LAUSR

‘Temporal trends in routine predischarge pulse oximetry screening: 6 years’ experience in a UK regional neonatal unit’ is yet another important publication on the value of this newborn screening test by the team from Birmingham. They set clear goals to evaluate the impact of pulse oximetry screening (POS) in their neonatal unit and to compare the impact with their previous 3year experience. Additionally, they aimed to address the concerns of the National Screening Committee (NSC), which has not yet recommended routine use of POS in the UK. At the Birmingham Women’s Hospital (BWH), POS continues to detect newborns with critical congenital heart disease (CCHD) not detected by antenatal testing prior to discharge from the hospital. Additionally, they have had no newborns suffer cardiovascular collapse from CCHD or other conditions detected by POS on their postnatal ward prior to discharge. They have also shown that with experience, improvements in POS are possible. Since their first series, the number of positive screens in healthy admissions (no disease) has decreased from 21% to 2.4% of the total test positive newborns and the CCHD detection rate has increased from 94% to 98.7%. Our team recently published a manuscript demonstrating the value of POS at Holy Cross Hospital (HCH), a community hospital in Maryland. We showed similar results in detecting newborns with CCHD before discharge from the postnatal ward, but differences in the screening algorithm resulted in different rates of test positive cases and a different spectrum of diagnoses. The ‘best’ algorithm for POS is not known. The current algorithm endorsed by the American Academy of Pediatrics (AAP) and in use in much of the USA was a compromise between the perceived strengths and weaknesses of the algorithm recommended by Granelli et al and that used in the BWH. A closer look at differences between the algorithms and the recent recommended changes to the US algorithm provides insight into factors to consider when selecting an algorithm for implementation. The BWH algorithm is similar to the AAPendorsed HCH algorithm in that both require screening both the right hand and foot. They differ in the cutoff percentage saturation for a pass, percentage saturation difference between the upper and lower extremity, in the number of rescreens and screening age (table 1). With a higher threshold to pass (both extremities >95%), tighter tolerated saturation difference between extremities, fewer rescreens and earlier screening time, the rate of positive screens is much higher in the BWH (0.5%) algorithm than that used in the HCH (0.05%). The higher rate of positive screens in the BWH study may also be influenced by the inclusion of newborns as young as 34 weeks’ gestation. The BWH definitions of heart disease are also different than those used in the USA; they separate out some conditions defined as ‘critical’ in the USA and apply the term ‘serious’ if they require an operation after 30 days. In the USA, we do not separately define newborns detected at birth but who are electively operated on after 30 days. The BWH algorithm detected 17 newborns (8 critical and 9 serious) with major CHD and 5 with significant CHD out of 49 375 babies tested; whereas, the HCH study detected fewer babies, 12 with CCHD and 9 with noncritical CHD out of 64 780 babies tested. This may be due to the later testing time and that perhaps some babies are diagnosed after symptoms develop before POS is performed at 24 hours of life. Both studies identified additional newborns with secondary conditions, such as respiratory problems, persistent pulmonary hypertension of the newborn and sepsis (table 2). The BWH study had a higher proportion of newborns identified with noncardiac disease (86.9%) than the HCH study (32%). This also may be due to the earlier time in which POS was performed or to differences in the prenatal detection of CCHD in the BWH population. The concerns raised by the NSC with respect to POS for CCHD seem insignificant when compared with the benefits that have been shown in this study. The investigators in the BWH cited NSC concerns with cost, insufficient evidence of improved outcomes and the potential harms of unnecessary investigations and admissions, longer hospital stay and parental anxiety. The cost issue has been previously addressed and POS has been found to be costeffective, more costeffective than the screening for some of the rarer metabolic conditions. Improved outcomes from POS have been inferred by the study in the USA which showed a 33% reduction in mortality from CCHD in states with mandatory POS. It is hard to argue that earlier diagnosis before cardiovascular collapse would not improve outcomes when the converse, newborns with cardiovascular collapse prior to surgical correction are known to have poorer outcomes. Certainly, earlier diagnosis cannot worsen outcomes. Both the BWH and HCH studies dispel any concerns about unnecessary investigations and admissions. Only 2% of the BWH newborns with positive studies did not have some cause of abnormally low oxygen saturation and the others had either CHD or important respiratory, transitional circulation or infections requiring treatment. Parental anxiety has not been shown to be an issue with POS. In the USA, parent advocacy became a significant contributor to the approval of the advisory committee on newborn screening.