Background Strategies to prevent heat loss in the delivery room after very preterm birth have been proven effective in randomised controlled trials. Nevertheless, we hypothesise that hypothermia at admission to neonatal care is still common and contributes to mortality after very preterm birth.

Methods The EPICE cohort included all births between 22+0 and 31+6 weeks of gestation in 19 regions from11 European countries in 2011–2012. We studied infants surviving to admission to neonatal care (n = 7577). The association between temperature at admission and in-hospital mortality was analysed using logistic regression. The final model adjusted for gestational age, small for gestational age (SGA), Apgarscore <7 at 5 min, infant sex and region of birth.

Results Of 6639 infants with data on body temperature at admission, 1670 infants (25%) were hypothermic (<36.0° C); 6% had temperatures <35°C, 7% between 35.0 and 35.4°and 12% between 35.5 and 35.9°. Body temperature at admission was inversely related to mortality. The crude odds ratio (OR) (95% confidence interval [CI]) for mortality was 5.81(4.27–7.92) when temperature was <35°C; 3.32 (2.35–4.69) at 35.0–35.4°; and 1.61 (1.18–2.19) at 35.5–35.9°compared to normothermic infants (36.5–37.5°C). After adjustment, temperatures below 35.5°C remained significantly associated with mortality, 1.94 (1.32–2.83) at <35°C and 1.91(1.30–2.82) at 35–35.4°C compared to normothermic infants.

Conclusion Hypothermia after very preterm birth contributes to mortality in modern perinatal care settings in Europe. Further studies should investigate if evidence-based heat loss prevention strategies have been implemented.