Michael A. Golden

Bone Formation in Carotid Plaques A Clinicopathological Study

Background and Purpose— Bone formation and dystrophic calcification are present in carotid endarterectomy plaques. The clinical significance of these findings is unknown. The purpose of this study was to determine whether bone formation and extensive dystrophic calcification are associated with stable plaques and protective against ischemic vascular events. Methods— Carotid endarterectomy plaques were collected from 142 patients (94 men) with carotid stenosis. The specimens were evaluated for lamellar bone formation, dystrophic calcifications, inflammatory infiltrates, neovascularization, and histological type or grade of plaque according to a standard AHA grading system. Immunohistochemical staining was performed to identify vascular endothelial cells in neovascularization (factor VIII) and lymphocytes. Clinical data, including history of cerebrovascular and cardiovascular events, were recorded at the time of surgery. Results— Patients with calcification of carotid plaques had fewer symptoms of stroke and transient ischemic attack (P =0.042) than those without calcification. Stroke and transient ischemic attack occurred less frequently in patients with plaques with large calcific granules (P =0.021). Of the patients, 13% had lamellar bone formation, which directly correlated with the presence of sheetlike calcifications (P =0.0001) and inversely correlated with ulcerated lesions (P =0.048). The presence of bone also correlated with diabetes (P <0.01) and coronary artery disease (P <0.01). Of the 20 patients with bone, 6 had a history of stoke and transient ischemic attack (P =0.5). Conclusions— The results indicate that bone formation tends to occur in heavily calcified carotid lesions devoid of ulceration and hemorrhage. Patients with extensive calcification of the carotid plaques are less likely to have symptomatic disease.

Selective evaluation and management of coronary artery disease in patients undergoing repair of abdominal aortic aneurysms. A 16-year experience.

Reduction of cardiac mortality associated with abdominal aortic aneurysm (AAA) repair remains an important goal. Five hundred consecutive urgent or elective operations for infrarenal nonruptured AAA were reviewed. Patients were divided into three groups based on preoperative cardiac status: group I (n = 260, 52%), no clinical or electrocardiographic (ECG) evidence of coronary artery disease (CAD); group II (n = 212, 42.2%), clinical or ECG evidence of CAD considered stable after further evaluation with studies such as dipyridamole-thallium scanning, echocardiography, or coronary arteriography; group III (n = 28, 5.6%), clinical or ECG evidence of CAD considered unstable after further evaluation. Group I had no further cardiac evaluation and groups I and II underwent AAA repair without invasive treatment of CAD. Group III underwent repair of cardiac disease before (n = 21) or coincident with (n = 7) AAA repair. In all instances, perioperative fluid volume management was based on left ventricular performance curves constructed before operation. The 30-day operative mortality rate for AAA repair in all 500 patients was 1.6% (n = 8). There was one (0.4%) cardiac-related operative death in group I, which was significantly less than the five (2.4%) in group II (p less than 0.02). Total mortality for the two groups were also significantly different, with one group I death (0.4%) and seven group II deaths (3.3%), (p less than 0.02). These data support the conclusions that (1) the leading cause of perioperative mortality in AAA repair is myocardial infarction, (2) correction of severe or unstable CAD before or coincident with AAA repair is effective in preventing operative mortality, (3) patients with known CAD should be investigated more thoroughly to identify those likely to develop perioperative myocardial ischemia so that their CAD can be corrected before AAA repair, and (4) patients with no clinical or ECG evidence of CAD rarely die of perioperative myocardial infarction, and thus selective evaluation of CAD based on clinical grounds in AAA patients is justified.

Platelet factor 4 localization in carotid atherosclerotic plaques: correlation with clinical parameters

Emerging evidence supports a role for platelets in the progression of atherosclerosis in addition to an involvement in thrombotic vascular occlusion. Platelet Factor 4 (PF4), a chemokine released by activated platelets, stimulates several pro-atherogenic processes. Therefore, we examined the localization of PF4 and the homologous protein, Neutrophil Activating Protein-2 (NAP-2) in lesions representing the evolution of human atherosclerotic plaques. Carotid plaques from 132 patients with critical carotid stenosis and 6 autopsy specimens were studied. Clinical, histologic and immunohistochemical data were analyzed using a chi(2)-test. PF4 was detected in the cytoplasm of luminal and neovascular endothelium, in macrophages and in regions of plaque calcification. The presence of PF4 in macrophages and neovascular endothelium correlated with lesion grade (p = 0.004; p = 0.044). Staining of macrophages for PF4 correlated with the presence of symptomatic atherosclerotic disease (p = 0.028). In early lesions, PF4 was commonly found in macrophages of early lesions (Grade I/II), whereas NAP-2 was rarely present. In conclusion, correlation between PF4 deposition, lesion severity and symptomatic atherosclerosis suggests that persistent platelet activation may contribute to the evolution of atherosclerotic vascular lesions. These studies support the rationale for the chronic use of anti-platelet therapy in patients at risk for developing symptomatic atherosclerosis.

Magnetic resonance angiography of peripheral runoff vessels

Recent improvements in magnetic resonance imaging techniques have made magnetic resonance angiography (MRA) a very useful adjunct to invasive angiography. Fifty-five limbs in 51 patients with occlusive peripheral vascular disease were studied with both MRA and contrast arteriography. The magnetic resonance and contrast arteriograms were read by radiologists and surgeons and separate interventional plans were based on each study. The MRA findings differed significantly from those of conventional arteriography in 26 limbs (48%). In every case MRA visualized all of the same vessels and hemodynamic stenoses seen on the contrast arteriogram. In 48% of the cases, however, MRA revealed additional findings. Thus the discrepancies in the two studies were always the result of the failure of the arteriogram to reveal all of the patent vessels seen on MRA. The additional information provided by MRA resulted in alteration of the interventional plan in 11 cases (22%). In nine cases (18%) target vessels suitable for use in a limb-salvage procedure were identified by MRA, although they had been missed by conventional arteriography. In all of these cases, intraoperative arteriograms confirmed the suitability of these vessels for use in technically successful bypass procedures. In two cases (4%) additional information provided by MRA identified a target runoff vessel for bypass grafting that proved to be a better alternative than the one that would have been chosen on the basis of contrast arteriography.(ABSTRACT TRUNCATED AT 250 WORDS)

Perigraft air, fever, and leukocytosis after endovascular repair of abdominal aortic aneurysms.

BACKGROUND The postimplantation syndrome of fever and leukocytosis after endovascular repair of infrarenal aortic aneurysms has not been previously characterized and its etiology is not known. METHODS We studied the first 12 patients who underwent successful endovascular repair of infrarenal aortic aneurysms with Dacron-covered stent-grafts, as part of an ongoing phase II clinical trial. Sepsis syndrome evaluations (physical examination, urinalysis, chest radiograph, urine cultures, and blood cultures) were performed for all patients with postoperative temperature (T) greater than 101.4 degrees F. Computed tomography scans of the abdomen were performed, as part of the clinical protocol, on postoperative days 2 and 30. RESULTS Fever (T > 101.4 degrees F) was seen in 8 of 12 (67%) patients (P < 05). An additional 2 of 12 (17%) patients had low-grade fevers (100.3 degrees F, 100.6 degrees F). Only 2 of 12 (17%) patients remained afebrile postoperatively. Leukocytosis with counts over 11,000 white blood cells (WBC)/dL was observed in 7 of 12 (58%) patients (P < 05). Sepsis evaluations failed to identify any source of infection in 11 of 12 (97%) patients. Computed tomography scan evidence of perigraft air was noted in 8 of 12 (67%) patients. All patients were afebrile, had normal white blood cell counts, and were discharged within 1 week postoperatively. There has been no evidence of graft infection after 1 to 6 months of follow-up. CONCLUSIONS Fever and leukocytosis after stent-graft repair of aortic aneurysms does not represent evidence of systemic or graft infection and is not clearly related to nonspecific causes of postoperative fever and leukocytosis. Moreover, the finding of early postoperative perigraft air is not necessarily an indication of graft infection even when concurrently present with fever and leukocytosis.

Evolving experience with thoracoabdominal aortic aneurysm repair at a single institution

Fifty-seven patients underwent repair of atherosclerotic thoracoabdominal aortic aneurysms between 1978 and 1990. Five patients had urgent surgery for rupture. The 30-day operative mortality rate for the entire group was 18% (10 patients). Before July 1987, 19 patients (group 1) were operated on by use of a technique previously described. In these earlier patients the peritoneum was routinely entered, the diaphragm was divided radially, and no heparin was given. Among patients in group 1 there was a 30-day operative mortality rate of 42% (8 patients), and morbidity included myocardial infarction 4 (21%), respiratory failure 9 (47%), renal failure 12 (63%), bleeding requiring reoperation 4 (21%), and intestinal ischemia 3 (16%). Since July 1987 a standardized approach to all elective thoracoabdominal aortic aneurysms has been used in 38 patients (group 2). This method uses a left thoracoabdominal incision, circumferential division of the hemidiaphragm, retronephric totally extraperitoneal aortic exposure, single lung anesthesia, full heparinization, the graft inclusion technique, and liberal use of visceral endarterectomy. Patients in group 2 sustained a 30-day operative mortality rate of 5% (2 patients) and morbidity included myocardial infarction 2 (5%), respiratory failure 10 (26%), renal failure 11 (29%), bleeding requiring reoperation 1 (3%), paraplegia 6 (16%), and paraparesis 4 (11%). Modern surgery for repair of thoracoabdominal aortic aneurysm results in acceptably low operative mortality rates. Spinal cord ischemia remains an unresolved source of morbidity.

Antisense basic fibroblast growth factor gene transfer reduces neointimal thickening after arterial injury

PURPOSE To determine whether synthesis of endogenous basic fibroblast growth factor (bFGF) after arterial injury is critical to the intimal thickening response, intraluminal adenoviral gene transfer of an antisense bFGF (Ad.ASbFGF) transgene was used to inhibit the subsequent synthesis of bFGF protein after injury. METHODS Sprague-Dawley rats underwent balloon catheter carotid artery injury and in vivo gene transfer. Isolated segments of rat common carotid artery were infected with an adenoviral vector encoding an antisense bFGF transcript at concentrations of 2 x 10(9), 1 x 10(10), or 1 x 10(11) pfu/ml. Control rats were treated with either a control adenovirus encoding the beta-galactosidase gene, (Ad.lacZ), at 1 x 10(10), or 1 x 10(11) pfu/ml, or phosphate-buffered saline solution (vehicle). Two weeks after injury the rats were killed and perfusion-fixed. Cross-sectional areas of the carotid arterial intima and media were measured by planimetry, and the intima/media ratio (I/M) was calculated for each vessel. RESULTS The mean I/M for each Ad.ASbFGF group and controls were compared and the significance assessed by analysis of variance. At two weeks after injury, the highest dose of Ad.ASbFGF, 1 x 10(11) pfu/ml, resulted in a near total inhibition of thickening (I/M = 0.14 +/- 0.04, mean +/- SEM) when compared with phosphate-buffered saline solution alone (I/M = 0.99 +/- 0.07), or Ad.lacZ 1 x 10(10) pfu/ml (I/M = 1.01 +/- 0.10) control treatments (p < 0.01). A tenfold lower dose of Ad.ASbFGF, 1 x 10(10) pfu/ml, also caused significant reduction in intimal thickening (I/M = 0.39 +/- 0.07, p < 0.01). Treatment with 2 x 10(9) pfu/ml Ad.ASbFGF did not significantly limit intimal thickening (I/M = 0.72 +/- 0.12). CONCLUSIONS Inhibition of bFGF synthesis in vivo using an antisense RNA strategy significantly inhibits intimal thickening after arterial balloon injury. This study suggests that continued bFGF synthesis contributes to intimal thickening after arterial injury, and that antisense bFGF may represent an effective strategy in limiting restenosis after angioplasty.

Barriers to Endovascular Aortic Aneurysm Repair: Past Experience and Implications for Future Device Development:

Despite improvements in endovascular aortic aneurysm repair (EVAR) devices and techniques, significant anatomic constraints still preclude successful EVAR in a large number of patients. The authors sought to identify the current barriers to EVAR and examine their evolution over time. Patients were evaluated for potential endovascular repair by computed tomography angiography (CTA) with or without supplemental conventional arteriograms. The patient population was separated into 2 groups (A and B) based on early and late time periods in the experience with EVAR, corresponding to the availability of various devices. Group A (early) consisted of the Guidant Ancure, Medtronic Talent, and AneuRx devices and comprised patients presenting between April 1997 through June 2000. Group B (late) consisted of the Medtronic AneuRx, Cook Zenith, Edwards Lifepath, Gore Excluder, and Endologix PowerLink devices and comprised patients presenting between July 2000 and December 2003. Patient demographics and anatomic reasons for rejection were recorded in a database for statistical analysis. In total, 547 patients were evaluated (463 men, 84 women). Of these, 346 patients (63%; 312 men, 34 women) were deemed suitable candidates for EVAR and 201 (37%; 151 men, 50 women) were rejected. There was no significant difference in the overall rate of rejection in the early vs the late time period (34% A, 41% B, p= 0.08), but the number of exclusion criteria per patient decreased over time; patients rejected for EVAR had an overall average of 1.6 exclusion criteria (Group A, 1.9; Group B, 1.2). The reasons for rejection did significantly change over time. Specifically, rejection on the basis of inadequate arterial access, presence of extensive iliac artery aneurysms, or an inadequate proximal neck decreased. A disproportionate number of women were excluded throughout the study: Group A, 56% of women compared to 30% of men (p= 0.0003); Group B, 63% of women compared to 36% of men (p= 0.0022). Women were more likely than men to have inadequate arterial access routes. In addition, patients with high operative risk were also more likely to be excluded from EVAR, a finding that persisted over time. Anatomic constraints continue to pose significant challenges to aortic endografting. Progress has been made in that technological advances have conquered some of the previous anatomic challenges, chiefly those of arterial access and treatment of concomitant iliac aneurysm disease. However, the overall rate of rejection for EVAR remains the same. The chief anatomic barriers continue to be the difficult aortic neck and management of branched vascular segments.

Endovascular AAA Repair in Patients with Renal Insufficiency: Strategies for Reducing Adverse Renal Events

Vascular imaging, usually employing nephrotoxic contrast agents is relied upon for all aspects of endovascular AAA repair causing some to consider renal insufficiency a relative contraindication. We sought to determine if endovascular AAA evaluation and repair could be successfully accomplished by minimally or non-nephrotoxic modalities. Records and results for 98 consecutive patients undergoing endovascular AAA repair were reviewed. Patients requiring dialysis preoperatively were excluded (N=3). The average volume of iodinated contrast agent employed for intraoperative imaging was 152 cc (35-420 cc). Twenty patients (20%) had baseline renal insufficiency (serum creatinine > or =1.3 mg/dl). A rise in serum creatinine above baseline was observed in 23 (24%) patients following repair; for 15 (16%) this was permanent. Creatinine rise occurred in patients with both normal (15) and abnormal (8) baseline values (P=0.09). Rise in creatinine was independent of contrast volume employed and of the use of infrarenal vs suprarenal device fixation (P>0.05). Two (2%) patients required permanent dialysis, one of which had a normal baseline creatinine and unclear etiology for renal failure, the other had a baseline creatinine of 2 and required device placement over an accessory renal artery. Strategies to minimize the use of nephrotoxic contrast for patients with renal insufficiency included the use of MRA, rather than contrast-CT for pre and postoperative imaging (7, 35%) and use of Gadolinium rather than iodinated contrast for performance of intraoperative arteriography (5, 25%). Endovascular grafts were successfully designed and implanted based upon MRA as the sole preoperative imaging modality in every case in which it was attempted (7). Mortality was not significantly different between those with and without abnormal baseline renal function (P>0.05). Adverse events (access failures, arterial injuries, blood loss, endoleaks) were not significantly correlated with baseline renal insufficiency, rise in creatinine from baseline, use of MRA or intraoperative Gadolinium angiography (P>0.05).Pre- and postoperative evaluation and performance of endovascular AAA repair can be accomplished in patients with renal insufficiency without increasing the rate of mortality or adverse events employing a strategy which minimizes the use of nephrotoxic contrast agents, relying upon Gadolinium arteriography and MRA. Endovascular grafts can be successfully planned and followed employing MRA as the sole imaging modality.

The fate of bypass grafts to angiographically occult runoff vessels detected by magnetic resonance angiography

PURPOSE Magnetic resonance angiography (MRA) is a noninvasive vascular imaging technique that is more sensitive than contrast arteriography (CA) for the detection of patent distal runoff vessels. This technique has facilitated performance of MRA-directed bypass procedures for patients who were believed not to be bypass candidates because of the absence of a suitable target vessel on the preoperative CA. The fate of bypasses to these angiographically occult runoff vessels is unknown, however, and it has been proposed that patients with angiographically occult runoff may have aggressive occlusive disease, rendering bypass procedures ultimately futile. METHODS Between April 1992 and February 1995, 212 autogenous vein infrageniculate bypasses were performed for limb-salvage indications, 22 (12%) to angiographically occult runoff vessels. Results of bypasses performed to angiographically occult vessels were compared with those of bypasses to CA-detected runoff vessels. Life-table analysis of graft-patency and limb-salvage rates was performed. RESULTS The accuracy of the MRA-predicted patency of angiographically occult vessels was confirmed in every case by the operative findings. Life-table analysis revealed no significant difference in primary graft patency (p > 0.05) or limb-salvage (p > 0.05) rates between patients with bypasses to runoff vessels seen by MRA alone. At 35 months after surgery, the primary graft patency rate was 68% for bypasses to CA-detected vessel bypass and 67% for MRA-detected vessels. The limb salvage rate was 83% for CA-detected vessel bypass patients and 78% for patients with angiographically occult runoff. CONCLUSIONS MRA can accurately identify patent runoff vessels not visualized by CA. Results of bypasses performed to angiographically occult runoff vessels are similar to those of bypasses performed to vessels detected by CA. MRA should be performed in patients in whom CA fails to reveal runoff vessels suitable for use in a limb-salvage procedure. The greater sensitivity of MRA may facilitate successful bypass surgery and improve the overall limb-salvage rate.