Objective of the study was to compare number of X-rays performed for neonatal central line tip positioning when real-time US is used compared to X-ray only, and to assess
consequences on position accuracy, irradiation and cost. Retrospective monocentric cohort study conducted at Evelina London Children’s Hospital Neonatal Unit over 6
months. Study was conducted during implementation of US for line tip localisation with formulation of US protocol. Tip position on X-ray was reviewed by one neonatologist and
one radiologist and inter-rater agreement calculated. Criteria for good, satisfactory or inadequate position of the tip were defined. Estimated effective radiation dose and cost for
each X-ray was determined. Two hundred seventy-four lines were inserted (nPICC, UVC, UAC). Eighty-three lines were scanned with US (US group); 191 lines were not (no-US
group). Number of X-rays performed was significantly lower in the US group: 1.19 vs. 1.5 (p 0.001), related to a significantly lower percentage of lines requiring multiple X-rays
(38.7% no-US group vs. 19.9% US group; p 0.004). Accuracy was higher in US group with more lines at cavoatrial junction (p 0.05) and was significantly increased with
US use for lines inserted from lower limbs (22.9% and 76.2%, p 0.001). Inter-rater agreement was strong (k > 0.8). US group received lower mean radiation dose (p < 0.001)
and cost related to X-ray was significantly reduced (p 0.001).
Conclusion: Real-time US use for line tip positioning in the NICU significantly decreased the number of X-rays performed and was associated with better-positioned lines,
decreased irradiation and cost.