Infant vitamin K injections are associated with an increased risk for acute lymphoblastic leukaemia.

Neonatal vitamin K administration and childhood cancer in the north of England: retrospective case-control study.

L Parker, Michael J Cole, A W Craft, E N Hey

BMJ

Jan 17, 1998

Sources

0:00 / 0:00

Sign in to access this feature

Create a free account or sign in to use AI summaries, listen to articles, download PDFs, and save to your library.

2,016 views

Abstract

OBJECTIVE: To explore the possible association between intramuscular vitamin K given to neonates and the subsequent development of childhood cancer. DESIGN: Retrospective case-control study on the basis of hospital records. SETTING: The former Northern Health region of England. SUBJECTS: 685 children who were born and lived in the region and who developed cancer before their 15th birthday, and 3442 controls also born between 1960 and 1991 and matched only for date and hospital of birth. The notes of a further 701 index cases were untraceable. MAIN EXPOSURE MEASURE: Administration of intramuscular vitamin K versus no exposure to vitamin K. RESULTS: There was no association between the administration of vitamin K and the development of all childhood cancers (unadjusted odds ratio 0.89; 95% confidence interval 0.69 to 1.15) or for all acute lymphoblastic leukaemia (1.20; 0.75 to 1.92), but there was a raised odds ratio for acute lymphoblastic leukaemia developing 1-6 years after birth (1.79; 1.02 to 3.15). No such association was seen in a separate cohort-based study not dependent on case note retrieval in which the rates of acute lymphoblastic leukaemia in children born in hospital units where all babies received vitamin K were compared with those born in units where less than a third received prophylaxis. CONCLUSIONS: It is not possible, on the basis of currently published evidence, to refute the suggestion that neonatal intramuscular vitamin K administration increases the risk of early childhood leukaemia. Any association may have been masked in earlier studies that did not use controls matched for time and locality by other unidentified factors affecting the spatiotemporal variations in incidence of leukaemia.