Skip to main content

Evaluation of Lower Extremity Bypass Grafts

Color Doppler imaging paired with duplex ultrasound is the ideal imaging modality for the evaluation of arterial bypass grafts. The modality is inexpensive, reproducible and non-invasive. The lower extremity bypass graft is also ideally placed for the successful application of Doppler ultrasound imaging. The typical infra-inguinal bypass graft lies relatively superficial in the subcutaneous tissues of the leg. The proximal anastomosis is readily accessible to imaging since it often originates from the common femoral artery. The distal anastomoses can, on occasion be difficult to evaluate, especially if it lies at the more distal superficial femoral artery or at the peroneal artery in the mid calf.

Use of color Doppler imaging has shown to benefit patients having had bypass grafting. Before its' implementation, bypass graft occlusions occurred without warning. With the introduction of ultrasound based surveillance protocols, the incidence of abrupt occlusions has dramatically decreased. Developing stenoses that used to lead to graft occlusions are now detected early and corrected. In the following sections, we will review the logistics of bypass graft surveillance and discuss its implementation.

Types of lower extremity bypass grafts

There are two broad types of lower extremity bypass grafts used in most clinical practices: those made of synthetic material and native veins harvested from the patient.

The synthetic bypass grafts are typically made of polytetrafluoroethylene (also referred as "Gore-Tex"). They have shown themselves to be relatively reliable for bypass operations where the distal anastomosis is placed in the above the knee popliteal artery. Their patency rates (likelihood of staying open) for below-the-knee applications are relatively low (figure 1).


Figure 1: This PTFE bypass graft is occluded. The sharp double line appearance (arrow) of the wall is typical of grafts made of PTFE. Absence of blood flow signals within the graft lumen confirms total occlusion of the graft.

To view an enlargement, click on the image.

The alternative to using native synthetic materials is to harvest a patient's own vein and to re-introduce it as a substitute conduit for the occluded or diseased arterial segment. Reversed vein grafts are created by removing, reversing and connecting the greater saphenous vein to bypass the diseased artery segment, although the lesser saphenous veins, the basilic and cephalic veins can also be used. The advantage is a vein segment that has been minimally manipulated and that by reversing the vein, blood flow will not be obstructed by the vein valves. The disadvantage is a potential mismatch in the size of the vein and the artery to which it is anastomosed. This is mostly a problem at the distal anastomosis where a large diameter vein is anastomosed to a relatively small diameter artery (figure 2).


Figure 2: Distal anastomosis of a reversed saphenous vein bypass graft. The vein lumen (arrowheads) is at least twice the diameter of the recipient tibial artery (arrows). This size mismatch is the cause of a relative velocity mismatch between graft blood flow velocities and the native artery.

To view an enlargement, click on the image.

A surgical alternative is to leave as much as possible of the native artery "in-place", or "in-situ" technique. This can only be done for the greater saphenous or the lesser saphenous veins. In the case of the greater saphenous vein, the vein can be used for above-the-knee bypass operations as well as can be extended all the way to the calf arteries. For the lesser saphenous vein, only operations bypassing diseased distal popliteal or proximal tibio-peroneal arteries are possible. In both cases, the vein valves would normally prevent blood flow towards the feet. The surgeon has to disrupt the integrity of the vein valves, in order for the vein graft to work properly. This can be done with the use of various surgical instruments. In addition, the communicating veins that are present between the superficial and deep veins also need to be ligated. If this is not done, the vein graft can effectively develop arterio-venous fistula through these still open perforating veins. One advantage to this approach is a size match between vein and artery at the anastomoses (figure 3).


Figure 3: Distal anastomosis of an in-situ bypass graft. The size of the graft (arrow) matches the diameter of the tibial artery (arrowhead).

To view an enlargement, click on the image.

Source of graft dysfunction

Manipulation of the vein conduit that can cause ischemic or mechanical injury to the endothelial lining of the vein and promote the development of fibro-intimal hyperplasia, a proliferation of smooth muscle cells that leads to the growth of stenotic lesions. These lesions progressively grow until they obstruct normal blood flow and ultimately cause occlusion of the bypass graft.

In synthetic bypass grafts, these lesions are likely to develop at the proximal or distal anastomoses whereas in native vein grafts, the stenotic lesions can be located anywhere in the vein conduit. The basic principle of graft surveillance is to detect developing stenoses due to fibro-intimal hyperplasia before they become severe enough to cause graft occlusion (figure 4).


Figure 4: This vein graft has absent blood flow signals. This is consistent with a total occlusion. This occlusion was an incidental finding in the early post-operative period with the patient remaining asymptomatic.

To view an enlargement, click on the image.

Color Doppler and duplex sonography monitoring: general principles

Bandyk et al. first showed that a peak-systolic velocity below 40 or 45 cm/sec can be used to identify grafts with severe stenoses and compromised volume blood flow. Such a measurement of absolute vein graft blood flow velocity can identify the more severely diseased grafts but will miss stenotic lesions before they have become severe enough to restrict volume blood flow (figure 5). An alternate strategy is used to detect lesions early enough so that interventions can be performed and preserve vein graft patency.


Figure 5a: The blood flow velocities in this bypass graft are severely depressed. They indicate the presence of a high grade stenosis somewhere in the graft, likely at a distal location. The reciprocating pattern indicates blood return from a compliant but occluded graft.

To view an enlargement, click on the image.



Figure 5b: The depressed blood flow velocities are accompanied by a slow rise time and low resistance pattern. This is consistent a proximal high grade stenosis in the graft.

To view an enlargement, click on the image.


The diagnosis and grading of the severity of stenotic lesions is done as follows:

Color Doppler imaging is used to survey the length of the bypass graft and to localize possible lesions; duplex ultrasound is then used to obtain Doppler waveforms at the site of possible abnormal blood flow patterns (figure 6a). The velocity information on the Doppler waveforms is then used to grade the severity of the lesions. The velocity parameter best suited to grade peripheral arterial lesions in bypass grafts is the peak-systolic velocity ratio. The peak-systolic velocity ratio is derived as follows. The elevated peak-systolic velocity is obtained at the site of a suspected stenosis as identified on the color Doppler image (figure 6b). The Doppler sample gate is then displaced just proximal to this focus and a second peak-systolic velocity measurement is obtained (figure 6c). These two peak-systolic velocities are then compared: the elevated peak-systolic velocity is divided by the peak-systolic velocity in the "normal" segment. The resultant ratio is indicative of the severity of the stenotic lesion. A ratio of 2 or above indicates the presence of a 50% or greater stenosis. A velocity ratio of 4 or above indicates a stenotic lesion of 75% or greater. In our case (figure 6), the velocity ratio is 329/59.7, and is greater than 4 (5.5). The peak-systolic velocity ratio corresponding to a 75% diameter stenosis has varied over the years: originally, a velocity ratio of 3.0 was considered to be indicative of a 75% stenosis. Subsequently, a ratio of 3.5 or 3.7 was judged indicative of 75% stenosis. Finally, a value of 4.0 or more is considered a robust indicator of a greater than 75% diameter stenosis. An alternate diagnostic criterion is an absolute velocity cut-point of 200 cm/sec. The problem with adopting an absolute cut-point is the possibility that, overall, volume blood flow might be severely depressed in the graft or arterial segment being studied. The blood flow velocity may therefore not increase above an absolute velocity threshold, even at the site of a very severe stenotic lesion.

End-diastolic velocities, if they are elevated above 100 to 120 cm/sec, indicate the presence of a severe stenotic lesion. However, end-diastolic velocities are often not elevated despite the presence of a severe stenosis (figure 6b).

Because of these various issues, the peak-systolic velocity ratio is judged as the most reliable diagnostic criterion to use for grading the severity of stenotic lesions in a bypass graft.

The absence of flow signals is indicative of total occlusions with a high level of accuracy.


Figure 6a: This color Doppler image shows an area of aliasing in the mid portion of a femoro-tibial bypass graft. This is indicative of a site that requires additional evaluation with Doppler waveform analysis.

To view an enlargement, click on the image.



Figure 6b: The site of maximal blood flow velocity corresponds to a blood flow velocity of 329 cm/sec.

To view an enlargement, click on the image.



Figure 6c: A Doppler blood flow velocity estimate 2 cm above this site shows a blood flow velocity of 59.7 cm/sec.

To view an enlargement, click on the image.



Figure 6d: Within 1 cm of the stenosis, blood flow velocities are significantly decreased but still show evidence of some flow disturbance. The flow disturbance associated with a vein graft stenosis can typically resolve within 2 cm and the Doppler tracing return to a normal appearance (figure 6a).

To view an enlargement, click on the image.

Natural history of native vein graft dysfunction

Different mechanisms are responsible for possible bypass graft failure during the post-operative period. In the case of in-situ vein grafts, failures that are seen within one month of surgery are normally due to technical errors. These errors include poor suture line placement, the opening of unsuspected venous channels in the in-situ grafts, poor selection of anastomotic sites, and retained vein valves. Fibro-intimal or fibrotic lesions typically develop between 6 months to 2 years after surgery. The lesions are located either at the anastomosis or within the graft conduit, most often at the site of a vein valve. Late failures beyond this two year time interval are likely secondary to continued progression of the atherosclerotic process in the native vessels proximal and distal to the anastomoses. The bypass graft can also develop atherosclerotic type lesions over time.

Vein graft surveillance protocols for native vein grafts

Color Doppler imaging is used to systematically survey the length of the bypass graft and to detect possible stenotic lesions before they progress to become flow restrictive.

The region of the proximal anastomosis is evaluated as follows: A Doppler velocity sample is taken in the native artery above the anastomosis (figure 7). This insures that inflow is sufficient to support the graft. Alterations in this waveform (loss of flow reversal in early diastole) can also be used to detect a lesion in the inflow artery. Such lesions may be due to native arterial disease or might have been caused during the surgical intervention when the graft was placed. This lesion can then be graded according to a velocity ratio. Obtaining a reference value for the peak-systolic velocity ratio above the site of the lesion might not be possible if the site is in the iliac artery. Using an absolute value in peak-systolic velocity above 200 cm/sec can then serve as a diagnostic criterion for the presence of a greater than 50% stenosis. The region of the proximal anastomosis is then evaluated. A Doppler sample is taken in the anastomosis proper and in the first 1 cm of the bypass graft (figure 8). The anastomosis can sometimes be artificially enlarged. This "patulous" appearance is purposeful since it permits possible ingrowths of hyperplasic tissue without necessarily causing a stenosis (figure 9). The Doppler velocity estimate must then be compared to the velocity measured in the first 4 to 5 cm of the bypass graft (Figure 10 a and b). This takes into account size differences between native artery and graft, especially when a small diameter graft has been anastomosed to a large diameter artery. In such a case, the step up in velocity from large artery to small graft could be misinterpreted as a stenosis. The simple difference in diameters between the graft and of the artery can account for a velocity increase in the graft that is without clinical importance. A more appropriate baseline should be established by comparing the proximal graft velocity to that in the graft conduit 5 cm downstream.

The graft surveillance protocol continues by displacing the ultrasound probe along the course of the graft (figure 10 c). If no step-up in blood flow velocities are detected, a sample should be take at every 10 cm (roughly 4 transducer lengths). If there is a zone of blood flow disturbance on the color Doppler image, the peak-systolic velocity ratio is calculated by dividing the peak-systolic velocity measured at the site of flow disturbance by that measured in the portion of the graft 2 to 4 cm proximal. Peak-systolic velocity ratios of 2 or more correspond to 50% diameter stenosis. Peak-systolic velocity ratios of 3 to 4 correspond to 75% diameter stenosis.


Figure 7: This velocity sample is taken in the lower external iliac artery, just above the anastomosis of a femoro-popliteal bypass graft. The proximal graft anastomosis is identified by the arrow.

To view an enlargement, click on the image.



Figure 8: This Doppler sample is taken within the proximal anastomosis. Blood flow velocities are low normal and there is no evidence of a stenosis. The Doppler waveform appears distorted. This is common to the proximal anastomosis where the anastomotic connection is made larger than either he native artery or the graft conduit (see figure 9).

To view an enlargement, click on the image.



Figure 9: This proximal common femoral artery anastomosis is protuberant (also called "patulous"). The size of the outpouching that results after the creation of the anastomosis is highly variable. Doppler tracings within this area are unreliable. Doppler samples immediately before and immediately after the anastomosis are better representative of blood flow dynamics.

To view an enlargement, click on the image.



Figure 10a: The native superficial femoral artery is sampled proximal to an anastomosis of a superficial femotal to tibial artery bypass graft.

To view an enlargement, click on the image.



Figure 10b: The blood flow velocities at the anastomosis and in the first few cm of the bypass graft show no significant change.

To view an enlargement, click on the image.



Figure 10c: The full length of the graft is interrogated with color Doppler imaging. Sampling of the Doppler waveform is done every 10 cm if no region of abnormal blood flow patterns is identified.

To view an enlargement, click on the image.



Figure 10d: The distal conduit is sampled a few cm proximal to the bypass graft.

To view an enlargement, click on the image.



Figure 10e: The region of the anastomosis proper is also being interrogated. No abnormal elevations in blood flow velocities are detected.

To view an enlargement, click on the image.



Figure 10f: The native artery distal to the anastomosis is also interrogated to insure that no unsuspected lesions are present in the native run-off.

To view an enlargement, click on the image.

Graft surveillance then continues until the distal anastomosis is identified (figure 10 d, e, and f). Blood flow velocity samples should be taken in the distal graft proper, at the anastomosis where the graft touches the native artery and 1 to 2 cm downstream. Stenosis often occur at the anastomosis proper (figure 11 a, b and c). It is also good practice to sample the native artery for a distance of 4 to 6 cm further downstream. This serves two purpose: it establishes a baseline against which to compare the distal graft velocities and it ensures that there are no outflow lesions are developing in the native artery (figure 12).


Figure 11a:The distal conduit of this graft has peak systolic velocities of 51 cm/sec.

To view an enlargement, click on the image.



Figure 11b: Blood flow velocities increase near the distal anastomosis due to the beginning of some tapering likely due to developing fibro-intimal hyperplasia.

To view an enlargement, click on the image.



Figure 11c: The stenosis sits at the anastomosis proper with peak velocities of 509 cm/sec.

To view an enlargement, click on the image.



Figure 12a:Blood flow velocities in this distal bypass graft anastomosis show no evidence of significant stenosis.

To view an enlargement, click on the image.



Figure 12b: This color Doppler image shows the tibial artery distal to the anastomosis (arrow). The site of aliasing indicates the presence of a stenosis.

To view an enlargement, click on the image.


  Figure 12c: Blood flow velocities are sampled in the native artery, distal to the anastomosis but proximal to the site of abnormality on the color Doppler image. These are somewhat elevated at 125 cm/sec)

  Figure 12d: Peak systolic velocities are elevated (295 cm/sec) at the point of abnormal color Doppler signals. This indicates a focal stenosis of 50 to 75% in the native artery distal to the graft.

The role of graft surveillance is well defined: detection of significant lesions and their correction in order to prevent graft occlusion. The cost-benefit of ultrasound based surveillance programs is difficult to evaluate. A significant lesion is one associated with increased blood flow velocities causing a peak-systolic velocity ratio of 3 to 4 (or more). Typical surveillance is done every 6 months, with the first visit being at 3 months following surgery. There has been evidence that early bypass graft lesions can be detected as early as 1 to 3 months following surgery. It would seem logical to then focus on this group that accounts for 30 to 50% of patients since they form a subset of grafts at high risk of occluding over the following year. Unfortunately, there are false positives as well as false negatives. This makes it necessary to apply the surveillance protocol to all patients who have undergone a vein bypass placement. While the use of surveillance protocols has improved graft patency rate, further improvements in efficacy should be possible. One proposed strategy would then be as follows: look, with color Doppler, imaging for early lesions. If the graft has an early lesion, than follow this graft more closely, every 3 months. If the graft has no stenosis on two serial visit, consider extending the follow-up period. Cases with borderline lesions that have velocity ratios of 3 to 4 should be followed closely. A velocity ratio of 4 should be considered as a cut-point for intervention.

Repeat intervention, most often surgical revision of the developing stenosis, is indicated since these lesions, if left alone, ultimately progress to cause bypass graft occlusions.

Immediate post-operative period:

Peri-graft hematomas can often be seen following graft placement. While they may temporary affect flow dynamics in the graft, they are normally of no consequence for long term graft patency (figure 13).


Figure 13: This hypoechoic collection just superior to a bypass graft is consistent for a peri-operative hematoma. It does not directly compromise blood flow dynamics in the graft proper.

To view an enlargement, click on the image.

Arterio-venous fistulas are often seen after placement of an in-situ vein bypass graft (figure 14). Intraoperative Doppler ultrasound can detect most of these fistula and guide their ligation. However, some of these lesions are missed. Post-operative detection of sites of A-V communication between the in-situ superficial veins and the deep veins can then be done with Doppler ultrasound. Identification with ultrasound is oten al that is needed before surgical correction.


Figure 14: This side branch from a greater saphenous in situ bypass graft contains low resistance velocities consistent with a fistula. This side branch was later ligated.

To view an enlargement, click on the image.

Synthetic vascular bypass grafts

There is normally no need for a formal surveillance protocol following the placement of synthetic bypass grafts. In the first and second years following operation, a stenosis can develop at anastomosis because of fibro-intimal hyperplasia. This is likely due in part to the tendency for blood flow eddies to develop as the graft tapers into the anastomosis and affects the wall of the native artery An anastomotic stenosis will typically cause a marked increase in the Doppler velocity signals sampled at the anastomosis or beyond. An abrupt increase in blood flow velocity of 100% can be due to the geometry of the anastomotic. As discussed in the case of a reversed vein graft, this is due to the mechanics of connecting a relatively large diameter graft to a smaller artery. This geometry can cause up to a doubling in blood flow velocity without being indicative of a pathologic lesion. There are no studies addressing the actual incidence and significance of this finding. Serial monitoring of these sites of disturbed flow may be used with the premise that an increase in velocity over a few months is indicative of a developing stenosis. Although this process is similar to that described for vein grafts, there does not appear to be any benefit for a systematic surveillance program.


Color Doppler and duplex sonography are practical tools that have proven advantages in the post-procedure monitoring of vascular bypass grafts. Knowledge of the type of graft placed, the site of the anstomoses, the date of placement and any interventions in then time since placement are essential to the performance of bypass vein graft surveillance.


Bandyk DF, Cato RF, Towne JB. A low flow velocity predicts failure of femoropopliteal and femorotibial bypass grafts. Surgery 1985; 98:799-809.

Bandyk DF, Jorgensen RA, Towne JB. Intraoperative assessment of in situ saphenous vein arterial bypass grafts using pulsed Doppler spectral analysis. Arch Surg 1986; 121:292-299.

Buth J, Disselhoff B, Sommeling C, Stam L. Color-flow duplex criteria for grading stenosis in infrainguinal vein grafts. J Vasc Surg 1991; 14:716-728.

Cohen JR, Mannick JA, Couch NP, Whittemore AD. Recognition and management of impending vein-graft failure. Importance for long-term patency. Arch Surg 1986; 121:758-759.

Fasih T, Rudol G, Ashour H, Mudawi A, Bhattacharya V. Surveillance versus nonsurveillance for femoro-popliteal bypass grafts. Angiology 2004; 55:251-6.

Ferris BL, Mills JL, Sr., Hughes JD, Durrani T, Knox R. Is early postoperative duplex scan surveillance of leg bypass grafts clinically important? Journal of Vascular Surgery 2003; 37:495-500.

Gibson KD, Caps MT, Gillen D, Bergelin RO, Primozich J, Strandness DE, Jr. Identification of factors predictive of lower extremity vein graft thrombosis. Journal of Vascular Surgery. 2001; 33:24-31.

Golledge J, Wright I, Lane IF. Comparison of clinical follow-up and duplex surveillance of infrainguinal vein bypasses. Cardiovascular Surgery. 1996; 4:766-70.

Grigg MJ, Nicolaides AN, Wolfe JH. Detection and grading of femorodistal vein grafts stenoses: duplex velocity measurements compared with angiography. J Vasc Surg 1988; 8:661-666.

Idu MM, Blankestein JD, de Gier P, Truyen E, Buth J. Impact of a color-flow duplex surveillance program on infrainguinal vein graft patency: A five-year experience. J Vasc Surg 1993; 17:42 -53.

Idu MM, Buth J, Hop WC, Cuypers P, van de Pavoordt ED, Tordoir JM. Factors influencing the development of vein-graft stenosis and their significance for clinical management. European Journal of Vascular & Endovascular Surgery. 1999; 17:15 -21.

Ihnat DM, Mills JL, Dawson DL, et al. The correlation of early flow disturbances with the development of infrainguinal graft stenosis: a 10-year study of 341 autogenous vein grafts. Journal of Vascular Surgery. 1999; 30:8-15.

Johnson BL, Bandyk DF, Back MR, Avino AJ, Roth SM. Intraoperative duplex monitoring of infrainguinal vein bypass procedures. Journal of Vascular Surgery. 2000; 31:678-90.

Mills JL, Sr., Wixon CL, James DC, Devine J, Westerband A, Hughes JD. The natural history of intermediate and critical vein graft stenosis: recommendations for continued surveillance or repair. Journal of Vascular Surgery 2001; 33:273-8; discussion 278-80.

Mills JL, Harris EJ, Taylor LM, Jr., Beckett WC, Porter JM. The importance of routine surveillance of distal bypass grafts with duplex scanning: a study of 379 reversed vein grafts. J Vasc Surg 1990; 12:379-389.

Mills JL, Bandyk DF, Gathan V, Esses GE. The origin of infrainguinal vein graft stenosis: a prospective study based on duplex surveillance. J Vasc Surg 1995; 21:16 -25.

Mohan C, Hoballah J, Schueppert M, et al. Should all in situ saphenous vein bypasses undergo permanent duplex surveillance? Arch Surg 1995; 130:483-8.

Nielsen TG. Natural history of infrainguinal vein bypass stenoses: early lesions increase the risk of thrombosis. European Journal of Vascular & Endovascular Surgery. 1996; 12:60-4.

Nielsen TG, Jensen LP, Schroeder TV. Early vein bypass thrombectomy is associated with an increased risk of graft related stenoses. European Journal of Vascular & Endovascular Surgery. 1997; 13:134-8.

Polak JF, Donaldson MC, Dobkin GR, Mannick JA, O'Leary DH. Early detection of saphenous vein arterial bypass graft stenosis by color-assisted duplex sonography: a prospective study. AJR 1990b; 154:857-861.

Sanchez LA, Suggs WD, Veith FJ, Marin ML, Wengerter KR, Panetta TF. Is surveillance to detect failing polytetrafluoroethylene bypasses worthwhile: twelve-year experience with 91 grafts. J Vasc Surg 1993; 18:981-990.

Teo NB, Mamode N, Murtagh A, Breslin P, Pollock JG. Effectiveness of surveillance of infrainguinal grafts. European Journal of Surgery. 2001; 167:605-9.

Whittemore AD, Clowes AW, Couch NP, Mannick JA. Secondary femoropopliteal reconstruction. Ann Surg 1981; 193:35-42.


Enjoy The Savings

IAME's Unlimited CME Plan is now the internet's best value for online CME in ultrasound. Just $75!

Your CME credits are available at any time in your Online CME Control Panel. They are automatically transferred to the ARDMS/APCA CME Bank and RSNA's CME Gateway (when you include your credentials).

"Your CME course site is easy to navigate. Overall I am a happy customer."

Mary Hamer MD, Lake City, Florida