The Need to Assess Benefits and Not Just Risks of 100% Oxygen for Newborn Resuscitation: In Reply

Abstract

Rozycki's letter regarding our recent commentary1 brings to attention some important aspects that should be considered when we discuss which oxygen concentration should be used for newborn resuscitation. Rozycki is completely right when he points out that most of the ∼1800 newborn children that up to now have been included in the 6 randomized or pseudorandomized studies performed in this field come from countries with low income. However, we have to consider the fact that most asphyxiated newborn infants are born in low-income countries. Therefore, we think that the material is relevant. A separate analysis of the children recruited from Spain, a country with significantly lower perinatal mortality than the United States, also shows a significant reduced mortality in newborns who are resuscitated with 21% compared with 100% O2.2 It has already been 13 years since our pilot study3 and 8 years since the large international multicenter study Resair 24 were published. There has been ample time for other investigators from, for instance, the United States to set up their own studies. Why this has not been done we do not know.The increased oxidative stress we have found in infants resuscitated with 100% O2 has immediate consequences as seen in heart and kidney.5,6 It is reasonable that these changes and similar changes in other organs such as the brain, adrenal glands, etc, so far not investigated clinically, could contribute to increased morbidity and mortality. We do not know at present if there will be any long-term consequences on organ function; however, the burden of proof that negative short-term and long-term effects do not appear should be on those, such as American Academy of Pediatrics, who promote the use of 100% O2.7Regarding the 30-second delay in response to resuscitation in infants exposed to 100% O2, this has been confirmed in several studies.2 We believe that such a delay may have several important consequences. Confronted with a lack of response, it may increase the aggressiveness of the resuscitation team. Because the new guidelines give only 30 seconds to move from one stage of resuscitation to the next, a 30-second delay in time to first breath may result in more children going through advanced cardiopulmonary resuscitation as chest compressions and drug administration. Therefore, this delay, in our opinion, is not an innocent byproduct of the hyperoxic therapy.We agree that the 2 studies demonstrating a significant association between brief oxygen exposure at birth and childhood cancer8,9 should be interpreted with caution, and we have done that. Still, the finding is interesting and should be concerning to everyone interested in this topic.We totally agree with Rozycki that follow-up studies are needed. The only one performed thus far is our own, which showed that surviving newborn infants resuscitated with room air at birth are doing as well at 18 to 24 months of age as those who were resuscitated with 100% O2.10 We, as well as others, have emphasized the need for more such follow-up studies.We will also underline that we have never promoted the use of room air for newborn resuscitation, and we are open to the possibility that some term or near-term newborns may need supplemental O2 (even 100%) for a short period of time. However, we are strongly convinced that the routine use of 100% O2 is detrimental. The data showing this are in fact so convincing that the Canadian Paediatric Society11 and the Swedish Neonatal Society (Mats Blennow, MD, PhD, Swedish Neonatal Society, written communication, 2006), for instance, have decided that newborn resuscitation should begin with room air. Also, US and other investigators recommend beginning newborn resuscitation with room air.12–14