Abstracts of Current Literature

CAROTID INTERVENTIONS

Iyer SS, White CJ, Hopkins LN, andBEACH Investigators. Carotid arteryrevascularization in high-surgical-riskpatients using the carotid WALL-STENT and FilterWire EX/EZ 1-yearoutcomes in the BEACH PivotalGroup. J Amer Coll Cardiol 2008; 51:427–434.

• OBJECTIVES: The multicenter, sin-gle-arm BEACH (Boston Scientific EPI:A Carotid Stenting Trial for High-RiskSurgical Patients) evaluated outcomes inhigh-surgical-risk patients with carotidartery stenosis treated with the CarotidWALLSTENT plus FilterWire EX/EZEmboli Protection System (Boston Scien-tific, Natick, Massachusetts). BACK-GROUND: Carotid artery stent (CAS)placement offers a less invasive alterna-tive for high-risk surgical carotid endar-terectomy (CEA) patients. METHODS:The trial enrolled 480 pivotal patientswho were candidates for carotid revas-cularization but considered high surgicalrisk due to pre-specified anatomic cri-teria and/or medical comorbidities.The primary end point (all stroke,death, or Q-wave myocardial infarc-tion [MI] through 30 days; non–Q-wave MI through 24 h; and ipsilateralstroke or neurologic death through 1year) was compared with a proportion-ally weighted objective performancecriterion (OPC) of 12.6% for publishedsurgical endarterectomy results in sim-ilar patients, plus a pre-specified non-inferiority margin of 4%. RESULTS:Among pivotal patients, 41.2% were athigh surgical risk due to comorbid riskfactors, and 58.8% due to anatomic riskfactors; 76.7% were asymptomatic withflow-limiting carotid stenosis �80%.At 1 year, the composite primary endpoint occurred in 8.9% (40 of 447), witha repeat revascularization rate of 4.7%.With an upper 95% confidence limit of11.5% for the primary composite endpoint, the BEACH trial results met thepre-specified criteria for noninferiorityrelative to the calculated OPC plusnoninferiority margin (16.6%) for his-torical surgical CEA outcomes in simi-lar patients (p � 0.0001 for noninferi-ority). CONCLUSIONS: The BEACHtrial results demonstrate that CAS withthe WALLSTENT plus FilterWire em-

bolic protection is non-inferior (equiv-

alent or better than) to CEA at 1-year inhigh-surgical-risk patients.Authors’ Abstract

Sayeed S, Stanziale SF, Wholey MH,Makaroun MS. Angiographic lesioncharacteristics can predict adverse out-comes after carotid artery stenting. JVasc Surg 2008; 47:81–87.

• BACKGROUND: Carotid arteryangioplasty and stenting (CAS) is anevolving and increasingly common en-dovascular treatment for carotid arterystenosis. Risk factors associated with anincreased incidence of adverse peripro-cedural neurologic outcomes are beingrecognized. The goal of this study was todetermine if certain angiographic lesioncharacteristics were predictive of higherrisks of adverse outcomes. METHODS:A total of 421 patients who underwent429 carotid artery stenting proceduresbetween June 1996 and June 2005 forsymptomatic or asymptomatic carotidstenosis, and in whom preoperative ca-rotid angiograms and follow-up recordswere available for review, were selectedfrom a prospectively maintained data-base. Demographic data and proceduralvariables were recorded, including thepresence or absence of the use of a cere-bral protection device. Angiograms werereviewed for the following carotid lesioncharacteristics: length of lesion, percent-age of stenosis, ostial involvement, le-sion ulceration, calcification, and pres-ence of contralateral carotid occlusion.Periprocedural stroke and 30-day ad-verse event rates (stroke, myocardial in-farction, and death) were recorded foreach patient. RESULTS: The periproce-dural all-stroke rate was 3.7%. Octoge-narians had a higher incidence of 30-dayadverse events at 10.0% vs 3.8% (P �.029). The incidence of periproceduralstroke was increased in lesions �15 mmlong, at 17.0% (8 of 47) vs 2.1% (8 of 382;P � .001), and in ostial centered lesions,7.1% (11 of 154) vs 1.8% (5 of 275; P �.007). Multivariate regression also iden-tified these two variables as indepen-dently associated with 30-day strokerate: lesion length �15 mm (odds ratio[OR], 6.38; 95% confidence interval [CI],35 to 17.29) or ostial involvement (OR,3.12; 95% CI, 3.12 to 8.36). Other vari-ables, including lesion calcification, ul-ceration, degree of stenosis, or presenceof contralateral occlusion, were not asso-ciated with adverse outcomes. When

studied separately, the use of cerebral

protection devices in 241 patients (56%)did not change our observed correla-tions between angiographic characteris-tics and adverse procedural events.CONCLUSIONS: Certain lesion charac-teristics on angiography, such as lengthand ostial location, can predict adverseoutcomes. The indication for CASshould be carefully evaluated in thesecases.Authors’ Abstract

CLINICAL CARE

Shorr AF, Jackson WL, Sherner JH,Moores LK. Differences betweenlow-molecular-weight and unfraction-ated heparin for venous thromboem-bolism prevention following ischemicstroke: a metaanalysis. Chest 2008;133:149–155.

• BACKGROUND: Venous thrombo-embolism (VTE) remains a major causeof morbidity following stroke. The opti-mal form of pharmacologic prophylaxisfollowing stroke is unknown. METH-ODS: We identified randomized trialscomparing unfractionated heparin (UFH) tolow-molecular-weight heparin (LMWH)for VTE prevention in ischemic stroke pa-tients. We focused on the risk for VTE,pulmonary embolism (PE), bleeding, andmortality as a function of the type of agentused for prophylaxis. Findings werepooled with a random-effects model. RE-SULTS: We identified three trials includ-ing 2,028 patients. Two of the studies wereblinded, two studies relied on enoxaparin,while one study utilized certoparin. In twostudies, UFH was administered threetimes a day, while it was administeredtwice daily in the remaining study. Theuse of LMWH was associated with a sig-nificant risk reduction for any VTE (oddsratio [OR], 0.54; 95% confidence interval[CI], 0.41 to 0.70; p � 0.001). Limiting theanalysis to proximal VTEs also indicatedthat LMWHs were superior (OR withLMWH vs UFH, 0.53; 95% CI, 0.37 to 0.75;p � 0.001). LMWH use led to fewer PEs aswell (OR, 0.26; 95% CI, 0.07 to 0.95; p �0.042). There were no differences in ratesof overall bleeding, intracranial hemor-rhage, or mortality based on the type ofagent employed. Restricting the analysisto the reports employing enoxaparin didnot alter our findings. CONCLUSIONS:The prophylactic use of LMWH comparedto UFH following ischemic stroke is asso-ciated with a reduction in both VTE and

PE. This benefit is not associated with an

785

786 • Abstracts May 2008 JVIR

increased incidence of bleeding. Broaderuse of LMWH for VTE prevention afterischemic stroke is warranted.Authors’ Abstract

Calligaro KD, Toursarkissian B,Clagett GP. Guidelines for hospitalprivileges in vascular and endovascu-lar surgery: Recommendations of theSociety for Vascular Surgery. J VascSurg 2008; 47:1–5.

• The Clinical Practice Council of theSociety for Vascular Surgery (SVS) wascharged with providing an updated con-sensus on guidelines for hospital privi-leges in vascular and endovascular sur-gery. One compelling reason to updatethese recommendations is that vascularsurgery as a specialty has continued toevolve with a significant shift towardsendovascular therapies. The Society forVascular Surgery is making the followingfour recommendations concerning guide-lines for hospital privileges for vascularand endovascular surgery. First, anyoneapplying for new hospital privileges toperform vascular surgery should havecompleted an Accreditation Council forGraduate Medical–accredited vascularsurgery residency and should obtainAmerican Board of Surgery certification invascular surgery within 3 years of comple-tion of their training. Second, we reaffirmand provide updated recommendationsconcerning previous established guide-lines for peripheral endovascular proce-dures, thoracic and abdominal aortic en-dograft replacements, and carotid arteryballoon angioplasty and stenting for train-ees and already credentialed physicianswho are adding these new procedures totheir hospital credentials. Third, we en-dorse the Residency Review Committeefor Surgery recommendations regardingopen and endovascular cases duringvascular residency training. Fourth, weendorse the Inter-societal Commissionfor Accreditation of Vascular Laborato-ries (ICAVL) recommendations for non-invasive vascular laboratory interpreta-tions and examinations to become aregistered physician in vascular inter-pretation (RPVI) or a registered vasculartechnologist (RVT).Authors’ Abstract

NONINVASIVECARDIOVASCULAR IMAGING

Lal BK, Hobson RW, Tofighi B, Kapa-dia I, Cuadra S, Jamil Z. Duplex ul-trasound velocity criteria for thestented carotid artery. J Vasc Surg2008; 47:63–73.

• OBJECTIVES: Ultrasound velocitycriteria for the diagnosis of in-stent re-

stenosis in patients undergoing carotid

artery stenting (CAS) are not well estab-lished. In the present study, we testwhether ultrasound velocity measure-ments correlate with increasing degreesof in-stent restenosis in patients under-going CAS and develop customized ve-locity criteria to identify residual steno-sis �20%, in-stent restenosis �50%, andhigh-grade in-stent restenosis �80%.METHODS: Carotid angiograms per-formed at the completion of CAS werecompared with duplex ultrasound(DUS) imaging performed immediatelyafter the procedure. Patients were fol-lowed up with annual DUS imaging andunderwent both ultrasound scans andcomputed tomography angiography(CTA) at their most recent follow-upvisit. Patients with suspected high-gradein-stent restenosis on DUS imaging un-derwent diagnostic carotid angiograms.DUS findings were therefore availablefor comparison with luminal stenosismeasured by carotid angiograms or CTAin all these patients. The DUS protocolincluded peak-systolic (PSV) and end-diastolic velocity (EDV) measurementsin the native common carotid artery(CCA), proximal stent, mid stent, distalstent, and distal internal carotid artery(ICA). RESULTS: Of 255 CAS proce-dures that were reviewed, 39 had con-tralateral ICA stenosis and were ex-cluded from the study. During a meanfollow-up of 4.6 years (range, 1 to 10years), 23 patients died and 64 were lost.Available for analysis were 189 pairs ofultrasound and procedural carotid an-giogram measurements; 99 pairs of ul-trasound and CTA measurements dur-ing routine follow-up; and 29 pairs ofultrasound and carotid angiograms mea-surements during follow-up for sus-pected high-grade in-stent restenosis�80% (n � 310 pairs of observations,ultrasound vs carotid angiograms/CTA). The accuracy of CTA vs carotidangiograms was confirmed (r2 � 0.88) ina subset of 19 patients. Post-CAS PSV (r2

� .85) and ICA/CCA ratios (r2 � 0.76)correlated most with the degree of ste-nosis. Receiver operating characteristicanalysis demonstrated the following op-timal threshold criteria: residual stenosis’20% (PSV �150 cm/s and ICA/CCAratio �2.15), in-stent restenosis �50%(PSV �220 cm/s and ICA/CCA ratio�2.7), and in-stent restenosis �80%(PSV 340 cm/s and ICA/CCA ratio�4.15). CONCLUSIONS: Progressivelyincreasing PSV and ICA/CCA ratioscorrelate with evolving restenosis withinthe stented carotid artery. Ultrasoundvelocity criteria developed for native ar-teries overestimate the degree of in-stentrestenosis encountered. These changespersist during long-term follow-up and

across all grades of in-stent restenosis

after CAS. The proposed new velocitycriteria accurately define residual steno-sis �20%, in-stent restenosis �50%, andhigh-grade in-stent restenosis �80% inthe stented carotid artery.Authors’ Abstract

Zhou W, Felkai DD, Evans M, etal. Ultrasound criteria for severe in-stent restenosis following carotid ar-tery stenting. J Vasc Surg 2008; 47:74–80.

• PURPOSE: In-stent restenosis (ISR)is a known complication following ca-rotid artery stenting (CAS). However,ultrasound criteria determining ISR arenot well established. We evaluated alter-native ultrasound velocity criteria for�70% ISR in our institution. METHODS:Clinical records of 256 patients undergo-ing 282 consecutive CAS proceduresover a 42-month period were reviewed.Follow-up ultrasounds were availablefor analysis in 237 patients. Selective an-giograms and repeat interventions wereperformed for �70% ISR. Ultrasoundcriteria including peak systolic velocity(PSV), end diastolic velocity (EDV), andinternal carotid to common carotid ar-tery ratios (ICA/CCA) were examined.The sensitivity, specificity, positive pre-dictive value (PPV), and negative pre-dictive value (NPV) were calculated forPSV (200, 250, 300, 350, and 400 cm/s),EDV (70, 80, 90, 100 cm/s), and CCA/ICA (3, 3.5, 4, 4.5, 5). RESULTS: Twenty-two carotid angiograms were performedand 18 lesions had confirmations of�70% ISR in 11 patients including priorCEA in five patients and neck irradiationin two patients. Receiver operator char-acteristics (ROC) was analyzed for PSV,EDV, and CCA/ICA ratio. For 70% orgreater angiographic ISR, PSV � 300cm/s correlated to a 94% sensitivity, 50%specificity, 90% positive predictive value(PPV), and 67% negative predictivevalue (NPV); EDV � 90 cm/s correlatedto an 89% sensitivity, 100% specificity,100% PPV, and 67% NPV; and ICA/CCA � 4 had a 94.4% sensitivity, 75%specificity, 94% PPV, and 75% NPV. Asignificant color flow disturbance wasdetected in one patient who did not meetthe aforementioned ultrasound velocitycriteria. Further statistical analysisshowed that an EDV of 90 cm/s pro-vided the best discriminant value. CON-CLUSIONS: Our study demonstratedthat PSV � 300 cm/s, EDV � 90 cm/s,and ICA/CCA � 4 correlated well with�70% ISR. Although still rudimentary,these velocity criteria combined withcolor flow patterns can reliably predictsevere ISR in our vascular laboratory.However, due to the relatively infre-quent cases of severe ISR following CAS,

a multicentered study is warranted to

Abstracts • 787Volume 19 Number 5

establish standard post-CAS ultrasoundsurveillance criteria for severe ISR.Authors’ Abstract

Ren X, Wang W, Zhang X, Pu Y, JiangT, Li C. Clinical study and compari-son of magnetic resonance angiogra-phy (MRA) and angiography diagno-sis of blunt vertebral artery injury.J Trauma 2007; 63:1249–1253.

• BACKGROUND: Two-dimensionaltime-of-flight (2D TOF) magnetic reso-nance angiography (MRA) is the majormethod for detecting vertebral artery in-jury (VAI). But there is still controversyover MRA’s accuracy in detecting VAI ofvarying degrees, a technique for whichthere are no animal studies found in theliterature. The purpose of this study wasto evaluate the value of MRA for detect-ing vertebral artery injury in cervical spi-nal trauma patients and to conduct acomparative study of MRA and angiog-raphy of blunt vertebral artery injury inan animal experiment. METHODS: Inclinical study, 319 patients with closecervical spinal trauma underwent 2DTOF MRA prospective examination. Thestrike-induced flexion injury model ofthe cervical spine was constructed basedon animal experiment with 14 adultdogs. X-ray studies were performed im-mediately after injury. 2D TOF MRA andangiography were formed within 24 to36 hours. RESULTS: Vertebral artery in-jury was detected by 2D TOF MRA in 52of the 319 patients. Of the 52 patients,there were 51 unilateral vertebral arteryinjuries, including 22 injuries on the leftvertebral artery and 29 on the right ver-tebral artery. One patient sustained bi-lateral vertebral artery injury. Sevendogs had unilateral VAI (5 on the left, 2on the right), and two dogs had narrower-than-normal left vertebral arteries onMRA examination. Angiography showedocclusion exactly in seven dogs with uni-lateral vertebral artery injury detected byMRA and no abnormal findings were de-tected in five dogs without vertebral arteryinjury on MRA. In two dogs with incom-plete left vertebral artery flow-related en-hancement on MRA, angiography showedocclusion in one dog and no vertebral in-jury with normal image in the other.CONCLUSIONS: 2D TOF MRA is an ef-fective detection method of blunt vertebralartery injury but it might not be able todifferentiate spasm, a small disruption ofintima, from others under certain condi-tions.Authors’ Abstract

Anderson DR, Kahn SR, Rodger MA,et al. Computed tomographic pulmo-nary angiography vs ventilation-perfu-sion lung scanning in patients with

suspected pulmonary embolism. A

randomized controlled trial. JAMA2007; 298:2743–2753.

• CONTEXT: Ventilation-perfusion(V̇/Q̇) lung scanning and computedtomographic pulmonary angiography(CTPA) are widely used imaging proce-dures for the evaluation of patients withsuspected pulmonary embolism. Venti-lation-perfusion scanning has beenlargely replaced by CTPA in many cen-ters despite limited comparative formalevaluations and concerns about CTPA’slow sensitivity (ie, chance of missingclinically important pulmonary embuli).OBJECTIVES: To determine whetherCTPA may be relied upon as a safe al-ternative to V̇/Q̇ scanning as the initialpulmonary imaging procedure for ex-cluding the diagnosis of pulmonary em-bolism in acutely symptomatic patients.DESIGN, SETTING, AND PARTICI-PANTS: Randomized, single-blindednoninferiority clinical trial performed at4 Canadian and 1 US tertiary care cen-ters between May 2001 and April 2005and involving 1417 patients consideredlikely to have acute pulmonary embo-lism based on a Wells clinical modelscore of 4.5 or greater or a positive D-dimer assay result. INTERVENTION:Patients were randomized to undergoeither V̇/Q̇ scanning or CTPA. Patientsin whom pulmonary embolism was con-sidered excluded did not receive anti-thrombotic therapy and were followedup for a 3-month period. MAIN OUT-COME MEASURE: The primary out-come was the subsequent developmentof symptomatic pulmonary embolism orproximal deep vein thrombosis in pa-tients in whom pulmonary embolismhad initially been excluded. RESULTS:Seven hundred one patients were ran-domized to CTPA and 716 to V̇/Q̇ scan-ning. Of these, 133 patients (19.2%) in theCTPA group vs 101 (14.2%) in the V̇/Q̇scan group were diagnosed as havingpulmonary embolism in the initial eval-uation period (difference, 5.0%; 95% con-fidence interval [CI], 1.1% to 8.9%) andwere treated with anticoagulant therapy.Of those in whom pulmonary embolismwas considered excluded, 2 of 561 pa-tients (0.4%) randomized to CTPA vs 6of 611 patients (1.0%) undergoing V̇/Q̇scanning developed venous thromboem-bolism in follow-up (difference, –0.6%;95% CI, –1.6% to 0.3%) including onepatient with fatal pulmonary embolismin the V̇/Q̇ group. CONCLUSIONS: Inthis study, CTPA was not inferior toV̇/Q̇ scanning in ruling out pulmonaryembolism. However, significantly morepatients were diagnosed with pulmo-nary embolism using the CTPA ap-proach. Further research is required to

determine whether all pulmonary em-

boli detected by CTPA should be man-aged with anticoagulant therapy.Authors’ Abstract

Ropers U, Ropers D, Pflederer T, etal. Influence of heart rate on the di-agnostic accuracy of dual-source com-puted tomography coronary angiogra-phy. J Amer Coll Cardiol 2007;50:2393–2398.

• OBJECTIVES: We evaluated the in-fluence of heart rate on image qualityand diagnostic accuracy of dual-sourcecomputed tomography (DSCT) coronaryangiography. BACKGROUND: Multide-tector computed tomography (MDCT)coronary angiography has demonstratedan inverse relationship between heartrate and image quality. Dual-source CTprovides a higher temporal resolution.METHODS: One hundred patients werestudied by DSCT (DEFINITION, Sie-mens Medical Solutions, Forchheim,Germany). A contrast-enhanced volumedataset was acquired (two tubes, 120 kV,400 mAs/rot, collimation 64 � 0.6 mm).Datasets were evaluated concerning thepresence of significant coronary stenosesand validated against invasive coronaryangiography. RESULTS: In 44 patientswith a heart rate �65 beats/min, 566 of616 coronary segments were evaluable(92%), whereas in 56 patients with aheart rate �65 beats/min, 777 of 778 cor-onary segments were evaluable (100%, p� 0.001). On a per-patient basis, 93% ofpatients (�65 beats/min) and 100% ofpatients (�65 beats/min) were consid-ered evaluable. By classifying unevalu-able segments as positive for stenosis,per-patient sensitivity was 95% (19 of 20)for heart rates �65 beats/min and 100%(22 of 22) for heart rates �65 beats/min.Specificity was 87% (21 of 24) versus 76%(26 of 34), and overall diagnostic accu-racy was 91% (40 of 44) versus 86% (48 of56). None of these differences were sta-tistically significant. Similarly, no differ-ence in diagnostic accuracy was found inper-vessel and -segment analyses. CON-CLUSIONS: In 100 patients studiedwithout beta-blocker pre-medication,DSCT demonstrated slightly lower per-segment evaluability for high heart ratesbut no decrease in diagnostic accuracyfor the detection of coronary artery ste-noses.Authors’ Abstract

ONCOLOGIC INTERVENTIONS

Khan MR. Poon RTP, Ng KK, etal. Comparison of percutaneous andsurgical approaches for radiofre-quency ablation of small and mediumhepatocellular carcinoma. Arch Surg

2007; 142:1136–1143.

788 • Abstracts May 2008 JVIR

• HYPOTHESIS: Radiofrequency ab-lation (RFA) for hepatocellular carci-noma (HCC) can be performed by per-cutaneous or surgical approach. Tumorsize is an important consideration whiledeciding the treatment approach. DE-SIGN: Case series with prospective datacollection. SETTING: A tertiary referralcenter. PATIENTS: A total of 228 pa-tients who underwent RFA of small (� 3cm; n � 155) and medium (3.1-5 cm; n �73) HCC by percutaneous or surgical ap-proach. MAIN OUTCOME MEASURES:Complete ablation rate, post-RFA com-plications, treatment-related mortality,and overall and disease-free survival.RESULTS: In patients with small HCC,the complete ablation rate was 95% withboth approaches (P � .99). Complicationrate (P � .001) and hospital stay (P �.001) were higher with the surgical ap-proach. One-year and 3-year survivalrates were 91% and 71%, respectively, inthe percutaneous group, and 89% and57%, respectively, in the surgical group(P � .30). In patients with medium HCC,the complete ablation rate was similarbetween the surgical and the percutane-ous groups (92% vs 95%; P � .48), andthe complication rate was also comparable(P � .17). The 1-year and 3-year survivalrates were 92% and 68%, respectively, inthe surgical group, significantly superiorto the corresponding rates of 81% and 42%in the percutaneous group (P � .03). CON-CLUSIONS: In patients with small HCC,the percutaneous approach achieved sim-ilar tumor control with lower morbiditycompared with the surgical approach andshould be the preferred approach pro-vided that tumor location is suitable. Formedium HCC, the surgical approachseems to achieve better overall survivaland may be a preferred option.Authors’ Abstract

Cho YK, Chung JW, Kim JK. Com-parison of 7 staging systems for pa-tients with hepatocellular carcinomaundergoing transarterial chemoembo-lization. Cancer 2007; 112:352–361.

• BACKGROUND: Many liver stag-ing systems have been proposed for pa-tients with hepatocellular carcinomaafter locoregional therapy; however,controversies persist regarding whichsystem is the best. In this study, the au-thors compared the performance of 7staging systems in a cohort of patientswith hepatocellular carcinoma who un-derwent transarterial chemoemboliza-tion. METHODS: In total, 131 patientswith hepatocellular carcinoma who un-derwent transarterial chemoemboliza-tion between August 1998 and February2005 were included in the study. Demo-graphic, laboratory, and tumor charac-

teristics were determined at diagnosis

and before therapy. At the time of cen-sorship, 109 patients had died (83.2%).Predictors of survival were identified byusing the Cox proportional hazardsmodel. The likelihood-ratio chi-squarestatistic and the Akaike Information Cri-terion were calculated for 7 prognosticsystems to evaluate their discriminatoryability. Comparisons of the survival ratebetween each stage were performed toevaluate the monotonicity of the gradi-ents using Kaplan-Meier estimation andthe log-rank test. RESULTS: The 5-yearsurvival rate for the entire cohort was13.6%. The independent predictors ofsurvival were serum albumin level (�3.4g/dL), the presence of ascites, serum�-fetoprotein level (�60 ng/mL), andportal or hepatic vein tumor thrombosis(P � .001, P � .001, P � .004, and P �.000, respectively). The Cancer of theLiver Italian Program classification sys-tem was superior to the other 6 prognos-tic systems regarding discriminatoryability and the monotonicity of the gra-dients. CONCLUSIONS: In this compar-ison of many staging systems, the Can-cer of Liver Italian Program systemprovided the best prognostic stratifica-tion for a cohort the patients with hepa-tocellular carcinoma who underwenttransarterial chemoembolization.Authors’ Abstract

Tse RV, Hawkins M, Lockwood G, etal. Phase I study of individualizedstereotactic body radiotherapy forhepatocellular carcinoma and intrahe-patic cholangiocarcinoma. J Clin On-col 2008; 26:657–664.

• PURPOSE: To report outcomes of aphase I study of individualized stereo-tactic body radiotherapy treatment(SBRT) for unresectable hepatocellularcarcinoma (HCC) and intrahepatic chol-angiocarcinoma (IHC). PATIENTS ANDMETHODS: Patients with unresectableHCC or IHC, and who are not suitablefor standard therapies, were eligible forsix-fraction SBRT during 2 weeks. Radi-ation dose was dependent on the vol-ume of liver irradiated and the estimatedrisk of liver toxicity based on a normaltissue complication model. Toxicity riskwas escalated from 5% to 10% and 20%,within three liver volume–irradiatedstrata, provided at least three patientswere without toxicity at 3 months afterSBRT. RESULTS: Forty-one patientswith unresectable Child-Pugh A HCC (n� 31) or IHC (n � 10) completed six-fraction SBRT. Five patients (12%) hadgrade 3 liver enzymes at baseline. Themedian tumor size was 173 mL (9 to1,913 mL). The median dose was 36.0 Gy(24.0 to 54.0 Gy). No radiation-inducedliver disease or treatment-related grade

4/5 toxicity was seen within 3 months

after SBRT. Grade 3 liver enzymes wereseen in five patients (12%). Two patients(5%) with IHC developed transient bili-ary obstruction after the first few frac-tions. Seven patients (five HCC, twoIHC) had decline in liver function fromChild-Pugh class A to B within 3 monthsafter SBRT. Median survival of HCC andIHC patients was 11.7 months (95% CI,9.2 to 21.6 months) and 15.0 months(95% CI, 6.5 to 29.0 months), respec-tively. CONCLUSION: Individualizedsix-fraction SBRT is a safe treatment forunresectable HCC and IHC.Authors’ Abstract

Bandi G, Hedican SP, NakadaSY. Current practice patterns in theuse of ablation technology for themanagement of small renal masses atacademic centers in the United States.Urology 2008; 71:113–117.

• OBJECTIVES: To determine thecurrent practice patterns in the use ofablation technology for the managementof small renal masses at academic cen-ters in the United States. METHODS: Anemail survey was sent to 112 academicurologists subspecializing in minimallyinvasive management of renal cancer.The survey consisted of 13 questions and4 clinical scenarios pertaining to the useof ablation technology. The responseswere then tabulated and analyzed to de-termine practice trends. RESULTS: Theoverall response rate was 62%. Ablationwas offered by 93% of the academic urol-ogy centers and cryoablation was morefrequently used (79%) than radiofre-quency ablation (55%). Lack of sufficientefficacy data was the most prevalent rea-son (80%) for not offering ablation. Themaximum size limit for offering ablationwas 4 cm by 55% and 3 cm by 34% of therespondents. A collaborative approachusing both radiologist and urologist wasmost commonly used (51%). Most urol-ogists (68%) used both laparoscopic andpercutaneous technique, depending onthe tumor and adjacent organ location.Intraoperative ultrasound was univer-sally used during the laparoscopic tech-nique and was usually performed by theurologist (95%). Computed tomographicscan was the most frequently used im-aging modality for percutaneous abla-tion (78%) and for surveillance of recur-rent disease (81%). In a younger, healthypatient, most urologists recommend extir-pative approach for the management of asmall renal mass, whereas laparoscopic-assisted ablation was most commonly rec-ommended for an elderly patient with co-morbidities. CONCLUSIONS: Our surveysuggests that laparoscopic and percutane-

ous ablation is offered by the majority of

Abstracts • 789Volume 19 Number 5

academic centers for carefully selected pa-tients.Authors’ Abstract

PERIPHERAL ARTERIALINTERVENTIONS—STENTS

Marroquin OC, Selzer F, MulukutlaSR, et al. A comparison of bare-metal and drug-eluting stents for off-label indications. NEJM 2008; 358:342–352.

• BACKGROUND: Recent reportssuggest that off-label use of drug-elutingstents is associated with an increased in-cidence of adverse events. Whether theuse of bare-metal stents would yield dif-ferent results is unknown. METHODS:We analyzed data from 6551 patients inthe National Heart, Lung, and Blood In-stitute Dynamic Registry according towhether they were treated with drug-eluting stents or bare-metal stents andwhether use was standard or off-label.Patients were followed for 1 year for theoccurrence of cardiovascular events anddeath. Off-label use was defined as usein restenotic lesions, lesions in a bypassgraft, left main coronary artery disease,or ostial, bifurcated, or totally occludedlesions, as well as use in patients with areference-vessel diameter of less than 2.5mm or greater than 3.75 mm or a lesionlength of more than 30 mm. RESULTS:Off-label use occurred in 54.7% of allpatients with bare-metal stents and48.7% of patients with drug-elutingstents. As compared with patients withbare-metal stents, patients with drug-eluting stents had a higher prevalence ofdiabetes, hypertension, renal disease,previous percutaneous coronary inter-vention and coronary-artery bypassgrafting, and multivessel coronary ar-tery disease. One year after intervention,however, there were no significant dif-ferences in the adjusted risk of death ormyocardial infarction in patients withdrug-eluting stents as compared withthose with bare-metal stents, whereasthe risk of repeat revascularization wassignificantly lower among patients withdrug-eluting stents. CONCLUSIONS:Among patients with off-label indica-tions, the use of drug-eluting stents wasnot associated with an increased risk ofdeath or myocardial infarction but wasassociated with a lower rate of repeatrevascularization at 1 year, as comparedwith bare-metal stents. These findingssupport the use of drug-eluting stentsfor off-label indications.

Authors’ Abstract

PERIPHERAL ARTERIALINTERVENTIONS—STENT-GRAFTS

Schermerhorn ML, O’Malley AJ, Jhav-eri A, Cotterill P, Pomposelli F,Landon BE. Endovascular vs. openrepair of abdominal aortic aneurysmsin the Medicare population. NEJM2008; 358:464–474.

• BACKGROUND: Randomized tri-als have shown reductions in periopera-tive mortality and morbidity with endo-vascular repair of abdominal aorticaneurysm, as compared with open sur-gical repair. Longer-term survival rates,however, were similar for the two pro-cedures. There are currently no long-term, population-based data from thecomparison of these strategies. METH-ODS: We studied perioperative rates ofdeath and complications, long-term sur-vival, rupture, and reinterventions afteropen as compared with endovascular re-pair of abdominal aortic aneurysm inpropensity-score–matched cohorts ofMedicare beneficiaries undergoing re-pair during the 2001–2004 period, withfollow-up until 2005. RESULTS: Therewere 22,830 matched patients undergo-ing open repair of abdominal aortic an-eurysm in each cohort. The average ageof the patients was 76 years, and approx-imately 20% were women. Perioperativemortality was lower after endovascularrepair than after open repair (1.2% vs.4.8%, P�0.001), and the reduction inmortality increased with age (2.1% dif-ference for those 67 to 69 years old vs.8.5% for those 85 years or older,P�0.001). Late survival was similar inthe two cohorts, although the survivalcurves did not converge until after 3years. By 4 years, rupture was morelikely in the endovascular-repair cohortthan in the open-repair cohort (1.8% vs.0.5%, P�0.001), as was reintervention re-lated to abdominal aortic aneurysm(9.0% vs. 1.7%, P�0.001), although mostreinterventions were minor. In contrast,by 4 years, surgery for laparotomy-re-lated complications was more likelyamong patients who had undergoneopen repair (9.7%, vs. 4.1% among thosewho had undergone endovascular re-pair; P�0.001), as was hospitalizationwithout surgery for bowel obstruction orabdominal-wall hernia (14.2% vs. 8.1%,P�0.001). CONCLUSIONS: As com-pared with open repair, endovascular re-pair of abdominal aortic aneurysm is as-sociated with lower short-term rates ofdeath and complications. The survivaladvantage is more durable among olderpatients. Late reinterventions related toabdominal aortic aneurysm are morecommon after endovascular repair but are

balanced by an increase in laparotomy-

related reinterventions and hospitaliza-tions after open surgery.Authors’ Abstract

VENOUS INTERVENTIONS—THROMBOEMBOLIC DISEASE

Ageno W, Becattini C, Brighton T,Selby R, Kamphuisen PW. Cardio-vascular risk factors and venousthromboembolism. A meta-analysis.Circulation 2008; 117:93–102.

• BACKGROUND: The concept thatvenous thromboembolism (VTE) andatherosclerosis are 2 completely distinctentities has recently been challenged be-cause patients with VTE have moreasymptomatic atherosclerosis and morecardiovascular events than control sub-jects. We performed a meta-analysis toassess the association between cardio-vascular risk factors and VTE. METH-ODS AND RESULTS: Medline and EM-BASE databases were searched toidentify studies that evaluated the prev-alence of major cardiovascular risk fac-tors in VTE patients and control subjects.Studies were selected using a priori de-fined criteria, and each study was re-viewed by 2 authors who abstracteddata on study characteristics, study qual-ity, and outcomes. Odds ratios orweighted means and 95% confidence in-tervals (CIs) were then calculated andpooled using a random-effects model.Statistical heterogeneity was evaluatedthrough the use of �2 and I2 statistics.Twenty-one case-control and cohortstudies with a total of 63 552 patientsmet the inclusion criteria. Comparedwith control subjects, the risk of VTEwas 2.33 for obesity (95% CI, 1.68 to3.24), 1.51 for hypertension (95% CI, 1.23to 1.85), 1.42 for diabetes mellitus (95%CI, 1.12 to 1.77), 1.18 for smoking (95%CI, 0.95 to 1.46), and 1.16 for hypercho-lesterolemia (95% CI, 0.67 to 2.02).Weighted mean high-density lipoproteincholesterol levels were significantlylower in VTE patients, whereas no dif-ference was observed for total and low-density lipoprotein cholesterol levels.Significant heterogeneity among studieswas present in all subgroups except forthe diabetes mellitus subgroup. Higher-quality studies were more homoge-neous, and significant associations re-mained unchanged. CONCLUSIONS:Cardiovascular risk factors are associ-ated with VTE. This association is clini-cally relevant with respect to individualscreening, risk factor modification, andprimary and secondary prevention ofVTE. Prospective studies should furtherinvestigate the underlying mechanismsof this relationship.

Authors’ Abstract

790 • Abstracts May 2008 JVIR

Muñoz FJ, Mismetti P, Poggio R, et al,for the RIETE Investigators. Clinicaloutcome of patients with upper-ex-tremity deep vein thrombosis. Resultsfrom the RIETE Registry. Chest 2008;133:143–148.

• BACKGROUND: There is little in-formation on the clinical outcome of pa-tients with upper-extremity deep veinthrombosis (DVT). METHODS: RIETE isan ongoing registry of consecutive pa-tients with objectively confirmed, symp-tomatic, acute DVT or pulmonary embo-lism (PE). In this analysis, we analyzedthe demographic characteristics, treat-ment, and 3-month outcome of all pa-

tients with DVT in the arm. RESULTS:

Of the 11,564 DVT patients enrolled, 512patients (4.4%) had arm DVT. They pre-sented less often with clinically overt PE(9.0% vs 29%; odds ratio, 0.24; 95% con-fidence interval [CI], 0.18 to 0.33) thanthose with lower-limb DVT, but their3-month outcome was similar. Of the 512patients with arm DVT, 196 patients(38%) had cancer and 228 patients (45%)had catheter-related DVT. During fol-low-up, those with cancer DVT had anincreased incidence of major bleeding(4.1% vs 0.9%; odds ratio, 4.4; 95% CI, 1.2to 21), recurrent venous thromboembo-lism (6.1% vs 2.8%; odds ratio, 2.2; 95%CI, 0.91 to 5.6; p � 0.04), and death (22%

vs 3.5%; odds ratio, 7.8; 95% CI, 4.0 to

16). Thirty patients had the compositeevent of recurrent DVT, symptomaticPE, or major bleeding. They were signif-icantly older, more often had cancer, andpresented more frequently with symp-tomatic PE on hospital admission. Onmultivariate analysis, only cancer pa-tients with arm DVT had an increasedrisk for the composite event (odds ratio,3.0; 95% CI, 1.4 to 6.4). CONCLUSIONS:At presentation, patients with arm DVThave less often clinically overt PE thanthose with lower-limb DVT, but their3-month outcome is similar. Among pa-tients with arm DVT, those with cancerhave the worse outcome.

Authors’ Abstract