Oxygen for Cyanotic Neonates: Friend or Foe?

Oxygen toxicity was first described by Paul Bert in 1878. He demonstrated that high levels of oxygen are toxic to plants and insects, and to the CNS of worms, mollusks, birds and other animals. During World War II, thousands of experiments were performed on deep sea divers in the USA, which revealed unique toxicities of breathing high concentration oxygen. Even as these seminal studies were being conducted, oxygen was widely used for the care of babies, especially those born prematurely. The exuberant use of oxygen for infants revealed certain toxicities. Now it has been almost 60 years since the first controlled trial appeared suggesting that too much oxygen could damage the eyes of preterm infants [1], and more than 50 years since the first large landmark trial in neonatal medicine confirmed that although oxygen was a ‘good thing’ it was quite possible to have ‘too much of a good thing’ [2,3].

In recent years, research has supported more judicious use of oxygen when resuscitating newborn infants. There is growing evidence that exposure to high-concentration oxygen, even briefly, may be harmful [4]. More importantly, in both term and preterm infants there is a reduction in neonatal mortality when using room air versus 100% oxygen [5]. Although oxygen has been given to babies more than any other medicinal product in the last 70 years [3], it is remarkable how little we know regarding how much oxygen babies actually need, or how much is safe to administer. Furthermore, there are no evidence-based guidelines for appropriate oxygen saturation ranges to optimize neonatal outcomes.

This is especially true for the initial management of an infant with cyanosis of uncertain etiology. Typically, the initial and appropriate management of the cyanotic infant is to administer high-concentration oxygen and then investigate the cause of cyanosis. The most likely causes of profound cyanosis unresponsive to oxygen supplementation in infants are congenital heart disease (CHD) and persistent pulmonary hypertension of the newborn (PPHN). The first priority is to establish adequate tissue perfusion and oxygenation by ensuring optimal cardiac output and arterial oxygen content. This is critical to avoid hypoxic–ischemic neurological injury and metabolic acidosis.

The optimal management of CHD and PPHN with regard to oxygen administration can be distinctly unique. Supplemental oxygen may be detrimental for infants with complex cardiovascular shunts, critically obstructed systemic circulation, and single ventricle physiology because it may increase pulmonary blood flow at the expense of systemic and cardiac perfusion [6]. For infants with CHD, limiting pulmonary blood flow is usually accomplished by titrating assisted ventilation support and weaning supplemental oxygen to room air with targeted systemic oxygen saturations of 70–85% [6].

Serious CHD is the most common cause of congenital anomalies occurring in approximately 1.4 per 1000 live-births [7–9]. In a population-based study in the state of Alabama, USA, 23% of infants with serious CHD were not diagnosed prenatally or during the initial hospitalization after birth [10]. Other large observational studies report up to 55% missed diagnosis of CHD in the immediate newborn period [11,12]. Furthermore, as many as 25% of neonates who die in the first postnatal week from ductal-dependent heart defects are undiagnosed at the time of death [13]. The reality is that the presenting signs of CHD may be subtle and are commonly missed. CHD may not be suspected until there are overt signs of cardiopulmonary failure after birth.

While oxygen restriction makes physiological sense for a subset of infants with CHD, there are some cyanotic infants who may benefit from more liberal oxygen administration, particularly those with PPHN. CHD and PPHN cannot be reliably differentiated clinically [6]. The incidence of PPHN is 1.9 per 1000 live-births [14], which is very similar to that of CHD. PPHN occurs when there is increased pulmonary vascular resistance, when there are structural abnormalities of the vasculature, or when there is pulmonary hypoplasia. Clinically, PPHN can be seen in newborn infants with sepsis or pneumonia, with aspiration syndromes and with congenital diaphragmatic hernia. Oxygen, a potent pulmonary vasodilator, is a cornerstone of a comprehensive cardiopulmonary treatment strategy for PPHN [15].

A recent report suggests that when transporting infants with suspected CHD to a tertiary referral center, they may be cautiously weaned to room air to maintain saturations of more than 70% [16]. While this report presents encouraging data showing no differences in metabolic or lactic acidosis for infants in low-, moderate- or high-saturation groups during stabilization and transport, we must proceed cautiously before adopting their recommendation: “Oxygen should be cautiously weaned to room air in patients with a strong suspicion of CHD.” Indeed, this practice may be harmful to patients with profound cyanosis not caused by CHD.

While it may be appropriate to limit oxygen exposure for infants with known CHD, the best approach to the cyanotic infant of unknown etiology is uncertain. Because of the known toxicities of supplemental oxygen, even with brief exposure, designing randomized, controlled trials of lower versus higher oxygen saturation targets for ill infants with cyanosis should be a priority. Central to any study of oxygen saturation targets in infants are short-term physiological measures, but also critically important are long-term neurodevelopmental outcomes.

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.

Additional information

Funding

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.