Risk Factors for Early and Late Iliac Limb Occlusions of Stent Grafts Extending to the External Iliac Artery after Endovascular Abdominal Aneurysm Repair

Abstract

Iliac limb stent grafting to the external iliac artery (EIA) is a commonly performed procedure in various situation during endovascular abdominal aneurysm repair (EVAR). However, this procedure is associated with a risk of graft occlusion. We evaluated risk factors affecting occlusion among patients who underwent iliac limb stent-graft extension to the EIA. We compared occluded limbs with patent limbs during the follow-up period using variables, including anatomical values, demographics, and other factors. Thereafter, we divided the occluded limbs into early and late occlusion subgroups. The main and subgroup analyses used the same variables. Survival analysis was performed to evaluate time-dependent risk factors for late limb occlusion. From 2007 to 2016, 766 iliac limbs from 383 patients who underwent EVAR were initially included in our analysis. Among them, 134 iliac limbs underwent limb extension to the EIA. The limb extension was a significant risk factor for occlusion (hazard ratio=6.34, P<0.001). Occlusion occurred in 10 patients who underwent iliac limb extension. The size of common iliac artery (CIA) was associated with occlusion. The most significant factor was iliac bifurcation diameter (patent vs. occluded limbs, 21.6±7.6 vs. 27.5±9.5mm, P=0.005). Subgroup analysis revealed that the CIAs of the early occlusion subgroup were generally more tortuous (1.16±0.33 vs. 1.47±0.25, P=0.091) and longer (53±24 vs. 74±9, P=0.01) than those of the patent limb group. In addition, the EIA diameters of the late occlusion subgroup were narrower than those of the patent limb group (10.9±1.6mm vs. 9.1±0.8mm, P=0.011). Using the log-rank test, those patients with an EIA diameter narrower than 10.1mm had a higher risk for late limb occlusion (log-rank χ2=5.73, P=0.017) and the patients who did not take at least a single antiplatelet agent had a significantly higher chance of limb occlusion (log-rank χ2=11.029, P=0.001). In addition, the patients who did not take a statin had a higher risk for late limb occlusion (log-rank χ2=7.41, P=0.007). Among patients who underwent EVAR with iliac limb extension, the CIA length affected early limb occlusion and predisposed patients to vessel injury or stent-graft kinking, and there was the possibility that CIA tortuosity was associated with a higher risk. The late occlusion subgroup had narrower EIAs than the patent limb group. Appropriate antiplatelet and statin therapy is expected to play a key role in the prevention of late limb occlusion after EVAR.