Presented at the Neonatal Society 2015 Spring Meeting.
Mawson I1, Babu P, Fox G, Simpson J
Neonatal and Cardiology Departments, Evelina London Children’s Hospital, Guy’s and St Thomas’ NHS Foundation Trust.
Background: The aim was to investigate the sensitivity of newborn pulse oximetry (Pulsox) screening for each type of critical congenital heart disease (CCHD). The hypothesis was that Pulsox screening sensitivity varies depending on the type of CCHD being identified and depending on the threshold used for further investigation. A meta-analysis in 2012 described Pulsox screening’s sensitivity as moderate, with high specificity and low false positive rates (1). Due to small numbers of infants with CCHD in studies little data is available on Pulsox sensitivity by diagnosis.
Methods: Retrospective review of admission pre-ductal oxygen saturations of infants with antenatally diagnosed CCHD delivered at a cardiac neonatal unit between 2010- 2014. Saturations were recorded at median of <1hr (range <1-9hrs) after delivery. Data was stratified by CCHD diagnosis and analysed according to three different oxygen saturation thresholds, ≤90%, ≤92%, and ≤95%. Sensitivities (with 95% confidence intervals) of Pulsox screening using each threshold were calculated using Microsoft Excel for the whole data set and for each CCHD diagnosis. Calculations were repeated for a group representing infants well enough for postnatal ward care.
Results: 276 neonates were identified. 208 (78.2%) were clinically well, admission to the neonatal unit occurred purely due to antenatal CCHD diagnosis. In this group Pulsox was more sensitive using ≤95% threshold at 71.6% (65.5- 77.8%) compared to ≤92% (52.3% (45.6-59.2%)) and ≤90% (46.2% (39.4-52.9%)). The 95% confidence intervals (95%CI) for Pulsox sensitivities using a threshold of ≤90% and ≤92% did not overlap with the 95%CI of the Pulsox sensitivity using ≤95%. Pulsox sensitivity was increased by 37.5-52.7% using ≤95% threshold compared to ≤90% for hypoplastic left heart (HLH), tetralogy of Fallot (TOF) and interrupted aortic arch (IAA). Using the highest threshold (≤95%) Pulsox sensitivity for aortic stenosis (AS) was 20% (0.0-55.1%), pulmonary stenosis was 36.4% (7.9-64.8%), and coarctation was 41.5% (26.4-56.5%) whereas sensitivity for pulmonary atresia (PA) at threshold ≤90% was 66.7% (40.0-93.3), TGA 83.9% (74.7-93.0%) and TAPVD 100.0% (100.0-100.0%).
Conclusion: Pulsox sensitivity is influenced by CCHD diagnosis and is highest using ≤95% threshold (overall and especially for HLH, IAA and TOF). In coarctation, AS and PS, the minority of cases were detected irrespective of saturation threshold. TGA, PA and TAPVD would be detected in the majority regardless of the threshold.
Corresponding author: firstname.lastname@example.org
1. Thangaratinam S, Brown K, Zamora J, et al. “Pulse oximetry screening for critical congenital heart defects in asymptomatic newborn babies: a systematic review and meta-analysis.” The Lancet 2012;379:2459-64