeutic anticoagulation (n = 8 [25%]), documented pulmonary embolism and absolute contraindication to anticoagulation (n = 5 [16%]), documented DVT and absolute contraindication to anticoagulation (n = 2 [6%]), prophylactic insertion with documented DVT and therapeutic anticoagulation (n = 5 [16%]), and documented DVT with relative contraindication to anticoagulation (n = 10 [31%]). Of the 32 patients with inferior vena cava filters, 17 were not given anticoagulants (7 absolute contraindications, 10 relative contraindications), and 15 were given anticoagulants. Insertion of a percutaneous inferior vena cava filter in patients who were not given anticoagulants was followed by the development of phlegmasia cerulea dolens in four patients (24%), which was bilateral in two patients; one patient eventually died. No patients treated with inferior vena cava filter and anticoagulation had development of phlegmasia.

CONCLUSIONS

Percutaneous inferior vena caval interruption effectively prevents pulmonary embolism in patients with DVT but does not impact the underlying thrombotic process and in fact may contribute to progressive thrombosis in patients who are not given anticoagulants. Anticoagulation with intravenous heparin in safe and effective therapy for DVT in most patients. We believe that percutaneous insertion of vena cava filters should not replace anticoagulation in routine proximal DVT, and those patients who require an inferior vena cava filter for failure of anticoagulation should continue to receive heparin to treat the primary thrombotic process. We caution that relative contraindications to anticoagulation should be carefully scrutinized before recommending vena cava interruption as a primary therapy for DVT.

MR and spiral/helical CT imaging of lower extremity occlusive disease

Magnetic resonance (MR) angiography and spiral CT angiography are promising new imaging modalities for evaluating patients with lower extremity arterial occlusive disease. Both techniques are less invasive than conventional angiography, and MR angiography has the additional advantages of not requiring iodinated contrast media or ionizing radiation. The basic principles of MR angiography and spiral CT angiography are reviewed with an emphasis on three-dimensional display techniques. This is followed by a discussion of their clinical applicability toward the diagnosis and treatment planning of lower extremity arterial occlusive disease.

Pathobiology of plaque modeling and complication

The artery wall adapts to changes in wall tension and wall shear stress by means of enlargement and changes in both thickness and composition. The intima may participate in these changes, and these compensatory adaptive-reactive modifications continue in the presence of atherogenesis. Further understanding of the interaction of the evolving plaque with the artery wall and the associated effects of the physical forces associated with the circulation should provide new insights into the nature of plaque instability and into the outcome of direct interventions.

An automated three-dimensional particle tracking technique for the study of modeled arterial flow fields

An automated three-dimensional particle tracking technique has been developed to study particle motion in modeled flow fields. A high speed video recording system. Kodak Ektapro 1000, with two cameras arranged relatively orthogonally is used for this technique. The particle tracking data are compared to theoretical Poiseuille flow and to laser Doppler data from an axisymmetric stenosis model. The particle tracking data are in good agreement with both theoretical and laser Doppler data, and at least 79 percent of the particle paths were determined successfully. Fluid dynamic properties derived by this technique are: 3-D particle paths, velocity, and particle residence time.

Fluid wall shear stress measurements in a model of the human abdominal aorta: oscillatory behavior and relationship to atherosclerosis

Clinically significant atherosclerosis in the human aorta is most common in the infrarenal segment. This study was initiated to test the hypothesis that flowfield properties are closely related to the localization of plaques in this segment of the arterial system. Wall shear stress was calculated from magnetic resonance velocity measurements of pulsatile flow in an anatomically accurate model of the human abdominal aorta. The wall shear stress values were compared with intimal thickening from 15 post-mortem aortas measured by quantitative morphometry of histological cross sections obtained at standard locations. Wall shear stress oscillated in direction throughout most of the infrarenal aorta, most prominently in the distal region. The time-averaged mean wall shear stress (-1.7 to 1.4 dyn/cm2) was lowest near the posterior wall in this region. These hemodynamic parameters coincided with the locations of maximal intimal thickening. Statistical correlation between oscillatory shear and intimal thickness yielded r = 0.79, P < 0.00001. Low mean shear stresses correlated nearly as well (r = -0.75, P < 0.00005). Comparison of our data with surface maps of Sudan Red staining and early lesions as reported by others revealed similar conclusions. In contrast, pulse and maximum shear stresses did not correlate with plaque localization as has been shown for other sites of selective involvement by atherosclerosis (r < 0.345). Simulated exercise conditions markedly changed the magnitude and pattern of wall shear stress in the distal abdominal aorta. These results demonstrate that in the infrarenal aorta, regions of low mean and oscillating wall shear stresses are predisposed to the development of plaque while regions of relatively high wall shear stress tend to be spared.

Quantitative morphologic study of intimal thickening at the human carotid bifurcation: I. Axial and circumferential distribution of maximum intimal thickening in asymptomatic, uncomplicated plaques

The spatial distribution of intimal thickening was determined for each of 42 carotid bifurcations removed at autopsy from patients with no clinical or anatomic evidence of cerebrovascular disease. Both right and left specimens were available for six of the individuals. Each bifurcation was removed intact and included a 1.5-2.9-cm length of the common carotid artery and a 1.5-2.5-cm length of the internal carotid artery. The specimens were restored to in situ length, fixed under conditions of controlled-pressure perfusion at 100 mmHg, filled with a radio-opaque mixture, radiographed and sectioned at 0.5-cm intervals. Computer assisted contour tracing of projected images of histologic sections was used to determine intimal thickness, intimal cross sectional area and lumen area within each of eight equal 45 degrees polar sectors with 0 degree indexed at the flow divider, 90 degrees at the outside wall, 180 degrees opposite the flow divider and 270 degrees at the inner side wall. Intima occupied 0.9-42% of the area encompassed by the internal elastic lamina, i.e. the potential lumen area if no intimal thickening were present, but there was no lumen narrowing on lateral X-ray projections. Intimal thickening was eccentric at each level of section but the circumferential location of maximum intimal thickness (MIT) shifted in a continuous helix from level to level. At the common carotid artery level 1.0 cm proximal to the bifurcation, MIT tended to be at the flow-divider side at 15 +/- 59 degrees. Immediately proximal to the flow divider, MIT was at the lateral side wall. In the mid-sinus region of the internal carotid artery MIT was opposite the flow divider at 179 +/- 64 degrees. At the distal internal carotid just beyond the sinus, MIT was at the inner side wall. The distal internal carotid was minimally involved or free of intimal thickening. Comparison of right and left bifurcations revealed that the helical spatial distribution of MIT was in mirror-image symmetry for the two sides. The findings correspond closely with previous demonstrations of a helical flow pattern in the region of the bifurcation. Although locations of MIT just proximal and just distal to the bifurcation are similar and tend to be at the 'far wall', individual differences in the shifts of MIT with axial location should be taken into account when sites of interrogation by non-invasive clinical methods are selected for detection of intimal thickening.

Quantitative morphologic study of intimal thickening at the human carotid bifurcation: II. The compensatory enlargement response and the role of the intima in tensile support

Arteries enlarge where intimal plaques form, tending to preserve lumen cross sectional area but causing an increase in mural tangential tension due to the increase in radius. To characterize the compensatory enlargement process at the carotid bifurcation and to evaluate the possible contribution of intima thickening to mural tensile support during the enlarging process, we assessed the relationships among intimal thickening, artery size and estimated tensile stress at 9 sequential axial levels in 42 human carotid bifurcations obtained during post-mortem examinations of 36 adults with no clinical or anatomical evidence of cerebrovascular disease. Right and left bifurcations were available for 6 patients. The arteries were fixed under conditions of controlled pressure distention and histologic sections were prepared at 0.5 cm axial intervals. We determined vessel radius (r), intima thickness (IT), media thickness (MT), intima area (IA), lumen area (LuA) and the area encompassed by the internal elastic lamina (IELA), i.e. the lumen area if there were no intimal thickening. Although IT, IA and r were greatest in the proximal sinus region, there was a highly significant linear relationship between IA and IELA at each axial level; correlation coefficients ranged from 0.64 to 0.97 with P < 0.001 at each level. Stenosis (IA/IELA x 100) ranged from 10.8 +/- 8.0% at the common carotid level immediately proximal to the bifurcation angle to 22.3 +/- 17.9% at the level immediately distal to the angle, but LuA remained nearly constant at each level regardless of IA.(ABSTRACT TRUNCATED AT 250 WORDS)

Hemodynamic stress and experimental aortoiliac atherosclerosis

PURPOSE

Human aortic atherosclerosis is predominantly localized to the infrarenal aorta where flow is bidirectional and wall shear stress oscillates. Similar flow patterns have been related to carotid atherosclerosis. The thoracic aorta is usually spared, where flow and shear stress are unidirectional. We hypothesized that because heart rate and systemic blood pressure modulate flow velocity and shear stress oscillation, both these hemodynamic forces may enhance aortoiliac atherogenesis.

METHODS

Eighteen male cynomolgus monkeys were fed an atherogenic diet for 6 months (mean serum cholesterol = 535 +/- 35 mg/dl). Heart rate was determined with 24-hour electrocardiographic telemetry at monthly intervals and blood pressure was measured by direct arterial cannulation. The product of mean heart rate and mean blood pressure was used to define hemodynamic stress for each animal. Atherosclerotic lesion formation at three standard thoracic aortic sites was quantitatively compared with lesion formation at five standard infrarenal aortoiliac locations with computer-assisted morphometry.

RESULTS

There was significantly more plaque in the aortoiliac segment than in the thoracic aorta (12.4% +/- 9.0% vs. 6.4% +/- 4.5% area stenosis, p = 0.02). No correlation was found between the degree of serum lipid elevations and lesion formation in either aortic location. Mean heart rate was 113 +/- 18 beats/min (87 to 163 beats/min) and mean blood pressure was 85 +/- 19 mm/Hg (62 to 130 mm Hg). Heart rate and blood pressure alone were not significantly related to lesion formation. A significant correlation was, however, found between hemodynamic stress and maximum lesion thickness (r = 0.47, p < 0.05) in the aortoiliac region but not in the thoracic aorta (r = 0.19, p > 0.10).

CONCLUSIONS

This study demonstrates that heart rate and blood pressure exert a mutually potentiating effect on aortoiliac atherosclerosis but not on thoracic aortic atherosclerosis. Regional differences in aortic atherosclerosis may therefore be attributable to the interaction between these hemodynamic forces and the local flow patterns specific to each aortic location. Additional investigation of these hemodynamic factors in relation to human aortic atherosclerosis is warranted.

Shear stress at a compliant model of the human carotid bifurcation

To investigate the role of a compliant wall to the near wall hemodynamic flowfield, two models of the carotid bifurcation were constructed. Both were of identical internal geometries, however, one was made of compliant material which produced approximately the same degree of wall motion as that occurring in vivo while the other one was rigid. The inner geometries were formed from the same mold so that the configurations are directly comparable. Each model was placed in a pulsatile flow system that produced a physiologic flow waveform. Velocity was measured with a single component Laser system and wall shear rate was estimated from near wall data. Wall motion in the compliant model was measured by a wall motion transducer and the maximum diameter change varied between 4-7 percent in the model with the greatest change at the axis intersection. The mean shear stress in the compliant model was observed to be smaller by about 30 percent at most locations. The variation in peak shear stress was greater and occasionally reached as much as 100 percent with the compliant model consistently having smaller positive and negative peaks. The separation point was seen to move further upstream in the compliant cast. The modified flowfield in the presence of a compliant wall can then be important in the hemodynamic theory of atherogenesis.

The role of fluid mechanics in the localization and detection of atherosclerosis

Fluid dynamics research over the past twenty years has contributed immensely to our knowledge of atherosclerosis. The ability to detect localized atherosclerotic plaques using noninvasive ultrasonic methods was advanced significantly by investigations into the nature and occurrence of velocity disturbances created by arterial stenoses, and diagnosis of carotid bifurcation disease using a combination of ultrasonic imaging and Doppler measurement of blood velocity is now quite routine. Since atherosclerotic plaques tend to be localized at sites of branching and artery curvature and since these locations would be expected to harbor complex flow patterns, investigators postulated that fluid dynamics might play an initiating role in atherogenesis. Several fluid dynamic variables were proposed as initiating factors. Investigations were undertaken during the 1980s in which fluid dynamic model experiments with physiologic geometries and flow conditions were employed to simulate arterial flows and in which morphometric mapping of intimal thickness was performed in human arteries. Correlations between fluid dynamic variables and intimal thickness revealed that atherosclerotic plaques tended to occur at sites of low and oscillating wall shear stress; and these observations were reinforced by studies in a monkey model of atherosclerosis. Concomitantly, it was realized that arteries adapt their diameters so as to maintain wall shear stress in a narrow range of values around 15 dynes/cm2, findings which were based both on observations of normal arteries and on animal studies in which flow rates were manipulated and arterial diameter adaptation was measured.(ABSTRACT TRUNCATED AT 250 WORDS)

Effect of coarctation on matrix content of experimental aortic atherosclerosis: relation to location, plaque size and blood pressure

Cynomolgus monkeys were fed an atherogenic diet for 6 months following surgically produced high-grade (n = 10) or mild (n = 16) mid-thoracic aortic coarctation. A diet-control (DC) group (n = 13) was fed the diet without coarctation. High-grade coarctation (HGC) resulted in 74.1% +/- 8.3% stenosis by aortography prior to sacrifice and was associated with systolic brachial blood pressures of 143.3 +/- 26.0 mmHg and gradients across the stenoses of 36.8 +/- 23.6 mmHg. Mild coarctation (MC) resulted in stenoses of 50.9% +/- 12.9%, brachial systolic pressures of 119.4 +/- 25.7 and gradients of 12.5 +/- 15.2 mm Hg (P < 0.01, P = 0.03 and P < 0.005, respectively, compared with HGC). When total plaque cross-sectional area exceeded 0.8 mm2, the entire arterial circumference was usually involved. HGC resulted in complete sparing or minimal plaque formation in sections distal to the stenoses compared with proximal sections (P < 0.001). There were no significant differences between MC and DC animals in plaque location or size. Matrix content increased with plaque area regardless of degree of stenosis or sampling level (P < 0.01), but lesions with more than 75% matrix content were more numerous in distal than in proximal sections despite their smaller size. The number of plaques with greater than 75% matrix content was increased proximal to HGC (P < 0.04). Thus, distal location and plaque size were independent determinants of plaque matrix content and matrix content was increased proximal to HGC regardless of plaque size. Attempts to evaluate effects of various regimens and interventions on plaque composition need to take location and plaque size, as well as blood pressure differences, into account.

Use of CT for diagnosis of traumatic rupture of the thoracic aorta

CT imaging of traumatic aortic rupture has been both advocated and disparaged in the current literature as a reliable diagnostic modality. In a retrospective review of blunt chest trauma patients at our institution evaluated by both thoracic CT and arteriography, we found a 17% false negative rate and a 39% false positive rate. Although we feel CT is not sufficiently sensitive at present to evaluate traumatic rupture of the aorta directly, it is an invaluable adjunctive imaging modality for stable blunt chest trauma patients with equivocal chest radiographs or arteriograms.

Hemodynamic patterns in two models of end-to-side vascular graft anastomoses: effects of pulsatility, flow division, Reynolds number, and hood length

Flow behavior in models of end-to-side vascular graft anastomoses was studied under steady and pulsatile flow conditions. Models were constructed to simulate geometries employed in experimental studies on intimal thickening in a canine model. Reynolds numbers, division of flow in the outflow tracts and the pulsatile waveform employed were taken from measurements obtained in the canine model. Flows in the scaled-up, transparent models were visualized with white, neutrally buoyant particles which were photographed under laser illumination and also recorded on video tape under bright incandescent light. Strong, three-dimensional helical patterns which formed in the anastomotic junction were prominent features of the flow fields. Regions of low wall shear, oscillatory wall shear and long particle residence time were identified from the flow visualization experiments. Comparisons with the limited qualitative data available on intimal thickening in vascular graft anastomoses suggest a relation between localization of vascular intimal thickening and those surfaces experiencing low shear and long particle residence time.

Experimental atherosclerosis at the carotid bifurcation of the cynomolgus monkey. Localization, compensatory enlargement, and the sparing effect of lowered heart rate

We have characterized plaque localization, the extent of compensatory artery enlargement, and the effect of heart rate in experimental atherosclerosis at the carotid bifurcation of the cynomolgus monkey. We altered heart rate by sino-atrial node ablation (SNA) and then fed the animals an atherogenic diet for 6 months. Heart rate was measured at four time points by 24-hour telemetry. Of nine animals with SNA, heart rate was reduced significantly in six (from 148 +/- 11 to 103 +/- 20 beats/min, p < 0.001) and was unchanged in three. Sham-operated monkeys had no significant change in heart rate. On the basis of comparison with the preoperative mean for all 17 animals (136 +/- 22 beats/min), animals were separated into a low-heart-rate (LHR) group (111 +/- 16 beats/min, n = 12) and a high-heart-rate (HHR) group (150 +/- 16 beats/min, n = 5). Blood pressure, serum cholesterol level, and body weight did not differ for the two groups. As in the human, plaques formed predominantly in the proximal portion of the internal carotid artery at the lateral wall opposite the flow divider. Plaque cross-sectional area increased progressively from the relatively uninvolved, adjacent common carotid artery to the mid-sinus region of the internal carotid artery and decreased from the mid-sinus region to the internal carotid artery beyond the sinus. Plaque distribution was the same for the LHR and HHR groups, but lesion area and percent stenosis were greater for the HHR group than for the LHR animals (2.01 +/- 1.19 compared with 0.76 +/- 0.42 mm2 for lesion area [p < 0.02] and 30.7 +/- 4.4% compared with 15.2 +/- 7.3% for stenosis [p < 0.002]).(ABSTRACT TRUNCATED AT 250 WORDS)

Micro-architecture and composition of artery walls: relationship to location, diameter and the distribution of mechanical stress

PURPOSE

We reviewed the structural basis of the mechanical properties of the arterial wall, in order to establish a coherent micro-anatomical basis for the differences in compliance among different arteries and a framework for assessing changes in the mechanical properties of specific individual arteries in relation to changing physical stresses.

DATA IDENTIFICATION

The data and concepts presented here were derived from both earlier and ongoing work. Features that assure stability and integrity in relation to blood flow (wall shear stress) and pressure (mural tensile stress) were examined. Particular attention was paid to the morphogenetic and biosynthetic means by which arteries adapt to normal or abnormal modifications of these forces, particularly in relation to growth, location in the arterial tree and geometric configuration.

RESULTS AND CONCLUSIONS

Thickness, composition and architecture of the artery wall, including thickness and composition of the intima, are normally determined by the stresses imposed by pressure and flow. Vessel radius is closely associated with flow, so that a normal baseline level of mean shear stress of about 15 dyn/cm2 is maintained or restored. Wall thickness and composition are determined by wall tension in relation to pressure and radius. Baseline levels of tensile stress differ with location but appear to be similar for homologous vessels. Changes in flow that modify the radius also modify wall tension. Changes in wall thickness and composition are likely to cause changes in compliance, due to altered flow and/or pressure patterns; these changes in compliance may be adaptive rather than destructive. Changes in the compliance of specific arteries over time may be used to evaluate the progression and severity of the conditions underlying these changes.

Pulsatile flow visualization in the abdominal aorta under differing physiologic conditions: implications for increased susceptibility to atherosclerosis

The infrarenal abdominal aorta is a common site for clinically significant atherosclerosis. As has been shown in other susceptible locations, vessel geometry, flow division rates, and pulsatility may result in hemodynamic conditions which influence the preferential localization of disease in the abdominal aorta segment. Pulsatile flow visualization was performed in a glass model of the aorta constructed from measurements of angiograms and cadaver aortas. Flow rates and pulsatile waveforms were varied to reflect typical physiological conditions. Under normal resting conditions, the flow patterns in the infrarenal aorta were more complex than those in the suprarenal location. Time varying vortex patterns appeared at the level of the renal arteries and propagated through the infrarenal aorta into the common iliac arteries. A region of oscillating velocity direction extended from the renal arteries to the aortic bifurcation along the posterior wall. Dye became trapped along the posterior wall, requiring several cardiac cycles for clearance. In contrast, there was rapid clearance of the dye in the anterior aorta. Under postprandial conditions, the flow patterns in the aorta were basically unchanged. Simulated exercise conditions created laminar hemodynamic features very different from the resting conditions, including a decrease in dye residence time. This study reveals significant time-dependent variations in the hemodynamics of the abdominal aorta under differing physiologic conditions. Hemodynamic factors such as low wall shear stress, oscillating shear direction, and high particle residence time may be related to the clinically seen preferential plaque localization in the infrarenal aorta.

Anastomotic intimal hyperplasia: mechanical injury or flow induced

All anastomotic intimal thickening may not be the same, and the underlying mechanism(s) regulating the different types may vary. We investigated the localization of experimental anastomotic intimal thickening in relation to known biomechanical and hemodynamic factors. Bilateral iliofemoral saphenous vein and polytetrafluoroethylene grafts were implanted in 13 mongrel dogs. The distal end-to-side anastomotic geometry was standardized, and the flow parameters were measured. After 8 weeks, seven of 10 animals (group I) with patent grafts were killed and the anastomoses fixed by perfusion. Histologic sections from each anastomosis were studied with light microscopy, and regions of intimal thickening were identified and quantitated with use of oculomicrometry. To characterize the anastomotic flow patterns, transparent silicone models were constructed from castings of the distal anastomosis of three animals (group II), and flow was visualized with use of helium-neon laser-illuminated particles under conditions simulating the in vivo pulsatile flow parameters. Histologic sections revealed two separate and distinct regions of anastomotic intimal thickening. The first, suture line intimal thickening, was greater in polytetrafluoroethylene anastomoses (0.35 +/- 0.23 microns) than in vein anastomoses (0.15 +/- 0.03 microns, p less than 0.05). The second distinct type of intimal thickening developed on the arterial floor and was the same in polytetrafluoroethylene (0.11 +/- 0.11 microns) and vein anastomoses (0.12 +/- 0.03 microns). Model flow visualization studies revealed a flow stagnation point along the arterial floor resulting in a region of low and oscillating shear where the second type of intimal thickening developed. High shear and short particle residence time were observed along the hood of the graft, an area devoid of intimal thickening.(ABSTRACT TRUNCATED AT 250 WORDS)

Cryopreserved saphenous vein allogenic homografts: an alternative conduit in lower extremity arterial reconstruction in infected fields

Autologous saphenous veins are considered the best arterial substitute for lower extremity revascularization in infected fields. The search continues for a vascular conduit in instances when an autologous biologic grafting is not feasible. Herein we report our experience with eight patients in whom cryopreserved saphenous vein allogenic homografts were used in 10 lower extremity arterial reconstructions for limb salvage with coexisting infection. Six patients with eight prosthetic grafts including four femoropopliteal, two femorotibial, a femorofemoral, and a femoroperoneal graft required complete or partial graft excision as a result of overt infection. The two remaining patients included one with an infected femoral pseudoaneurysm and another with extensive chemical burns. All cryopreserved saphenous vein allogenic homografts were of identical match to the ABO/Rh blood groupings of the recipient patients. No immunosuppressive drugs were administered after operation. Mean follow-up was 9.5 months (range, 6.0 to 14.0 months). One patient died 5 weeks after operation with a patent graft. Two grafts occluded during follow-up; in one graft, patency was restored with thrombectomy alone. The remaining seven arterial reconstructions continue to be patent with no evidence of aneurysmal dilation with complete eradication of the primary infection. These preliminary findings suggest that cryopreserved saphenous vein allogenic homografts can serve as interim conduits for lower extremity arterial reconstruction to preserve limb viability when autogenous conduits are unsatisfactory or unavailable. Further definitive reconstruction may thereafter be necessary once sepsis is eradicated and sufficient wound healing is achieved.

Aneurysmal enlargement of the aorta during regression of experimental atherosclerosis

We explored the relationship between regression of diet-induced atherosclerosis and aneurysmal enlargement of the aorta in cynomolgus monkeys. Atherosclerotic plaques were induced in 17 monkeys by feeding them a diet containing 2% cholesterol and 25% peanut oil for 6 months (group I, n = 6; group III, n = 6) or 12 months (group II, n = 5). Regression was induced in group III by feeding a regression diet consisting of 0.25% cholesterol and 15% corn oil in a standard chow diet, for 6 months after the 6-month induction period. Serum cholesterol was 788 +/- 80 mg/dl after 6 months of induction, 508 +/- 53 mg/dl after the 12-month induction period, and 198 +/- 15 mg/dl in the regression group at 12 months. Aortas were fixed in situ under conditions of controlled pressure perfusion, and transverse sections of the unopened vessels were taken at standard levels in the midthoracic and abdominal aortic segments. The area encompassed by the internal elastic lamina was taken as a measure of artery size. Plaques were abundant in abdominal and thoracic sections after the 6- and 12-month induction periods, and no significant difference was observed in lumen area or artery size between the groups. The ratio of abdominal to thoracic aortic plaque area was markedly reduced in the regression group (0.3 +/- 0.2 for regression compared with 0.6 +/- 0.3 for 6-month induction and 1.3 +/- 0.2 for 12-month induction animals; p less than 0.05 for both). A twofold increase was observed in abdominal aortic lumen area in the regression group (10.0 +/- 1.5 mm2 for regression compared with 5.6 +/- 0.7 mm2 for the 6-month and 4.2 +/- 0.7 mm2 for the 12-month induction groups; p less than 0.05 for both) as well as a twofold increase in internal elastic lamina area (10.5 +/- 1.5 mm2 compared with 6.0 +/- 0.7 mm2 for the 6-month and 5.9 +/- 0.8 mm2 for the 12-month induction group; p less than 0.05 for both). Aortic enlargement in the regression group was accompanied by a reduction in media thickness in the abdominal aorta. No significant vessel enlargement or alteration in media thickness occurred in the thoracic aorta. One of six regression animals (17%) had a threefold enlargement of the abdominal aorta and was thought to have a manifest aneurysm.(ABSTRACT TRUNCATED AT 400 WORDS)

The effect of hypercholesterolemia on early atherosclerotic lesions initiated by fibrinopeptide B

Hypercholesterolemia and thrombosis have been implicated as factors in the development of atherosclerosis. Fibrinopeptide B (FPB) is a short chain peptide cleaved from fibrinogen during the production of fibrin. FPB is a known chemoattractant and has been shown to produce experimental atherosclerotic lesions in association with hypercholesterolemia. The present study was designed to examine the role of hypercholesterolemia in this process and to study the time course of the development of these lesions. Twelve New Zealand White rabbits were placed on an atherogenic diet and had suture carrying either FPB, fibrinopeptide A (FPA), or saline (controls) implanted in the adventitia of the femoral arteries and were sacrificed at 14 days. An equal number of animals were left on a standard diet and underwent similar treatment. Eleven animals were treated as the hypercholesterolemic group but were sacrificed at 2, 4, and 7 days. The thickness of the intima was measured adjacent to the suture in the animals sacrificed at 14 days, and the hypercholesterolemic FPB sites were thicker (12.23 mu +/- 6.60) than either hypercholesterolemic FPA (6.06 mu +/- 3.72), saline (4.94 mu +/- 1.42), or the normocholesterolemic FPB (5.99 mu +/- 4.61), FPA (3.89 mu +/- 2.20), or saline (3.97 mu +/- 1.83) (P less than 0.05 for all groups). Transmission electron microscopy of the hypercholesterolemic FPB group showed evidence of macrophages, actively secreting smooth muscle cells with newly deposited elastin, and foam cells by 7 days. We conclude that FPB attracts or stimulates macrophages and smooth muscle cells and that the resultant cellular and extracellular proliferation favors early atherosclerotic lesion formation in the presence of hypercholesterolemia.

Postoperative vasospasm after antegrade mesenteric revascularization: a report of three cases

Vascular reconstruction for chronic intestinal ischemia can be accomplished by endarterectomy or aortomesenteric bypass. In our practice, antegrade bypasses from the supraceliac aorta to the celiac axis and superior mesenteric artery are currently the most frequently used techniques. Such reconstructions often use multiple or bifurcated large diameter vascular prostheses and have demonstrated excellent long-term patency. Despite these salutory results, we have noted an unusual perioperative response in three of these patients, which is the subject of this report. All three patients underwent uncomplicated elective mesenteric revascularization with grafts (diameter greater than or equal to 6 mm) originating in the supraceliac aorta. Indications for operation included (1) history of postprandial pain, (2) documentation of weight loss, and (3) angiographic evidence of advanced atherosclerotic disease with appropriate collateral development. Episodes of abdominal pain occurred 5 to 20 days after operation when normal food intake was reinstituted. In two patients immediate angiograms revealed patent grafts with diffuse mesenteric vasospasm. Treatment with intravenous hyperalimentation and nifedipine for 10 days resulted in complete resolution of symptoms. In the third patient, symptoms were totally relieved by temporary reduction in oral intake and administration of nifedipine. A later angiogram revealed a patent graft. All patients have remained asymptomatic and regained normal weight. This pattern of postrevascularization pain has not been seen in our patients undergoing revascularization with small (i.e., venous) conduits originating in the infrarenal aorta. The cause appears to be a heightened myogenic response of a "protected" vascular bed when suddenly exposed to the high perfusion pressure and blood flow of large caliber antegrade conduits. Prophylaxis with calcium channel blockers and use of smaller diameter grafts (5 mm) may avoid this disturbing syndrome.

Hypertension sustains plaque progression despite reduction of hypercholesterolemia

To assess the effect of hypertension on diet-induced coronary artery plaques after a return to a nonatherogenic diet, 10 cynomolgus monkeys were fed an induction regimen containing 2% cholesterol and 25% peanut oil for 6 months and then were subjected to midthoracic aortic coarctation to induce hypertension. The animals were then fed a nonatherogenic "prudent" ration for 6 additional months (hypertension-regression group). Twelve additional monkeys were fed the atherogenic diet for 6 months; six were killed (lesion-induction control group) and six were changed to the prudent diet for 6 additional months without coarctation (normotension-regression control group). At the end of the induction period, cholesterol levels averaged 744 +/- 178 mg/dl for the 22 animals and were similar for the three groups throughout the induction period. For the animals restored to the nonatherogenic diet (hypertension-regression and normotension-regression groups), serum cholesterol levels fell to 486 +/- 252 mg/dl at 1 month, to 341 +/- 162 mg/dl at 2 months, and to 234 +/- 78 mg/dl at 6 months. There was no significant difference between the hypertensive and normotensive animals. Six months after coarctation, blood pressure proximal to the coarctations for the hypertension-regression group ranged from 100/60 to 220/145 mm Hg with a mean of 166/103 +/- 36/28 mm Hg. Cross-sectional area of coronary plaques was somewhat lower for the normotension-regression control group compared with the lesion-induction control group, but the difference was not significant. Plaque area was, however, markedly greater in the hypertension-regression group than in either the lesion-induction or the normotension-regression groups (p less than 0.05 for each) despite progressive reduction in hyperlipidemia.(ABSTRACT TRUNCATED AT 250 WORDS)

True aneurysmal disease in the hand and upper extremity

Ten patients with true aneurysmal disease of the hand and forearm vessels were treated at our institution between 1981 and 1990. Pseudoaneurysms resulting from penetrating trauma or mycotic aneurysms were specifically excluded. Eight patients were male, two were female; mean patient age was 38 years (range 26 to 72 years). A history of repetitive occupational or recreational trauma was elicited in five patients. All patients presented with painful masses or neurologic symptoms due to nerve compression. Ischemic changes were evident in five patients due to thrombosis or distal embolization. Arteriography and transcutaneous Doppler ultrasound aided in documentation of flow characteristics and planning for operative intervention. Three patients underwent excision and ligation once collateral flow was demonstrated to be adequate and reconstruction was not felt to be feasible. Seven patients underwent resection with vein graft reconstruction. Immediate postoperative and interval patency rates were 100%. No digital amputations were required even in those patients presenting with severe distal ischemia.