Objective: To evaluate the predictive abilities of pulse oximetry screening (POS) for critical congenital heart disease (CRIT.CHD) at two different hospital settings in Thailand.
Study design: The study was conducted in healthy newborns at Ramathibodi Hospital (RH), a university hospital and Maharat Nakhon Ratchasima Hospital (MH), a regional hospital. Positive POS was defined as oxygen saturation (SpO2) <95% or difference between pre- and postductal SpO2 >3%.
Results: Of 11 407 live births, 10 603 (92.9%) newborns were enrolled with a follow-up rate at 1 month of 78.3%. Incidence of CRIT.CHD (per 1000 live births) at RH and MH were 5.7 and 2.7, respectively. POS could detect three newborns who would have had a missed diagnosis. Sensitivity of POS for CRIT.CHD at RH was 82.3% vs 100% at MH. Overall specificity was 99.9% and false-positive rate was 0.009%. Combination of POS and physical examination (PE) enhanced detection ability to 100% at both hospitals.
Conclusion: POS combined with PE improved detection of CRIT.CHD. Routine POS is useful in personnel-limited settings.