Outcomes of endovascular treatment for stenosis occurring after cephalic vein transposition and graft interposition.

Objective: To evaluate the outcome of endovascular treatment after surgical treatment for cephalic arch stenosis (CAS) in the brachiocephalic fistula (BCF) and to analyze the factors influencing patency.

Methods: We conducted a retrospective review of patients undergoing cephalic vein transposition (CVT) or graft interposition (GIP) for CAS from January 1, 2017, to December 31, 2019. A total of 73 patients with restenosis were included in this study. Patients were classified into cephalic transposition (BCF-CVT) (n = 49) and graft interposition (BCF-GIP) (n = 24) groups. We calculated the postintervention primary and secondary patency of endovascular treatment by using the Kaplan-Meier analysis and analyzed variables associated with loss of postintervention patency.

Results: Six-month and 12-month postintervention primary patency rates of endovascular treatment for restenosis were 56.7% and 15.6% and secondary patency rates were 89.7% and 72.1%, respectively. In the BCF-CVT group, 6-month and 12-month postintervention primary patency was 56.8% and 17.6% and secondary patency was 93.3% and 79.4%, respectively. In the BCF-GIP group, 6-month and 12-month postintervention primary patency was 56.5% and 8.7% and secondary patency was 85.7% and 56.3%, respectively. There was no significant difference in postintervention primary patency between the two groups (P = .79). However, the BCF-CVT group demonstrated higher postintervention secondary patency (P = .034). The BCF-GIP group had a higher number of stenosis sites (P < .01). There was no significant predictor of reduced postintervention primary patency. The only adverse variable of postintervention secondary patency was BCF-GIP (hazard ratio, 3.14; 95% CI, 1.06-9.34; P < .05).

Conclusions: Endovascular treatment is still the acceptable option for stenosis occurring after surgical treatment for CAS. CVT provides higher postintervention secondary patency than GIP.