Background Ex situ normothermic liver perfusion (NMP) in a blood-based perfusate is associated with a risk of microbe growth resulting in life-threatening post-transplant sepsis. Antibiotics are widely used, but the pharmacokinetics of these agents are unknown as is their efficacy. We wished to assess the perfusate concentrations of the meropenem and fluconazole that we use, and to audit the incidence of infection with this antimicrobial therapy. Methods Fluconazole and meropenem (100mg each) were added to the perfusate before NMP began, and serial samples taken and assayed for drug concentrations. Perfusate cultures were available from 210 of the 242 perfusions performed between between 1st February 2018 and 6th April 2023; these were reviewed. Results Following administration of 100mg fluconazole, levels fell slightly from a median 24.9mg/L at 1 hour to 22.6mg/L at 10 hours. In contrast, meropenem concentrations fell over time, from a median 21.8mg/L at 1 hour to 9.4mg/L at 10 hours. There were 4 significant microorganisms grown in the perfusions, including three Candida species and an Enterococcus faecium. All the Candida infected livers were transplanted with no adverse consequences, the recipients being treated with anidulafungin upon identification of the infecting organism; the Enterococcus infected liver was not transplanted. Conclusions Serious infection is a risk with NMP, but appears to be mitigated with a protocol combining fluconazole and meropenem. This combination may not be appropriate in areas where resistance is prevalent. Routine culture of NMP perfusate cultures is essential to identify breakthrough organisms early and enable recipient treatment.