Screening for congenital heart disease with newborn pulse oximetry – Authors' reply

Abstract

1 Ewer AK, Middleton LJ, Furmston AT, et al. Pulse oximetry screening for congenital heart defects in newborn infants (PulseOx): a test accuracy study. Lancet 2011; 378: 785–94. to intensive care before screening took place compared with none in our study. We confi rm that all babies in our study were asymptomatic at testing, and the individual case highlighted by OstmanSmith and de-Wahl Granelli was not receiving prostaglandin or oxygen. We disagree that it is impossible to have saturations of 100% and 97% with hypoplastic left heart syndrome. Notably, since the initial examination for this baby was normal, the second test result would have passed according to the protocol described by de-Wahl Granelli and colleagues. We cannot deduce the value of screening where physical examination precedes pulse oximetry from our study, since the results of physical examination are likely to have been aff ected by the prior knowledge of the saturation result. We disagree that all cyanosed babies would be detected clinically, since the ability of clinicians to detect cyanosis is notably poor—in the non-screening regions in de-Wahl Granelli and colleagues’ study, 44% of babies with transposition of great arteries were discharged without diagnosis. Our study, combined with the evidence of others, supports the introduction of routine pulse-oximetry screening for neonates. Introduction of pulseoximetry screening will particularly improve detection rates in regions where antenatal diagnosis rates are lower. The optimum timing of screening requires careful consideration; delayed screening is likely to minimise false-positive rates, whereas earlier screening should improve timely diagnosis.