Choice of bypass graft material for lower-limb arterial bypasses


A person with severely diseased arteries in one or both legs can experience pain on walking (intermittent claudication), pain at rest, or death of tissues in the leg. When the main thigh artery has a long blockage, the best option is to insert a bypass to carry the blood from an artery with good blood flow to the affected artery below the blockage. Bypass is intended to improve walking, or to save limbs that might otherwise require amputation. The different types of material available to create the bypass include the person's own vein (autologous vein), human umbilical vein, and the prosthetic materials polytetrafluoroethylene (PTFE) or Dacron, alone or with the blood thinning agent heparin bonded to the inside of the graft. Bypass grafts extending to below the knee are not as effective at remaining patent (open) with good blood flow as those above the knee. The aim of this review was to determine the most effective type of material to use for above-knee and below-knee bypass grafts.

Study characteristics and key results

We identified 19 randomised controlled trials that included a total of 3123 people. Of these people, 2547 were given above-knee bypass grafts and 576 were given bypass grafts below the knee. The evidence in our review is current until 13 March 2017. From our analysis, we found that grafts made from a person's own vein had a better primary patency (blood flow) rate than the prosthetic materials PTFE or Dacron for above-knee bypass grafts. Meanwhile, Dacron (and possibly also human umbilical vein) achieved better blood flow (patency) than PTFE. We also found that Dacron with supporting rings around it (designed to prevent external compression) showed worse patency than non-supported Dacron when used in grafts above the knee.

Adding a 'cuff' of vein did not improve the patency of PTFE for grafts extending to below the knee. The included trials provided few results on how long people's limbs survived following the bypass procedure. There was not much consistency between the trials (and sometimes within the trials) with regards to people taking additional medications such as antiplatelets or anticoagulants, and this might have affected the results.

Quality of the evidence

The overall quality of the evidence ranged from very low to moderate. Issues which affected the quality of the evidence included differences in the design of the trials, and differences in the types of grafts they compared. These differences meant we were often only able to combine and analyse small numbers of participants and this resulted in uncertainty over the true effects of the graft type used.

Authors' conclusions: 

There was moderate-quality evidence of improved long-term (60 months) primary patency for autologous vein grafts when compared to prosthetic materials for above-knee bypasses. In the long term (two to five years) there was low-quality evidence that Dacron confers a small secondary patency benefit over PTFE for above-knee bypass. Only very low-quality data exist on below-knee bypasses, so we are uncertain which graft type is best. Further randomised data are needed to ascertain whether this information translates into an improvement in limb survival.

Read the full abstract...

Femoro-popliteal bypass is implemented to save limbs that might otherwise require amputation, in patients with ischaemic rest pain or tissue loss; and to improve walking distance in patients with severe life-limiting claudication. Contemporary practice involves grafts using autologous vein, polytetrafluoroethylene (PTFE) or Dacron as a bypass conduit. This is the second update of a Cochrane review first published in 1999 and last updated in 2010.


To assess the effects of bypass graft type in the treatment of stenosis or occlusion of the femoro-popliteal arterial segment, for above- and below-knee femoro-popliteal bypass grafts.

Search strategy: 

For this update, the Cochrane Vascular Information Specialist searched the Vascular Specialised Register (13 March 2017) and CENTRAL (2017, Issue 2). Trial registries were also searched.

Selection criteria: 

We included randomised trials comparing at least two different types of femoro-popliteal grafts for arterial reconstruction in patients with femoro-popliteal ischaemia. Randomised controlled trials comparing bypass grafting to angioplasty or to other interventions were not included.

Data collection and analysis: 

Both review authors (GKA and CPT) independently screened studies, extracted data, assessed trials for risk of bias and graded the quality of the evidence using GRADE criteria.

Main results: 

We included nineteen randomised controlled trials, with a total of 3123 patients (2547 above-knee, 576 below-knee bypass surgery). In total, nine graft types were compared (autologous vein, polytetrafluoroethylene (PTFE) with and without vein cuff, human umbilical vein (HUV), polyurethane (PUR), Dacron and heparin bonded Dacron (HBD); FUSION BIOLINE and Dacron with external support). Studies differed in which graft types they compared and follow-up ranged from six months to 10 years.

Above-knee bypass

For above-knee bypass, there was moderate-quality evidence that autologous vein grafts improve primary patency compared to prosthetic grafts by 60 months (Peto odds ratio (OR) 0.47, 95% confidence interval (CI) 0.28 to 0.80; 3 studies, 269 limbs; P = 0.005). We found low-quality evidence to suggest that this benefit translated to improved secondary patency by 60 months (Peto OR 0.41, 95% CI 0.22 to 0.74; 2 studies, 176 limbs; P = 0.003).

We found no clear difference between Dacron and PTFE graft types for primary patency by 60 months (Peto OR 1.67, 95% CI 0.96 to 2.90; 2 studies, 247 limbs; low-quality evidence). We found low-quality evidence that Dacron grafts improved secondary patency over PTFE by 24 months (Peto OR 1.54, 95% CI 1.04 to 2.28; 2 studies, 528 limbs; P = 0.03), an effect which continued to 60 months in the single trial reporting this timepoint (Peto OR 2.43, 95% CI 1.31 to 4.53; 167 limbs; P = 0.005).

Externally supported prosthetic grafts had inferior primary patency at 24 months when compared to unsupported prosthetic grafts (Peto OR 2.08, 95% CI 1.29 to 3.35; 2 studies, 270 limbs; P = 0.003). Secondary patency was similarly affected in the single trial reporting this outcome (Peto OR 2.25, 95% CI 1.24 to 4.07; 236 limbs; P = 0.008). No data were available for 60 months follow-up.

HUV showed benefits in primary patency over PTFE at 24 months (Peto OR 4.80, 95% CI 1.76 to 13.06; 82 limbs; P = 0.002). This benefit was still seen at 60 months (Peto OR 3.75, 95% CI 1.46 to 9.62; 69 limbs; P = 0.006), but this was only compared in one trial. Results were similar for secondary patency at 24 months (Peto OR 4.01, 95% CI 1.44 to 11.17; 93 limbs) and at 60 months (Peto OR 3.87, 95% CI 1.65 to 9.05; 93 limbs).

We found HBD to be superior to PTFE for primary patency at 60 months for above-knee bypass, but these results were based on a single trial (Peto OR 0.38, 95% CI 0.20 to 0.72; 146 limbs; very low-quality evidence). There was no difference in primary patency between HBD and HUV for above-knee bypass in the one small study which reported this outcome.

We found only one small trial studying PUR and it showed very poor primary and secondary patency rates which were inferior to Dacron at all time points.

Below-knee bypass

For bypass below the knee, we found no graft type to be superior to any other in terms of primary patency, though one trial showed improved secondary patency of HUV over PTFE at all time points to 24 months (Peto OR 3.40, 95% CI 1.45 to 7.97; 88 limbs; P = 0.005).

One study compared PTFE alone to PTFE with vein cuff; very low-quality evidence indicates no effect to either primary or secondary patency at 24 months (Peto OR 1.08, 95% CI 0.58 to 2.01; 182 limbs; 2 studies; P = 0.80 and Peto OR 1.22, 95% CI 0.67 to 2.23; 181 limbs; 2 studies; P = 0.51 respectively)

Limited data were available for limb survival, and those studies reporting on this outcome showed no clear difference between graft types for this outcome. Antiplatelet and anticoagulant protocols varied extensively between trials, and in some cases within trials.

The overall quality of the evidence ranged from very low to moderate. Issues which affected the quality of the evidence included differences in the design of the trials, and differences in the types of grafts they compared. These differences meant we were often only able to combine and analyse small numbers of participants and this resulted in uncertainty over the true effects of the graft type used.