Pulse oximetry screening of newborns: Not only a screening test for congenital heart disease


  • Gary F Sholler

    Corresponding author
    1. SCHN Cardiac Service/Heart Centre for Children, Children's Hospital at Westmead, Sydney Children's Hospitals Network, Sydney, New South Wales, Australia
    2. Sydney Medical School, University of Sydney, Sydney, New South Wales, Australia
    • Correspondence: Associate Professor Gary F Sholler, Heart Centre for Children, Children's Hospital at Westmead, Locked Bag 4001, Westmead, NSW 2145, Australia. Fax: 61 2 9845 2163; email: gary.sholler@health.nsw.gov.au

    Search for more papers by this author

  • Conflict of interest: The author has no conflict of interest; however, he assisted in accessing data for the paper of Bhola, Kluckow and Evans.

A strong argument is being made internationally for pulse oximetry newborn screening to be introduced for early diagnosis of major congenital heart disease (mCHD) in the newborn; however, despite its importance, this may be too narrow a perspective. Pulse oximetry can identify newborns with some (but not all) forms of mCHD.[1-3] Early diagnosis also reduces morbidity and mortality in some mCHD lesion groups.[4] Several studies have now confirmed that screening patients in their newborn populations identifies mCHD prior to the onset of symptoms in some. The benefits of early diagnosis are already well demonstrated in the literature reviewing outcomes after prenatal detection of mCHD with obstetric anomaly screening and fetal cardiology input.[4, 5]

The American Academy of Pediatrics have supported pulse oximetry screening for the identification of mCHD,[3] and the US Department of Health and Human Services has recommended oximetry screening and tasked the CDC and NIH with responsibilities related to this. The CDC[6] recommends that after a confirmed positive screening result, ‘any infant with a positive screen should have a diagnostic echocardiogram, which would involve an echocardiogram within the hospital or birthing center [sic], transport to another institution for the procedure, or use of telemedicine for remote evaluation'. This care cascade, also proposed in other publications, does not formally engage medical assessment and triage prior to echocardiography.

Other factors will influence the applicability, efficacy and identification rate of a pulse oximetry screening programme. Locally, the likelihood of fetal diagnosis of major neonatal cardiac lesions exceeds 60% in many Australian states,[4, 7] reducing the chance of an undiagnosed newborn presentation of mCHD. The case identification rate in most reported series suggests that the additional load for paediatric cardiology programmes is small.[8, 9] The identification rate of mCHD in Bhola et al.'s paper[9] is 0.02% and consistent with local prenatal detection rates of major lesions. Further, to satisfy the Wilson and Jungner criteria[10] for screening for other countries and communities, practical considerations will come into play. These include the distances and resources involved in transfer after a positive finding, the availability of reliable oximeters, expert neonatal cardiac imaging, cardiac assessment, and in some countries with limited resources, downstream treatment options.

Families value the confidence that a major abnormality may be identified early; however, for those babies with false positives and possible need for relocation for additional assessment, this may be associated with important issues of parental anxiety and consequential impacts on bonding.[4, 11, 12] It is also imperative that programmes involved with oximetry screening make it clear to parents and professionals that a negative test does not rule out all serious illness. For example, some cardiac lesions may not be associated with early desaturation (e.g. tetralogy of Fallot) or may not be associated with desaturation at all (e.g. later onset coarctation of the aorta, significant but not critical aortic stenosis, ventricular septal defect, etc). This clarity should minimise the risk of late presentation of heart abnormality based on parental misunderstanding regarding the certainty of oximetry screening.

The paper by Bhola et al.'s[9] not only offers further confirmation of the efficacy of newborn pulse oximetry screening but also offers a variation on some published triage paths and diagnostic opportunities. Importantly, the authors highlight the benefits of early diagnosis of significant, non-cardiac illness and propose that a positive result is not an immediate trigger for echocardiography but rather should prompt careful, expert medical evaluation. Expanding the longstanding strategies of vital signs assessment for newborns (pulse, respiration, temperature) to include oximetry could be seen more as an augmented ‘well-being evaluation’, with the associated significant benefit of identifying some forms of mCHD. The authors also recommend clinical review as part of the triage process, which should allow for early and appropriate initiation of treatment for non-cardiac disease and should minimise unnecessary patient transfers. Whether or not oximetry should be deemed screening or part of augmented vital sign evaluation remains an open question, and in either case confers benefit. If it is considered in the context of screening, the identification of all illnesses (cardiac and non-cardiac) in Bhola et al.'s[9] is 0.08% and approximates the projected rate of identification of about 0.1% in current NHMRC Australian newborn screening.[13] Irrespective of the institution of statewide or national screening, several large Australian obstetric centres have already developed neonatal oximetry programmes.[9] Caution will also be needed, should screening be introduced, to avoid resource deployment away from important pre-natal diagnosis strategies.

Adding oximetry as an additional means of expanded vital sign evaluation of newborn well-being is sensible and will identify some cases of newborn mCHD prior to decompensation. Recognising oximetry to be more than only screening for mCHD permits more unwell newborns with other conditions to benefit from early treatment and ameliorates the impact for families carried by the stigma of oximetry being a cardiac investigation, while offering the same benefits. Careful evaluation of the costs, manpower, expertise and tools is still required before wholesale Australian screening can be implemented; however, collaborative development region-appropriate protocols is important and should involve relevant medical and nursing disciplines, and government public health agencies. Tertiary cardiac programmes, especially given local Australasian prenatal detection rates for mCHD, are not likely to see a major impact but will see a substitution of a small additional inflow of stable newborns with cardiac abnormality rather than unstable babies with late recognition of their lesion.