www.elsevier.com/locate/semvascsurg

Available online at www.sciencedirect.com

Interpretation of arterial duplex testing of lower-extremity arteries and interventions

Kelley D. Hodgkiss-Harlowa, and Dennis F. Bandykb,n

aDivision of Vascular Surgery, Kaiser Permanente Foundation, San Diego, CAbDivision of Vascular and Endovascular Surgery, University of California, San Diego School of Medicine, SulpizioCardiovascular Center, 7404 Medical Center Drive, Mail Code 7403, La Jolla, CA 92037

a b s t r a c t

Arterial duplex testing is used to evaluate patients with lower-limb arterial occlusive or

aneurysmal disease to provide clinicians with detailed information on location, extent, and

severity of disease. It is possible to detect disease from the visceral aorta to the tibial

arteries. Duplex testing is interpreted in conjunction with limb-pressure measurements to

accurately categorize arterial hemodynamics and functional impairment. Understanding

the features of duplex-acquired velocity spectra recordings is fundamental to accurate

diagnostic testing, including the characteristic spectral features of “normal” versus

“abnormal” lower-limb arterial flow, hemodynamic changes associated with stenosis or

occlusion, and the status of distal limb or foot perfusion. Scanning can provide an arterial

map of occlusive or aneurysm lesions analogous to an angiogram. Testing is accurate

before and after intervention for the detection of stenosis; a common failure mode after

bypass grafting or peripheral angioplasty. The detection of high-grade stenosis in an

arterial repair allows for pre-emptive treatment before thrombosis occurs and improves

long-term patency.

& 2013 Elsevier Inc. All rights reserved.

1. Introduction

Duplex ultrasound is an integral component of diagnostictesting for the evaluation of lower-extremity arterial disease,including after intervention (bypass grafting and angioplasty)[1,2]. Testing provides objective information about blood flow(pulsed Doppler spectral analysis) and anatomy (B-mode andcolor Doppler imaging) for the accurate classification ofocclusive and aneurysm disease. Modern ultrasound instru-mentation affords assessment of blood flow using one ofseveral techniques (eg, color Doppler imaging, pulsed Dopplerspectral analysis, power Doppler imaging, or B-flow imaging)and high-resolution B-mode imaging of artery anatomy,including three-dimensional vessel reconstruction and eval-uation of atherosclerotic plaque morphology. Test

interpretation ranges from normal to clinically relevant dis-ease categories of mild, moderate, or severe ischemia, andshould be performed in conjunction with indirect physiologictesting (eg, systolic blood pressure measurement and pulsevolume plethysmography) to accurately categorize arterialhemodynamics and functional impairment. When peripheralarterial disease (PAD) is identified, duplex imaging can beused to map the arterial tree for occlusive or aneurysmlesions, analogous to an angiogram [3]. After intervention,duplex testing can detect changes in functional patency of abypass graft, arterial repair, or angioplasty sites by identi-fication of stenosis. Accurate interpretation of velocity spec-tra recordings obtained by duplex scanning is fundamental tosuccessful application of this diagnostic technique. Thisreview details the characteristic features of normal and

0895-7967/$ - see front matter & 2013 Elsevier Inc. All rights reserved.http://dx.doi.org/10.1053/j.semvascsurg.2013.11.002

nCorresponding author.E-mail address: dbandyk@ucsd.edu (D.F. Bandyk).

S E M I N A R S I N V A S C U L A R S U R G E R Y 2 6 ( 2 0 1 4 ) 9 5 – 1 0 4

abnormal duplex-acquired velocity spectra necessary for testinterpretation and disease classification.

2. Lower-extremity duplex testing

To perform detailed arterial mapping, duplex ultrasoundinstrumentation for peripheral testing requires the use ofcurved and linear array transducers with 3.5- to 7-MHzimaging frequencies for appropriate depth penetration andhigh-resolution imaging of the abdominal and extremityarteries. A 90-degree imaging angle should be used to meas-ure vessel diameter, identifying intima-medial thickening,and assessing atherosclerotic plaque composition. There aretwo types of Doppler ultrasound displays: a color flowDoppler image showing mean flow velocity distribution dis-played as a color-encoded map superimposed on the gray-scale B-mode tissue image; and a spectral Doppler imageshowing the time-varying flow velocity distribution within aselected sample volume. The latter display provides quanti-tative information as to the peak velocity during the pulsecycle and the spectral content, that is, the range of velocitiesat each point in time. To obtain reproducible informationfrom pulsed spectral Doppler recordings, a Doppler beamangle of r60 degrees relative to the transducer insonationbeam and the artery wall should be used.The normal pulsed Doppler velocity spectra recorded from

a peripheral lower- or upper-extremity artery has the featuresof multiphasic or triphasic waveform with a narrow spectralwidth (range of velocities) throughout the pulse cycle, indi-cating red blood cells are moving at a similar speed anddirection in a nondisturbed or laminar flow pattern (Fig. 1).The velocity spectra waveform with each cardiac pulsereflects blood acceleration during systolic, an early diastolicflow reversal caused by the propagated pressure pulse waveand its reflection from a higher downstream resistance,followed by late antegrade diastolic flow. There might beatherosclerotic plaque imaged, but lumen reduction at therecording site and proximal is o50% diameter reducing (DR).Spectral broadening with peak systolic velocity (PSV) increasein the pulsed Doppler signal indicates “disturbed” flow orturbulence and can be recorded centerstream at bifurcations,regions of focal diameter change, and sites of stenosis.For interpretation of PAD severity, the duplex-acquired

velocity spectra parameters of acceleration time, pulsatilityindex (PI), and maximum spectra velocity measured at PSVand end-diastole are used (Table 1). Changes in these wave-form parameters allow detection of segmental, hemody-namic significant occlusive disease. The PSV measurementis reproducible and the most common velocity spectra

parameter used in the interpretation of normal artery flow,critical limb ischemia, and for the grading of arterial stenosis.Normal PSV in lower-limb arteries is in the range of 55 cm/sat the tibial artery to 110 cm/s at the common femoral artery(Table 2). The end-diastole velocity measurement is used inconjunction with PSV for evaluating high-grade stenosis(>70% DR) with values >40 cm/s indicating a pressure-reducing stenosis.The PI is calculated by dividing the peak-to-peak velocity

spectra shift by the mean velocity. The PI of normal periph-eral arteries is >4.0 (femoral artery, >6; popliteal artery >8). PIvalues r4 reflect proximal inflow or occlusive disease, and

Fig. 1 – Duplex image with normal, multiphasic velocityspectra recorded from the common femoral artery (top)and superficial femoral artery. Recording made using a60-degree Doppler beam angle and pulsed Dopplersample volume is positioned in the flow centerstreamwhere color flow pixels indicate highest mean velocity.

Table 1 – Velocity spectra waveform parameters used for duplex test interpretation.

Testing area PSV EDV AT PI Mean flow velocity

Peripheral artery X X X XBypass graft surveillance X X X XPeripheral angioplasty X X X

AT, systolic acceleration time; EDV, end-diastolic velocity; PI, pulsatility index; PSV, peak systolic velocity.

S E M I N A R S I N V A S C U L A R S U R G E R Y 2 6 ( 2 0 1 4 ) 9 5 – 1 0 496

changes in PI or spectral waveform damping are diagnostic ofmultilevel occlusive disease (Fig. 2). Division of the distalartery PI by the proximal artery PI calculates the “dampingfactor” with the normal value of Z0.9 and a value o0.9diagnostic of occlusive disease.The systolic acceleration time during systole can also be

used to diagnose occlusive disease proximal to pulsed Dop-pler recording site. A normal value is o133 ms [4]. As thesystolic acceleration time increases to >200 ms, spectrawaveform develops a rounded upslope (termed tardus parvas)configuration due to the prolonged time to PSV. The presenceof damped, low-velocity blood flow in an extremity artery orbypass graft indicates a proximal pressure-reducing lesionwith regional systolic blood pressure o60% of normal (Fig. 3).The diagnostic accuracy of systolic acceleration time can beinfluenced by cardiac conditions (eg, cardiomyopathy, aorticvalve disease), but downstream occlusive disease has mini-mal influence on diagnostic sensitivity.Arterial stenosis is recognized with color flow imaging by a

reduction in the color-encoded flow lumen, imaging a high-

velocity flow region with color bar aliasing and developmentof a mosaic flow pattern in the lumen signifying turbulentflow. At the site of a high-grade (>75% DR) stenosis, the real-time color Doppler flow will appear as a whitened, color-desaturated “flow jet” with mosaic color flow extending forseveral vessels diameters downstream, corresponding topost-stenotic turbulence.The definition of a significant or “critical” arterial stenosis is

a lesion that is associated with a resting systolic pressuregradient of >15 mm Hg and reduces resting volume flow. Inperipheral arterial circulation, this correlates with Z50% DRstenosis or >75% cross-sectional area reduction. Assessing thePSV changes that occur from proximal to, within, and distal toan arterial stenosis, duplex testing can estimate hemodynamicsignificance and predict the degree of lumen DR with specifiedranges, eg, o50% DR, >50% DR, and >70% to 75% DR (Fig. 4,Table 3). Velocity spectra characteristics of a >50% DR arterialstenosis include an elevated PSV >180 cm/s, systolic spectralbroadening indicating highly disturbed flow, ie, post-stenoticturbulence, with simultaneous forward and retrograde velocityspectra during systole. The ratio of PSV (Vr) across a stenosis isa useful parameter to grade stenosis severity, with a Vr >2indicating >50% DR and a VR >4 correlating with >70% DR.Typically, a pressure-, and flow-reducing arterial stenosis isassociated with monophasic waveform with PSV >250 to 300cm/s, a Vr across the stenosis >3.5, and end-diastolic velocity>40 cm/s. Downstream of a significant pressure-reducingarterial stenosis, the spectral waveform should appeardamped and monophasic, with prolongation of accelerationtime and a decrease in PSV to below normal levels. As stenosisseverity increases to >90% DR, the volume flow through thestenosis trends toward zero, which can produce a PSV at thestenosis in a minimally elevated range (100 to 200 cm/s) andlow velocity (o10 cm/s) “trickle” flow downstream. Athero-sclerotic plaque associated with >50% stenosis might becalcified and produce an acoustic shadow that interferes withmeasurement of residual artery diameter or cross-section areareduction on transverse imaging. Correlation studies betweenduplex testing and angiogram measurements have foundmeasurement of PSV and Vr are the best predictors ofperipheral arterial stenosis severity when expressed as %DR.

3. Lower limb arterial duplex testing

The extent of duplex mapping can be individualized based onthe indication for arterial testing. Screening for aneurysm andor imaging to identify tibial artery calcification are appropriatetest indications in selected patients. Evaluation of the patientwith exertional leg pain, ie, diagnosis of claudication, and/orsigns of PAD, such as absent pulses, dependent rubor, andnonhealing ulcer, can localize site(s) of arterial occlusivedisease. Lower-limb testing should include imaging of theabdominal aorta and iliac arteries for aneurysm and thecommon femoral arteries interrogated for the presence ofnormal, ie, triphasic, velocity spectra. Incomplete vessel imag-ing can occur due to bowel gas, obesity, poor patient cooper-ation, or acoustic shadowing caused by plaque calcification.The findings of duplex mapping are recorded in a sche-

matic of the extremity arterial tree, analogous to an

Table 2 – Mean (7standard deviation) artery diametersand peak systolic velocities in healthy subjects.

Artery Diameter (cm) Velocity (cm/s)

Infrarenal aorta 2 7 0.2 55 7 12Common iliac 1.5 7 0.18 70 7 18External iliac 0.8 7 0.13 115 7 21Common femoral 0.8 7 0.14 114 7 24Superior femoral 0.6 7 11 90 7 14Popliteal 0.5 7 0.1 68 7 14Tibial arteries 0.3 7 0.4 55 7 10

Fig. 2 – Segmental duplex-acquired velocity spectra recordedfrom a patient with multilevel femoropopliteal and tibialartery occlusive disease. The multiphasic common femoralartery flow indicates no significant proximal occlusivedisease. Abnormal monophasic velocity spectra recordedfrom the popliteal and tibial artery signify segmentalocclusive lesions. Progressive damping of the popliteal andtibial artery spectral waveforms establishes the diagnosis ofmultilevel infrainguinal disease. Note the decrease in peaksystolic velocity (PSV) and low pulsatility index (PI) at theankle level.

S E M I N A R S I N V A S C U L A R S U R G E R Y 2 6 ( 2 0 1 4 ) 9 5 – 1 0 4 97

arteriogram, with notation of site(s) of aneurysm or occlusivedisease and measurements of velocity spectra at nondiseasedarterial segments (common femoral, superficial femoral,popliteal, and tibial arteries) and at sites of stenosis. Thelength of an arterial occlusion can be estimated based on thelocation of exit and re-entry collaterals. Duplex mappingallows classification of atherosclerotic occlusive disease inthe aortoiliac, femoral-popliteal, and popliteal-tibial arterialsegments based on TransAtlantic InterSociety Consensusguidelines for grading lesions from A through D based onlesion length and morphology. Endovascular therapy is rec-ommended for TransAtlantic InterSociety Consensus A (sin-gle lesion o3 cm) and B (single lesion 3 to 5 cm or tandemlesions o3 cm) lesions because clinical results are equivalentto surgical intervention with less morbidity. More advancedTransAtlantic InterSociety Consensus C and D lesions, suchas long-segment (>5 to 10 cm) occlusions or multiple

stenoses, might also be amenable to endovascular therapy,but the patency and re-intervention rates are not superior tosurgical revascularization; causing most surgeons to indi-vidualize intervention based on patient risk factors predictiveof procedural morbidity.Duplex-acquired velocity spectra recorded from nonstenotic

artery segments should be correlated with segmental orankle-brachial index (ABI) systolic pressure measurements.Changes in spectral waveform, PI, and PSV develop as the ABIdeceases from normal (>0.9), to moderate ischemia (0.5 to0.85), to severe ischemia levels (o0.5) (Fig 3). Severe limbischemia is present when the peripheral artery spectral wave-form is damped, the pulsatility index is o1.5, and PSV is low(o20 to 30 cm/s). The correlation of velocity spectra with ABIis useful in estimating regional systolic pressure based on thePSV and PI of segmental duplex spectra waveform recordedfrom nonstenotic artery segments in diabetic or renal

Fig. 3 – Normal (left) and abnormal (right) velocity spectra recording from tibial artery at ankle. Multiphasic, normal peaksystolic velocity (PSV) (>50 cm/s) predicts normal ankle systolic pressure. Monophasic, low PSV tibial artery flow predictsresting ankle-brachial index of o0.6.

Fig. 4 – Duplex categories (normal, o50%, >50%, 70% to 75%) of peripheral artery stenosis based on velocity spectra waveforminterpretation. PSV, peak systolic velocity.

S E M I N A R S I N V A S C U L A R S U R G E R Y 2 6 ( 2 0 1 4 ) 9 5 – 1 0 498

transplantation patients with heavily calcified, incompressi-ble arteries precluding cuff-derived systolic pressure measure-ment. In patients with mild claudication (ABI >0.8), tibialartery velocity spectra can be triphasic, but after exercisetreadmill testing, a monophasic, damped waveform develops,corresponding to the decrease in ankle systolic pressure.The overall accuracy of peripheral duplex testing compared

with contrast angiography in grading PAD lesions is >80%, butvaries relative to arterial segment scanned, and is decreasedin limbs with multilevel disease. In blinded comparisons,duplex grading of atherosclerotic occlusive disease (o50%DR, >50% DR, occlusion) in the aorto-iliac segment had adiagnostic sensitivity of 81% to 91% and specificity of 90% to99%; for the femoropopliteal segment, sensitivity ranged from67% to 91% and specificity 94% to 99% [4,5]. Multilevel diseasereduces diagnostic accuracy primarily in interpretation ofpercent DR for the distal tandem stenosis with overall accu-racy decreasing from >90% to 63% in the aortoiliac segmentand from 93% to 83% in the femoropopliteal tract. Distal to anocclusion or high-grade stenosis, a stenosis identified by colorDoppler imaging should have a Vr >2.5 for interpretation of>50% DR. The diagnostic accuracy of arterial duplex is suffi-cient that several vascular groups have performed infraingui-nal revascularization based solely on duplex arterial mapping,which identified suitable proximal and distal anastomoticsites, confirmed a patent artery to the foot as well as anadequate diameter saphenous vein as a bypass conduit [6].

3.1. Femoral artery duplex scan after cannulation

Duplex scanning is the preferred diagnostic modality foridentification of pseudoaneurysm and arteriovenous fistuladevelopment after catheter-based interventions. Test

indications can include pain, pulsatile mass, or bruit at anarterial access site. Duplex features of femoral false aneurysminclude flow outside the artery, presence of a track or “stalk”from the puncture site to the aneurysm sac, and a character-istic “to and fro” flow pattern in the stalk corresponding toblood flow into the aneurysm during systole and sac emptyingduring diastole. Imaging of a false aneurysm stalk indicatessuitable anatomy for ultrasound-guided thrombin injection tocause thrombus. Using B-mode and color Doppler imaging, aneedle is positioned within the sac and thrombin slowlyinjected. Confirmation of sac thrombosis as well as normalarterial and venous flow at the arterial puncture site com-pletes the procedure. If an arteriovenous fistula is present atthe access site, a high-velocity flow jet (PSV >300 cm/s)between the artery and vein will be identified and the externaliliac artery velocity spectra will have elevated PSVs, a low-resistance flow signal proximal to the arteriovenous fistula,and a triphasic (high resistance) signal distal.

3.2. Duplex testing after arterial intervention

A surveillance program after lower-limb arterial interventionis recommended, but the extent of testing, including thefrequency of duplex testing, remains controversial. Afterlower-limb open arterial repair (bypass graft, endarterectomy)or endovascular therapy, duplex imaging can be a usefulcomponent of clinical assessment that should also includepatient query for new symptoms of limb ischemia andmeasurement of ABI. The frequency of testing should beindividualized to the patient, type of arterial intervention,and initial post-repair duplex scan findings. An appropriatesurveillance schedule would be within 2 weeks of the proce-dure and, if normal, 3 to 6 months later. Patients treated for

Fig. 5 – Duplex images of normal, multiphasic velocity spectra recording from a polytetrafluoroethylene prostheticfemoropopliteal bypass graft (left) and superficial femoral artery stent (right). Peak systolic velocity is in normal(450 cm/s) range.

Table 3 – Duplex classification of peripheral artery occlusive disease.

Stenosis category Peak systolic velocity (cm/s) Velocity ratio Distal artery spectral waveform

o20% o150 o1.5 Triphasic, normal PSV20% to 49% 150�200 1.5�2 Triphasic, normal PSV50% to 75% 200�300 2�4 Monophasic, reduced PSV>75% >300 End-diastolic velocity >40 >4 Damped, monophasic, reduced PSVOcclusion No flow, length of occlusion estimated based on distance from exit and

re-entry collateralsDamped, monophasic reduced PSV

PSV, peak systolic velocity.

S E M I N A R S I N V A S C U L A R S U R G E R Y 2 6 ( 2 0 1 4 ) 9 5 – 1 0 4 99

critical limb should have testing every 3 months during thefirst year. Color Doppler imaging of the arterial repair,including adjacent inflow and outflow arteries, is performedwith the hemodynamics characterized by duplex-derived PSVmeasurements along the reconstruction. The presence ofmultiphasic, normal PSV flow at the ankle correlates with anormal ABI and indicates the absence of angioplasty orbypass graft stenosis (Fig. 5). The focus of duplex surveillanceis on the identification and repair of critical stenosis definedby duplex velocity spectra criteria of PSV >300 cm/s and PSVratio across the stenosis >3.5—correlating with >70% DRstenosis (Fig. 6) [7]. In some instances, the only indication ofa “failing” bypass is the identification of a low (o40 cm/s)graft PSV. Low graft flow velocity predicts thrombosis andrequires a detailed search for an occlusive lesion, includingthe use of angiographic imaging if duplex did not detection agraft stenosis. When conducted appropriately, a duplexsurveillance program should result in an intervention graftfor detected stenosis of approximately 20% within the firstyear and a failure rate of o5% per year.

Interpretation of initial duplex scan is based on color flowmapping for occlusive or aneurysm lesions and velocityspectra changes associated with stenosis. Sites of stenosis(PSV >180/cm/s, Vr >2) are interpreted as abnormal, andpredict the arterial repair is prone to development of myoin-timal hyperplasia and more likely to require revision than ifthe duplex testing is normal. Residual repair-site abnormal-ities should be scanned more frequently (at 6- to 8-weekintervals) to detect stenosis progression, which can occur inthe absence of symptoms, especially in patients treated forcritical limb ischemia. In patients with lower-limb arterialbypass grafts, the measurement of mean graft flow velocity,calculated as the mean PSV recorded from two to threenonstenotic graft sites, correlates with volume flow and,when low (o40 cm/s), signifies a graft at risk for thrombosis.A >30 cm/s reduction of mean graft velocity (MGV) on serialscans indicates development of a hemodynamically signifi-cant stenosis that should be repaired. Graft flow velocitymight be o40 cm/s in large-caliber (>6 mm diameter) graftsor bypasses to a pedal or isolated tibial artery.

Fig. 6 – Surveillance algorithm after bypass grafting and endovascular therapy. ABI, ankle-brachial index; PSV, peak systolicvelocity; Vr, velocity ratio.

Table 4 – Risk stratification for graft thrombosis based on vascular laboratory testing data, including peak systolic velocityat stenosis, velocity ratio at stenosis, graft flow velocity, and change in ankle-brachial index.

Category High-velocity criteria, PSV(cm/s)

Velocityratio

Low-velocity criteria, GFV(cm/s)

ABIchange

I Highest riska (>70% stenosis with low graft flow) >300 >3.5 o45 or staccato graft flow >0.15II High riska (>70% stenosis without change or normal

graft flow)>300 >3.5 >45 o0.15

III Moderate riskb (50% to 70% stenosis with normal graftflow)

180�300 >2.0 >45 o0.15

IV Low risk (normal bypass or o50% stenosis withnormal graft flow)

o80 o2.0 >45 o0.15

ABI, ankle-brachial index; GFV, graft flow velocity; PSV, peak systolic velocity.a 40% to 50% likelihood of stenosis progression or graft thrombosis within 3�6 monthsb 20% to 30% of early (o 3 months) lesions regress, 10%�20% of lesions remain stable, 40% to 50% progress to >70% stenosis.

S E M I N A R S I N V A S C U L A R S U R G E R Y 2 6 ( 2 0 1 4 ) 9 5 – 1 0 4100

If color Doppler imaging identifies a stenosis, measure-ments of PSV and Vr are performed, as well as recordinglesion length and graft/vessel diameter. Lesions with duplex-derived velocity spectra of a high-grade stenosis (PSV >300cm/s, end-diastolic velocity >20 cm/s, Vr across the stenosis>3.5) correlate with a >70% DR stenosis and should berepaired [8]. The application of these threshold criteria hasbeen shown to correctly identify arterial repairs (bypassgrafts, angioplasty sites) at risk for thrombosis. If a policy ofno intervention is followed, the incidence of thrombosis is inthe range of 25% during the subsequent 6-month interval[9,10].The risk of infrainguinal bypass thrombosis is predicted by

using the combination of high- and low-velocity duplexcriteria and decreases in ABI (Table 4) [11]. In the highest-risk group (category I), development of a pressure-reducingstenosis has produced low flow in the graft, which, if itdecreases below the “thrombotic threshold velocity,” willresult in thrombosis. Prompt repair of category I lesions isrecommended and category II lesions can be scheduled forelective repair within 1 to 2 weeks. A category III stenosis(PSV of 150 to 300 cm/s, Vr o3.5) is not pressure or flowreducing in the resting limb. Serial scans at 4- to 6-weekintervals are recommended to determine the hemodynamicprogression course of these lesions [12,13]. An importantfeature of the “graft-threatening” stenosis is its propensityto progress in severity, reduce graft flow, and form surfacethrombus; events that can precipitate thrombosis. Usingserial duplex scans, a category III stenosis that does notprogress can be distinguished from the progressive lesionthat needs to be repaired. The majority (approximately 80%)of bypass grafts will have no stenosis identified, ie, a categoryI scan. For these patients, surveillance at 6-month intervals isrecommended. Category I scan but graft flow velocity o40cm/s indicates a “low-flow” bypass and increased risk forthrombosis, based on the concept of “thrombotic thresholdvelocity,” which is lower in autologous vein than prostheticbypasses. Prescribing an anticoagulation regimen of sodiumwarfarin to maintain the prothrombin time at an interna-tional normalized ratio of 1.6 to 2 and antiplatelet therapy(aspirin, 81 mg/d or clopidogrel bisulfate, 75 mg/d) can reducethe incidence of low-flow vein or prosthetic polytetrafluoro-ethylene (MGV o 60 cm/s) bypass graft thrombosis [14].Using duplex ultrasound surveillance, it can be anticipated

that approximately 20% to 30% of infrainguinal bypass graftand peripheral angioplasty sites will have a >70% stenosisidentified within the first year. The likelihood of stenosisdeveloping is influenced by a number of factors; but the mostpredictive is the presence of a residual stenosis (PSV >180 cm/s, Vr >20) in the repair. An arterial intervention with residual

stenosis should be monitored at 1- to 2-month intervals forstenosis progression. After lower-limb vein bypass grafting,risk factors for graft stenosis include small vein caliber (4mm), presence of venovenous anastomosis (spliced vein), useof alternative venous conduits (eg, arm vein, lesser saphe-nous vein, greater saphenous vein remnants), prior intra-operative graft revision, or a history of early graftthrombectomy. The majority of stenosic lesions producingbypass graft or angioplasty failure are focal (o2 cm in length)and suitable for endovascular treatment. More extensive, ie,longer segment, stenosis or early-appearing (o3 months)lesions might require surgical repair (bypass graft) or stentangioplasty (balloon angioplasty, atherectomy). Timely inter-vention for repair-site stenosis is associated with improvedstenosis-free patency in the range of 70% to 80% at 1 year[7,13]. For infrainguinal vein bypass, patency at 2 years wasidentical for surgical (63%) and endovascular (63%) repair ofduplex-detected stenosis, and overall assisted graft patencyby life-table analysis was 91% at 1 year and 80% at 3 years.Based on the costs of duplex surveillance, the salvage of 10%of arterial intervention, ie, thrombosis prevented, is costeffective. Many vascular groups believe duplex surveillanceshould be “part of the service” after infrainguinal vein bypassgrafting. It should be emphasized that the benefit of surveil-lance is highly dependent on the durability and morbidity ofprocedures used to correct “graft stenosis.” Importantly,duplex surveillance with intervention for >70% stenosis issafe, with mortality rate o0.5%, early failure rate o1%, andintermediate (3-month) failure rate o15%.

4. Summary

Arterial duplex scanning is a versatile and accurate diagnos-tic method for the diagnosis, screening, and management oflower-limb vascular disease. Although resolution of com-puted tomography angiography continues to improve, pro-viding detailed artery anatomy and three-dimensional vesselreconstruction, these advancements cannot replace duplexultrasound because of its noninasive nature, lower cost, andportability. Also, an important advantage of duplex ultra-sound is the hemodynamic information afforded to assesslimb perfusion before and after intervention. The interpreta-tion of arterial duplex testing requires knowledge of anatomy,hemodynamics, and the spectrum of artrerial interventionsused to treat occlusive and aneurysmal disease. This exper-tise can be demonstrated by obtaining certification from theAmerican Registry for Diagnostic Medical Sonography (www.ARDMS.org) as a registered physician in vascular laboratoryinterpretation.

S E M I N A R S I N V A S C U L A R S U R G E R Y 2 6 ( 2 0 1 4 ) 9 5 – 1 0 4 101

Questions regarding arterial duplex test interpretation

1. The duplex scan image of the right groin after a cardiac catheterization procedure demonstrates:

A. Femoral artery stenosisB. Femoral false aneurysmC. Arteriovenous fistulaD. Femoral artery dissection

Answer:

2. The duplex scan of the right femoral artery demonstrates:

A. false aneurysmB. arteriovenous fistulaC. cavernous venous aneurysmD. femoral aneurysm rupture

Answer:

3. The duplex scan image of the external iliac artery stent demonstrates:

S E M I N A R S I N V A S C U L A R S U R G E R Y 2 6 ( 2 0 1 4 ) 9 5 – 1 0 4102

A. o50% stenosisB. 50% to 70% stenosisC. o70% stenosisD. o90% stenosis

Answer:

4. The duplex scan images of a femoral-popliteal vein bypass show the graft is:

A. Low risk for failureB. Moderate risk for failureC. High risk for failureD. Cannot determine failure risk without an angiogram

Answer:

5. The duplex scan images of the superficial femoral artery indicate:

A. Stenosis with exit collateral arteryB. Focal arterial dissectionC. Arteriovenous fistulaD. False aneurysm

Answer:

Answers to questions:

1. C2. A3. C4. C5. C

S E M I N A R S I N V A S C U L A R S U R G E R Y 2 6 ( 2 0 1 4 ) 9 5 – 1 0 4 103

r e f e r e n c e s

[1] Strandness DE Jr. Duplex Scanning in Vascular Surgery.Philadelphia: Lippincott Williams & Wilkins; 2002. (A com-prehensive review of clinical applications for duplexscanning.)

[2] Zwiebel WJ, Pellerito JS, Introduction to Vascular Ultrasonog-raphy, 5th ed. Philadelphia: Elsevier Saunders; 2005. (Anintroductory textbook for vascular technologists that pro-vides background physics, hemodynamics, and testing pro-cedures for duplex scanning.)

[3] Cossman DV, Ellison JE, Wagner WH, et al. Comparison ofcontrast arteriography to arterial mapping with color-flowduplex imaging in the lower extremities. J Vasc Surg1989;10:522–9.

[4] Moneta GL, Yeager RA, Antonovic R, et al. Accuracy oflower extremity arterial duplex mapping. J Vasc Surg 1992;15:275–84.

[5] Moneta GL, Yeager RA, Lee RW, Porter JM. Noninvasivelocalization of arterial occlusive disease: a comparison ofsegmental Doppler pressures and arterial duplex mapping. JVasc Surg 1993;17:578–82.

[6] Hatsukami TS, Primozich JF, Zierler RE, et al. Color Dopplerimaging of infrainguinal arterial occlusive disease. J VascSurg 1992;16:527–33.

[7] Gonsalves C, Bandyk DF, Avino A, Johnson BL. Duplex featuresof vein graft stenosis and the success of percutaneous trans-luminal angioplasty. J Endovasc Surg 1999;6:66–72.

[8] Ascher E, Marks NA, Hingorani AP, et al. Duplex-guidedendovascular treatment of occlusive and stenotic lesions ofthe femoral popliteal arterial segment: a comparative studyin the first 253 cases. J Vasc Surg 2006;38:251–8.

[9] Caps MT, Cantwell-Gab K, Bergelin RO, Strandness DE Jr. Veingraft lesions: time of onset and rate of progression. J VascSurg 1995;22:466–75.

[10] Bandyk DF, Johnson BL, Gupta AK, Esses GE. Nature and mana-gement of duplex abnormalities encountered during infraingui-nal vein bypass grafting. J Vasc Surg 1996;24:430–8.

[11] Gupta AK, Bandyk DF, Cheanvechai D, Johnson BL. Naturalhistory of infrainguinal vein graft stenosis relative to bypassgrafting technique. J Vasc Surg 1997;25:211–25.

[12] Westerband A, Mills JL, Kistler S, et al. Prospective validation ofthreshold criteria for intervention in infrainguinal vein graftsundergoing duplex surveillance. Ann Vasc Surg 1997;11:44–8.

[13] Armstrong PA, Bandyk DF, Wilson JS, Shames ML, JohnsonBL, Back MR. Optimizing infrainguinal arm vein bypasspatency with duplex ultrasound surveillance and endovas-cular therapy. J Vasc Surg 2004:724–31.

[14] Brumberg SR, Back MR, Armstrong PA, et al. The relativeimportance of graft surveillance and warfarin therapy in infrain-guinal prosthetic bypass failure. J Vasc Surg 2007;46:1160–6.

S E M I N A R S I N V A S C U L A R S U R G E R Y 2 6 ( 2 0 1 4 ) 9 5 – 1 0 4104