Michel S. Makaroun

The care of patients with an abdominal aortic aneurysm: The Society for Vascular Surgery practice guidelines

The Clinical Practice Council of the Society for Vascular Surgery charged a writing committee with the task of updating practice guidelines, initally published in 2003, for surgeons and physicians who are involved in the preoperative, operative, and postoperative care of patients with abdominal aortic aneurysms (AAA). This document provides recommendations for evaluating the patient, including risk of aneurysm rupture and associated medical co-morbidities, guidelines for selecting surgical or endovascular intervention, intraoperative strategies, perioperative care, long-term follow-up, and treatment of late complications. Decision making related to the care of patients with AAA is complex. Aneurysms present with varying risks of

Five-year results of endovascular treatment with the Gore TAG device compared with open repair of thoracic aortic aneurysms

OBJECTIVES Report the results of a phase II multicenter, prospective trial comparing endovascular treatment of descending thoracic aneurysm (TEVAR) with the TAG device to surgical controls after 5 years of follow-up. METHODS The Gore TAG trial compared the TAG endograft patients (n = 140) with standard open surgical controls (n = 94) with enrollment from September of 1999 to May of 2001. An additional 51 patients were enrolled in 2003 after revision of the endograft. Follow-up consisted of patient visits, computed tomography (CT) scans and x-rays at 1, 6, and 12 months and yearly. Significant sac size change was defined as >or=5 mm increase or decrease from the 1 month baseline measurement. Migration was defined as >or=10 mm cranial or caudal movement of the device inside the aorta. Significance was determined as P <or= .05. RESULTS At 5 years, aneurysm-related mortality was lower for TAG patients at 2.8% compared with open controls at 11.7% (P = .008). No differences in all-cause mortality were noted, with 68% of TAG patients and 67% of open controls surviving to 5 years (P = .43). Major adverse events at 5 years were significantly reduced in the TAG group; 57.9% vs 78.7% (P = .001). Endoleaks in the TAG group decreased from 8.1% at 1 month to 4.3% at 5 years. Five TAG patients have undergone major aneurysm-related re-interventions at 5 years (3.6%), including one arch aneurysm repair for type 1 endoleak and migration, one open conversion and five endovascular procedures for endoleaks in three patients. There were fewer secondary procedures not directly related to aneurysm repair in the TAG vs the open repair group at 5 years, 15.0% vs 31.9%, (P = .01). For TAG patients, sac size at 60 months decreased in 50% and increased in 19% compared with the 1-month baseline. Comparison with the modified low-porosity device at 24 months showed sac increase in 12.9% of original vs 2.9% in modified grafts (P = .11). At 5 years, there have been no ruptures, one migration, no collapse, and 20 instances of fracture in 19 patients, all before the revision of the TAG graft. CONCLUSIONS In anatomically suitable patients, TAG treatment of thoracic aneurysms is superior to surgical repair at 5 years. Although sac enlargement is concerning, early modified device results indicate this issue may be resolved.

Compromised cerebral blood flow reactivity is a predictor of stroke in patients with symptomatic carotid artery occlusive disease

PURPOSE The purpose of this study was to determine whether the hemodynamic consequences of extracranial carotid disease correlate with the risk of subsequent cerebral infarction. METHODS In 95 patients with symptoms who had greater than or equal to 70% stenosis (31 patients) or who had occlusion (64 patients) of the ipsilateral carotid artery, cerebral blood flow was measured by the stable xenon/computed tomography technique both at baseline and after vasodilatory challenge with intravenous acetazolamide. Patients were stratified into group 1, 43 patients with no more than a 5% decrease in flow in any vascular territory, and group 2, 52 patients with greater than a 5% decrease in one or more vascular territories after an acetazolamide challenge. RESULTS In group 2, 15 (28.9%) of 52 patients had a new stroke, but only one (2.3%) of 43 patients in group 1 did (p = 0.0005). Of patients with total carotid occlusion 10 (26%) of 38 in group 2 and none (0%) of 26 in group 1 had a new stroke (p = 0.003). Of patients with greater than or equal to 70% stenosis, five (36%) of 14 in group 2 and only one (6%) of 17 in group 1 had a stroke (p = 0.067). CONCLUSION The loss of cerebral reactivity in patients with symptoms who had greater than or equal to 70% carotid stenosis or occlusion is an important predictor of impending cerebral infarction.

Mechanical Properties and Microstructure of Intraluminal Thrombus From Abdominal Aortic Aneurysm

Accurate estimation of the wall stress distribution in an abdominal aortic aneurysm (AAA) may prove clinically useful by predicting when a particular aneurysm will rupture. Appropriate constitutive models for both the wall and the intraluminal thrombus (ILT) found in most AAA are necessary for this task. The purpose of this work was to determine the mechanical properties of ILT within AAA and to derive a more suitable constitutive model for this material. Uniaxial tensile testing was carried out on 50 specimens, including 14 longitudinally oriented and 14 circumferentially oriented specimens from the luminal region of the ILT, and 11 longitudinally oriented and 11 circumferentially oriented specimens from the medial region. A two-parameter, large-strain, hyperelastic constitutive model was developed and used to fit the uniaxial tensile testing data for determination of the material parameters. Maximum stiffness and strength were also determined from the data for each specimen. Scanning electron microscopy (SEM) was conducted to study the regional microstructural difference. Our results indicate that the microstructure of ILT differs between the luminal, medial, and abluminal regions, with the luminal region stronger and stiffer than the medial region. In all cases, the constitutive model fit the experimental data very well (R2>0.98). No significant difference was found for either of the two material parameters between longitudinal and circumferential directions, but a significant difference in material parameters, stiffness, and strength between the laminal and medial regions was determined (p<0.01). Therefore, our results suggest that ILT is an inhomogeneous and possibly isotropic material. The two-parameter, hyperelastic, isotropic, incompressible material model derived here for ILT can be easily incorporated into finite element models for simulation of wall stress distribution in AAA.

Color-flow duplex ultrasound scan versus computed tomographic scan in the surveillance of endovascular aneurysm repair

OBJECTIVE The purpose of this study is to compare both computed tomographic scan (CT) and color flow duplex ultrasound scanning (CDU) as surveillance modalities for clinically significant endoleaks and to evaluate concordance in abdominal aortic aneurysm (AAA) diameter measurements in patients after endovascular aneurysm repair (EVAR) in a busy hospital vascular laboratory. METHODS We conducted a retrospective review of all patients who underwent endovascular repair of abdominal aortic aneurysms between February 1996 and November 2002 and had same-day CT and CDU studies. Ninety-seven patients enrolled in phase II clinical studies of Ancure devices had long-term follow-up with both modalities. The other patients underwent simultaneous studies, usually only at the 1-month postoperative visit. Peripheral vascular studies were performed by two certified vascular technicians; all CT scans were reviewed by one vascular surgeon. CT was used as the standard against which the sensitivity, specificity, negative predictive value, and positive predictive value of CDU in endoleak detection was determined. Statistics were performed by using the paired t test; a P value <.05 was considered significant. Kappa statistic was used to assess the correlation between CDU and CT in identifying endoleaks. The correlation between CT and CDU in AAA size measurements as well as in serial size measurements was also determined. RESULTS Four hundred ninety-five same-day CT and CDU examinations were reviewed in 281 patients. Patients had an average follow-up of 34.6 months (range, 1 to 72 months). Thirty-five leaks were identified among the patients studied (12.4% overall). In comparison with CT, diagnosis of endoleak with ultrasound scanning was associated with a sensitivity of 42.9%, specificity of 96.0%, positive predictive value of 53.9%, and negative predictive value of 93.9%. The correlation between the two modalities was modest (kappa statistic 0.427). The minor axis transverse diameter as measured by ultrasound and CT scans (4.81 +/- 1.1 cm on CT and 4.55 +/- 1.1 cm on ultrasound) correlated closely (r =.93, P <.001.) Seventy percent of paired studies differed by < or =5 mm. Changes in aneurysm size throughout follow-up were -.29 +/-.71 cm on CT scan -.34 +/-.57 cm on duplex ultrasound scan. The correlation coefficient was.65 (P <.001). There was no significant difference in the change as measured by either modality on the paired t test. CONCLUSIONS Although CDU demonstrates a high degree of correlation with CT scan in determining aneurysm size change over time, it has a low sensitivity and positive predictive value in endoleak detection. In the hospital vascular laboratory at a large tertiary care center, CDU cannot effectively replace CT scan in surveillance after EVAR.

Does hostile neck anatomy preclude successful endovascular aortic aneurysm repair

OBJECTIVES Poor outcomes have been reported with endovascular aneurysm repair (EVAR) in patients with hostile neck anatomy. Unsupported endografts with active fixation may offer certain advantages in this situation. We compared EVAR results using the Ancure (Guidant) endograft in patients with and without hostile neck anatomy. METHODS Records of EVAR patients from October 1999 to July 2002 at a tertiary care hospital were retrospectively reviewed from a division database. Patients with elective open abdominal aortic aneurysm (AAA) repair during the same period were reviewed to determine those unsuitable for EVAR. Hostile neck anatomy, assessed by computer tomography (CT) scans and angiograms, was defined as one or more of the following: (1) neck length </=10 mm, (2) focal bulge in the neck >3 mm, (3) >2-mm reverse taper within 1 cm below the renal arteries, (4) neck thrombus > or =50% of circumference, and (5) angulation > or =60 degrees within 3 cm below renals. RESULTS Three hundred and twenty-two patients underwent EVAR with an average follow-up of 18 months. Patients in Phase II trials (n = 41), repaired with other graft types (n = 48), or without complete anatomic records (n = 27) were excluded. Demographics and co-morbidities were similar in the 115 good-neck (GN) and 91 bad-neck (BN) patients except for age (mean, 72.9 years GN vs 75.7 BN; P = 0.13), gender (11% female GN vs 22% BN; P =.04); neck length (mean, 21.8 mm GN vs 14.4 mm BN: P <.001), and angulation (mean, 22 degrees GN vs 40 degrees BN; (P <.001). Perioperative mortality (0 GN vs 1.1% BN), late mortality (5.2% GN vs 4.4% BN), all endoleaks (19.1% GN vs 17.6% BN), proximal endoleaks (0.8% GN vs 2.1% BN), and graft migration (0 for both groups) did not reach statistical significance. Neck anatomy precluded EVAR in 106 of 165 (64%) patients with open AAA. CONCLUSIONS Unsupported endografts with active fixation can yield excellent results in treating many medically compromised patients with hostile neck anatomy. Nonetheless, an unsuitable neck remains the most frequent cause for open abdominal AAA.

A randomized trial of carotid artery stenting with and without cerebral protection

BACKGROUND The use of a distal filter cerebral protection device with carotid artery stenting is commonplace. There is little evidence, however, that filters are effective in preventing embolic lesions. This study examined the incidence of embolic phenomenon during carotid artery stenting with and without filter use. METHODS This was a prospective, randomized, single-center study of carotid artery stenting with or without a distal cerebral protection filter. A 1:1 scheme was used to randomize 36 carotid artery stenting procedures in 35 patients. Diffusion-weighted magnetic resonance imaging (DW MRI) 24 hours after stenting was used to assess the occurrence of new embolic lesions. Blinded observers calculated lesion number and volume. RESULTS The mean age was 78.6 +/- 7.0 in the cerebral protection group compared with 74.1 +/- 8.7 in the no cerebral protection group (P = .92). Despite similar average age, the percentage of octogenarians was higher in the cerebral protection group (61.1% vs 22.2%; P = .04). Two procedures in the cerebral protection group were not successful. One was completed without protection because of inability to track the filter, and the second was aborted because of severe tortuosity with a later carotid endarterectomy. New MRI lesions were noted in 72% of the cerebral protection group compared with 44% in the no cerebral protection group (P = .09). The average number of lesions in these patients was 6.1 and 6.2, respectively, with mean DW MRI lesion size of 16.63 mm(3) vs 15.61 mm(3) (P = .79 and .49, respectively). Four strokes occurred (11%), two in each group, in patients aged 75, 80, 82, and 84 years. The only major stroke occurred in the no cerebral protection group. CONCLUSIONS The use of filters during carotid artery stenting provided no demonstrable reduction of microemboli, as expected. Routine use of cerebral protection filters should undergo a more critical assessment before mandatory universal adoption.

A biomechanics-based rupture potential index for abdominal aortic aneurysm risk assessment: Demonstrative application

Abstract:  Abdominal aortic aneurysms (AAAs) can typically remain stable until the strength of the aortic wall is unable to withstand the forces acting on it as a result of the luminal blood pressure, resulting in AAA rupture. The clinical treatment of AAA patients presents a dilemma for the surgeon: surgery should only be recommended when the risk of rupture of the AAA outweighs the risks associated with the interventional procedure. Since AAA rupture occurs when the stress acting on the wall exceeds its strength, the assessment of AAA rupture should include estimates of both wall stress and wall strength distributions. The present work details a method for noninvasively assessing the rupture potential of AAAs using patient‐specific estimations the rupture potential index (RPI) of the AAA, calculated as the ratio of locally acting wall stress to strength. The RPI was calculated for thirteen AAAs, which were broken up into ruptured (n= 8 and nonruptured (n= 5) groups. Differences in peak wall stress, minimum strength and maximum RPI were compared across groups. There were no statistical differences in the maximum transverse diameters (6.8 ± 0.3 cm vs. 6.1 ± 0.5 cm, p= 0.26) or peak wall stress (46.0 ± 4.3 vs. 49.9 ± 4.0 N/cm2, p= 0.62) between groups. There was a significant decrease in minimum wall strength for ruptured AAA (81.2 ± 3.9 and 108.3 ± 10.2 N/cm2, p= 0.045). While the differences in RPI values (ruptured = 0.48 ± 0.05 vs. nonruptured = 0.36 ± 0.03, respectively; p= 0.10) did not reach statistical significance, the p‐value for the peak RPI comparison was lower than that for both the maximum diameter (p= 0.26) and peak wall stress (p= 0.62) comparisons. This result suggests that the peak RPI may be better able to identify those AAAs at high risk of rupture than maximum diameter or peak wall stress alone. The clinical relevance of this method for rupture assessment has yet to be validated, however, its success could aid clinicians in decision making and AAA patient management.

Thrombin injection versus compression of femoral artery pseudoaneurysms

OBJECTIVE The compression of femoral artery pseudoaneurysms is a time consuming, painful, and sometimes unsuccessful procedure. Thrombin injection has been advocated as a superior alternative. In this study, we compare our experiences with both techniques. METHODS All the records of femoral artery false aneurysms that were treated in the vascular laboratory from January 1996 to April 1999 were retrospectively reviewed. Treatment with ultrasound scan-guided compression was compared with treatment with dilute thrombin injection (100 U/mL). RESULTS Both groups had similar demographics and aneurysm sizes (P >.2). Of the pseudoaneursyms, 88% were caused by cardiac catheterization and the others were the results of femoral artery access for cardiac surgery (6%), arteriography (5%), and renal dialysis (1%). Compression was successful in 25 of 40 patients (63%). Nine persistent aneurysms necessitated operation, and six were treated successfully with thrombin injection. Primary thrombin injection successfully obliterated 21 pseudoaneurysms in 23 patients. Overall, 27 of 29 pseudoaneurysms were treated successfully with thrombin injection (93%). Thrombosis occurred within seconds of the thrombin injection and required, on average, 300 units of thrombin (100 to 600 units). The patients who underwent successful compression required an average of 37 minutes of compression (range, 5 to 70 minutes) and required analgesia on several occasions. No patients in the thrombin group required analgesia or sedation. Neither group had complications. A cost analysis shows that thrombin treatment results in considerable savings in vascular laboratory resource use but not in overall hospital expenditures. CONCLUSION Ultrasound scan-guided thrombin injection is a safe, fast, and painless procedure that completely obliterates femoral artery pseudoaneurysms. The shift from compressive therapy to thrombin injection reduces vascular laboratory use and is less expensive, although it does not significantly impact hospital costs.

Effects of wall calcifications in patient-specific wall stress analyses of abdominal aortic aneurysms.

It is generally acknowledged that rupture of an abdominal aortic aneurysm (AAA) occurs when the stress acting on the wall over the cardiac cycle exceeds the strength of the wall. Peak wall stress computations appear to give a more accurate rupture risk assessment than AAA diameter, which is currently used for a diagnosis. Despite the numerous studies utilizing patient-specific wall stress modeling of AAAs, none investigated the effect of wall calcifications on wall stress. The objective of this study was to evaluate the influence of calcifications on patient-specific finite element stress computations. In addition, we assessed whether the effect of calcifications could be predicted directly from the CT-scans by relating the effect to the amount of calcification present in the AAA wall. For 6 AAAs, the location and extent of calcification was identified from CT-scans. A finite element model was created for each AAA and the areas of calcification were defined node-wise in the mesh of the model. Comparisons are made between maximum principal stress distributions, computed without calcifications and with calcifications with varying material properties. Peak stresses are determined from the stress results and related to a calcification index (CI), a quantification of the amount of calcification in the AAA wall. At calcification sites, local stresses increased, leading to a peak stress increase of 22% in the most severe case. Our results displayed a weak correlation between the CI and the increase in peak stress. Additionally, the results showed a marked influence of the calcification elastic modulus on computed stresses. Inclusion of calcifications in finite element analysis of AAAs resulted in a marked alteration of the stress distributions and should therefore be included in rupture risk assessment. The results also suggest that the location and shape of the calcified regions--not only the relative amount--are considerations that influence the effect on AAA wall stress. The dependency of the effect of the wall stress on the calcification elastic modulus points out the importance of determination of the material properties of calcified AAA wall.