Vitamin E (α-toxopherol) is widely used, at different dosages and schedules and in different formulations, in neonatal intensive care units to prevent intraventricular hemorrhage,retrolental fibroplasia, bronchopulmonary dysplasia and hemolytic anemia of preterm infants. As part of a wider project to assess the effect of vitamin E, the present study was designed to determine whether the only intramuscular formulation available today in Italy and in other European countries (olive oil solution), and widely used in all Italian neonatal intensive care units according to a standard schedule, achieves and maintains suggested therapeutic levels, compared to the colloidal aqueous solution used in clinical trials in England and Canada, which contained the acetate ester of α-tocopherol and as used in the USA, which contained the free tocopherol, but is not yet available in Italy. Forty-four babies, of less than 32 weeks’ gestation, 670-1,800 g birth weight, were randomly allocated to one of the two formulations. We gave 20 mg/kg of vitamin E intramuscularly on 3 consecutive days starting within 8 h of birth (day 0), 24 and 48 h later (days 1 and 2). Blood, plasma, red blood cells after transfusions and milk were sampled up to the sixth day of life. Clinical data were collected up to discharge from the neonatal intensive care unit. Plasma concentrations of vitamin E (free tocopherol) averaged 1.1 at 24 h and 3.3 mg/dl at 72 h after the colloidal aqueous solution, and were about 6 times lower after the oil solution. Plasma vitamin E levels did not rise above baseline after injection of the olive oil preparation. The acetate ester of vitamin E was measured only after use of the colloidal aqueous preparation, and the highest plasma concentration averaged 1.01 mg/dl 72 h after injection. These findings have far-reaching implications related to the current clinical practice in Italy of prophylaxis with intramuscular vitamin E, using the scantily bioavailable olive oil formulation currently on the market.