Neonatal birth fractures: a retrospective tertiary maternity hospital review.

We aimed to identify the incidence and types of neonatal birth fractures in a single tertiary maternity hospital in the United Kingdom and to find possible associated factors, including all live births born between 2000 and 2016. We reviewed hospital records and imaging of all neonates who had any imaging done to identify birth fractures. We identified 87,461 consecutive live births. Sixty-six sustained a fracture during delivery: 46 clavicle-, 13 humerus-, four skull-, one femoral-, one rib- and one tibial fracture. Five neonates with a clavicle or humeral fracture had an Erb's palsy. Sixty-five fractures were in singletons. Twenty-five fractures were diagnosed after discharge. Binary logistic regression analysis with R-Studio showed a significant association between 'Fracture' and 'Birthweight' ( < .0005), 'Delivery Mode' (Forceps:  < .001, Ventouse:  < .0004) and 'Gestation' ( < .0005) but not with 'Sex', 'Day' and 'Time' of delivery, 'Number of deliveries per day', 'Singleton/Multiple Births' and 'Breech'. The incidence of birth fractures (0.075%) was low with 24 hours obstetrician support on site in comparison to published data. We recommend to include data on neonatal birth injuries in addition to the existing clinical safety markers for delivery units.Impact statement Most birth fractures affect the clavicle with a large variation in published incidences from 0.035% to 3.2%. High birthweight is the most frequently identified risk factor. An increased risk for out of hours deliveries (16.00-8.00) and inverse association between fracture rate and level of experience and academic qualification have also been reported. Between 14% and 39% of fractures are diagnosed after discharge but many studies are based on birth certificate and discharge diagnoses coding only. This is the first study on neonatal birth fractures from the United Kingdom and the only study for which radiological investigations of all neonates were reviewed. Our fracture rate of 0.075% for all fractures is therefore most likely the most accurate, showing no significant difference in the fracture risk between our six defined time intervals and days of the week, with experienced midwifes managing many high risk pregnancies and an obstetrician being present on site all the time. Our findings support to use data on neonatal birth injuries as one indicator to assess the quality and safety of maternity units.