The approval of gadofosveset trisodium as the first imaging agent for magnetic resonance angiography (MRA) by the US Food and Drug administration (FDA) in December 2008 served as a motive to review a rare serious adverse reaction possibly related to the use of gadolinium based contrast agents. Nephrogenic systemic fibrosis (NSF) is a disabling and potentially fatal disease characterized by thickening and hardening of the skin, especially of the extremities. It is restricted to patients with renal insufficiency (estimated glomerular filtration rate [eGFR] < 30ml/min/ 1,73 m2). Systemic manifestations may involve skeletal muscles, lungs and myocardium. Progressive fibrosis may restrict the movement of joints, leading to contractures limiting mobility of the patients. Since the first report by Cowper and co-workers in 2000 as "skleromyxedema-like cutaneous disease in renal-dialysis patients", hundreds of cases have been published. In 2006 a possible relationship to gadolinium based contrast agents used for magnetic resonance imaging was established, especially when applied in high doses, although the exact pathophysiologic mechanisms remain unclear. Gadolinium based contrast agents bind gadolinium (Gd3+) ions into a chelate complex (GBC). It has been proposed, that due to prolonged exposure in patients with renal insufficiency Gd3+ ions may increasingly be released from the chelate. There are several theories how these unbound Gd3+ ions may initiate fibrotic changes. Macrophages, that engulf these ions, may release profibrotic cytokines by which circulating fibrocytes are attracted to the tissue. Free Gd3+ ions may also activate tissue transglutaminase, which in turn activates transforming growth factor beta (TGF-beta) produced by dendritic cells. Active TGF-beta attracts further dendritic cells and on the other hand stimulates skin fibroblasts to produce extracellular matrix proteins (e.g. collagen), whereas matrix degrading enzymes are decreased. The chemical structures of the chelates account for different stabilities of the GBC agents. Most cases of NSF were reported after exposure to gadodiamide or gadopentetate dimeglumine, which represent linear, non-ionic imaging agents. In contrast, macrocyclic, ionic imaging agents have rarely been associated with this disease. Although the recently FDA approved gadofosveset trisodium has a linear structure and a long plasma half life due to high protein binding, there have been no reports of NSF up to date. This may be due to the much lower doses required for this agent compared to gadodiamide (approx. 0,03 mmol/l compared to 0,2 mmol/). Since there is no effective therapy for NSF, the top priority is to avoid exposure to gadolinium based contrast agents in patients at increased risk. In case of renal insufficiency (eGFR < 30 ml/min/1,73 m2) they should only be given if clearly necessary. After a thorough risk-benefit-assessment, the lowest possible dose should be used.