Repeated oral vitamin K prophylaxis in West Germany: acceptance and efficacy

BMJ 1995; 310 doi: (Published 29 April 1995) Cite this as: BMJ 1995;310:1097
  1. Rudiger von Kries, paediatric epidemiologya,
  2. Alexandra Hachmeister, medical documentalista,
  3. Ulrich Gobel, head of paediatric haematology and oncologya
  1. a Kinderklinik der Heinrich-Heine-Universitat, D-40225 Dusseldorf, Federal Republic of Germany
  1. Correspondence to: Professor von Kries.
  • Accepted 7 March 1995

Owing to concern that neonatal vitamin K prophylaxis might cause childhood cancer,1 parenteral vitamin K prophylaxis for all newborn infants has been abandoned in the United Kingdom2 and Germany.3 In its place three oral doses of vitamin K (0.5 mg in the United Kingdom, 1 mg in Germany) are now recommended for healthy neonates (at birth and during days 4-10 and weeks 4-6). We investigated whether these recommendations are being followed and whether three oral doses of vitamin K are as effective as parenteral vitamin K for preventing late haemorrhagic disease of the newborn.

Methods and results

We drew a random sample of 100 obstetric units from all 995 such units in West Germany. In August 1993 we sent questionnaires to the consultants in charge of these 100 units asking about their use of vitamin K prophylaxis.

Surveillance for late haemorrhagic disease was carried out by sending monthly postcards to heads of all paediatric hospitals in Germany.4 The average response rate during the observation period (West Germany) was 81%. Data from the postcards were validated by questionnaires. All cases of unexpected bleeding from severe, proved vitamin K deficiency in weeks 2-26 observed in West Germany from April 1993 to March 1994 were included in the case definition.

Ninety eight of the 100 questionnaires were returned (one unit was closed, one consultant refused to collaborate). Of these 98 units, 92 gave oral vitamin K (1 mg) twice to all healthy newborn infants in the first week of life as recommended. In 18 units intramuscular vitamin K was given for “at risk” conditions—for example, to infants delivered surgically. In 84 responding units the parents were alerted to the need for a third oral dose of vitamin K (three or more times in 45 units, twice in 27, and once in 12).

Twenty infants met the case definition for late haemorrhagic disease of the newborn (table), of whom 10 had intracranial haemorrhage. Fifteen infants had been breast fed exclusively. In five cases no additional diagnosis was known before and none was detected after the bleeding. Six infants had known conditions for which additional vitamin K prophylaxis was not generally recommended. In one of these additional mild cholestasis was found at bleeding, and one died before tests for cholestasis could be performed. In nine cases cholestasis (which remained “idiopathic” in four) was diagnosed after the bleeding.

Of the 20 infants who met the case definition, two had not received any vitamin K prophylaxis. Thirteen infants had received vitamin K according to the recommendations, but in five cases the required third dose had been overlooked. The time between the last dose of vitamin K and the occurrence of bleeding ranged between eight and 98 (median 28) days. The information about vitamin K prophylaxis was obtained from the well baby check up book (15 cases) or hospital delivery records (first two doses) plus parents' recall of the third dose given by the paediatrician at the well baby check up at weeks 4-6 (four cases).


Acceptance of the new recommendations for vitamin K prophylaxis was high in German obstetric units, possibly because of the availability of an oral preparation. In addition, 84 of the 98 units that responded alerted parents to the need for the third oral dose. These efforts were effective: in 10 of 14 cases that occurred after week 6 the infants had received their third dose of vitamin K.

Even with a dose twice as high as that recommended in the United Kingdom, however,2 3 the efficacy of repeated oral vitamin K prophylaxis was unsatisfactory. Eighteen of 20 infants who met the case definition had received at least two oral doses of vitamin K and 13 of the 20 had received the recommended prophylaxis.

During the observation period a maximum of 550000 babies were given the recommended oral prophylaxis. Had they been given 1 mg vitamin K at birth the expected number of bleedings between weeks 2 and 12 would have been seven.5 The number observed was 10.

This study was performed with the help of Erhebungseinheit fur seltene padiatrische Erkrankungen in Deutschland (ESPED) and the collaboration of the heads of paediatric and obstetric hospitals and departments in Germany. The funding for the ESPED user fees was donated by Roche, Germany.

Late haemorrhagic disease in West Germany 1 April 1993 to 31 March 1994: relation to vitamin K prophylaxis

View this table:


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
View Abstract