• 
• 

Three Part Question

In [newborn babies] is [intramuscular Vitamin K prophylaxis] associated with [an increased risk of developing childhood cancers]?

Clinical Scenario

On the post natal ward and labour ward it is routine to consent parents for the administration of Vitamin K to newborn babies.

Vitamin K deficiency may develop in neonates within the first few days of life. In newborns, vitamin K nutrition is at risk for the following reasons:

1. Poor placental transfer of lipids

2. Immaturity of the neonatal liver (with respect to prothrombin synthesis)

3. Low quantities of vitamin K in breast milk

4. Sterility of the neonatal gut during first week of life

The neonate may thus be predisposed to the bleeding diathesis: "Vitamin K Deficiency Bleeding" (VKDB) (formerly known as haemorrhagic disease of the newborn). Bleeding may manifest as cutaneous, Gastrointestinal, intrathoracic, or, in the worst cases, intracranial bleeding.

In order to prevent these problems current recommendations state that all newborns should be given Vitamin K prophylaxis either intramuscularly (1mg) or via an oral regime (Department of Health UK CMO report, 1998). Oral regimens, if not strictly followed, may lead to late onset VKDB, whereas a singular IM injection is sufficient protection. During background reading about this practice, my attention was directed to concerns about Vitamin K prophylaxis administered intramuscularly (IM) and its association with the development of childhood cancers. This report aims to examine the literature relevant to the possible association of IM Vitamin K administration and childhood cancers.

Search Strategy

Systematic reviews – Cochrane Library: Using "Vitamin K AND Neonates AND Cancer"

Primary research – PubMed: The following search terminology was derived from the MeSH browser - "Vitamin K"[MeSH] AND "Infant, Newborn"[MeSH] AND "Neoplasms"[MeSH] (when examining "Vitamin K AND Neonates AND Cancer")

Secondary sources – included UK Department of Health Guidelines, a Policy Statement by the American Academy of Pediatrics Committee on Fetus and Newborn, and additional referenced papers from the British Medical Journal (BMJ) from the above.
As above

Search Outcome

Systematic reviews – Cochrane Library: Using "Vitamin K AND Neonates AND Cancer" identified only one systematic review whose subject was not relevant to the search.

Secondary sources – included UK Department of Health Guidelines, a Policy Statement by the American Academy of Pediatrics Committee on Fetus and Newborn, and additional referenced papers from the British Medical Journal (BMJ) from the above.

Search results – the PubMed strategy identified 65 papers of which 11 were used (2 of which were Guidance reports), and two additional papers referenced from these articles to the British Medical Journal.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
American Academy of Pediatrics Committee on Fetus and Newborn 2003 USA		Expert Committee Report (Level IV)	Concerns over possible causal association between intramuscular Vitamin K administration and childhood cancer	Committee came to conclusion that such a concern has not been substantiated. The report recommends intramuscular Vitamin K use to prevent VKDB of newborns. It questions the efficacy of oral regimen of Vitamin K in preventing late VKDB.
Passmore S, Draper G, Brownhill P, Kroll M 1998	597 cases with childhood cancer and matched controls were studied. Controls matched for sex, month of birth and hospital.	Case control study (Level III)	Odds ratios for childhood cancers with administration of IM Vitamin K	The association between cancer and IM Vitamin K was shown to be of borderline significance (Odds Ratio 1.44, P = 0.05). The association was strongest for leukaemia.	The Records for Vitamin K use were only available for 40% of cases so the authors assumed hospital policy on the matter was followed for all. The Odds Ratios were not not adjusted for effect of abnormal delivery, which could explain some of the findings.
Ansell P, Bull D, Roman E 1996	Cases = children (0-14yrs) diagnosed with leukaemia. Two controls per case (matched on sex, hospital, and year and month of birth) randomly selected from relevant hospitals	Case-control study designed to investigate associations between leukaemia and prenatal and neonatal exposures(Level III)	Adjusted odds ratios for leukaemia with intramuscular Vitamin K versus all other possibilities combined (none/oral/no record of vitamin K)	Adjusted Odds Ratio among those whose medical notes indicated administration of intramuscular Vitamin K was 1.0 (95% CI 0.5-1.9, based on 45 informative sets) for acute lymphoblastic leukaemia and 1.2 (0.7-2.3, 52 informative sets) for all leukaemias. No significant association demonstrated	Relatively small sample sizes (~100-300). Did not include multiple pregnancies, children with chromosomal abnormalities and those who died before discharge.
Von Kries R, Göbel U, Hachmeister A, Kaletsch U, Michaelis J 1996	272 children with leukaemia, nephroblastoma, neuroblastoma, rhabdomyosarcoma, and CNS tumours diagnosed between 30 days of age and 15 years of age. 334 population based controls without diagnoses of cancer	Population based case-control study Comparison of vitamin K exposure in children with leukaemia/other tumours with two control groups (Level III)	Odds ratios for association of childhood cancers and leukaemias with Parenteral vitamin K prophylaxis (intramuscular and subcutaneous) versus oral and no vitamin K prophylaxis.	No association between parenteral Vitamin K exposure and childhood cancer (including leukaemias) concluded: Odds Ratio 1.04 (95% CI 0.74 - 1.48). Observed Odds Ratio for leukaemia 0.98 (0.64 - 1.50)	IM and Subcutaneous routes both examined together as opposed to simply IM route of administraton. Small sample size.

Comment(s)

It was first shown in 1939 that treatment with vitamin K could abolish symptomatic prothrombin deficiency in the first week of life (Hey, 2003). Initially, selective prophylaxis for "at risk” (mostly preterm babies and babies having an operative delivery) was given, since cases of excessive use leading to haemolysis and kernicterus had been documented (particularly in the UK). Quite rapidly this policy of selective prophylaxis gave way to one of universal prophylaxis (as resurgence of VKDB occurred), with reports showing that policy in the UK varied widely from unit to unit (i.e. in methods and dose of administration). Universal IM Vitamin K prophylaxis was preferred to oral - not so much to manage the occasional case of early bleeding, but to prevent late bleeding in the breast-fed baby. However this was questioned when Golding et al initially postulated that babies given IM vitamin K might be more likely to develop cancer (particularly leukaemia) later in childhood (Golding et al., 1992). Klebanoff et al went on to dispute these findings, demonstrating that there was no significant association (Klebanoff et al., 1993) The difference for the opposing findings may in part have been due to the different study methods employed: Golding et al conducted a retrospective chart review, whereas Klebanoff et al based their investigations on a prospective case-control study that provided better quality data with documentation of all clinical events in the infants. Since then, successive additional case-control studies (conducted in similar manners on similar sample sizes) have been interpreted as finding no support for the original postulation by Golding et al (Golding et al., 1992, Ansell et al., 1996, Fear et al., 2003, McKinney et al., 1998, Passmore et al., 1998, Parker et al., 1998). Two studies (Passmore et al., 1998, Parker et al., 1998) have commented on a borderline significance in association of IM Vitamin K administration with childhood cancers – particularly that of acute lymphoblastic leukaemia (Passmore et al., 1998, Parker et al., 1998). It is for this reason that the Department of Health indicate in their summary that it is not possible to exclude a small risk in leukaemia, suggesting that the data is limited, hence further studies are required (Department of Health, 1998). Recent studies (backed by the UK Childhood Cancer Study) have added to the above, indicating the lack of significant association between IM Vitamin K administration and childhood cancers (Fear et al., 2003, Roman et al., 2002). In particular, Roman et al, performed a pooled analysis of 6 case control studies and concluded that there was no significant association between IM Vitamin K administration and childhood cancers. However, they too acknowledge that due to poor data small effects could not be conclusively dismissed (Roman et al., 2002). Additionally, two large scale studies based on entire populations of neonates born in Sweden and Denmark within defined time periods showed no elevated risk of childhood cancers (including leukaemias) in subjects that had received IM Vitamin K as opposed to oral regimens (Ekelund et al., 1993, Olsen et al., 1994). Lastly, both IM and subcutaneous modes were compared with oral administration in von Kries et al (1996). Again, no significant associations were elucidated (including leukaemias). Since it is not possible to completely rule out the associations with childhood cancers, particularly leukaemias, the Department of Health Guidelines also advocate usage of the oral regimens involving repetition, in order to achieve levels similar to those in IM Vitamin K administration. However, though the report by the Committee on Fetus and Newborn agrees the data does not support a significant association between IM Vitamin K prophylaxis and childhood cancer, it questions the efficacy of oral regimes in preventing late onset VKDB (American Academy of Pediatrics Committee on Fetus and Newborn, 2003). Therefore, further studies need to provide answers to a few essential remaining questions: Firstly, should Vitamin K prophylaxis be given orally as opposed to IM, as no studies have postulated a link between this method of administration and the development of childhood cancers? And secondly, if IM prophylaxis is to be continued, should it be reserved only for high-risk patients? To conclude, a number of trials have shown no significant association of IM Vitamin K prophylaxis with childhood cancers, yet several have refused to completely rule out the risk of association to leukaemias. The widespread use of Vitamin K prophylaxis and the recognized deleterious consequences when it is withheld make it unjustifiable to perform a large scale randomized control trial. However, it seems that only such a controlled trial could ever resolve the residual uncertainty.

Clinical Bottom Line

The available data does not support an increased risk of childhood cancers caused by IM Vitamin K prophylaxis given to neonates Studies have been unable to confidently exclude a small risk National Guidelines recommend all newborn babies to receive an appropriate Vitamin K regime to prevent VKDB

References

1. American Academy of Pediatrics Committee on Fetus and Newborn Controversies concerning vitamin K and the newborn Pediatrics 2003 Jul; 112(1 Pt 1): 191-2
2. Passmore S, Draper G, Brownhill P, Kroll M Case-control studies of relation between childhood cancer and neonatal vitamin K administration BMJ 316: 178 - 184
3. Ansell P, Bull D, Roman E Childhood leukaemia and intramuscular vitamin K: findings from a case-control study BMJ 1996; 313: 204 205
4. Von Kries R, Göbel U, Hachmeister A, Kaletsch U, Michaelis J Vitamin K and childhood cancer: a population based case-control study in Lower Saxony, Germany BMJ 1996; 313: 199 - 203