Presented at the Neonatal Society 2017 Summer Meeting.
Helenius K1,2, Longford N2, Lehtonen L1, Modi N2, Gale C2
1 Department of Paediatrics and Adolescent Medicine, Turku University Hospital and University of Turku, Finland
2 Section of Neonatal Medicine, Department of Medicine, Chelsea and Westminster campus, Imperial College London, London SW10 9NH, UK
Background: There is a well-recognised association between early postnatal transfer of extremely preterm infants and increased mortality and brain injury. We aimed to assess if this association is causally related. We hypothesised that transferred infants would have worse outcomes: death before discharge from neonatal care, severe brain injury (grade 3-4 IVH, cystic PVL, porencephalic cysts or posthaemorrhagic hydrocephalus) and the composite outcomes of death or severe brain injury, when comparing to non-transferred infants born in NICUs.
Methods: Infants born <28 weeks in England over Jan 1 2008 to Dec 31 2015 were identified in the National Neonatal Research Database held at Imperial College London. Infants were grouped by birth unit and transfer status at 48 hours of age into 4 groups: upward transfer (birth in SCBU or LNU, transfer to NICU); horizontal transfer (birth in NICU, transfer to another NICU); low intensity (birth in LNU, no transfer); reference (birth in NICU, no transfer). Infants were matched by propensity score and key variables, thus enabling the groups to be compared as in a randomised study. Ethical approval: East Midlands – Leicester South Research Ethics Committee (16/EM/0351).
Results: Of 18,213 infants, 17,577 were eligible: 2,915 in the upward transfer, 306 in the horizontal transfer, 2,668 in the low intensity, and 10,866 in the reference group. Compared to the reference group, infants in the upward transfer group had a higher rate of severe brain injury (OR 1.50, 95% CI 1.36, 1.66, number needed to treat, (NNT) 15 (95% CI 11, 23) and a lower rate of survival without severe brain injury (OR 0.80 (95% CI 0.74, 0.87), NNT 19 (95% CI 12, 53)); infants in the horizontal transfer group had a higher rate of mortality (OR 1.09, 95% CI 0.81, 1.41), severe brain injury (OR 1.22, 95% CI 0.88, 1.61) and a lower rate of survival without severe brain injury (OR 0.93, 95% CI 0.72, 1.18), but the results were not statistically significant; infants in the low intensity group had a higher rate of mortality (OR 1.30, 95% CI 1.17, 1.46, NNT 25 (95% CI 16, 65), a similar rate of severe brain injury (OR 0.99, 95% CI 0.88, 1.13) and a lower rate of survival without severe brain injury (OR 0.82, 95% CI 0.74, 0.90, NNT 25 (95% CI 15, 84).
Conclusion: Birth in low intensity neonatal facilities, and early postnatal transfer appears to be causally implicated in greater risk of death and severe brain injury in extremely preterm infants. Health care systems should focus on maximising these births in high intensity neonatal facilities rather than relying on upward postnatal transfer.
Corresponding author: kkhele@utu.fi