Bypass After Failed Endovascular Intervention Is Associated with an Increased Risk of Above Ankle Amputation Among Patients with Chronic Limb Threatening Ischaemia in a Randomised Trial Population.

Objective: One concern about the endovascular-first (ENDO) approach for chronic limb threatening ischaemia (CLTI) is whether bridges are burned for a secondary bypass (SB) if required in the future. This secondary analysis of a prospective randomised trial aimed to compare above ankle amputation rates in patients with CLTI treated with primary bypass (PB) compared with those treated with SB after an initial ENDO approach.

Methods: Data from the randomised unblinded Best Endovascular versus Best Surgical Therapy of Patients with CLTI (BEST-CLI) trial were analysed. Patients were included if they had CLTI and were considered as candidates for open or ENDO revascularisation with the primary outcomes being major adverse limb free event survival. There were two parallel cohorts based on whether single segment great saphenous vein (SSGSV) was (cohort 1) or was not (cohort 2) available. Primary bypass was compared with SB after index ENDO using the primary outcome of above ankle amputation with death as a competing risk. Multivariable and propensity matched analyses were performed.

Results: There were 665 PB and 158 SB in cohort 1 and 192 PB and 45 SB in cohort 2. Time to SB after ENDO occurred at a median of 28 days in all patients and median of 210 days in those who had a successful initial ENDO procedure. Unadjusted one year analysis showed SB to be associated with increased above ankle amputation (14% vs. 8.1%; p = .002) overall. Secondary bypass was associated with increased above ankle amputation in cohort 1 (13.5% vs. 7.4%; p = .003), whereas this was not statistically significant in cohort 2 (15.9% vs. 10.9%; p = .28). These findings were confirmed on multivariable analysis (adjusting for age, gender, wound ischaemia foot infection stage, randomisation strata, diabetes mellitus, end stage kidney disease, previous index infrainguinal reconstruction, and smoking history) for cohort 1, with SB associated with increased above ankle amputation (HR 1.72, 95% CI 1.08 - 2.73; p = .02), remaining true when restricting to SB after a technically successful ENDO (HR 2.21, 95% CI 1.26 - 3.86; p =.005). Results were similar on propensity matched analyses.

Conclusion: In patients with CLTI deemed suitable for either open or ENDO, SB was associated with worse limb salvage compared with PB, particularly in patients with available SSGSV.