RESUMEN
BACKGROUND: Infantile haemangioma is the most common tumour of infancy, but the association with pre-eclampsia is poorly understood. OBJECTIVES: We determined the relationship between variants of pre-eclampsia and risk of infantile haemangioma. METHODS: We carried out a retrospective cohort study of hospital data for all live births between 1989 and 2013 in Quebec, Canada. We identified 14 240 neonates with, and 1 930 564 without haemangioma before discharge, and determined whether early- or late-onset pre-eclampsia was documented on the maternal chart. We used log-binomial regression to compute prevalence ratios (PRs) and 95% confidence intervals (CIs) for the association between pre-eclampsia and infantile haemangioma, adjusted for maternal characteristics. RESULTS: The prevalence of any haemangioma was higher for pre-eclampsia than for no pre-eclampsia (81·3 vs. 72·9 per 10 000), with a PR of 1·15 (95% CI 1·06-1·25) after adjustment for maternal characteristics. Pre-eclampsia with onset before 34 weeks' gestation was associated with cutaneous (PR 2·32, 95% CI 1·68-3·21), noncutaneous (PR 3·66, 95% CI 2·49-5·37) and unspecified haemangioma (PR 2·49, 95% CI 1·77-3·49). However, the association between early-onset pre-eclampsia and haemangioma was attenuated once long neonatal length of hospital stays was accounted for. There was no association with late-onset pre-eclampsia after 34 weeks, and associations were weaker for other variants including severe pre-eclampsia and pre-eclampsia with low birthweight. CONCLUSIONS: Early-onset pre-eclampsia is associated with increased risk of haemangioma at birth, but detection bias due to longer hospital stays and closer follow-up may be part of the reason.