Published online January 2, 2007
PEDIATRICS Vol. 119 No. 1 January 2007, pp. 217 (doi:10.1542/peds.2006-2443)

This Article Extract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My File Cabinet Download to citation manager Request Permissions Citing Articles Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Rozycki, H. J. Search for Related Content PubMed PubMed Citation Articles by Rozycki, H. J. Related Collections Premature & Newborn Social Bookmarking                         What's this?

LETTER TO THE EDITOR

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

Henry J. Rozycki, MD
Neonatal-Perinatal Medicine
Virginia Commonwealth University
Richmond, VA 23298-0276

To the Editor.—

The commentary "Oxygen for Newborn Resuscitation: How Much Is Enough?" by Saugstad et al¹ provides a compendium of what these authors, in particular, have studied and published over the last 15 years. To their credit, they have diligently brought scientific examination to what has been the long-standing but unjustified practice of delivering 100% oxygen to asphyxiated newborns. Although I have a few concerns about their interpretation of the risk side of the risk/benefit equation, of greater concern is the absence of any information in their article about potential benefits. This makes it difficult to fully and fairly assess the practice.

The first piece of negative evidence cited by the authors is the reduction in neonatal mortality in room-air–resuscitated infants.² The vast majority of subjects in the studies included in the meta-analysis were from countries with significantly higher neonatal mortality rates. This would mean that the number needed to treat to demonstrate an equal reduction in the mortality rate would be much higher in countries with lower baseline mortality. Next, the effects of oxygen resuscitation on markers of oxidative stress and on biochemical markers of heart and kidney injury,³ although provocative, do not correlate with short-term or long-term effects on organ function or other morbidities. Regarding the delay in response to resuscitative efforts in the hyperoxia-exposed infants,^1,4 Kattwinkel⁵ pointed out that this may be an innocent byproduct of the therapy in which hyperoxia suppressed the infant's own drive.

The link between childhood cancer and oxygen exposure was demonstrated by a case-control study from Sweden.⁶ Retrospective analysis of the participants in the Collaborative Perinatal Project from 1959 to 1966 also showed a significant but small increased risk for childhood cancers after the first year of life if oxygen duration was >3 minutes, after adjustment for low 5-minute Apgar scores, an association the investigators themselves called "weak" and "ambiguous."⁷ In addition, there was a narrow window for oxygen exposure as a risk, >3 minutes but not, according to the Swedish study, after more prolonged postnatal oxygen therapy. Generally, findings from case-control studies and retrospective analyses of old databases can demonstrate association, but they are not as strong at defining a causal relationship.

Even without those caveats, to decide if these risks outweigh the benefits, one has to examine the other side of the equation, and that is what is missing from the Saugstad et al commentary. None of the referenced studies specifically addressed developmental outcome. The most extensive follow-up study examined a little over 40% of the original cohort and, with only 213 subjects, was not large enough to determine if oxygen provides any long-term benefit to asphyxiated infants and, if so, to what degree.⁸ If the purpose of using oxygen (at any fraction of inspired oxygen >0.21) at delivery is to prevent significant hypoxic brain injury, then information on the developmental follow-up of subjects from randomized, blinded clinical trails is necessary. Only with that kind of data can we properly assess the risks documented in the commentary within the complete scientific context.

REFERENCES

1. Saugstad OD, Ramji S, Vento M. Oxygen for newborn resuscitation: how much is enough? Pediatrics. 2006;118 :789 –792[Free Full Text]
2. Saugstad OD, Ramji S, Vento M. Resuscitation of depressed newborn infants with ambient air or pure oxygen: a meta-analysis. Biol Neonate. 2005;87 :27 –34[CrossRef][Web of Science][Medline]
3. Vento M, Sastre J, Asensi MA, Vina J. Room-air resuscitation causes less damage to heart and kidney than 100% oxygen. Am J Respir Crit Care Med. 2005;172 :1393 –1398[Abstract/Free Full Text]
4. Vento M, Asensi M, Sastre J, Lloret A, Garcia-Sala F, Vina J. Oxidative stress in asphyxiated term infants resuscitated with 100% oxygen [published correction appears in J Pediatr. 2003;142:616]. J Pediatr. 2003;142 :240 –246[CrossRef][Web of Science][Medline]
5. Kattwinkel J. Evaluating resuscitation practices on the basis of evidence: the findings at first glance may seem illogical. J Pediatr. 2003;142 :221 –222[CrossRef][Web of Science][Medline]
6. Naumberg E, Bellocco R, Cnattingious S, Jonzon A, Ekbom A. Supplementary oxygen and risk of childhood lymphatic leukemia. Acta Paediatr. 2002;91 :1328 –1333[CrossRef][Web of Science][Medline]
7. Spector LG, Klebanoff MA, Freusner JH, Georgieff MK, Ross JA. Childhood cancer following neonatal oxygen supplementation. J Pediatr. 2005;147 :27 –31[CrossRef][Web of Science][Medline]
8. Saugstad OD, Ramji S, Irani SF, et al. Resuscitation of newborn infants with 21% or 100% oxygen: follow-up at 18 to 24 months. Pediatrics. 2003;112 :296 –300[Abstract/Free Full Text]

---

PEDIATRICS (ISSN 1098-4275). ©2007 by the American Academy of Pediatrics

CiteULike    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

This Article Extract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My File Cabinet Download to citation manager Request Permissions Citing Articles Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Rozycki, H. J. Search for Related Content PubMed PubMed Citation Articles by Rozycki, H. J. Related Collections Premature & Newborn Social Bookmarking                         What's this?