Patterns of Delivery Following in Utero Transfer for Extremely Preterm Birth: A Retrospective Cohort Study From a London Tertiary Centre

AIM: To evaluate delivery patterns following in utero transfer (IUT) to a London tertiary neonatal intensive care unit, examining gestational age, referral indication, transfer-to-delivery interval, and maternal length of stay. METHODS: Retrospective cohort study of women accepted for IUT at St George's Hospital, London (2021-2025), between 22 + 0-27 + 6 weeks' gestation (or 28-29 weeks with multiples/expected birth weight < 800 g) for threatened extremely preterm birth. Data were extracted from BadgerNet, maternity records, and delivery suite databases; Fisher's exact test was used for categorical comparisons. RESULTS: Of 126 IUT episodes, 53 (42%) delivered at the tertiary centre; median transfer-to-delivery interval was 1 day (mean 2.02). Delivery rates averaged ~10%/day over the first 5 days. Transfers from outside the local network delivered within 24 h more often (18%). Among 73 (58%) who did not deliver, median length of stay was 4 days. Delivery was less likely at 22-24 weeks (32% vs. 50% at ≥ 25 weeks) and significantly associated with sepsis (p = 0.01), chorioamnionitis (p < 0.001), pathological CTG (p = 0.006), and Doppler abnormalities (p = 0.02). CONCLUSION: Over half of women undergoing IUT did not deliver during admission. Infection and fetal compromise indications were most strongly associated with delivery, supporting liberalised IUT acceptance and capacity planning.