The introduction of synthetic vascular grafts has drastically changed the field of vascular surgery. With the broadened treatment possibilities also came to risk of complications related to these prostheses, of which graft infection remains one of the most detrimental in this population with established cardiovascular disease. A variety of techniques with several graft materials have been proposed as the optimal management but no clear superiority has been shown. The objective of the study was to analyse the long-term outcome and determine possible risk factors in patients treated by cryopreserved arterial allograft (CAA) implantation for vascular graft infection.

This retrospective study describes the use of these CAA in the treatment of 142 patients with vascular graft infection in a single centre over 18 years (2000 to 2018). Patients with femoropopliteal and femorodistal vascular graft infections were not included in this cohort, as it is the hospital's policy to treat infections at these levels by cryopreserved great saphenous vein implantation. The outcomes of interest included 30 day and in-hospital mortality, overall and graft related survival, reintervention free survival and primary patency up to 5 years. Short-term outcomes were presented by percentage and analysed with chi-square tests. Long-term outcomes analysed by Kaplan-Meier curves and Cox regression for evaluating risk factors. These long-term outcomes at 1, 3 and 5 years follow up were described separately in the aortoiliac and peripheral group. Cox regressions were performed in the pooled study population at 1- and 2-year follow up.

The study included 98 aortoiliac, 21 common femoral, 9 axillofemoral and 14 femorofemoral revascularisations. The in-hospital and 30-day mortality were respectively 12.1% and 15.7% in the aortoiliac group and both 2.3% in the peripheral group. In the aortoiliac group the 1-, 3- and 5-year outcomes were as follows: resp. 75.1%, 59.6% and 51.9% overall survival, 84.5%%, 82.7% and 82.7% graft related survival, 81.8%, 68.7% and 53.3% for primary patency and 68.2%, 51.6% and 45.9% for reintervention free survival. In the peripheral group the 1-, 3- and 5-year outcomes were: resp. 68.3%, 51.6% and 17.2% for overall survival, 91.8%, 91.8% and 76.5% for graft related survival, 72.4%, 67.6% and 59.1% for primary patency and 58.1%, 50.1% and 37.5% reintervention free survival. The overall survival was lower for patients receiving antibiotic treatment of 3 weeks or less after hospital discharge (HR 4.51, 95%CI [1.264; 16.089], p=.020 at 2 years). The in-hospital mortality was higher in patients with an ASA-score of 4 (OR 8.00, 95%CI [1.501; 42.652], p=.012). Patients with positive cultures from their explanted prostheses showed an increased risk of re-intervention (p=.002), as well as patients presenting with an aorto-enteric fistula (AEF) (HR 3.49, 95%CI [1.581; 7.683], p=.002 at 2 years). Despite the higher re-intervention risk, patients presenting with an AEF did not have a decreased overall or graft related survival at 2 years follow up (resp. HR 2.26, 95%CI [0.295; 17.377] and HR 0.73, 95% CI [0.154; 3.416]). When necessary, partial graft removal did not alter the patients' overall or graft related survival (resp. HR 1.34, 95%CI [0.423; 4.232] and HR 2.92, 95%CI [0.845; 10.096]), nor did it increase the risk of re-intervention (HR 0.99, 95%CI [0.438; 2.222] at 2 years). ABO incompatibility between CAA and acceptor did not compromise the grafts primary patency (HR 2.02, 95%CI [0.691; 5.924] at 2 years).

CAA use is one of the key alternatives in the management of vascular graft infections. Although the overall survival in this frail group remains low, the graft related survival highlights the potential of these allografts. One of the treatment cornerstones is effective and prolonged antibiotic treatment, both pre-and post-operative, to minimise the risk of re-intervention and optimise long-term survival. The presence of an AEF does not diminish the patient's chances of survival but given the long-term risk of re-interventions, such patients require a more thorough follow-up. Partial graft removal, under certain conditions, can be performed without impairing any long-term results. As stated, cryopreservation alters the antigenicity of the arterial allograft, making ABO matching obsolete.