435 - Assessing Radiation Risk Associated with X-Ray Confirmation of Umbilical Catheters in the NICU

Publication Number: 3420.435

Tess A. Korte, Saint Louis University School of Medicine, St. Louis, MO, United States; William R. Carter, George Washington University School of Medicine and Health Sciences, Potomac, MD, United States; Amit Mathur, Saint Louis University School of Medicine, Chesterfield, MO, United States; Srirupa Hari Gopal, Saint Louis University School of Medicine, St Louis, MO, United States

Background: Cumulative radiation exposure to neonates admitted into the NICU is understudied. Umbilical Catheter (UC) placement at admission and within the first 24 hours of life, requires confirmation with X-Rays, often done repeatedly to ensure safe placement.

Objective: In preparation for launching a point of care ultrasound (POCUS) program we wanted to quantify the radiation exposure associated with confirmation of UC placement in neonates and risk factors associated with increased exposure at the time of line placement and within 24 hours of life (HOL).

Design/Methods: A retrospective review was conducted from September 2024 - March 2025, with a cohort of 61 neonates admitted to the NICU. Neonates ≤ 34 weeks' gestation, ≤ 1500 grams birth weight, and who had received an UC placement after NICU admission were included; those without line placement or who died in the delivery room were excluded. Demographic information and the number of UC-related X-rays (number and millisievert-mSv) received at the time of placement and in the 1st 24 hours of life (HOL) were collected from electronic and radiology records. Number of X-rays were analyzed by gestational age (GA) groups ( < 28, 28-32, >32 weeks) and birth weight (BW) groups ( < 1000, 1000-1500 g). Normality and variance were assessed (Shapiro-Wilk, Levene), followed by ANOVA or Kruskal-Wallis tests. Pearson correlations and Poisson linear regression estimated associations, with incidence rate ratio (IRR), 95% CIs, and significance (p < 0.05). Results were visualized using forest plots and predicted count graphs.

Results: The median (CI) number of X-rays associated with initial umbilical line placement was 3, 0.06 mSv (radiation dose per x-ray-0.02 mSv) and the median (CI) within 24 hours was 4, 0.08 mSv (Table 1). There was no significant difference between in the number of X-Rays between the GA and BW groups either at line placement or at 24 HOL (Figure 2). Pearson correlations showed a trend towards lower BW (r=-0.136, p=0.29) (Figure 3A) and lower GA associated with higher X-rays at 24 HOL (r = -0.328, p = 0.0099) (Figure 3 B). Upon applying Poisson linear regression model, for line placement, neither GA nor BW predicted X-ray counts. At 24 HOL, neonates >32 weeks GA had significantly fewer X-rays (IRR ≈ 0.65, p < 0.001) compared to < 28 and 28-32 weeks (Figure 3C).

Conclusion(s): Radiation exposure related to UC placement is significant, especially in neonates with extreme prematurity and low birth weight. Alternative methods to confirm UC placement should be considered, such as POCUS could potentially decrease this exposure.

Table 1. Median number of X-rays at placement and 24 HOL in various GA and BW subgroups.
Gestational Age-GA, Birth Weight-BW

Figure 2. Box plot showing number of x-rays at line placement (bottom panels) and 24 hours of life (HOL) (top panels) with birth weight (left) and gestational age (right) groups.


Figure 3: A: Pearson's correlation plot shows BW versus number of X-rays at 24 HOL. B: Pearson's correlation plot shows GA versus number of X-rays at 24 HOL. C: Poisson regression model showing incidence rate ratio.