Critical congenital heart defects (CCHDs) affect around 2 in 1 000 newborns, and babies with CCHD often have low blood oxygen concentrations, which can be measured by non-invasive pulse oximetry (PO) using a sensor placed on the newborn's hand or foot (Plana et al, 2018). The earliest reports of PO screening were published in 1995 (Valmari, 2007) and a Cochrane review of 436 758 participants from 19 studies showed that PO screening is highly specific, moderately sensitive and has a low false-positive rate, prompting the conclusion that ‘[c]urrent evidence supports the introduction of routine screening for CCHD in asymptomatic newborns before discharge from the well-baby nursery’ (Plana et al, 2018).
Yet a Public Health England ([PHE], 2019) patient information leaflet – last updated on 5 September 2019 – makes no mention of PO. An explanation is provided by the UK National Screening Committee (NSC), which advises ministers and the NHS in the four UK countries on population screening and helps to implement screening programmes.
According to a November 2019 consultation report, the UK NSC's (2019a) February meeting determined ‘that there was insufficient evidence on the evidence of benefits to outweigh the harms associated with screening, and insufficient evidence to demonstrate cost-effectiveness’. In December 2019, the UK NSC (2019b) noted that it has been considering screening all babies with PO since 2012; that around 40% of units use PO and 60% do not use PO; and ‘the UK NSC agreed that further research is needed before they can make a recommendation on the use of PO as an extra test in the newborn and infant physical exam (NIPE)’. Among the UK NSC's (2019a) concerns were false positive results, and that a positive PO result would generate parental anxiety and more tests for non-symptomatic conditions; a longer hospital stay; and possible transfer to the neonatal unit (NNU).
Responding, Ewer et al (2019) reported that out of 32 597 babies they screened, 114 (0.35%) who tested positive were admitted to NNU, of which eight had a CCHD. However, 82 of the babies admitted to NNU (72% of the total admitted) had a significant non-cardiac illness. Although strictly speaking, false positives they were not trivial conditions: congenital pneumonia, persistent pulmonary hypertension of the newborn, culture positive and culture negative sepsis, meconium aspiration, pneumothorax, transient tachypnoea of the newborn and respiratory distress syndrome, all of which needed treatment, and ‘only 22 babies admitted to NNU (0.07% of all babies screened) were healthy’ (Ewer et al, 2019). It seems reasonable for Ewer et al (2019) to urge the UK NSC ‘to review their decision not to introduce routine newborn PO screening for CCHD in light of our conclusions.’
Dr Katherine Wood, UK Neonatal Grid Trainee Group representative, reported the unanimous opinion of trainees that PO screening should be included in the NIPE, adding: ‘False positive results have been raised in the NSC report as one of the reasons for deciding against PO screening. False positive results are commonly due to non-cardiac conditions, but the majority of these conditions also benefit from early detection and treatment … we strongly believe that no baby should be sent home with hypoxaemia … [and] the concern that the NSC (2019a) has about increasing parental anxiety is unfounded’.
Similarly, in August 2019, the Congenital Cardiac Nurses felt that the UK NSC's (2019a) decision would disadvantage babies throughout the UK: ‘In our daily work, we see the consequences of late diagnosis and we feel the addition of PO will help reduce this and improve outcomes for these babies’.
A dilemma arises from those on one side who see the practical benefit of PO, and on the other side the UK NSC's demand for greater evidential stringency before a firm decision can be made. This has the makings of a moral dilemma with the PO intervention associated with both good and a perceived harm for a newborn. Utilitarianism – the greatest benefit for the greatest number – seems an attractive basis for resolution, but in the resource-limited context of the NHS to what extent can doing what is morally right take precedence over what is cost-effective and wholly evidence-based?