The historical term ''hemorrhagic disease of the newborn,'' which is used as a synonym for vitamin K deficiency bleeding (VKDB) in infancy, preferably should be abandoned, since neonatal bleeding is often not due to vitamin K (VK) deficiency and VKDB may occur after the neonatal period. VKDB is a form of bleeding that is caused by reduced activity of VK-dependent coagulation factors (II, VII, IX, X), has normal or even increased activity of VK-independent coagulation factors, and responds to VK. Acarboxy proteins are present. In a bleeding infant a prolonged one-stage prothrombin time (a decreased Quick value, which means prothrombin time expressed as percent of normal) in association with a normal (or increased) fibrinogen level and platelet count is almost diagnostic of VKDB. The diagnosis is proven, if administration of VK is followed by a shortening of the prothrombin time (after only 30 minutes) or cessation of bleeding. Classification is by age of onset as early, classic, and late form of VKDB. Rare cases of VKDB occur also after week 15; therefore the upper age limit should be 6 months and not 3 months. In idiopathic VKDB the cause (other than breast-feeding) is unknown. In secondary VKDB additional factors can be demonstrated, such as poor intake or absorption of VK and increased consumption of VK. Most often cholestasis is present. Postnatal VK provides effective protection from the classic and late form of VKDB. The classic form can be prevented equally well by a single oral dose of 1 mg VK after birth as by parenteral VK. Parenteral prophylaxis appears to be more effective in preventing the late form, especially in patients with hepatobiliary disease. The protection achieved by single oral prophylaxis can be im proved by repeated oral administration. In addition to reducing the incidence of VKDB, VK prophylaxis also reduces the proportion of intracranial hemorrhages and increases the age at onset of bleeding, parenteral prophylaxis being more effective than single oral prophylaxis in this respect, Because of the potential risks (cancer?) associated with extremely high levels of VK (20,000-fold the normal value for healthy newborns) and the possibility of injection injury, parenteral VK has been questioned as the first choice of prophylaxis for normal neonates. These disadvantages do not apply to oral prophylaxis. The major disadvantage of oral prophylaxis, namely, its lesser reliability in terms of intake and absorption, could be largely overcome by repeated administration. Although VK prophylaxis seems to be necessary only for breast-fed infants, breast-feeding should be promoted. As VK is involved not only in coagulation but also in carboxylation with multiple effects, care should be taken to avoid any excessive deviation from the physiologic conditions, that prevail in the fully breast-fed healthy mature infant, with low VK levels in the postnatal period. Repeated (daily or weekly) oral doses of VK are closer to physiologic conditions than single parenteral bolus doses, which expose neonates to excessively high VK levels. The incidence of intracranial VKDB can be reduced if the grave significance of warning signs is recognized (icterus, failure to thrive, feeding problems, minor bleeding, diseases with cholestasis) or if a simple test for acarboxy proteins (similar to the Guthrie test) would be applicable. Whether or not the more reliable absorption of the new mixed micellar preparation of VK could reduce the protective oral dose of Vg prophylaxis has to be evaluated.
