Routine screening of blood oxygen levels before discharge from hospital improves the detection of life threatening congenital heart disease in newborns and may save lives, according to a study published on bmj.com today.
The low false positive rate of pulse oximetry screening* and the reduced need for treatment because of a timely diagnosis also makes this a cost effective intervention, say the authors.
About 1 babies per 1000 live births have an immediately life threatening heart abnormality, because a fetal blood vessel called the ductus arteriosus - which bypasses the baby's non-functioning lungs when in the uterus and normally closes off soon after birth - remains partly open. The current screening technique of a routine clinical examination shortly after birth fails to detect many of these babies because duct-dependent heart disease often lacks heart murmurs. Indeed, 30% of such infants leave hospital without their condition being diagnosed, which leads to higher rates of complications such as circulatory collapse with organ damage and sometimes death.
Pulse oximetry screening has been suggested for early detection of congenital heart disease, but its effectiveness is unclear.
Professor Östman-Smith and colleagues assessed the introduction of universal oximetry screening in one region of Sweden (West Götaland) and examined the diagnostic accuracy for detection of duct dependent heart disease compared to other regions using clinical examination alone. Nearly 40,000 babies born between 1 July 2004 and 31 March 2007 were screened with a pulse oximeter before routine physical examination.
The authors found that in apparently well babies ready for discharge a combination of clinical examination and pulse oximetry screening had a detection rate of 82.8% for duct-dependent heart disease. The detection rate of physical examination alone was 62.5%. Pulse oximetry also had a substantially lower false positive rate (0.17%) compared to physical examination alone (1.90%).
However, some babies had been detected before discharge examination, meaning that the introduction of pulse oximetry screening in West Götaland improved the total detection of duct dependent heart disease to 92%. This was significantly higher than the 72% detection rate in other regions not using the screening technique. Thus the risk of leaving hospital with an undiagnosed duct dependent circulation was 8% in West Götaland versus 28% in the other regions.
Babies discharged from hospital without diagnosis had higher mortality than those diagnosed in hospital (18% v 0.9%). In addition, no babies died in West Götaland from undiagnosed heart disease, but there were five deaths in the other regions.
Interestingly, improved detection was achieved by a maximum of just five minutes of extra nursing time per baby.
The authors conclude: "Such screening seems cost neutral in the short term, but the probable prevention of neurological morbidity and reduced need for preoperative neonatal intensive care suggest that such screening will be cost effective long term."
In an accompanying editorial, Professor Keith Barrington from the University of Montreal in Canada, says that in light of this new evidence on the effectiveness of universal pulse oximetry screening as a low risk and low cost strategy for improving the detection of critical congenital heart disease, "serious consideration should be given to its introduction wherever neonatal cardiac surgery is available."