Safety and efficacy of an endovascular-first approach to acute limb ischemia

Background: The optimal techniques to manage acute limb ischemia (ALI) remain unclear. Previous reports have suggested that the decreased morbidity and mortality of endovascular approaches are mitigated by the limited technical success rates relative to open or hybrid approaches for ALI. However, these data failed to include newer technologies that might improve the technical success rates. We, therefore, sought to describe the current outcomes for an endovascular-first approach to ALI.

Methods: We performed a single-center, single-arm, retrospective cohort study of consecutive patients with ALI from 2015 to 2018. Technical success, limb salvage, survival, patency, and length of stay were quantified using Kaplan-Meier (KM) analysis. Cox regression analysis was used to identify the predictors of amputation-free survival.

Results: During the 3 years, 60 consecutive patients (39 men [65%]; median age, 65 years) presented with ALI. The Rutherford class was I in 15 patients (25%), IIa in 23 (38%), IIb in 13 (22%), and III in 9 patients (15%). Of the 60 patients, 34 had a history of previous failed ipsilateral revascularization (56%), including open bypass for 8 (13%), endovascular for 8 (13%), and both open and endovascular intervention for 18 (30%). The endovascular-first approach procedures included catheter-directed thrombolysis only (n = 19; 3%), catheter-directed thrombolysis plus aspiration and/or rheolytic thrombectomy (n = 19; 32%), and aspiration and/or rheolytic thrombectomy (n = 16; 26%). Six patients (10%) underwent covered stent placement only. The underlying occlusive process was most often thrombosis of a previous bypass graft or stent in 32 patients (53%), followed by native vessel thrombosis in 15 (25%). ALI had resulted from embolism in 13 patients (21.7%), including 2 (3%) with embolization to occlude a previous bypass graft or stent. Technical success was achieved in 58 patients (97%), with open conversion required in two patients (3%). At 30 days postoperatively, 52 patients (87%) survived, and 53 (88%) had successful limb salvage. Five patients (8%) had required four-compartment fasciotomy. No major hemorrhagic complications developed. The median length of stay overall and in the intensive care unit was 9 days (interquartile range, 4-14 days) and 2 days (interquartile range, 1-5 days), respectively. At 1 year, the KM estimates were as follows: amputation-free survival, 58% ± 0.08%; limb salvage, 74.3% ± 0.07%; and survival, 73.3% ± 0.07%. The 1-year KM estimates for primary and secondary patency were 39.4% ± 0.08% and 78.2% ± 0.07%, respectively. On multivariable Cox regression analysis, only age independently predicted for death and/or amputation at the last follow-up (hazard ratio, 1.06; 95% confidence interval, 1.01-1.10; P = .01).

Conclusions: The current endovascular approaches to ALI have high technical success rates. Survival, limb salvage, perioperative complications, and length of stay were similar to those from previous reports of historical open cohorts. Further prospective, appropriately powered, multicenter cohort studies are warranted to evaluate the efficacy of endovascular vs open approaches to ALI.