Proposal of a Multicentric-Based Score for Graft Patency in Below-knee Femoropopliteal Bypass With Heparin-Bonded ePTFE Graft in Patients With Critical Limb Ischemia

Objectives This study retrospectively created a predictive score for primary patency (PP) in patients with critical limb ischemia (CLI) operated on with the use of a heparin-bonded expanded polytetrafluoroethylene (ePTFE) bypass graft (Hb-ePTFE) in a large multicenter registry. Methods Over a 13-year period, ending in December 2014, a Hb-ePTFE graft was implanted in 683 patients undergoing below-knee revascularization for CLI in seven Italian vascular hospitals. Data concerning these interventions were retrospectively collected in a multicenter registry with a dedicated database. Follow-up results were analyzed in terms of PP: univariate and multivariable analyses with Kaplan-Meier estimates were used to identify potential significant predictors of such end point at 3 years, and then a predictive risk score was constructed by dividing the β-coefficient of each significant predictor at multivariable analysis by 0.25 and by rounding off to the nearest integer value. A qualitative assessment of the Kaplan-Meier survival estimates for each integer score was performed, and subgroups of risk were stratified on the basis of the primary end point. Comparison between subgroups was performed with log-rank test. Results Median duration of follow-up was 34 months (range, 1-123; standard deviation, 27.8 months). Overall, estimated 3-year PP rate was 51.3% (standard error, 0.02). At multivariate analysis, female gender, reintervention following the failure of previous open or endovascular intervention, and the need for distal tibial anastomosis were predictors of poorer PP. The integer score ranged from 0 to 6. Kaplan-Meier analysis for PP in each score group identified four subgroups with significant differences at 3 years: score 0 (PP, 63%), score 2 (PP, 49%; P = .003, log-rank 6.1, in comparison with score 0), score 4 (PP, 39%; P = .05, log-rank 2.2, in comparison with score 2), and score 6 (PP, 25%; P = . 01, log-rank 2.7 in comparison with score 4). Conclusions A category of patients with CLI treated with the indexed graft with excellent patency rates does exist, thus suggesting a primary role for Hb-ePTFE in such patients. A prospective validation of such a score is necessary.