The current pulse-oximetry screening (POS) protocol for detection of critical congenital heart defects (CCHDs) is recommended only for newborns in well-infant and intermediate care nurseries, and there is no evidence-based protocol for infants discharged from the neonatal intensive care unit (NICU). The objectives of this study were to examine the efficacy of the current screening protocol in a NICU setting and to determine the impact of a unit protocol on the use of POS. Charts of 250 infants previous (group 1) and 250 infants after (group 2) the protocol implementation were reviewed. The results of screening test and preductal and postductal SpO₂ were recorded for screened infants. A predischarge SpO₂ value was recorded if screening was not performed. No infant in group 1 had POS. All eligible infants in group 2 received screening and passed. Preductal and postductal oxygen saturations in preterm infants at discharge were similar to saturations in late preterm and term infants. These results show that oxygen saturations at discharge in late preterm and term infants requiring admission to the NICU are similar to infants with no morbidities and that the current POS protocol can be safely used for these infants at discharge from the NICU. This study also confirms that preductal and postductal oxygen saturations at discharge in preterm infants are not different from those in late preterm and term infants. A unit protocol is likely to be more effective than relying on individual providers to ensure that all infants undergo POS for detection of CCHD.