NONINVASIVECARDIOVASCULAR IMAGING

Contrast Material-enhanced MRAOverestimates Severity of Carotid Ste-nosis, Compared with 3D Time-of-Flight MRA. T. C. Townsend, D. Sa-loner, et al. J Vasc Surg 2003; 38:36–40.

• Objective: Non-contrast-enhancedmagnetic resonance angiography (MRA)carotid imaging with the time-of-flight(TOF) technique compares favorablywith angiography, ultrasound, and ex-cised plaques. However, gadoliniumcontrast-enhanced MRA (CE-MRA) hasalmost universally replaced TOF-MRA,because it reduces imaging time (25 sec-onds vs 10 minutes) and improves sig-nal-to-noise ratio. In our practice wefound alarming discrepancies betweenCE-MRA and TOF-MRA, which was theimpetus for this study. Study Design: Tocompare the two techniques, we mea-sured stenosis, demonstrated on three-dimensional images obtained at TOFand CE-MRA, in 107 carotid arteries in58 male patients. The measurementswere made on a Cemax workstationequipped with enlargement and mea-surement tools. Measurements to 0.1mm were made at 90 degrees to the flowchannel at the area of maximal stenosisand distal to the bulb where the bordersof the internal carotid artery lumen werejudged to be parallel (North AmericanSymptomatic Carotid EndarterectomyTrial criteria). Experiments with carotidphantoms were done to test the contri-bution of imaging software to imagequality. Results: Twelve arteries were oc-cluded. In the remaining 95 arteries,compared with TOF-MRA, CE-MRAdemonstrated a greater degree of steno-sis in 42 arteries, a lesser degree of ste-nosis in 14 arteries, and similar (�5%)stenosis in 39 arteries (P �.02, �2 analy-sis). The largest discrepancies were ar-teries with 0% to 70% stenosis. In thosearteries in which CE-MRA identified agreater degree of stenosis than shownwith TOF-MRA, mean increase was 21%for 0% to 29% stenosis, 36% for 30% to49% stenosis, and 38% for of 50% to 69%stenosis. The carotid phantom experi-ments showed that the imaging param-eters of CE-MRA, particularly the planeon which frequency encoding gradientswere applied, reduced signal acquisitionat the area of stenosis. Conclusions: Col-lectively these data demonstrate that CE-MRA parameters must be retooled if the

method is to be considered reliable fordetermination of severity of carotid ar-tery stenosis. CE-MRA is an excellentscreening technique, but only TOF-MRAshould be used to determine degree ofcarotid artery stenosis.Authors’ Abstract

PERIPHERAL ARTERIALINTERVENTIONSDiagnosis

Pathogenesis in Acute Aortic Syn-dromes: Aortic Dissection, IntramuralHematoma, and Penetrating Athero-sclerotic Aortic Ulcer. K. J. Macura,F. M. Corl, et al. Am J Roentgenol 2003;181:309–316.

• Acute aortic syndromes refer to thespectrum of aortic emergencies that in-clude aortic dissection, intramural he-matoma, penetrating atherosclerotic ul-cer of the aorta, aortic aneurysm leakand rupture, and traumatic aortic tran-section.Authors’ Abstract

Stent-grafts

Technique and Results of Transfemo-ral Superselective Coil Embolizationof Type II Lumbar Endoleak. K. Ka-sirajan, B. Matteson, et al. J Vasc Surg2003; 38:61–66.

• Objective: This study was under-taken to describe the technique of trans-femoral superselective coil embolizationof type II endoleak and its influence onabdominal aortic aneurysm diameter.Methods: Over 23 months, 104 aorticstent grafts were deployed to excludeabdominal aortic aneurysms, at an aca-demic medical center. Increase in aneu-rysm diameter and perigraft findings oncontrast material-enhanced computedtomography scans prompted arteriogra-phy. Procedures were performed solelyby vascular surgeons in a surgical an-giography suite. In 7 patients aneurysmaccess was via the iliolumbar branchesof the internal iliac artery, and in 1 pa-tient aneurysm access was via the infe-rior mesenteric artery through the arc ofRiolan from the superior mesenteric ar-tery. Coaxial catheters were placed togain access to the aneurysm (8F to 5F to3F, or 5F to 3F). A 3F Tracker18 was themost distal catheter through which anassortment of 0.018 microcoils were de-ployed within the aneurysm, and theorigin of the feeding vessels when pos-

sible. Results: Aneurysm diameter in-creased 0.48 � 0.2 cm over 10.8 � 5months before superselective coil embo-lization. In 6 of 8 patients superselectivecoil embolization embolization resultedin a mean decrease in aneurysm diame-ter of 1.3 � 1.2 cm over 9 � 3.2 months.Failure was presumed due to inability toreach the aneurysm sac in 1 patient andwas associated with oral anticoagulationin 1 other patient. Conclusion: Properidentification of the source of type II en-doleak and its complete occlusion, com-bined with aneurysm sac coiling, mayresult in prompt decrease in aneurysmsize.Authors’ Abstract

Five-year Interim Comparison of theGuidant Bifurcated Endograft withOpen Repair of Abdominal Aortic An-eurysm. W. S. Moore, J. S. Mat-sumura, et al. J Vasc Surg 2003; 38:46–55.

• Objective: This study was under-taken to compare 1-year and 5-year re-sults of endovascular repair of abdomi-nal aortic aneurysm (AAA) with theGuidant/EVT bifurcated graft systemwith results of open repair. Methods: Thiswas a prospective, nonrandomized, con-current controlled study that comparedresults of endovascular versus open re-pair of AAA. The Phase II study with theEGS delivery system included 268 pa-tients in 18 US medical centers; and thePhase III trial with the Ancure deliverysystem incuded 305 patients in 21 USinstitutions. Data were internally and ex-ternally audited and subjected to peri-odic review by the US Food and DrugAdministration. The control group of111 patients were excluded from endo-vascular repair with a tube graft becauseof anatomic considerations, but wereotherwise comparable to the experimen-tal group. Patients in the control groupunderwent conventional open surgicalrepair concurrently with patients whounderwent EGS repair in 18 US institu-tions. Results: Five hundred thirty-one of573 patients (92.7%) underwent success-ful implantation of the Guidant/EVT bi-furcated endograft. The combined majormorbidity and mortality in the endograftgroup was 28.8%, compared with 44.1%in the open control group. Additionalbenefits in the endograft group includedshorter hospital stay (2 days vs 6 days),less surgical blood loss (400 mL vs 800mL), and less intensive care unit use

Abstracts of Current Literature

1586

(33% vs 94%). These early results arereported on an intent-to-treat basis; in allpatients an attempt was made to treatwith the endovascular graft, includingthose patients in whom conversion tostandard open repair was necessary dur-ing the primary procedure. Three hun-dred nineteen patients were selected forlong-term follow-up to 5 years, on thebasis of date of implantation; ie, patientswith the earliest implantations were fol-lowed up for 5 years. The primary pur-pose of long-term follow-up was to ob-tain data on long-term efficacy of thegraft; thus only patients in whom im-plantation was successful were selected.No patient has experienced an aneurysmrupture to date. Survival (Kaplan- Meiermethod) in the experimental group was68.1%, compared with 77.2% in the con-trol group (P � NS). At 60 months,74.4% of patients (32 of 43) were free ofendoleak. There were no type I or typeIII endoleaks remaining. Aneurysm sacdiameter decreased or remained stablein 97.6% of patients (41 of 42) and in-creased in only 1 patient. During thecourse of long-term follow-up, post-pro-cedural conversion to open repair wasrequired in only 9 patients (2.8%). Con-clusion: The EVT/Guidant bifurcatedgraft is effective in preventing AAA rup-ture, and long-term survival is compara-ble to that with open repair.Authors’ Abstract

Effectiveness and Cost of Screeningfor Abdominal Aortic Aneurysm: Re-sults of a Population Screening Pro-gram. A. B. Wilmink, C. R. Quick, etal. J Vasc Surg 2003; 38:72–77.

• Objectives: We undertook this studyto calculate the cost per life-year gainedin the first round of a screening programfor abdominal aortic aneurysm (AAA)and to estimate the costs in a subsequentround. Methods: This was an interven-tion study, with follow-up for rupturedaneurysms. Men older than 50 yearswere screened for asymptomatic AAA.Outcome measures included cost perlife-year saved and number of menneeded to be screened to save one life.Results: The incidence of ruptured AAAwas 2.6 per 10,000 person- years in thescreening group and 7.1 per 10,000 per-son-years in the control group. Screen-ing is estimated to have prevented 10.8ruptured AAA and 8 deaths per year,gaining 51 life-years per year for thestudy population, and to have reducedthe incidence of ruptured AAA by 64%(95% CI, 42%–77%). Each life-yeargained during the first screening roundcost $1107. To save one life, 1000 menneed to be screened and 5 elective oper-ations performed. We predict that a sec-ond round of screening can be cost neu-

tral. Conclusions: The cost-effectivenessof screening for AAA compares favor-ably with screening programs for otherdisorders in adults.Authors’ Abstract

Embolization

Ovarian Artery: Angiographic Appear-ance, Embolization and Relevance toUterine Fibroid Embolization. J. P.Pelage, W. J. Walker, et al. CardiovascIntervent Radiol 2003; 26:227–233.

• Purpose: To describe the angio-graphic appearance of the ovarian arteryand its main variations that may be rel-evant to uterine fibroid embolization.Methods: The flush aortograms of 294women who had been treated by uterineartery embolization for fibroids were re-viewed. Significant arterial supply to thefibroid, and the origin and diameter ofidentified ovarian arteries were re-corded. In patients with additionalembolization of the ovarian artery, thefollow-up evaluation also included hor-monal levels and Doppler imaging of theovaries. Results: A total of 75 ovarianarteries were identified in 59 women (bi-laterally in 16 women and unilaterally in43 women). All ovarian arteries origi-nated from the aorta below the level ofthe renal arteries with a characteristictortuous course. Fifteen women had atleast one enlarged ovarian artery sup-plying the fibroids. Fourteen women(14/15, 93%) presented at least one of thefollowing factors: prior pelvic surgery,tubo-ovarian pathology or large fundalfibroids. Conclusion: We advocate the useof flush aortography in women withprior tubo-ovarian pathology or surgeryor in cases of large fundal fibroids. In thecase of an ovarian artery supply to thefibroids, superselective catheterizationand embolization of the ovarian arteryshould be considered.Authors’ Abstract

Digital Subtraction Fluoroscopy toEnhance Visualization during UterineFibroid Embolization: A TechnicalNote. R. T. Andrews and C. A. Binkert.Cardiovasc Intervent Radiol 2003; 26:296–297.

• We describe a simple but underuti-lized technique for improving visualiza-tion during transcatheter embolizationusing particulate agents. The techniqueis of distinct utility in uterine fibroidembolization, during which non-targetembolization can be of particular clinicalsignificance.Authors’ Abstract

VENOUS INTERVENTIONSThromboembolic Disease

Low-molecular-weight Heparin versusa Coumarin for the Prevention of Re-current Venous Thromboembolism inPatients with Cancer. A. Y. Lee,M. N. Levine, et al. N Engl J Med 2003;349:146–153.

• Background: Patients with cancerhave a substantial risk of recurrentthrombosis despite the use of oral anti-coagulant therapy. We compared the ef-ficacy of a low-molecular-weight hepa-rin with that of an oral anticoagulantagent in preventing recurrent thrombo-sis in patients with cancer. Methods: Pa-tients with cancer who had acute, symp-tomatic proximal deep-vein thrombosis,pulmonary embolism, or both were ran-domly assigned to receive low-molecu-lar-weight heparin (dalteparin) at a doseof 200 IU per kilogram of body weightsubcutaneously once daily for five toseven days and a coumarin derivativefor six months (target international nor-malized ratio, 2.5) or dalteparin alone forsix months (200 IU per kilogram oncedaily for one month, followed by a dailydose of approximately 150 IU per kilo-gram for five months). Results: Duringthe six-month study period, 27 of 336patients in the dalteparin group had re-current venous thromboembolism, ascompared with 53 of 336 patients in theoral-anticoagulant group (hazard ratio,0.48; P�0.002). The probability of recur-rent thromboembolism at six monthswas 17% in the oral-anticoagulant groupand 9% in the dalteparin group. No sig-nificant difference between the daltepa-rin group and the oral-anticoagulantgroup was detected in the rate of majorbleeding (6% and 4%, respectively) orany bleeding (14% and 19%, respec-tively). The mortality rate at six monthswas 39% in the dalteparin group and41% in the oral-anticoagulant group.Conclusions: In patients with cancer andacute venous thromboembolism, dalte-parin was more effective than an oralanticoagulant in reducing the risk of re-current thromboembolism without in-creasing the risk of bleeding.Authors’ Abstract

TRAUMA

Pathogenesis in Acute Aortic Syn-dromes: Aortic Aneurysm Leak andRupture and Traumatic Aortic Tran-section. K. J. Macura, F. M. Corl, etal. Am J Roentgenol 2003; 181:303–307.

• This pictorial essay focuses on thepathophysiology of enlargement andrupture of the atherosclerotic aortic an-eurysm and on mechanisms involved intraumatic aortic transection related to

Abstracts • 1587Volume 14 Number 12

deceleration injury. The aortic wall iscomposed of three layers: the inner layerof intima, the middle layer of media, andthe outer layer of adventitia. Multiplemechanisms are involved in the disrup-tion of the aortic wall layers, leading tovarious acute aortic syndromes. This pic-torial essay focuses on the distinction ofa typical aortic dissection from an intra-mural hematoma and penetrating ath-erosclerotic ulcer.Authors’ Abstract

Occupant- and Collision-related RiskFactors for Blunt Thoracic Aorta In-jury. G. McGwin, Jr., J. Metzger, et al.J Trauma 2003; 54:655–660; discussion660–662.

• Background: Blunt thoracic aortic in-jury (BAI) is a rare and highly lethalinjury. We sought to identify occupantand collision characteristics associatedwith motor vehicle collision (MVC)-re-lated BAI. Methods: The 1995 to 2000 Na-tional Automotive Sampling Systemdata files were used. The National Auto-motive Sampling System is a nationalprobability sample of passenger vehiclesinvolved in police-reported tow-awayMVCs. The risk of BAI was calculatedaccording to specific occupant (e.g., age,seat belt use) and collision (e.g., delta-V[estimated change in velocity], vehicularintrusion) characteristics. The associa-tion between BAI and these characteris-tics was calculated using risk ratios(RRs) and associated 95% confidence in-tervals (CIs). Results: Specific occupantand collision characteristics demon-strated independent association withBAI. Occupant characteristics includedage � 60 (RR, 3.6; 95% CI, 2.5–5.2), seatbelt use (RR, 0.3; 95% CI, 0.2–0.5), andbeing a front-seat occupant (RR, 3.1; 95%CI, 1.5–6.3). Frontal and near-side MVCswere associated with an increased risk(RR, 3.1; 95% CI, 1.9–5.1; and RR, 4.3;95% CI, 2.6–7.2, respectively) relative toother collision types. Collisions with adelta-V � 40 km/h (RR, 3.8; 95% CI,2.6–5.6) or that produce extensive vehi-cle crush (� 40 cm) (RR, 4.1; 95% CI,2.7–6.3) or intrusion (� 15 cm) (RR, 5.0;95% CI, 3.5–7.3) also increase the risk ofBAI. Conclusion: The risk factors for BAIidentified in this study support gener-ally accepted etiologic mechanisms forthis injury.Authors’ Abstract

Operative Management and Outcomesin 103 AAST-OIS Grades IV and VComplex Hepatic Injuries: TraumaSurgeons Still Need to Operate, butAngioembolization Helps. J. A.Asensio, G. Roldan, et al. J Trauma2003; 54:647–653; discussion 653–654.

Background: American Association forthe Surgery of Trauma (AAST) OrganInjury Scale (OIS) grades IV and V com-plex hepatic injuries are highly lethal.Our objectives were to review experi-ence and identify predictors of outcomeand to evaluate the role of angioemboli-zation in decreasing mortality. Methods:This was a retrospective 8-year study ofall patients sustaining AAST-OIS gradesIV and V hepatic injuries managed op-eratively. Statistical analysis was per-formed using univariate and multivari-ate logistic regression. The mainoutcome measure was survival. Results:The study included 103 patients, with amean Revised Trauma Score of 5.61 �2.55 and a mean Injury Severity Score of33 � 9.5. Mechanism of injury was pen-etrating in 80 (79%) and blunt in 23(21%). Emergency department thoracot-omy was performed in 21 (25%). AASTgrade IV injuries occurred in 51 (47%)and grade V injuries occurred in 52(53%). Mean estimated blood loss was9,414 mL. Overall survival was 43%. Ad-justed overall survival rate after emer-gency department thoracotomy patientswere excluded was 58%. Results strati-fied to AAST-OIS injury grade were asfollows: grade IV, 32 of 51 (63%); gradeV, 12 of 52 (23%); grade IV versus gradeV (p � 0.001) odds ratio, 2.06; 95% con-fidence interval, 2.72 (1.40–3.04). Logis-tic regression analysis identified as inde-pendent predictors of outcome RevisedTrauma Score (adjusted p � 0.0002), an-gioembolization (adjusted p � 0.0177),direct approach to hepatic veins (ad-justed p � 0.0096), and packing (adjustedp � 0.0013). Conclusion: Improvements inmortality can be achieved with an ap-propriate operative approach. Angioem-bolization as an adjunct procedure de-creases mortality in AAST-OIS grades IVand V hepatic injuries.Authors’ Abstract

HEPATOBILIARY

Impact of Liver Biopsy Size on Histo-logical Evaluation of Chronic ViralHepatitis: The Smaller the Sample,the Milder the Disease. G. Colloredo,M. Guido, et al. J Hepatol 2003; 39:239–244.

• Background/Aims: In chronic viralhepatitis, liver biopsy is performed forassessing disease activity and fibrosis. Inthis study, we evaluated the impact ofthe size of liver biopsy on the gradingand staging. Methods: We selected 161liver biopsies from patients with chronictypes B and C hepatitis on the basis oftheir length (�3 cm) and width (1.4mm). Ishak scoring system was used forgrading and staging. The score wasblindly repeated reducing the length of

the specimen from �3 to 1.5 cm and to 1cm long and width from 1.4 to 1 mm.Results: Reducing the length of the bi-opsy led to an increase of cases withmild grades: 49.7% in �3 cm, 60.2% in1.5 cm and 86.6% in 1 cm long specimens(P�0.001). Similarly, cases staged ashaving mild fibrosis significantly in-creased in the shorter specimens: 59% in�3 cm, 68.3% in 1.5 cm and 80.1% in 1cm long specimens (P � 0.001). As forthe width, both grade and stage weresignificantly underscored in the 1 mmsamples, regardless of their length. Con-clusions: Liver biopsy size strongly influ-ences the grading and staging of chronicviral hepatitis. The use of fine needlesshould be discouraged in this setting.Authors’ Abstract

SEDATION

Nurse-administered Propofol Sedationwithout Anesthesia Specialists in 9152Endoscopic Cases in an AmbulatorySurgery Center. J. A. Walker, R. D.McIntyre, et al. Am J Gastroenterol 2003;98:1744–1750.

• Objective: Narcotics and benzodiaz-epines are commonly used for sedationfor endoscopy in the United States.Propofol has certain advantages overnarcotics and benzodiazepines, but itsuse is often controlled by anesthesia spe-cialists. This report describes our experi-ence with dosage, safety, patient satis-faction, and discharge time with nurse-administered propofol sedation in 9152endoscopic cases. Methods: The studywas performed in a private practice am-bulatory surgery center in Medford, Or-egon. With the assistance of an anesthe-siologist, we developed a protocol foradministration of propofol in routine en-doscopic cases, in which propofol wasgiven by registered nurses under the su-pervision of endoscopists or gastroenter-ologists. We then applied the protocolwith 9152 patients. Results: There wereseven cases of respiratory compromise(three prolonged apnea, three laryngo-spasm, one aspiration requiring hospi-talization), all associated with upper en-doscopy. Five patients required maskventilation, but none required endotra-cheal intubation. There were seven co-lonic perforations (�1 per 1000 colonos-copies), of which three may haveinvolved forceful sigmoid disruption. Ofpatients who had previously receivednarcotic or benzodiazepine sedation,84% preferred propofol. Gastroenterolo-gists strongly preferred propofol. Themean time from completion of proce-dures to discharge in a sample of 100patients was 18 min. Conclusion: Nurse-administered propofol sedation in anambulatory surgery center was safe and

1588 • Abstracts December 2003 JVIR

resulted in high levels of patient satisfac-tion and rapid postprocedure recoveryand discharge.Authors’ Abstract

CLOSURE DEVICES

Femoral Endarteritis Associated withPercutaneous Suture Closure: NewTechnology, Challenging Complica-tions. H. Whitton Hollis, Jr. and T. F.Rehring. J Vasc Surg 2003; 38:83–87.

• Objective: Use of percutaneous su-ture closure devices after catheter-basedinterventions is increasing. We recentlyhave seen several severe femoral arterialwall infections after use of such devices.The purpose of this study was to exam-ine the incidence, comorbid associations,and management of femoral arterial in-fections associated with percutaneoussuture closure devices. Methods: We ret-rospectively reviewed all infectiouscomplications that occurred after 2223consecutive cardiac catheterization pro-cedures performed over 12 months in a

university-affiliated community teach-ing hospital. Outcome variables in-cluded demographics, procedural de-tails, infection, type of arterialreconstruction required, mortality, andlimb loss. Results: During this study, 822patients received percutaneous suturedevices. Infection developed in 6 pa-tients (0.7%). The incidence of diabetesin the population undergoing percutane-ous suture closure was 219 of 822 pa-tients (26.6%). Three comorbid condi-tions, noted in multiple patients withinfectious complications, included dia-betes mellitus, obesity, and placement ofa percutaneous suture closure devicewithin the past 6 months. Invasive fem-oral endarteritis developed in 4 patients.Gram-positive cocci predominated in 4patients. In 1 patient with polymicrobialinfection catastrophic complications de-veloped, including multiple anastomoticruptures and hemorrhage. A newmethod of repair that incorporated dou-ble-thickness everted saphenous veinwas used in 2 patients, and safe arterialclosure was achieved. There was 1 late

fatality on postoperative day 36. Limbsalvage was achieved in all patients.Conclusions: Femoral endarteritis com-plicating percutaneous suture closure isa challenging new problem for vascularsurgeons and can result in catastrophiccomplications. Customary techniquesthat use saphenous vein patch or inter-position grafting are not adequate in allcircumstances. Successful outcome re-quires operative exploration in patientswith suspected infection. Removal of thepercutaneous suture closure device anddebridement to normal arterial wall isrecommended in all patients with sus-pected femoral endarteritis, based onpositive intraoperative Gram stains orabnormal appearance of the adjacentfemoral artery. Early success with an au-tologous bolstered repair is reported.Caution is advised when considering theuse of a percutaneous suture closure de-vice in patients with comorbid condi-tions including diabetes, obesity, andpreviously implanted devices.Authors’ Abstract

Abstracts • 1589Volume 14 Number 12