Cryopreserved arterial allografts (CAA) have been described as a valuable alternative reconstruction material in vascular infections, offering resistance to infection, reduced operation time, and better compatibility. Up to an 18% early rupture and occlusion rate has been reported in aortic procedures. 1 However, peripheral graft infections (PGI) have their own specificities. Controversy remains over the optimal conduit when autologous superficial veins are not available, as seen in the recent ESVS guidelines which list CAA among eligible materials without prioritisation. 2 The aim of this study was to evaluate the outcomes of CAA use in PGI with particular focus on the early graft specific complications (GSC, i.e. rupture and thrombosis). Patients who underwent CAA implantation for a PGI from February 2008 to December 2018 were identified. Diagnosis of PGI was based on clinical, biological, ultrasound, and computed tomography findings. Indications for graft removal were group 3e5 infections according to the Samson modification of the Szilagyi system (deep wound infections with skin/surrounding tissues necrosis, graft exposure, or rupture) and systemic sepsis. Revascularisation was performed in patients with ischaemia at presentation, or in whom the initial graft was implanted for a Rutherford 3e6 chronic limb ischaemia or intended to bypass an iliac vessel. Autologous great saphenous vein (GSV) was always assessed and CAA considered only when the GSV was insufficient. External iliac and superficial femoral CAA were harvested from brain deceased donors, immersed in antibiotics, and cryopreserved either in vapour phase of liquid nitrogen (À150 C) after controlled rate freezing, or at À80 C. No ABO matching was performed at the study centre. Broad spectrum intravenous antibiotics were administered peri-operatively, followed by adjusted oral therapy for four to six weeks. No immunosuppressive therapy was given post-operatively. The recorded variables were age, sex, comorbidities, ASA score, Rutherford stage at initial operation, infection stage, cryopreservation method (À80 C vs. À150 C), microbiology and operative elements (duration, ex situ/in situ, urgent/elective). The primary endpoint was GSC (rupture or thrombosis). The secondary endpoints were all cause mortality, limb amputation, and re-infection. Descriptive and survival analyses were performed. The association of each variable with the outcome was estimated with a Cox proportional hazards model. Thirty-eight patients (men, 89%; mean age 69.5 AE 10.3 years) were included. The median time to PGI diagnosis was 36.5 (13e3650) days. Reconstruction was performed urgently in 11 (29%) patients. In six (15%) patients, a venous
