OBJECTIVE To evaluate whether desferrioxamine decreases ischemia and perfusion injury aggravated by cold storage (CS) in a rat liver perfusion model. METHODS Isolated rat livers were kept in CS in University of Wisconsin Solution for 20 h at 4 °C, then exposed to 25 min of warm ischemia (WI) at 37 °C followed by 2 h of warm perfusion (WP) at 37 °C with oxygenated (95% oxygen and 5% carbon dioxide) Krebs-Henseleit buffer. Desferrioxamine (DFO), an iron chelator, was added at different stages of storage, ischemia and perfusion: in CS only, in WI only, in WP only, in WI and perfusion, or in all stages. Effluent samples were collected after CS and after WI. Perfusate samples and bile were collected every 30 min (0, 0.5, 1, 1.5 and 2 h) during liver perfusion. Cellular injury was assessed by the determination of lactate dehydrogenase (LDH) and aspartate aminotransferase (AST) in the effluent and perfusate samples. Total iron was analysed in the perfusate samples. After WP, the liver was collected for the determination of liver swelling (wet to dry ratio) and liver morphological examination (hematoxylin and eosin staining). RESULTS Increased CS time caused increased liver dysfunction during WP. After 2 h of WP, liver injury was indicated by increased release of AST (0.5 h CS: 9.4 ± 2.2 U/g liver vs 20 h CS: 45.9 ± 10.8 U/g liver, P < 0.05) and LDH (0.5 h CS: 59 ± 14 U/g liver vs 20 h CS: 297 ± 71 U/g liver, P < 0.05). There was an associated increase in iron release into the perfusate (0.5 h CS: 0.11 ± 0.03 μmoL/g liver vs 20 h CS: 0.58 ± 0.10 μmoL/g liver, P < 0.05) and reduction in bile flow (0.5 h CS: 194 ± 12 μL/g vs 20 h CS: 71 ± 8 μL/g liver, P < 0.05). When DFO was added during WI and WP following 20 h of CS, release of iron into the perfusate was decreased (DFO absent 0.58 ± 0.10 μmoL/g liver vs DFO present 0.31 ± 0.06 μmoL/g liver, P < 0.05), and liver function substantially improved with decreased release of AST (DFO absent 45.9 ± 10.8 U/g liver vs DFO present 8.1 ± 0.9 U/g liver, P < 0.05) and LDH (DFO absent 297 ± 71 U/g liver vs DFO present 56 ± 7 U/g liver, P < 0.05), and increased bile flow (DFO absent 71 ± 8 μL/g liver vs DFO present 237 ± 36 μL/g liver, P < 0.05). DFO was also shown to improve liver morphology after WP. Cellular injury (the release of LDH and AST) was significantly reduced with the addition of DFO in CS medium but to a lesser extent compared to the addition of DFO in WP or WI and perfusion. There was no effect on liver swelling or bile flow when DFO was only added to the CS medium. CONCLUSIONS DFO added during WI and perfusion decreased liver perfusion injury aggravated by extended CS.