Nissenson 1995

Methods	Study design: RCT Study duration: September 27, 1989 to January 10, 1992
Participants	Setting: 16 centres Country: USA Stage of CKD: ESKD undergoing PD Number: epoetin alfa (78), placebo (74) Mean age ± SD (years): epoetin alfa (46.8 ± 15.5), placebo (49.9±15.9) Sex (M/F): epoetin alfa (31/47), placebo (28/46) Other characteristics: baseline HCT < 30%; adequate iron stores Exclusion criteria: patients were excluded for any of the following reasons: systemic haematological disease that would interfere with the evaluation and interpretation of the data (e.g. sickle cell anaemia, thalassaemia, myelodysplastic syndromes, or haematologic malignancies); more than one documented episode of peritonitis within the past 4 months or clinical evidence of peritonitis within the past 30 days; likelihood of receiving a kidney transplant within the first 90 days on‐study; current drug addiction; consistent supine DBP of 100 mm Hg or higher; thrombocytopenla (platelet count less than 100,000/mm³); haemolytic anaemia, Coombs positive or negative; participation in any other clinical investigational drug or biologic study while participating in this study or within the past 30 days (including, but not limited to, antihypertensive and antibiotic studies); androgen therapy initiated in the preceding 4 wk or changes in dose of androgens in the preceding 4 wk; deferoxamine therapy during the prestudy period; uncontrolled seizure disorder
Interventions	Treatment group Epoetin alfa SC titrated to HCT 32‐38% for 3 months Control group Placebo SC titrated to HCT 32‐38% for 3 months Iron supplementation Oral or IV
Outcomes	Primary trial outcome HCT RBC transfusion requirements Outcomes extracted for meta‐analysis All‐cause mortality Hypertension
Notes	Funding: Amgen Trials registration: not applicable Contact with authors: not contacted
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	The randomisation sequence was designed to ensure that approximately equal numbers of patients were randomised. At each centre, treatment unit numbers were assigned consecutively by date of randomisation
Allocation concealment (selection bias)	Unclear risk	The randomisation sequence was designed to ensure that approximately equal numbers of patients were randomised. At each centre, treatment unit numbers were assigned consecutively by date of randomisation
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Double‐blind
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Not described
Incomplete outcome data (attrition bias) All outcomes	High risk	9/78 in epoetin alfa arm lost to follow‐up (11.5%) and 7/74 in placebo arm lost to follow‐up (9.5%). As the loss to follow‐up in the trial overall was > 10% this was judged as high risk
Selective reporting (reporting bias)	High risk	No data for major cardiovascular events
Other bias	Low risk	None apparent