Vascular Graft Infections With Candida: A Factor for Increased Mortality in in-situ Reconstructions.

BACKGROUND: The empiric antibiotic regimen started after deep cultures and explantation of the graft mostly do not cover antifungals. We retrospectively studied the outcome of candida compared to non-candida VGI and assessed whether these results could justify the addition of antifungals to the empiric antibiotics in the early postoperative period. METHODS: All patients treated for infected aorto(ilio)femoral graft with excision and reconstruction at the vascular department of University Hospitals Leuven between January 2010 and 2017 (n = 56) were studied retrospectively. Patients were allocated to the candida group (n = 10) or non-candida group (n = 46) according to the presence of Candida in deep culture isolates. RESULTS: All-cause mortality was significantly higher in the candida group compared to the non-candida group. All-cause 30-day mortality was 40% and 13% for both groups respectively (P = 0.066). At 5 years this was 90% and 46% respectively (P = 0.014). In the candida group 6 patients (60%) had to be revised in the operating room due to bleeding, compared to 5 patients (11%) in the non-candida group (P = 0.002).  Two patients (20%) and 5 patients (11%) had to be readmitted to the ICU, respectively. CONCLUSION: Survival of candida related VGI is significantly worse, especially in the first 5 postoperative months. This could justify the addition of an antifungal to the early empiric postoperative antibiotic cocktail, especially in patients with an aorto-enteric fistula. A cost-benefit analysis could be useful to evaluate the yield.