Vitamin K for Newborns

Newborns in the US are routinely given a Vitamin K injection at birth to prevent hemorrhagic disease of the newborn (HDN). Here's what I've found on this practice.

Some information from the AAP Pediatric Nutrition Handbook:

"Except for Vitamin K, most infants born to well-nourished mothers have adequate vitamin stores at birth. Vitamin K is necessary for blood coagulation. Proteins C, S, and Z participate in the clotting mechanism and bind calcium. The newborn infant is usually given Vitamin K soon after birth for prophylaxis against hemorrhagic disease of the newborn. Vitamin K should be given as a single intramuscular dose of 0.5 to 1 mg, or as an oral dose of 1 to 2 mg."
Food sources of Vitamin K: cow's milk, green leafy vegetables, pork, liver (cow's milk has 4.9 mg/liter of Vitamin K, compared to 2.1 mg/liter in human milk).
Major brands of formula (Enfamil, Gerber, Good Start, Similac, SMA) are all enriched to 54-55 mg of Vitamin K per liter of milk, so formula-fed infants are extremely unlikely to be deficient in this vitamin.

And from the Nelson Textbook of Pediatrics:

"Naturally occurring vitamin K is fat soluble; it is found in high concentrations in hog's liver, soybeans, and alfalfa and in smaller amounts in some vegetables, such as spinach, tomatoes, and kale."
In the body, vitamin K is naturally produced by intestinal bacteria, which of course the newborn lacks. Consequently, "suppression of intestinal bacteria by various antibiotics may be responsible for vitamin K deficiency, which results in dimunition of prothrombin." (Prothrombin is one of the proteins responsible for blood clotting when it is supposed to.)
Classic HDN sets in 2-7 days after birth. Usual sites of bleeding are gastrointestinal, ear-nose-throat, intracranial, circumcision, cutaneous, and injection sites. Without prophylaxis, approximately 2% of infants will develop classic HDN. There's also late HDN, occurring 1-6 months after birth, with the usual sites of bleeding being intracranial, gastrointestinal, cutaneous, ear-nose-throat, and injection sites.
"Administering 1 mg of natural oil-soluble vitamin K intramuscularly at the time of birth prevents the fall in vitamin K-dependent factors in full-term infants but is not uniformly effective in the prophylaxis of hemorrhagic disease of the newborn in premature infants." Options for treatment of HDN include IV infusion of vitamin K and transfusion of whole blood or plasma.
"The recommended drug and dose of vitamin K (IM) for prophylaxis in the United States has been safe and is not associated with an increased risk of cancer. Although oral vitamin K (birth, discharge, 3-4 wk: 1-2 mg) has been suggested as an alternative, the oral route is not universally accepted and the IM route remains the method of choice."

In 1993, the American Academy of Pediatrics (AAP) issued a policy statement on Controversies Concerning Vitamin K and the Newborn. Some excerpts:

"Vitamin K deficiency may cause unexpected bleeding (0.25% to 1.7% incidence) during the first week of life in previously healthy-appearing neonates (classic hemorrhagic disease of the newborn [HDN]). The efficacy of neonatal vitamin K prophylaxis (either oral or parenteral) in the prevention of classic HDN is firmly established."
"Late HDN, a syndrome defined as unexpected bleeding due to severe vitamin K deficiency in infants aged 2 to 12 weeks, occurs primarily in exclusively breast-fed infants who have received no or inadequate neonatal vitamin K prophylaxis....When a single dose of oral vitamin K has been used as neonatal prophylaxis, the rate has decreased to 1.4 to 6.4 per 100 000 births. Parenteral neonatal vitamin K prophylaxis prevents the development of late HDN, with the rare exception of infants with severe malabsorption syndromes. Oral regimens that have a similar efficacy as parenteral vitamin K in prevention of late HDN include the repeated administration of oral vitamin K1 (Germany) or K2 (Japan) at birth, 1 week, and 2 to 4 weeks."
"In 1990 Golding et al reported a study of a 1970 birth cohort in Britain in which they noted an unexpected association between childhood cancer and pethidine given in labor and the neonatal administration of vitamin K. Subsequently, Golding and others conducted a case-control study designed to examine the risk of cancer associated with intramuscular vitamin K administration among infants born in two hospitals in Avon between 1965 and 1987 and diagnosed with cancer between 1971 and 1989. They reported a significant association between intramuscular vitamin K and cancer when compared to no vitamin K or oral vitamin K. They recommended exclusive use of oral vitamin K."
"If intramuscular vitamin K doubles the incidence of childhood leukemia, a sharp increase should have been seen after 1961, when the American Academy of Pediatrics first recommended that all neonates be given vitamin K at birth. Since the 1940s the peak incidence of leukemia has been in children younger than 5 years of age. In this age-interval... no marked increase in incidence occurred between 1947 and 1950 and 1983 and 1984 among whites in five geographic areas of the United States."
"There is no evidence of an increase in childhood leukemia since the period of 1947 to 1950, well before intramuscular vitamin K at birth was first recommended for US children. An increase in diverse forms of childhood cancer, claimed to be a result of vitamin K administration at birth, was not seen even among children of Hiroshima and Nagasaki exposed to the atomic bomb, and such diversity has not been seen in adults exposed to other human carcinogens. The biological mechanism of carcinogenesis (induction of SCEs) proposed by Golding et al does not accord with other information from the literature about in vivo effects of vitamin K or other tests of carcinogenicity."

The Canadian Paediatric Society has their own position paper on Routine Administration of Vitamin K to Newborns. This position paper, issued in 1997, reverses a previous recommendation from the CPS that vitamin K be administered orally. Their current recommendation is for intramuscular administration. Again, some excerpts:

In 1961, the Committee on Nutrition of the American Academy of Pediatrics (AAP) recommended that vitamin K1 (hereafter referred to as vitamin K, the only form of vitamin K1 used in neonates) 0.5 to 1.0 mg be administered intramuscularly to all newborns shortly after birth to prevent this problem. In 1988, the Canadian Paediatric Society (CPS) indicated that 2.0 mg of vitamin K administered orally within 6 h of birth was an acceptable alternative. This was before the suggestion that the risk of childhood cancer increases after intramuscular vitamin K shortly after birth, a suggestion which has subsequently been shown invalid. Although other countries joined Canada in recommending the alternative oral administration of vitamin K, the AAP has continued to advocate sole use of the intramuscular route, noting that an approved oral form is not available. The CPS believes that, on the basis of available information, their recommendations should be modified.
Although no significant complications after 420,000 intramuscular injections of vitamin K to newborns were reported, the psychological effects of intramuscular injections on newborn infants and their parents are unknown. It has been reported that pain experienced during the neonatal period may have long term effects. However, the benefits of routine vitamin K administration have been clearly shown, and it is important that this be given in the most effective manner. The 1988 CPS recommendations aimed to obtain the benefit of vitamin K for newborns without incurring pain. These recommendations supported the oral route of administration of vitamin K with a formulation designed for parenteral use, a regimen reported to be effective, practical and economical.
An epidemiological study from Germany by von Kries showed a failure rate (occurrence of late HDNB) after intramuscular administration of 0.25 per 100,000 infants, compared with a rate of 1.4 per 100,000 infants after oral administration. In other countries in which oral administration is the primary form of vitamin K deficiency prophylaxis, the incidence of late HDNB varied – 1.5 (Britain), 6.0 (Sweden) and 6.4 (Switzerland) per 100,000 infants. Some of these infants could have had underlying disorders that affected vitamin K metabolism.

Women's Health UK have a very informative page: Vitamin K - Does my baby need it?. Excerpts:

Unfortunately, in about 1/3 of cases the vitamin K deficiency bleed occurs without prior warning or risk factor. It has been estimated that if vitamin K were only given to high risk babies, among the 800,000 or so annual births in the UK, there might be:
- 60-80 babies who suffer a bleed
- 15-20 babies suffering a bleed into the brain
- 4-6 babies who die from the bleed into the brain
- 10-20 babies who may be brain damaged because of the bleeding
Concerns about the safety of this arose in the early 1990's when 2 papers were published in the medical literature, suggesting an association between vitamin K injection and childhood leukaemia (a blood cancer). The papers looked at children with leukaemia and checked how many of them had received vitamin K injection compared to children without leukaemia. They found an increased risk of leukaemia by a factor of about 1.8.

Following this very unexpected finding, lots of other comparisons were made throughout the world to see if this was indeed a true increase in risk or just a chance happening in this group. Studies from the UK, USA, Germany and Sweden found no evidence to support these findings. A Danish study followed all children born in Denmark over a 40 year period from 1945 and compared the leukaemia rates in those receiving no vitamin K, vitamin K by mouth and vitamin K injection and found no difference.

More recently a Scottish study looked at over 400 children aged up to 14 years with cancers and found no association between vitamin K injection and any cancer. Two more studies were published this year containing a total of 4000 cases of childhood cancers, demonstrating no association with vitamin K usage.
After the initial concens about vitamin K injection and before the many more reassuring studies, the national policy in Germany and Australia changed over to the 'by mouth' or oral regime for vitamin K. This means giving one dose at delivery, one at 7 days of age, and a further one at 28 days to achieve the same protection as a single injection just after delivery. After this change in policy, an increase in the rate of vitamin K deficiency bleeding was found, affecting about 1/100,000 infants.

This failure of the oral vitamin K to work was probably in part due to not having a specific oral form of the drug available, as it is now. It is also, in some cases, due to a failure to complete the course. If the newer vitamin K oral formulation (Konakion MM) is taken as an extended course, at the correct times, it is almost certainly as successful in preventing bleeding problems as the single injection.

Gentlebirth.org, an online resource for midwives also has a page on Administration of Vitamin K to Newborns. Much of their infomation in anecdotal. Excerpts from their page:

Some very recent studies in The Lancet have associated increased clotting with twice the likelihood of death from bacterial meningitis. These higher clotting factors may increase the risk from all bacterial infections. Since the purpose of administration of vitamin K is to increase clotting factors, is it possible that this is also inadvertently increasing a newborn's susceptibility to infection? [Note, though, that the study cited was a study of genetic clotting abnormalities, not of vitamin K]
However, an injection creates an avenue of infection for a newborn with an immature immune system in an environment that contains the most dangerous germs. In addition, the possible trauma from the injection can jeopardize the establishment of breastfeeding, which does much more to protect the baby's health than vitamin K injections have ever been alleged to do. At the very least, the injection should be delayed until after the baby has learned to nurse.

I've sometimes wondered whether there's a connection between vitamin K administration and SIDS. Some studies have shown a lower incidence of SIDS among breastfed babies, and we know that breastmilk is lower in vitamin K. Who knows? Nobody, really. Why are we messing with delicate systems we don't understand?

Forced Medication of Healthy Newborns with Vitamin K in New York State discusses the legal requirement for this injection if you live in New York, and includes some statistics challenging the medical necessity for the injection.

As always, you should make up your own mind on what to do for your own infant, and consult with your healthcare provider on your decision. Our decision is that we're going to have the Vitamin K injection administered at birth. Balancing off a 2% risk of intracranial bleeding (possibly life-threatening) against an injection, this seems a clear choice to us. We don't find the reports of serious risks to Vitamin K administration to be convincing, and the increased incidence of late HDN in infants receiving oral Vitamin K is disturbing. If we were going to use the oral dose, we'd go with multiple rather than single administration, since it appears to take multiple doses to eliminate the chance of late HDN.