Clinical consequence of bare metal stent and stent graft failure in femoropopliteal occlusive disease

Objective: The optimal role for bare metal stents (BMS) or stent grafts (SG) in femoropopliteal occlusive disease (FPOD) is as of yet undefined. Understanding the clinical consequences of failure can help guide initial treatment decisions. The goal of this study was to define the nature, frequency, and risk factors for adverse clinical events related to BMS and SG failure in FPOD. Methods: This is a single-institution retrospective review of primary endovascular interventions for FPOD using either a BMS or SG, from September 2007 through October 2011. Patients were excluded if they had any previous lower extremity interventions. Patient demographics, indications for intervention, anatomic characteristics, procedural details, clinical outcomes, and reintervention details were reviewed. Clinical outcomes included the composite end point of any reintervention, amputation, or stenosis, acute limb ischemia (ALI), and the composite end point of major adverse limb events, which included a need for bypass, thrombolysis, or major amputation. Results: Seventy-one limbs were treated with BMS and 63 with SG. Although patient demographics were largely similar between cohorts, key differences included indication for intervention (percent claudication BMS vs SG, 34/71 (48%) vs 42/63 (67%); P < .05) and the TransAtlantic Inter-Society Consensus II classification of lesions in the claudicant subgroup (TransAtlantic Inter-Society Consensus D BMS vs SG, 4/34 (12%) vs 17/42 (40%); P < .01). Freedom from reintervention at 1 year was better in the SG group (75% vs 64%; hazard ratio, 0.46; 95% confidence interval, 0.25-0.78; P < .01). Freedom from major adverse limb events was not different between groups; however, SG thrombosis resulted in a more frequent need for thrombolysis. On multivariate analysis, treating with a BMS vs SG was a significant predictor for freedom from thrombolysis (hazard ratio, 0.53; confidence interval, 0.37-0.76; P < .01). ALI during follow-up was seen only in the SG group (nine vs zero events, log-rank; P < .02). Conclusions: Failure modes of BMS and SG used to treat FPOD differ, and the clinical consequences may not be benign. Claudicants may not revert back to claudication with treatment failure. Although the overall reintervention rate at 1 year is lower for SG compared to BMS, we observed a higher rate of ALI and need for thrombolysis with SG failure. In light of these differential risks of treatment failure, we believe that the use of SG as initial therapy for FPOD should be carefully deliberated and mandates close postoperative surveillance.