Since 1987, the NRP has recommended the use of 100% oxygen during all resuscitations. Empiric evidence as well as data from studies of perinatal physiology suggested that 100% oxygen would facilitate the postnatal fall in pulmonary artery pressure, thus assisting with the transition from fetal to newborn circulation in those newly-born infants who manifested signs of abnormal transition after delivery. However, some investigators have hypothesized that the use of 100% oxygen could increase oxidative stress in an already compromised infant and, therefore, could exacerbate post-asphyxial injury. Over the past 15 years, a number of animal studies and several clinical trials^[15,19,23] have examined the question of whether newborns could be successfully resuscitated after a perinatal insult by using 21% oxygen rather than 100% oxygen and, further, whether use of 21% oxygen might even improve outcomes after resuscitation.

Studies in newborn animal models (LOE 6) have, in general, found that the rates of successful resuscitation from an imposed asphyxial insult were equivalent regardless of the FiO2 used for resuscitation. Investigation of short-term outcomes (most <7 days) in most cases did demonstrate increases in biochemical markers of oxidative stress as well as alterations in neurotransmitter release and cerebral blood flow response in those animals that were resuscitated with 100% oxygen^[7,8,16,17]. Conversely, several studies found potentially adverse short-term biochemical changes in animals resuscitated with room air compared to those in which 100% oxygen was used^[20,21].

Extrapolation of the results of these animal studies to the clinical situation is complicated by a number of limitations. First, several different models were used to mimic changes associated with perinatal asphyxia. Second, no animal studies have been conducted to examine long-term outcomes, i.e., at weeks to months of age. Perhaps most important, virtually all of the studies were performed in animals that were several hours old or older, and therefore had already successfully transitioned to extrauterine life before being subjected to the insult. Thus, the question of whether the use of 100% oxygen has particular benefits when pulmonary artery pressures are still high, as they are just after birth, was not addressed by these studies. However, based on the success of resuscitation with 21% oxygen in animals, several clinical trials were carried out examining the use of 21% oxygen for resuscitation in newly born infants.

In 1998, Saugstad et al reported on a total of 609 newly born infants (from 10 centers) with asphyxia (defined as apnea or gasping at birth with heart rate <80 bpm), who were resuscitated with 21% oxygen on even dates and 100% oxygen on odd dates (not a true randomization procedure). Infants who did not respond after 90 seconds of resuscitation were switched over from the initial gas to the “other” gas. The primary outcome measure was death within the first 7 days and/or diagnosis of hypoxic-ischemic encephalopathy. A number of secondary outcomes were also investigated, including Apgar score at 5 minutes, heart rate at 90 seconds, and duration of resuscitation. The investigators found no significant difference in the primary outcome measure between the two groups; in addition, no differences were found in heart rate at 60 or 90 seconds after delivery or median 5-minute Apgar scores, or median time to first breath. However, infants resuscitated using 21% oxygen had a significantly shorter median time to first breath. Of note is the fact that equal numbers of infants were switched from room air to oxygen and from oxygen to room air due to perceived lack of response (LOE 2).

In the 2001 Vento et al report, term neonates with “perinatal asphyxia” were randomized to be resuscitated with either 21% or 100% oxygen; caregivers were blinded with regard to the identity of the resuscitation gas. Primary outcomes were Apgar scores at 1, 5, and 10 minutes, time to first cry, and time to regular respiration. In addition, a number of indices of oxidative stress were measured in blood at delivery, 72 hours postnatal age, and 28 days postnatal age. The authors found no differences in Apgar scores between the two groups. Time to first cry and time to sustained respiratory pattern were significantly shorter in the group resuscitated with 21% oxygen. The reduced-to-oxidized glutathione ratio at 28 days was no different from control in the 21% group, while it was significantly lower in the 100% oxygen group, suggesting that the latter group had experienced sustained oxidative stress (LOE 2).

In these studies, as well as several others^[15,24] (LOE 2) comparing the effects of using 21% and 100% oxygenation for resuscitation in the delivery room, no significant difference in mortality rates was observed between the two treatments. In addition, on follow-up at 18-24 months postnatal age^[18], Saugstad et al found no difference in outcomes. Two meta-analyses of the same five clinical trials^[5,22] (LOE 7) found minimal differences in outcome measures between the two treatment groups; however, in the 2004 paper^[5] the authors concluded that mortality rate was significantly decreased in the 21% oxygen group, while the 2005 paper^[22] concluded that there were insufficient data to recommend the use of either 21% or 100% oxygen for resuscitation of infants with perinatal asphyxia.

Overall, the results of the clinical studies suggest that resuscitation with 21% oxygen is as effective as, and might even be superior to, resuscitation with 100% oxygen. However, unresolved issues regarding study design, subject enrollment and randomization, diagnoses in those infants who died, and the number of severely asphyxiated infants included in each study group confounded interpretation of the clinical trials. As a result, the NRP recommendation that 100% oxygen be used when positive-pressure ventilation is required or in the presence of central cyanosis was not changed. Because the evidence suggests that 21% may be as effective as 100% oxygen, however, the 2005 guidelines also state that individual providers may take the option of starting with less than 100% oxygen. No evidence is available to support a recommendation for using a specific oxygen concentration between 21% and 100% at present. If an oxygen concentration <100% is used to initiate resuscitation, crossover to 100% oxygen is recommended if there is no improvement in 90 seconds. This time point was chosen because it was used in the clinical trials that included a crossover design^[15,23]; however, there have been no studies to determine the optimum timing for changing the oxygen concentration during resuscitation if an appropriate response (i.e., increasing heart rate) is not observed with the initial oxygen concentration.

Previously, the recommended dose of epinephrine was 0.1 to 0.3 ml/kg of a 1:10,000 solution (0.01 to 0.03 mg/kg), given via either the endotracheal tube or intravenous access, typically via an umbilical venous catheter. Because endotracheal access is generally available before intravenous access during resuscitation, the endotracheal route was recommended for initial doses. However, the 2005 Consensus on Science was that published studies in adult cardiac arrest patients^[13,14] found that endotracheal administration of epinephrine was of little or no benefit in restoring cardiac function. In addition, both animal^[4,10-12] (LOE 6) and human[9] (LOE 2) studies demonstrated that endotracheal epinephrine administration was associated with increased blood epinephrine concentrations and/or an effect on cardiovascular variables only when given at approximately 10 times the previously recommended dose of 0.01-0.03 mg/kg.

In 1991, using an adult pig model of ventricular fibrillation followed by external cardiac compressions, Crespo et al tested the effects of two different doses of epinephrine given via the endotracheal tube on plasma epinephrine levels and blood pressure. Animals that received 0.01 mg/kg (i.e., the previously recommended dose) via the endotracheal tube had no increase in plasma epinephrine levels. Animals treated with 0.1 mg/kg (i.e., 10x the previously recommended dose) had significantly higher plasma epinephrine levels than pigs who received no exogenous epinephrine or 0.01 mg/kg; however, the investigators found no significant increase in blood pressure associated with the higher epinephrine levels.

In Kleinman & Stonestreet’s 1991 study, radiolabeled epinephrine (0.01 mg/kg) or placebo was administered to newborn piglets via intracardiac, intravenous, or endotracheal route during resuscitation after ventricular fibrillation-induced cardiac arrest. The authors reported that intravascular epinephrine administration was associated with increased plasma epinephrine levels and increased carotid blood pressure, but that endotracheal administration was not associated with an increase in either epinephrine level or blood pressure.

In 1985, using an adult dog model of cardiac arrest due to electromechanical dissociation, Ralston et al determined dose-response curves for intravenous and intratracheal epinephrine. Outcomes did not differ significantly between the animals assigned to the two routes of administration, but the median effective dose of endotracheal epinephrine was approximately 10 times higher than the median effective dose for intravenous epinephrine (0.13 mg/kg compared to 0.014 mg/kg, respectively).

As is apparent from these studies and others, there are no data from randomized clinical studies in newly born infants. Most of the animal studies must also be interpreted cautiously because they were carried out in older animals and/or used a model, such as ventricular fibrillation, that produces cardiac arrest via a mechanism not usually encountered in the newborn. The few studies in newborn animals used animals that were already being ventilated and oxygenated or were hypoxic but not in frank cardiac arrest. In spite of these limitations, the consensus reached by the members of the neonatal ILCOR was that, given the lack of data to support the efficacy of endotracheal administration of epinephrine during resuscitation in newly born infants, the intravenous route is the optimal route of administration. Because there are limited data suggesting that doses of epinephrine 10 times higher than the intravenous dose might be effective if given via the endotracheal tube, the administration of one dose of epinephrine via the endotracheal tube at a higher dose (up to 0.1 mg/kg) than the intravenous dose could be considered while intravenous access is being obtained. However, there are no data that establish the safety or efficacy of this dose in newly born infants. Data in the adult and pediatric populations suggest that higher doses of intravenous epinephrine are associated with poorer outcomes after resuscitation; therefore, the use of epinephrine doses greater than 0.03 mg/kg IV is not recommended, and care must be taken to ensure that a dose of epinephrine prepared for endotracheal administration is not inadvertently administered via the intravenous route.

LAST MONTH’S Q & A    December 2006 - Volume 4 - Issue 4

In our December 2006 issue, we overviewed the development and revision of the NRP practice guidelines as published in the new AAP/AHA Neonatal Resuscitation Program textbook. Based on those data, the eNeonatal Review Team asked the December faculty a few questions:

Commentary & Reviews: Jane E. McGowan, M.D. Professor of Pediatrics Drexel University College of Medicine Philadelphia, PA

The eNeonatal Review Team asked the December faculty a few questions.
	Since Vain et al showed no decrease in the incidence of meconium aspiration syndrome after Intrapartum suctioning of the oropharynx, does this mean that obstetricians should never perform Intrapartum suctioning?
	The current recommendation states that “routine” Intrapartum suctioning is not recommended. There may be circumstances where there is material in the oropharynx that could cause mechanical obstruction during the baby’s first breaths, and brief suctioning with a bulb syringe may be useful in such cases. However, delaying delivery for prolonged suctioning and/ or suctioning vigorously enough to cause vagally-mediated bradycardia is not indicated whether or not meconium is present.
	If I do not choose to use 100% oxygen for resuscitation, what oxygen concentration should I start with?
	Unfortunately, there are no published data comparing the outcome of resuscitation using specific oxygen concentrations between 21% and 100%, although such studies are in progress. Thus, it is up to the individual practitioner to select an oxygen concentration that he/she deems appropriate for use during resuscitation of the newly-born infant. The most important aspect of resuscitation, regardless of the initial FiO2 used, is the ongoing evaluation of the response of the infant to resuscitative efforts. Thus, if the infant’s heart rate and/or respiratory effort do not improve after initiation of positive-pressure ventilation, regardless of oxygen concentration, further steps must be undertaken to achieve the desired response.
	Can I give more than one dose of endotracheal epinephrine during a resuscitation?
	As stated in the 5th edition of the Textbook of Neonatal Resuscitation, there are no studies demonstrating that administration of endotracheal administration of epinephrine at currently used doses has any effect on heart rate or blood pressure during neonatal resuscitation. Thus, if epinephrine administration is indicated, efforts should be directed at obtaining vascular access via one of several routes as described in the textbook. Obtaining vascular access should take priority over drawing up and administering multiple doses of epinephrine via the endotracheal tube as it is unlikely that subsequent doses will have any effect if there was no response to the initial dose.