Objectives: There is continued controversy over whether a failed distal bypass influences the level of amputation. This issue is important as the number of arterial bypass grafts undertaken for critical ischaemia is increasing, followed by an increasing number of failed grafts.

Setting: Teaching hospital. STUDY DESIGN AND MATERIALS: A prospective analysis of 109 consecutive femorocrural/pedal bypass grafts performed between June 1991 and January 1995 on patients presenting with severe critical lower limb ischaemia (CLI) to a single vascular unit. A further 43 amputations for non-reconstructible distal disease were also analysed.

Chief outcome measures: Mortality, amputation, rehabilitation, survival and knee salvage rates. The Kaplan-Meier method was used for comparison of factors associated with knee preservation.

Results: Primary amputees had a higher in-hospital mortality (18% vs. 10%) but similar 3 year survival rates (30%) compared with secondary amputees (36.6%). Patients with successful grafts showed a trend towards better survival (61.9% at 3 years) compared to amputees (38.6% at 42 months, p = 0.061). Below- to above-knee amputation ratio was similar in the two groups (0.85 in secondary vs. 0.95 in primary amputees). Factors significantly associated with knee salvage at 3 years were shown to be: the condition of the inflow (81.9% for good vs. 43.1% for impaired, p = 0.000) the state of the profunda femoris artery (good 93%, impaired 71%, occluded 37% p = 0.0001) and the graft material (vein 81.8% vs. PTFE 59.8%, p = 0.033). The presence of tissue loss (p = 0.0523) and secondary procedures (p = 0.0879) showed a trend to become significant. Multivariate and Cox regression analysis showed that the most important factors were the inflow (p = 0.001), the state of the profunda (p = 0.001), the graft material (p = 0.034) and previous revascularisation attempts (p = 0.019).

Conclusions: The factors which determine knee loss are a compromised inflow state, the presence of an inadequate profunda femoris, previous revascularisation attempts and the use of synthetic graft material. Most of these factors (with the exception of infection related to revascularisation) are present before reconstructive arterial surgery is performed and this study shows that failure of a distal graft does not affect the final amputation level.