Background. Nephrogenic systemic fibrosis may be caused by gadolinium (Gd)-containing magnetic resonance imaging contrast agents. Most reported cases were associated with one particular agent, gadodiamide. Yet, unidentified cofactors might explain why only a minority of renal failure patients exposed to gadodiamide develop nephrogenic systemic fibrosis. Methods. We conducted a case-control study of 19 histologically verified cases and 19 sex- and age-matched controls. All subjects had chronic renal failure when exposed to gadodiamide. Clinical, biochemical and pharmacological data were retrieved from medical records. Results. Cases had been exposed to a mean gadodiamide dose of 0.29 mmol/kg (range 0.18-0.50) shortly before first signs of nephrogenic systemic fibrosis. Controls had been exposed to 0.28 mmol/kg (0.13-0.49). Cumulative gadodiamide exposure while in chronic kidney disease stage 5 was significantly higher among cases compared with controls (0.41 vs 0.31 mmol/kg, P = 0.05) and among severe cases (n = 9) compared with non-severe cases (0.49 vs 0.33 mmol/kg, P = 0.02). Severe cases developed primarily among patients in regular haemodialysis therapy at exposure. Cases had higher serum concentrations of ionized calcium and phosphate than controls and tended to receive higher doses of epoietin-β than controls at time of exposure. Severe cases were treated with higher doses of epoietin-β than non-severe cases at exposure (10.8 vs 4.4 103 IU/week, P = 0.02). Conclusions. Increasing cumulative gadodiamide exposure, high-dose epoietin-β treatment, and higher serum concentrations of ionized calcium and phosphate increase the risk of gadodiamide-related nephrogenic systemic fibrosis in renal failure patients. Severe cases seem to develop primarily among patients in regular haemodialysis therapy at exposure.