PERIPHERAL ARTERIALINTERVENTIONS

Aquarius AE, Denollet J. Impairedhealth status and invasive treatmentin peripheral arterial disease: a pro-spective 1-year follow-up study. JVasc Surg 2005;41(3):436–442.

• OBJECTIVE: It has been argued thathealth status and quality of life (QOL)should be taken into account in the treat-ment policy of patients with peripheralarterial disease (PAD). In cardiac pa-tients, it has been shown that poor per-ceived health status is an independentpredictor of mortality and hospitaliza-tion. We therefore examined (1) the roleof health status, QOL, and clinical in-dexes of disease severity as determi-nants of invasive treatment in patientswith PAD and (2) the effect of invasivetreatment on health status and QOL.METHODS: At their first visit, patientscompleted the RAND 36-item HealthSurvey and World Health OrganizationQuality of Life assessment instrumentquestionnaires to assess health statusand QOL, respectively. During the1-year follow-up period, data concern-ing hospitalization were derived fromthe patients’ medical files. Furthermore,patients completed the RAND 36 andthe World Health Organization Qualityof Life assessment instrument again at1-year follow-up. The setting was a vas-cular outpatient clinic of a teaching hos-pital in Tilburg, The Netherlands; partic-ipants were 200 consecutive patientsnewly diagnosed with intermittent clau-dication, a common expression of PAD.Diagnosis was based on history, physicalexamination, treadmill walking distance,and ankle-brachial pressure indexes.Main outcome measures were (1) inva-sive treatment of PAD that took placeduring the 1-year follow-up, derivedfrom the patients’ medical files, and (2)health status and QOL after 1 year offollow-up. RESULTS: After 1 year of fol-low-up, 107 patients (53.5%) were eventfree, whereas 77 patients (38.5%) hadbeen hospitalized for invasive treatmentof PAD. Sixteen patients (8%) were hos-pitalized for other cardiovascular rea-sons. In a multivariate logistic regressionmodel, age (odds ratio [OR], 0.95; 95%confidence interval [CI], 0.91–0.99; P �.024), pain-free walking distance (OR,2.74; 95% CI, 1.05–7.17; P � .04), andphysical functioning (OR, 4.46; 95% CI,

1.79–11.12; P � .001) were independentpredictors of invasive treatment of in-termittent claudication. After 1 year offollow-up, patients who were treated in-vasively experienced a significant im-provement in their physical functioning(P � .004), role limitations due to emo-tional problems (P � .018), and bodilypain (P � .026). CONCLUSIONS: Pa-tients with poor self-reported physicalfunctioning, limited walking distance,and a younger age were likely to betreated invasively. The physician’s clini-cal judgment about when to interveneadequately reflects the patient’s ownopinion about his or her health status.Invasive treatment led to a significantimprovement in patients’ health status.These findings indicate the effectivenessof the strategy to include patients’ per-ceived physical functioning into the pro-cess of clinical decision-making.Authors’ Abstract

Angioplasty

Walker SR, Beasley A. Delayed re-canalization angioplasty after trau-matic occlusion of the superficial fem-oral and popliteal artery. J Vasc Surg2005;41(3):544–547.

• The management of patients pre-senting with limb-threatening ischemiaafter lower limb trauma poses particulardemands on vascular surgeons. We de-scribe two patients with lower limb ar-terial occlusions who presented withlimb-threatening ischemia 8 years and 3years after trauma to the lower limb.They were successfully managed by re-canalization angioplasty, with a goodearly clinical outcome after 5 monthsand 3 months, respectively. Recanaliza-tion angioplasty is a useful adjunct in themanagement of ischemic limbs aftertrauma.Authors’ Abstract

Kudo T, Chandra FA. The effective-ness of percutaneous transluminal an-gioplasty for the treatment of criticallimb ischemia: a 10-year experience. JVasc Surg 2005;41(3):423–435; discus-sion 435.

• OBJECTIVE: To determine the effi-cacy, safety, and long-term results, in-cluding continued clinical improvementand limb salvage, of percutaneous trans-luminal angioplasty (PTA) in patientswith critical limb ischemia (CLI). METH-ODS: From August 1993 to March 2004,

138 limbs in 111 patients with CLI (restpain in 62 [45%] and ulcer/gangrene in76 [55%]) were treated by PTA. In iliaclesions, stents were placed selectively forprimary PTA failure: residual stenosis(�30%) or pressure gradient (�5 mmHg). Stent placement was limited in in-frainguinal lesions. The most distal af-fected arteries treated with angioplastywere the iliac artery in 45 limbs (33%;iliac group), the femoropopliteal arteryin 41 limbs (30%; FP group), and tibialarteries in 52 limbs (37%; BK group). Allanalysis was performed according to anintent-to-treat basis. Reporting stan-dards of the Society for Vascular Surgeryand the International Society for Cardio-vascular Surgery were followed to eval-uate initial success, and late follow-upstatus was evaluated with the Kaplan-Meier method. Patency was evaluatedby using ultrasound scanning and ankle-brachial pressure index measurement.RESULTS: There was one (0.9%) periop-erative death. Twenty stents were placedselectively in 14 iliac arteries. Mean fol-low-up was 14.7 months (range, 1-75mo). Overall, initial technical and clinicalsuccess rates were 96.4% and 92.8%, re-spectively. The cumulative primary, as-sisted primary, and secondary patency;continued clinical improvement; andlimb salvage rates � SE at 5 years were31.4% � 10.4%, 75.5% � 5.7%, 79.6% �5.5%, 36.1% � 10.0%, and 89.1% � 4.0%,respectively. In each subgroup, the pri-mary, assisted primary, and secondarypatency; continued clinical improve-ment; and limb salvage rates at 3 yearswere 51.6%, 94.7%, 97.8%, 65.1%, and95.0%, respectively, in the iliac group;49.4%, 72.2%, 76.4%, 57.4%, and 92.7%,respectively, in the FP group; and 23.5%,41.8%, 46.1%, 51.1%, and 77.3%, respec-tively, in the BK group. Of the 12 pre-dictable variables, hypertension, multi-ple segment lesions, more distal lesions,and TransAtlantic Inter-Society Consen-sus classification type D were significantindependent risk factors for the out-comes (P � .05; univariate log-rank testand Cox regression multivariate analy-sis). CONCLUSIONS: PTA is a feasible,safe, and effective procedure for thetreatment of CLI. The high limb salvagerate is attributed to the high assisted pri-mary and secondary patency rates de-spite the low primary patency rate. An-gioplasty can be the primary choice for

Abstracts of Current Literature

1159

the treatment of CLI due to iliac andinfrainguinal arterial occlusive disease.Authors’ Abstract

CAROTID INTERVENTIONS

Parodi JC, Ferreira LM. Cerebral pro-tection during carotid stenting usingflow reversal. J Vasc Surg 2005;41(3):416–422.

• BACKGROUND: Carotid angio-plasty and stenting can be used in strokeprevention in high-risk patients. As em-bolic complications can occur during ca-rotid angioplasty and stenting, a devicewas developed to protect from cerebralembolization. METHODS: Between Sep-tember 1999 and May 2002, carotid an-gioplasty and stenting was performed in100 patients (84 men; mean age, 69.2 y)with symptomatic (26%) or asymptom-atic (74%) severe carotid artery stenosis.Wallstents were used in all cases withselective predilatation. Cerebral protec-tion devices (Parodi Anti-Emboli System[PAES], ArteriA, San Francisco, Calif)were used in all patients. All patientswere evaluated by a neurologist, bothbefore and after the procedure. Accord-ing to the criteria set forth by the largetrials, the occurrence of minor, major orfatal stroke and myocardial infarction(end points) within 30 days and fol-low-up were determined as end points.Data were collected prospectively. RE-SULTS: The overall perioperative strokeand death rate was 3% (1 noncorre-sponding minor stroke, 1 hemorrhagicstroke, and 1 cardiac event). Four pa-tients developed postoperative transientneurologic events (three related to he-modynamic instability and the fourthdue to postoperative embolization). Theoverall technical success rate for carotidangioplasty (protection device placed inposition percutaneously) was 99%.CONCLUSION: The efficacy and safetyof carotid angioplasty and stenting withPAES are confirmed. This innovativeprotection device may prevent the de-bris released by angioplasty from enter-ing the cerebral circulation. Further in-vestigation is warranted.Authors’ Abstract

Chimowitz MI, Lynn MJ. Compari-son of warfarin and aspirin for symp-tomatic intracranial arterial stenosis.N Engl J Med 2005;352(13):1305–1316.

• BACKGROUND: Atheroscleroticintracranial arterial stenosis is an impor-tant cause of stroke. Warfarin is com-monly used in preference to aspirin forthis disorder, but these therapies havenot been compared in a randomizedtrial. METHODS: We randomly assignedpatients with transient ischemic attack or

stroke caused by angiographically veri-fied 50 to 99% stenosis of a major intra-cranial artery to receive warfarin (targetinternational normalized ratio, 2.0–3.0)or aspirin (1300 mg per day) in a double-blind, multicenter clinical trial. The pri-mary end point was ischemic stroke,brain hemorrhage, or death from vascu-lar causes other than stroke. RESULTS:After 569 patients had undergone ran-domization, enrollment was stopped be-cause of concerns about the safety of thepatients who had been assigned to re-ceive warfarin. During a mean follow-upperiod of 1.8 years, adverse events in thetwo groups included death (4.3% in theaspirin group vs. 9.7% in the warfaringroup; hazard ratio for aspirin relative towarfarin, 0.46; 95% confidence interval,0.23–0.90; P � .02), major hemorrhage(3.2% vs. 8.3%, respectively; hazard ra-tio, 0.39; 95% confidence interval, 0.18–0.84; P � .01), and myocardial infarctionor sudden death (2.9% vs. 7.3%, respec-tively; hazard ratio, 0.40; 95% confidenceinterval, 0.18–0.91; P � .02). The rate ofdeath from vascular causes was 3.2% inthe aspirin group and 5.9% in the war-farin group (P � .16); the rate of deathfrom nonvascular causes was 1.1% and3.8%, respectively (P � .05). The primaryend point occurred in 22.1% of the pa-tients in the aspirin group and 21.8% ofthose in the warfarin group (hazard ra-tio, 1.04; 95% confidence interval, 0.73–1.48; P � .83). CONCLUSIONS: Warfa-rin was associated with significantlyhigher rates of adverse events and pro-vided no benefit over aspirin in this trial.Aspirin should be used in preference towarfarin for patients with intracranialarterial stenosis.Authors’ Abstract

Ascher E, Marks NA. Duplex-as-sisted internal carotid artery balloonangioplasty and stent placement: anovel approach to minimize or elimi-nate the use of contrast material. JVasc Surg 2005;41(3):409–415.

• BACKGROUND: Carotid arteryballoon angioplasty and stenting (CBAS)is emerging as an acceptable alternativeto carotid endarterectomy in selectedhigh-risk patients. Conversely, patientswith pre-existing renal impairment, dia-betes, or both may be harmed by thenephrotoxic contrast agents requiredduring CBAS. We attempted to limit oreliminate the use of contrast materialduring CBAS. METHODS: Eighteen pa-tients with severe carotid stenoses(�70%) underwent CBAS at our institu-tion over the last 12 months with duplexscan-assisted CBAS. Of these, 12 wereprimary procedures, and 6 were per-formed for carotid re-stenosis. Fourteenpatients (78%) were neurologically

asymptomatic. The average age of thesepatients was 75 � 11 years (range, 44–92y). Hypertension, chronic renal insuffi-ciency (serum creatinine level � or � 1.5mg/dL), coronary artery disease, diabe-tes, and smoking were present in 89%,67%, 59%, 33%, and 28% of patients, re-spectively. Preoperative duplex carotidmapping was performed in all cases. Allprocedures were performed with pa-tients under local anesthesia and lightsedation. RESULTS: An ATL HDI 5000scanner with the SonoCT feature wasused. The common femoral artery wascannulated with a single-entry needleunder direct ultrasound visualization.Fluoroscopy was used to assist passageof the guidewire into the aorta and thecommon carotid artery. In only fourcases (22%) was an aortic arch angio-gram obtained. Selective catheterizationof the internal and external carotid arter-ies was performed under ultrasoundguidance. The distal cerebral protectiondevice (17 cases) was placed under fluo-roscopic guidance. Balloon width andlength were chosen according to ultra-sound measurements. Balloon and stentdeployment were successfully achievedwith ultrasound guidance alone in allcases. Appropriate stent apposition andresolution of the stenosis was confirmedby duplex scanning in all cases. Five pa-tients (28%) were noted to have low(�100 mL/min) internal carotid arteryvolume flow after stent deployment(range, 20–88 mL/min; mean � SD, 50� 25 mL/min). The internal carotid ar-tery volume flow increased immediatelyafter Filterwire retrieval in all cases andranged from 136 to 400 mL/minute(mean, 245 � 107 mL/min). This in-crease was statistically significant (P �.02). No ipsilateral strokes or deaths oc-curred during follow-up from 1 to 12months (mean follow-up, 5 mo). CON-CLUSIONS: Duplex scan-assisted CBASis feasible and may reduce the need forintra-arterial contrast injection in se-lected patients deemed at high risk forrenal failure from nephrotoxic contrastmaterial. Additional advantages includedirect visualization of the puncture site,precise position of the balloon and stent,and B-mode and hemodynamic confir-mation of the adequacy of the technique.Authors’ Abstract

Vos JA, Vos AW. Impact of headmovements on morphology and flowin the internal carotid artery after ca-rotid angioplasty and stenting versusendarterectomy. J Vasc Surg2005;41(3):469–475.

• OBJECTIVE: Because stents cancause vessel angulation during move-ment, we hypothesized that internal ca-rotid artery (ICA) stents might lead to

1160 • Abstracts August 2005 JVIR

alterations of cerebropetal blood flow.This study assessed three-dimensionalanatomy and volumetric flow rate (VFR)in the ICA in various head positions bycomparing patients treated with carotidangioplasty and stenting (CAS) with pa-tients treated with carotid endarterec-tomy (CEA). METHODS: Three-dimen-sional time-of-flight magnetic resonanceangiography and magnetic resonanceflow quantification were performed onsix subjects after CAS (median age, 70 y)and on six subjects after CEA (medianage, 67 y). All investigations were per-formed in five head positions: neutral,bent forward, bent backward, andturned to the treated, ipsilateral side andto the contralateral side. Maximum-in-tensity projection reconstructions wereobtained to measure maximal angu-lation of the ICA in the forward,backward, ipsilateral, and contralateralpositions compared with neutral. Subse-quently, the plane perpendicular to theICA, 1 cm distal to the stent or 4 cmdistal to the carotid bifurcation (CEA pa-tients), was established. The VFRthrough this plane was measured foreach position, and the forward, back-ward, ipsilateral, and contralateral posi-tions were compared with neutral. RE-SULTS: In CAS patients, there was amedian change in ICA angulation of�10.2 degrees (interquartile range, �7.3degrees to �17.9 degrees ) in the for-ward position, compared with �0.2 de-grees (�1.0 degrees to �2.4 degrees ) inCEA patients (P � .016). In all other headpositions, there was no statistically sig-nificant difference in angulation change.There was no statistically significant dif-ference in VFR change between groupsin any of the head positions tested.CONCLUSIONS: There was a significantincrease in ICA angulation in CAS pa-tients if the head was bent forward; thiswas not observed in CEA patients. Thisangulation change did not lead to signif-icant acute changes in cerebropetalblood flow, but it might have chroniceffects not yet tested.Authors’ Abstract

RENOVASCULARINTERVENTIONS

Patel ST, Mills JL, Sr. The limita-tions of magnetic resonance angiogra-phy in the diagnosis of renal arterystenosis: comparative analysis withconventional arteriography. J VascSurg 2005;41(3):462–468.

• PURPOSE: Gadolinium-enhancedmagnetic resonance angiography (MRA)is commonly used as a screening modal-ity for the detection of renal artery ste-nosis. However, evidence supporting itsutility in clinical practice is lacking; few

rigorous studies have compared MRAwith contrast arteriography (CA). Aftermaking anecdotal clinical observationsthat MRA sometimes overestimated thedegree of renal artery stenosis, we de-cided to determine the interobservervariability, sensitivity, specificity, anddiagnostic accuracy of MRA comparedwith CA. METHODS: From September1999 to April 2003, we evaluated 68 renalarteries in 34 patients with clinically sus-pected renal artery stenosis using bothMRA and CA. All studies were indepen-dently reviewed by four blinded observ-ers. Renal arteries were categorized byMRA as normal, �50%, and � 50% ste-nosis/occlusion. The sensitivity, speci-ficity, and accuracy of MRA detection ofrenal artery stenosis were compared toCA as the gold standard. Interobservervariability (kappa) was also calculated.RESULTS: MRA demonstrated 87% sen-sitivity, 69% specificity, 85% accuracy,95% negative predictive value, and 51%positive predictive value for the diagno-sis of renal artery stenosis. Interobserveragreement was moderate for MRA(kappa � 0.53) and good for CA (kappa� 0.76). In 21 arteries (31%), MRA wasfalsely positive. CONCLUSIONS: In pa-tients with a high clinical suspicion ofrenal artery stenosis, MRA is 87% sensi-tive in the detection of �50% stenosis.However, MRA is relatively nonspecificcompared with CA and results in signif-icant overestimation of renal artery ste-nosis in nearly one third of patients. Toreduce unnecessary CA, cliniciansshould consider supplemental studies.Authors’ Abstract

VENOUS INTERVENTIONS

Castelli P, Caronno R. Endovasculartreatment for superior vena cava ob-struction in Behcet disease. J VascSurg 2005;41(3):548–551.

• Behcet disease is a heterogeneous,multisystem inflammatory disorder;vascular involvement has been consid-ered to result from systemic vasculitis,which most frequently affects veins andoccurs in 5% to 10% of these patients.However, superior vena cava (SVC) in-volvement is rare; it accounts for only6% of the cases. The aim of this article isto a report on a case of SVC recanaliza-tion through the use of fibrinolysis andself-expanding stents as treatment oflife-threatening SVC syndrome in ayoung male patient with Behcet disease.Authors’ Abstract

Thromboembolic disease

Kucher N, Koo S. Electronic alerts toprevent venous thromboembolism

among hospitalized patients. N EnglJ Med 2005;352(10):969–977.

• BACKGROUND: Prophylaxisagainst deep-vein thrombosis in hospi-talized patients remains underused. Wehypothesized that the use of a computer-alert program to encourage prophylaxismight reduce the frequency of deep-veinthrombosis among high-risk hospital-ized patients. METHODS: We devel-oped a computer program linked to thepatient data base to identify consecutivehospitalized patients at risk for deep-vein thrombosis in the absence of pro-phylaxis. The program used medical-record numbers to randomly assign 1255eligible patients to an interventiongroup, in which the responsible physi-cian was alerted to a patient’s risk ofdeep-vein thrombosis, and 1251 patientsto a control group, in which no alert wasissued. The physician was required toacknowledge the alert and could thenwithhold or order prophylaxis, includ-ing graduated compression stockings,pneumatic compression boots, unfrac-tionated heparin, low-molecular-weightheparin, or warfarin. The primary endpoint was clinically diagnosed, objec-tively confirmed deep-vein thrombosisor pulmonary embolism at 90 days. RE-SULTS: More patients in the interventiongroup than in the control group receivedmechanical prophylaxis (10.0% vs. 1.5%,P � .001) or pharmacologic prophylaxis(23.6% vs. 13.0%, P � .001). The primaryend point occurred in 61 patients (4.9%)in the intervention group, as comparedwith 103 (8.2%) in the control group; theKaplan-Meier estimates of the likelihoodof freedom from deep-vein thrombosisor pulmonary embolism at 90 days were94.1% (95% confidence interval,92.5–95.4%) and 90.6% (95% confidenceinterval, 88.7–92.2%), respectively (P �.001). The computer alert reduced therisk of deep-vein thrombosis or pulmo-nary embolism at 90 days by 41% (haz-ard ratio, 0.59; 95% confidence interval,0.43–0.81; P � .001). CONCLUSIONS:The institution of a computer-alert pro-gram increased physicians’ use of pro-phylaxis and markedly reduced the ratesof deep-vein thrombosis and pulmonaryembolism among hospitalized patientsat risk.Authors’ Abstract

Venous access

Fratino G, Molinari AC. Central ve-nous catheter-related complications inchildren with oncological/hematologi-cal diseases: an observational study of418 devices. Ann Oncol 2005;16(4):648–654.

• BACKGROUND: The use of in-dwelling central venous catheters (CVCs)

Abstracts • 1161Volume 16 Number 8

has become commonplace in the man-agement of children undergoing antican-cer treatment. Several types of CVC areavailable, while information on compli-cations observed in children is scarce.We describe the experience of two ter-tiary care centers in Italy that prospec-tively followed up three types of CVCused at both institutions over a 30-month period. PATIENTS AND METH-ODS: Between January 2000 and May2002, double-lumen (DL) or single-lu-men (SL) Hickman-Broviac (HB) cathe-ters, and single-lumen pressure-acti-vated safety valve (PASV) catheterswere used and prospectively evaluated.Four types of possible complication weredefined a priori: mechanical, thrombotic,malfunctioning and infectious. RE-SULTS: Four hundred and eighteenCVCs (180 SL-HB, 162 DL-HB and 76PASV) were inserted in 368 children, fora total of 107 012 catheter days at risk ofcomplication. At least one complicationoccurred while using 169 of the devices(40%): 46% of the DL-HB, 46% of thePASV and 33% of the SL-HB (P � .02)catheters. Subjects with hematologicalmalignancies or non-malignant diseaseshad significantly more complicationsthan those with solid tumors (P � .0001).Overall, 234 complications were docu-mented: 93 infectious [complication rateper 1000 catheter days at risk (CR) �0.87], 84 malfunctioning (CR � 0.78), 48mechanical (CR � 0.45) and nine throm-botic (CR � 0.08). SL-HB had statisticallyfewer infectious complications, whilePASV had more mechanical complica-tions. In a multivariate regressionmodel, the most significant risk factorsfor having a CVC complication were he-matological disease [relative risk (RR) �3.0; 95% confidence interval (CI) 1.8 to4.8] and age � 6 years at CVC insertion(RR � 2.5; 95% CI 1.5–4.1). As for thetype of CVC, compared with SL-HB, theDL-HB catheter had a statistically signif-icant two-fold increased risk of any com-plication (RR � 2.1; 95% CI 1.2–3.6),while the PASV catheter had a border-line RR of 1.8 (95% CI 1.0-3.6). Analysisby tumor type showed a higher risk ofany kind of complication in patientswith solid malignancies who had re-ceived a DL-HB catheter as comparedwith an SL-HB catheter (RR � 7.2; 95%CI 2.8–18.7). CONCLUSIONS: CVCsmay cause complications in up to 40% ofpatients, with type of CVC, underlyingdisease and patient age being the threemain factors that affect the incidence ofCVC-related complications. SL-HB cath-eters have the best performance.Authors’ Abstract

HEPATOBILIARY

Tateishi R, Shiina S. Percutaneousradiofrequency ablation for hepatocel-lular carcinoma. An analysis of 1000cases. Cancer 2005;103(6):1201–1209.

• BACKGROUND: Radiofrequencyablation (RFA) was introduced recentlyas a therapeutic modality for hepatocel-lular carcinoma (HCC), an alternative topercutaneous ethanol injection therapy(PEIT), which is coming into use world-wide. Previously reported treatment ef-ficacy and complication rates have var-ied considerably. METHODS: BetweenFebruary 1999 and February 2003, theauthors performed 1000 treatments ofRFA to 2140 HCC nodules in 664 pa-tients with a cooled-tip electrode at theUniversity of Tokyo Hospital (Tokyo, Ja-pan). Short-term and long-term compli-cations were analyzed by treatment andsession basis. Cumulative survival wasalso assessed in 319 patients who re-ceived RFA as primary treatment (naivepatients) and 345 patients who receivedRFA for recurrent tumor after previoustreatment including resection, PEIT, mi-crowave coagulation therapy, andtransarterial embolization (nonnaive pa-tients). RESULTS: A total of 40 majorcomplications (4.0% per treatment, 1.9%per session) and 17 minor complications(1.7% per treatment, 0.82% per session)were observed during the observationperiod until March 31, 2004. There wereno treatment-related deaths. Surgical in-tervention was required in one case eachof bile peritonitis and duodenal perfora-tion. The cumulative survival rates at 1,2, 3, 4, and 5 years were 94.7%, 86.1%,77.7%, 67.4%, and 54.3% for naive pa-tients, whereas the cumulative survivalrates were 91.8%, 75.6%, 62.4%, 53.7%,and 38.2% for nonnaive patients, respec-tively. CONCLUSIONS: The authorsconfirmed the safety and efficacy of RFAfor HCC in a large-scale series and long-term prognosis was satisfactory.Authors’ Abstract

CARDIAC/CORONARY

Sabatine MS, Cannon CP. Additionof clopidogrel to aspirin and fibrino-lytic therapy for myocardial infarctionwith ST-segment elevation. N EnglJ Med 2005;352(12):1179–1189.

• BACKGROUND: A substantial pro-portion of patients receiving fibrinolytictherapy for myocardial infarction withST-segment elevation have inadequatereperfusion or reocclusion of the infarct-related artery, leading to an increasedrisk of complications and death. METH-ODS: We enrolled 3491 patients, 18 to 75years of age, who presented within 12hours after the onset of an ST-elevation

myocardial infarction and randomly as-signed them to receive clopidogrel(300-mg loading dose, followed by 75mg once daily) or placebo. Patients re-ceived a fibrinolytic agent, aspirin, andwhen appropriate, heparin (dispensedaccording to body weight) and werescheduled to undergo angiography 48 to192 hours after the start of study medi-cation. The primary efficacy end pointwas a composite of an occluded infarct-related artery (defined by a Thromboly-sis in Myocardial Infarction flow gradeof 0 or 1) on angiography or death orrecurrent myocardial infarction beforeangiography. RESULTS: The rates of theprimary efficacy end point were 21.7% inthe placebo group and 15.0% in the clo-pidogrel group, representing an absolutereduction of 6.7% points in the rate and a36% reduction in the odds of the endpoint with clopidogrel therapy (95%confidence interval, 24–47%; P � .001).By 30 days, clopidogrel therapy reducedthe odds of the composite end point ofdeath from cardiovascular causes, recur-rent myocardial infarction, or recurrentischemia leading to the need for urgentrevascularization by 20% (from14.1–11.6%, P � .03). The rates of majorbleeding and intracranial hemorrhagewere similar in the two groups. CON-CLUSIONS: In patients 75 years of ageor younger who have myocardial infarc-tion with ST-segment elevation and whoreceive aspirin and a standard fibrino-lytic regimen, the addition of clopidogrelimproves the patency rate of the infarct-related artery and reduces ischemic com-plications.Authors’ Abstract

Cram P, Rosenthal GE. Cardiac re-vascularization in specialty and gen-eral hospitals. N Engl J Med2005;352(14):1454–1462.

• BACKGROUND: The emergence ofspecialty hospitals focusing on narrowprocedural areas has generated contro-versy, although little is known abouttheir quality. METHODS: We conducteda retrospective cohort study of 42,737Medicare beneficiaries who underwentpercutaneous coronary intervention(PCI) and 26,274 who underwent coro-nary-artery bypass grafting (CABG) dur-ing 2000 and 2001 in specialty cardiachospitals (15 for PCI and 15 for CABG)and general hospitals (82 for PCI and 75for CABG) in the same markets. Admin-istrative data were used to compare pa-tients’ characteristics, hospital proce-dural volumes, and patient outcomes.RESULTS: Patients undergoing PCI orCABG in specialty hospitals were lesslikely to have coexisting conditions thanthose being treated at general hospitalsand were less likely to have had an acute

1162 • Abstracts August 2005 JVIR

myocardial infarction (P � .001). Thebetter health of the patients at specialtyhospitals than of those at general hospi-tals was reflected by the lower meanpredicted risk of death (2.1% vs. 3.1% forPCI and 5.0% vs. 5.8% for CABG; P �.001 for each comparison). Mean vol-umes of PCI and CABG procedures in2000 and 2001 were higher in specialtyhospitals than in general hospitals (799vs. 375 PCI procedures, P � .001; and571 vs. 236 CABG procedures, P � .001).The unadjusted rate of death during theindex hospitalization or within 30 daysafter admission was lower in specialtyhospitals than in general hospitals (2.1%vs. 3.2% for PCI and 4.7% vs. 6.0% forCABG; P � .001 for both comparisons).In multivariate analyses adjusted for pa-tients’ characteristics, the odds ratio fordeath after PCI in specialty hospitals andgeneral hospitals was similar (0.89; 95%confidence interval, 0.69–1.15; P � .39),but the odds ratio for death after CABGwas lower in specialty hospitals than ingeneral hospitals (0.84; 95% confidenceinterval, 0.72–0.99; P � .05). In stratifiedanalyses comparing specialty and gen-eral hospitals with similar volumes, dif-ferences in mortality were not signifi-cant. CONCLUSIONS: The lowerunadjusted mortality rate after cardiacrevascularization in specialty cardiachospitals is accounted for by theirhealthier patients and higher proceduralvolumes.Authors’ Abstract

Ridker PM, Cook NR. A randomizedtrial of low-dose aspirin in the pri-mary prevention of cardiovascular dis-ease in women. N Engl J Med 2005;352(13):1293–1304.

• BACKGROUND: Randomized tri-als have shown that low-dose aspirindecreases the risk of a first myocardialinfarction in men, with little effect on therisk of ischemic stroke. There are fewsimilar data in women. METHODS: Werandomly assigned 39,876 initiallyhealthy women 45 years of age or olderto receive 100 mg of aspirin on alternatedays or placebo and then monitoredthem for 10 years for a first major car-diovascular event (ie, nonfatal myocar-dial infarction, nonfatal stroke, or deathfrom cardiovascular causes). RESULTS:During follow-up, 477 major cardiovas-cular events were confirmed in the aspi-rin group, as compared with 522 in theplacebo group, for a nonsignificant re-duction in risk with aspirin of 9% (rela-tive risk, 0.91; 95% confidence interval,0.80–1.03; P � .13). With regard to indi-vidual end points, there was a 17% re-duction in the risk of stroke in the aspi-rin group, as compared with the placebogroup (relative risk, 0.83; 95% confidence

interval, 0.69–0.99; P � .04), owing to a24% reduction in the risk of ischemicstroke (relative risk, 0.76; 95% confi-dence interval, 0.63–0.93; P � .009) and anonsignificant increase in the risk ofhemorrhagic stroke (relative risk, 1.24;95% confidence interval, 0.82–1.87; P �.31). As compared with placebo, aspirinhad no significant effect on the risk offatal or nonfatal myocardial infarction(relative risk, 1.02; 95% confidence inter-val, 0.84–1.25; P � .83) or death fromcardiovascular causes (relative risk, 0.95;95% confidence interval, 0.74–1.22; P �.68). Gastrointestinal bleeding requiringtransfusion was more frequent in the as-pirin group than in the placebo group(relative risk, 1.40; 95% confidence inter-val, 1.07–1.83; P � .02). Subgroup analy-ses showed that aspirin significantly re-duced the risk of major cardiovascularevents, ischemic stroke, and myocardialinfarction among women 65 years of ageor older. CONCLUSIONS: In this large,primary-prevention trial among women,aspirin lowered the risk of stroke with-out affecting the risk of myocardial in-farction or death from cardiovascularcauses, leading to a nonsignificant find-ing with respect to the primary endpoint.Authors’ Abstract

THORACIC

Maskell NA, Davies CW. U.K. Con-trolled trial of intrapleural streptoki-nase for pleural infection. N EnglJ Med 2005;352(9):865–874.

• BACKGROUND: Intrapleural fi-brinolytic agents are used in the drain-age of infected pleural-fluid collections.This use is based on small trials that didnot have the statistical power to evaluateaccurately important clinical outcomes,including safety. We conducted a trial toclarify the therapeutic role of intrapleu-ral streptokinase. METHODS: In thisdouble-blind trial, 454 patients withpleural infection (defined by the pres-ence of purulent pleural fluid or pleuralfluid with a pH below 7.2 with signs ofinfection or by proved bacterial invasionof the pleural space) were randomly as-signed to receive either intrapleuralstreptokinase (250,000 IU twice daily forthree days) or placebo. Patients receivedantibiotics and underwent chest-tubedrainage, surgery, and other treatmentas part of routine care. The number ofpatients in the two groups who had diedor needed surgical drainage at threemonths was compared (the primary endpoint); secondary end points were therates of death and of surgery (analyzedseparately), the radiographic outcome,and the length of the hospital stay. RE-SULTS: The groups were well matched

at baseline. Among the 427 patients whoreceived streptokinase or placebo, therewas no significant difference betweenthe groups in the proportion of patientswho died or needed surgery (with strep-tokinase: 64 of 206 patients [31%]; withplacebo: 60 of 221 [27%]; relative risk,1.14 [95% confidence interval, 0.85–1.54;P � .43), a result that excluded a clini-cally significant benefit of streptokinase.There was no benefit to streptokinase interms of mortality, rate of surgery, radio-graphic outcomes, or length of the hos-pital stay. Serious adverse events (chestpain, fever, or allergy) were more com-mon with streptokinase (7%, vs. 3% withplacebo; relative risk, 2.49 [95% confi-dence interval, 0.98–6.36]; P � .08).CONCLUSIONS: The intrapleural ad-ministration of streptokinase does notimprove mortality, the rate of surgery, orthe length of the hospital stay amongpatients with pleural infection.Authors’ Abstract

RISK FACTORS FOR VASCULARDISEASE

Hobbs SD, Wilmink AB. Assessmentof smoking status in patients with pe-ripheral arterial disease. J Vasc Surg2005;41(3):451–456.

• OBJECTIVE: To assess the utility ofa novel rapid urinary cotinine assay todetect and quantify the level of smokingin patients with peripheral arterial dis-ease. METHODS: This was a cross-sec-tional study in a vascular surgical out-patient department of a large teachinghospital. Participants were 100 consecu-tive subjects presenting to a hospital out-patient clinic with a diagnosis of inter-mittent claudication confirmed by apositive Edinburgh claudication ques-tionnaire and an ankle-brachial pressureindex of less than 0.9. Main outcomemeasures were patient-offered smokinghistory, exhaled breath carbon monox-ide levels, urinary cotinine levels as mea-sured by a novel rapid assay, and labo-ratory-measured creatinine-adjustedurinary cotinine levels. RESULTS: Fifty-five subjects declared that they were cur-rent smokers, 40 declared that they wereex-smokers, and 5 declared that theywere never-smokers. Of the 40 ex-smok-ers, 6 subjects (15%) had urinary cotininelevels greater than 500 ng/mL (regularsmokers), and a further 2 (5%) had uri-nary cotinine levels between 100 and 500ng/mL (light, irregular, or passivesmokers). The rapid urinary cotinine as-say had a sensitivity and specificity of100% and 98%, respectively, in its abilityto detect active smoking, and the degreeof smoking correlated well with labora-tory creatinine-corrected urinary cotin-ine levels (Spearman coefficient, 0.805; P

Abstracts • 1163Volume 16 Number 8

� .001). By contrast, exhaled carbonmonoxide had a sensitivity and specific-ity of 95% and 89%, respectively, andalthough it correlated well with urinarycotinine (Spearman coefficient, 0.782; P� .001), it was found on linear regressionanalysis to be unreliable in differentiat-ing light smokers from nonsmokers.CONCLUSIONS: Patient-offered smok-ing history is unreliable because there isno correlation between the patient-re-ported number of cigarettes smoked perday and urinary cotinine levels. Thenovel rapid assay for urinary cotininedescribed here is superior to exhaled car-bon monoxide measurement in detect-ing the level of smoking exposureamong patients with intermittent claudi-cation, and its results correlate well withlaboratory-measured cotinine.Authors’ Abstract

LaRosa JC, Grundy SM. Intensivelipid lowering with atorvastatin inpatients with stable coronary disease.N Engl J Med 2005;352(14):1425–1435.

• BACKGROUND: Previous trialshave demonstrated that lowering low-density lipoprotein (LDL) cholesterollevels below currently recommendedlevels is beneficial in patients with acutecoronary syndromes. We prospectivelyassessed the efficacy and safety of low-ering LDL cholesterol levels below 100mg per deciliter (2.6 mmol per liter) inpatients with stable coronary heart dis-ease (CHD). METHODS: A total of10,001 patients with clinically evidentCHD and LDL cholesterol levels of lessthan 130 mg per deciliter (3.4 mmol perliter) were randomly assigned to double-blind therapy and received either 10 mgor 80 mg of atorvastatin per day. Patientswere followed for a median of 4.9 years.The primary end point was the occur-rence of a first major cardiovascularevent, defined as death from CHD, non-fatal non-procedure-related myocardialinfarction, resuscitation after cardiac ar-rest, or fatal or nonfatal stroke. RE-SULTS: The mean LDL cholesterol levelswere 77 mg per deciliter (2.0 mmol perliter) during treatment with 80 mg ofatorvastatin and 101 mg per deciliter (2.6mmol per liter) during treatment with 10mg of atorvastatin. The incidence of per-sistent elevations in liver aminotransfer-ase levels was 0.2% in the group given 10mg of atorvastatin and 1.2% in the groupgiven 80 mg of atorvastatin (P � .001). Aprimary event occurred in 434 patients(8.7%) receiving 80 mg of atorvastatin, ascompared with 548 patients (10.9%) re-ceiving 10 mg of atorvastatin, represent-ing an absolute reduction in the rate ofmajor cardiovascular events of 2.2% anda 22% relative reduction in risk (hazardratio, 0.78; 95% confidence interval,

0.69–0.89; P � .001). There was no dif-ference between the two treatmentgroups in overall mortality. CONCLU-SIONS: Intensive lipid-lowering therapywith 80 mg of atorvastatin per day inpatients with stable CHD provides sig-nificant clinical benefit beyond that af-forded by treatment with 10 mg of ator-vastatin per day. This occurred with agreater incidence of elevated amino-transferase levels.Authors’ Abstract

Kistorp C, Raymond I. N-terminalpro-brain natriuretic peptide, C-reac-tive protein, and urinary albumin lev-els as predictors of mortality and car-diovascular events in older adults.Jama 2005;293(13):1609–1616.

• CONTEXT: B-type natriuretic pep-tides have been shown to predict cardio-vascular disease in apparently healthyindividuals but their predictive abilityfor mortality and future cardiovascularevents compared with C-reactive protein(CRP) and urinary albumin/creatinineratio is unknown. OBJECTIVE: To assessthe prognostic value of the N-amino ter-minal fragment of the prohormone brainnatriuretic peptide (NT-proBNP) versusCRP and urinary albumin/creatinine ra-tio in an older adult population. DE-SIGN, SETTING, AND PARTICIPANTS:A population-based prospective study of764 participants aged 50 to 89 years froma community in Copenhagen, Denmark,in which 658 participants providedblood and urinary samples and were ex-amined between September 1, 1998, andJanuary 24, 2000. Of these participants,626 without heart or renal failure wereenrolled. A subgroup of 537 had no his-tory of cardiovascular disease at base-line. During 5 years of follow-up (to De-cember 31, 2003), 94 participants diedand 65 developed a first major cardio-vascular event. MAIN OUTCOMEMEASURES: Risk of mortality and firstmajor cardiovascular event by baselinelevels of NT-proBNP, CRP, and urinaryalbumin/creatinine ratio levels. RE-SULTS: After adjustment for the cardio-vascular risk factors of age, sex, smok-ing, diabetes mellitus, hypertension orischemic heart disease, total cholesterol,and serum creatinine, the hazard ratio(HR) of mortality for values above the80th percentile of NT-proBNP was 1.96(95% confidence interval [CI], 1.21–3.19);for CRP, 1.46 (95% CI, 0.89–2.24); and forurinary albumin/creatinine ratio, 1.88(95% CI, 1.18–2.98). Additional adjust-ment for left ventricular systolic dys-function did not markedly attenuate thepredictive value of NT-proBNP (HR,1.82; 95% CI, 1.11–2.98). The absolute un-adjusted increase in mortality risk forparticipants with values above the 80th

percentile versus equal to or below the80th percentile was 24.5% for NT-proBNP, 7.8% for CRP, and 19.5% forurinary albumin/creatinine ratio. TheNT-proBNP levels were associated withfirst major cardiovascular events (nonfa-tal myocardial infarction, fatal coronaryheart disease, unstable angina, heart fail-ure, stroke, and transient ischemic at-tack) with an adjusted HR of 3.24 (95%CI, 1.80–5.79) versus 1.02 (95% CI, 0.56–1.85) for CRP and 2.32 (95% CI, 1.33–4.05) for urinary albumin/creatinine ra-tio when comparing participants withvalues above the 80th percentile withthose with values equal to or below the80th percentile. CONCLUSIONS: Mea-surements of NT-proBNP provide prog-nostic information of mortality and firstmajor cardiovascular events beyond tra-ditional risk factors. NT-proBNP was astronger risk biomarker for cardiovascu-lar disease and death than CRP was innonhospitalized individuals aged 50 to89 years.Authors’ Abstract

Wanhainen A, Bergqvist D. Risk fac-tors associated with abdominal aorticaneurysm: a population-based studywith historical and current data. JVasc Surg 2005;41(3):390–396.

• OBJECTIVES: To define risk factorsassociated with the development of ab-dominal aortic aneurysm (AAA) in apopulation-based case-control study.MATERIAL AND METHODS: Thirty-five patients found to have AAA atscreening were compared with 140 age-and sex-matched controls. Three distinctcomparisons were made. Current riskfactors in both cohorts were compared.Because data were also available for bothcohorts from a study conducted 12 yearspreviously, a historical comparison wasmade between risk factors identified atthe time of this initial evaluation. A lon-gitudinal comparison was made be-tween historical and current risk factors.RESULTS: Elevated low-density-li-poprotein cholesterol, total cholesterol,and triglyceride levels 12 years beforeAAA screening were associated withcurrent AAA with an odds ratio (OR) of2.3 (95% confidence interval [CI], 1.2–4.4); OR, 1.9 (95% CI, 1.3–2.8); and OR,1.9 (95% CI,1.2–3.1)/mmol/liter, respec-tively. Current variables assessed atAAA screening that were associatedwith AAA were: A history of atheroscle-rotic disease, OR, 3.8 (95% CI, 1.7–8.5);having a first-degree relative with AAA,OR, 4.4 (95% CI, 1.5–13.0); current smok-ing, OR, 5.2 (95% CI, 1.6–16.8); high-density-lipoprotein cholesterol level,OR, 0.1 (95% CI, 0.02–0.7)/mmol/liter;and high-sensitivity C-reactive protein

1164 • Abstracts August 2005 JVIR

(hsCRP) level, OR, 1.1 (95% CI, 1.01–1.2)/mg/liter. Hypertension and diabe-tes were not associated with AAA. Asignificant increase of hsCRP over time(12 y) was observed in AAA patients (P� .039) but not among controls. The

variables of a history of atherosclerosis,smoking, and family history of AAA ap-pear to interact synergistically to in-crease the prevalence of AAA (P � .001).CONCLUSION: Among traditional riskfactors for atherosclerosis, some were as-

sociated with AAA and others were not,indicating complex and partly differentcauses. Inflammation and heredity ap-pear to be important factors in the devel-opment of AAA.Authors’ Abstract

Abstracts • 1165Volume 16 Number 8