Erythropoietin has been shown to be effective both in the reversal of anaemia in patients with end-stage renal failure and to increase the volume of autologous red blood cells donated preoperatively as well as to decrease the units of homologous blood transfused. This review analyzes the side effects of erythropoietin reported in the literature for long-term administration (mainly in patients with end-stage renal failure) as well as for acute/short-term administration (in patients participating in an autologous,predeposit programme). The most important adverse events reported for long-term administration are as follows: (a) arterial hypertension; (b) cerebral convulsion/hypertensive encephalopathy; (c) thrombo-embolism; (d) iron deficiency; (e) influenza-like syndrome. The numbers given for these side effects are mainly taken from the first and dose-finding studies in patients with renal failure. These figures differ very much from the data given in controlled studies analyzing adverse events as well. Summarizing the results from controlled, multi-center trials in patients with end-stage renal failure or in AIDS patients, no significant differences have been observed between the control group and the patients treated with erythropoietin. The overall-incidence of side effects occurring in either group of these two studies was of approximately 83% and 95%, respectively. In contrast to these results the data published for the dose finding/treatment studies is approximately 30% for development of arterial hypertension, approximately 5% for occurrence of cerebral convulsion/hypertensive encephalopathy, approximately 10% for thrombo-embolic complications/clotting of vascular access, approximately 50% for development of iron deficiency, and approximately 10% for symptoms summarized as influenza-like syndrome. The incidence of side effects reported for acute/short-term treatment in patients on autologous transfusion is very much lower. Especially development of arterial hypertension and of cerebral convulsions, have so far not been reported during acute and short-term administration. The incidence of thrombo-embolic complications amounts to approximately 3% and does not differ between the control group and the erythropoietin group. Development of iron deficiency is to be expected if erythropoietin is effective to stimulate erythropoesis, as iron is incorporated to build up haemoglobin. And finally occurrence of an ''influenza-like syndrome'' is reported in approximately 5%. Erythropoietin has been proven safe and effective in longterm administration and it can be considered safe and effective for acute/short-term treatment in autologous transfusion as well.