Background Understanding kidney metabolism during perfusion is vital to further develop the technology as a preservation, viability assessment, and resuscitation platform. We reviewed the evidence on the use of labeled metabolites (tracers) to understand “on‐pump” kidney behavior. Methods PubMed, Embase, Web of Science, and Cochrane databases were systematically searched for studies evaluating metabolism of (non)radioactively labeled endogenous compounds during kidney perfusion. Results Of 5899 articles, 30 were included. All were animal studies [rat (70%), dog (13%), pig (10%), rabbit (7%)] perfusing but not transplanting kidneys. Perfusion took place at hypothermic (4–12°C) (20%), normothermic (35–40°C) (77%), or undefined temperatures (3%). Hypothermic perfusion used albumin or a clinical kidney preservation solution, mostly in the presence of oxygen. Normothermic perfusion was mostly performed with oxygenated crystalloids often containing glucose and amino acids with unclear partial oxygen tensions. Active metabolism of carbohydrate, amino acid, lipids, and large molecules was shown in hypothermic and normothermic perfusion. Production of macromolecules, such as prostaglandin, thromboxane, and vitamin D, takes place during normothermic perfusion. No experiments compared differences in metabolic activity between hypothermic and normothermic perfusion. One conference abstract showed increased anaerobic metabolism in kidneys donated after circulatory death by adding labeled glucose to hypothermically perfused human kidneys. Conclusions Tracer studies during kidney perfusion contribute to unraveling kidney metabolic behavior in pre‐clinical models. Whether findings are truly translational needs further investigation in large animal models of human kidneys. Furthermore, it is essential to better understand how ischemia changes this metabolic behavior.