© 2005 American Thoracic Society doi: 10.1164/rccm.2509002
Reevaluating Neonatal Resuscitation with 100% OxygenRainbow Babies and Children's Hospital Case Western Reserve University School of Medicine Cleveland, Ohio
University of Alabama at Birmingham Birmingham, Alabama The successful transition from fetal to neonatal life is a truly remarkable cascade of interrelated cardiorespiratory events. This period includes the onset of continuous respiratory efforts, replacement of lung fluid with air, and circulatory changes, including pulmonary vasodilation and closure of the ductus arteriosus and other fetal shunts. Typically, the transition is smooth without need for active intervention. However, not infrequently, skilled resuscitation is called on and fetal cardiorespiratory depression needs to be rapidly reversed. Fortunately, the pediatric community has risen to the challenge with multiple revisions and translations of the widely accepted Neonatal Resuscitation Program and the Texbook of Neonatal Resuscitation published by the American Academy of Pediatrics/American Heart Association (1). Traditionally, neonatal resuscitation has been performed with 100% oxygen. This recommendation was never evidence-based but seemed logical as the goal was to rapidly reverse fetal hypoxemia and acidosis and their consequences, notably pulmonary vasoconstriction. Over the last 15 years, the benefits of reoxygenation with 100% oxygen began to be challenged by Solås and his colleagues from Norway (2). Using both global hypoxemia and combined hypoxemia-ischemia models of perinatal asphyxia in the piglet, they have shown improved restoration of systemic and cerebral cortical perfusion with 100% oxygen versus room air resuscitation. However, this appears to occur at the price of greater biochemical oxidative stress in response to hyperoxic exposure, consistent with the widespread belief that hyperoxia during reoxygenation after asphyxia may introduce toxic free oxygen radicals and aggravate a preexisting problem (3). Of course, the piglet, like all animal models, has its limitations, including relative developmental maturity at birth. We have begun to explore hyperoxia-exposed rat pups as an alternate model for testing resuscitation strategies, as this animal is relatively immature at birth and is well suited for assessment of longer term neurobehavioral effects (4). Studies in human infants have been done concurrently, and these are clearly essential to resolve the dilemma of optimal neonatal resuscitation. A multicenter study enrolled over 600 predominantly term infants to receive either 21 or 100% oxygen during resuscitation, and observed no detrimental short- or long-term effects of 21% oxygen exposure (5, 6). Interestingly, a majority of infants enrolled in this study were from developing countries in which oxygen may be unavailable as a supplement to bag and mask ventilation. Meta-analyses of trials of room-air resuscitation also report no detrimental effects of room-air resuscitation. Furthermore, a reduction in mortality was observed in infants resuscitated with room air (7). However, because of the relatively small number of trials and their limitations, there should be caution in interpreting and applying these results. The other major contributors to this field have been Vento and associates from Spain (8). They have clearly demonstrated delayed onset of respiration in 100% oxygen versus room-air resuscitated infants, consistent with inhibition of peripheral chemoreceptors during hyperoxia. The hyperoxic infants also exhibited evidence of greater oxidative damage as reflected in a lower ratio of reduced to oxidized glutathione, which persisted over the first 28 days of life (9). In their current article in this issue of the Journal (pp. 1393–1398), Vento and associates substantially expand on their prior work by measuring biochemical markers of cardiac and renal tissue injury in response to 100% versus room-air resuscitation of asphyxiated term infants. Plasma cardiac troponin T and urinary N-acetyl-glucosaminidase measurements were both more elevated after hyperoxic versus normoxic resuscitation, suggestive of greater cardiac and renal injury, respectively (10). The authors attribute the former to the action of free radicals on a reperfusing myocardium, whereas the latter marker of renal tubular cell necrosis remained elevated for a longer duration after 100% resuscitation. So, where do we go from here? The current issue of the Journal provides compelling additional data from Vento and colleagues for a potential detrimental effect of hyperoxic resuscitation on oxidant-induced cardiac and renal injury (10). We already know that delayed onset of respiratory efforts is a consequence of 100% oxygen resuscitation. It is, therefore, appropriate to challenge the standard practice of routine neonatal resuscitation with 100% oxygen for term infants. Nonetheless, we do believe that further study, using both infants and animal models is warranted. Such efforts need to move away from the two extremes, 100 and 21% oxygen, if we are to determine the optimal method for administration of oxygen to these infants (11).
It has been suggested that a classic randomized trial evaluating a small effect size of reduction in mortality or brain damage from 24 to 21%, with a power of 80% and a two-sided FOOTNOTES Conflict of Interest Statement: None of the authors have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
of 0.05, would require 7,000 asphyxiated neonates (