The aim of this study was to verify if the adoption of a strict guideline would reduce the need for red blood cell transfusions in the first 28 days of life in very low birth weight preterm infants. Retrospective study of two cohorts of very low birth weight infants transfused according to neonatologist decision (Period 1) or according to a strict guideline for erythrocytes transfusion (Period 2). Clinical and hematological data of 80 premature infants transfused in both periods of the study were obtained by chart review. During the first 28 days of life, 45 (62.5%) of 72 premature infants born in the Period 1, and 44 (55.7%) of 79 newborns born in Period 2 received at least one erythrocyte transfusion; p = 0.40. Among patients transfused, the median number of transfusions was four per infant transfused (range: 1-13; mean: 4.6 +/- 3.2) in Period 1 and 3 (range: 1-18; mean: 4.0 +/- 3.5) in Period 2; p = 0.26. The median volume of erythrocytes administered per infant transfused in Period 1 was 40 ml kg(-1) (range: 10-170; mean: 48.8 +/- 38.3) and in Period 2 was 30 ml kg(-1) (range: 10-225; mean: 43.4 +/- 40.4), p = 0.41. Multiple linear regression analysis, after adjusting for birth weight, clinical risk index for babies, blood loss and days of ventilation, showed that the adoption of the strict guideline reduced the volume of erythrocytes transfused in 15.91 ml kg(-1) per infant transfused (95% CI: -24.69-7.14) p < 0.001. The adoption of a strict guideline reduced the need for red blood cells transfusions in very low birth weight infants.