PURPOSE: Vitamin K deficiency is associated with hemorrhagic disease of the newborn. Late hemorrhagic disease is often intracranial and may be fatal. Many countries recommend vitamin K prophylaxis after birth to prevent this hazard of vitamin K deficiency. Nevertheless, there are still controversies concerning the best way of providing effective prophylaxis. A recent article by Golding and colleagues has questioned the safety of the routine use of intramuscular vitamin K for the newborn. These authors reported a significantly increased rate of childhood cancer in infants who received intramuscular prophylaxis. So we compared the prophylactic effect of intramuscular, oral, and maternal administration of vitamin K on hemorrhagic disease of the newborn. METHODS: A total of 60 newborns, delivered spontaneously vaginally, in the Masan Fatima hospital from March to June, 1996, were enrolled. Neonated with intrapartum anoxia, liver disease or hereditary coagulation factor deficiencies, who received antibiotics were excluded. Mothers receiving any medication known to interferes with vitamin K metabolism(such as antiepileptics, antibiotics and anticonvulsions) were excluded. The newborns were randomly allocated to one of the four groups. A group was not supplied. B group received 1mg of vitamin K1 intramusculary, C group received 2mg of vitamin K1 orally. D group was given 20mg of vitamin K1 orally to their mothers at least 2days(range 2 to 7) before birth. Blood samples were collected from 48hrs to 72hrs after birth. PIVKA-II level was measured by enzyme-linked immunosorbent assay (EITEST-MONOP, Eisai Ltd), using a monospecific monoclonal antibody against PIVKA-II. The results obtained are expressed in arbitrary unit (AU) : 1AU corresponds to 1micro gram of purified prothrombin. (healthy adults have less than 0.13AU/ml). PT, PTT were measured simultaneously. RESULTS: 1) PIVKA-II was detected in 4 of 15 infants in group A, who were not supplied. None was detected in other groups. So PIVKA-II detection rate was significantly decreased in other groups compared with group A(p<0.05). 2) PT(sec) values were 12.74+/-0.91, 12.58+/-0.89, 12.36+/-1.04, 12.16+/-0.90 respectively, and there was no significant difference between groups. 3) PTT(sec) values were 52.41+/-13.26, 38.39+/-10.04, 42.67+/-7.01, 39.77+/-10.48 respectively and there was significant shortening in other groups compared with group A (p<0.05). CONCLUSION: Not only intramuscular administration but oral and maternal administration of vitamin K have prophylactic effect on hemorrhagic disease of the newborn. Prophylactic effect on the late hemorragic disease of the newborn requires further extensive study and evaluation.