First treated by HD in January 1994, this 39 year old diabetic patient was transferred to a dialysis unit using the study EPR in November 1995. He received intravenous (IV) iron in three 1 g courses between November 1995 and June 1996. A modest improvement in the hemoglobin level occurred, but transferrin saturation (TSAT, an important measure of iron deficiency), mean corpuscular volume (MCV), and serum albumin level (a predictor of mortality) remained low. Recognizing that conventional IV iron replacement had been ineffective, 1 g courses of IV iron were again given starting in November 1996. With appropriate feedback, the courses were repeated whenever measures of iron deficiency were low. Not only was there an excellent hemoglobin response, but two unexpected findings were apparent: MCV increased to well above the normal range, and serum albumin increased to a range predictive of relatively low mortality. These observations were then confirmed in a large patient cohort .