Perinatal asphyxia: abnormal cardiotocographic pattern, complicated birth or sentinel event requiring advanced resuscitation at birth and/or 5-minute Apgar score of 5 or less and/or cord blood pH of 7.00 or less or base deficit greater than 16mmol/L.

Perinatal HIE: neurologic impairment (encephalopathy) from the first hours post birth attributable to a hypoxic-ischaemic attack around the time of birth and associated with a history of perinatal asphyxia.

Although there are antepartum factors that may predispose or contribute to intrapartum adverse events, their presence alone is not sufficient to cause HIE, and the intrapartum period is considered the time when the risk is highest, with intrapartum events constituting the main or a contributing factor in the causal pathway of HIE.10 The aim of obstetric care is to identify and rescue the foetus at risk of hypoxic-ischaemic events through the recognition of signs suggestive of foetal hypoxia. The high rates of HIE in LMICs are probably determined, in part, by social factors that affect maternal health and by the circumstances and the setting of childbirth. Delivery of higher-quality prenatal care by qualified community-based midwives and the reinforcement of hospital and out-of-hospital maternal and neonatal care systems could achieve reductions in perinatal mortality and the incidence of HIE.11

The goal of neonatal resuscitation in newborns with perinatal asphyxia is to re-establish adequate tissue oxygenation and cerebral blood flow and to prevent conditions or events affecting the brain and other organs that could exacerbate the hypoxic-ischaemic insult. It is estimated that 15% of newborns in LMICs require basic resuscitation (stimulation, “sniffing” positioning of head, opening of airway and bag-mask ventilation with room air), which reduces intrapartum mortality by 40%, neonatal intrapartum-related mortality by 30% and early neonatal mortality by 38%.12

Competence in neonatal resuscitation acquired through training programmes is crucial to guarantee the safety and effectiveness of this intervention and to reduce morbidity and mortality secondary to perinatal asphyxia, which are particularly high in RL settings.12

The International Liaison Committee on Resuscitation (ILCOR) periodically publishes consensus-based recommendations that are the foundation of the three most widely used neonatal resuscitation guidelines and training programmes worldwide: the neonatal resuscitation programme, neonatal life support and Helping Babies Breathe (HBB), especially developed for LR settings. Table 1 summarises the key points of current recommendations, in reference to the 2020 ILCOR consensus,13 the 2021 European Resuscitation Council Guidelines,14 the 2023 American Heart Association and American Academy of Pediatrics guidelines15 and, in Spanish, the specific position statement of the Neonatal Resuscitation Group of the Sociedad Española de Neonatología (SENeo, Spanish Society of Neonatology)16 as well as, on the subject of basic resuscitation in RL settings, the second edition of the HBB from 2016.

The control of comorbid factors that can exacerbate hypoxic-ischaemic brain lesions is a cornerstone in the care of newborns with HIE and is particularly important in RL settings. The recommendations described in this section apply to the immediate postpartum period in a newborn with perinatal asphyxia, but can also be extended to the first few days of life in newborns with HIE. Monitoring and control of body temperature and blood glucose levels (Appendix A, Table S1) are key, in addition to other aspects presented in Table 2.

The identification HIE and grading its severity are key in the management of newborns with perinatal asphyxia (Table 3 and Appendix A S2).

The severity of HIE in the first 6h post birth is the key factor in the selection of candidates for TH. In the era before TH, the level of impairment associated with HIE was low if the encephalopathy was mild, intermediate if the encephalopathy was moderate and high if the encephalopathy was severe (practically all infants with severe HIE died in the neonatal period or had severe disability at 18–24 months).17–19 Similar outcomes were reported in RCTs that assessed the efficacy of TH.20–22 While the predictive value of the severity of HIE in the first 6h is moderate for the prediction of long-term outcomes (Appendix A Table S3), the severity assessed at 48–96h and at 7 days offers a more accurate prognosis,19,23–25 which reflects the dynamic nature (improvement or worsening) of HIE.18,19,23,24 When the encephalopathy does not improve in the first 72h, the probability of death or disability is much higher than when it improves within 24h post birth. In RL settings, the assessment of severity of HIE as well as changes in severity over time, using categorical or numerical scales,19,26 can be the most efficient approach for establishing neurodevelopmental risk groups. If the staff is insufficiently trained, it is best to seek consultation with local experts or through telemedicine.27

Hypoxic-ischaemic encephalopathy is the leading cause of neonatal seizures independently of the income and resources of the country.28 In HICs, 50%–85% of newborns with moderate or severe HIE develop clinical seizures, whether or not they are managed with cooling,22,23,29 compared to nearly 80% of patients with moderate HIE and more than 80% of patients with severe HIE in RL settings.5,9

The onset of seizures associated with HIE usually occurs after the first 6h post birth, with onset occurring between 6 and 24h post birth in half the patients and commonly after 12h.29–31 This suggests that when features suggestive of seizures are observed before 6h, they may not be actual seizures or may be seizures secondary to a prenatal hypoxic-ischaemic insult or with a different aetiology.

When neonatal seizures in infants with HIE are only identified based on clinical features (without the use of amplitude-integrated electroencephalography [aEEG]), the true seizure burden (seizure frequency and cumulative duration over a given period) is underestimated, as most neonatal seizures are subclinical and up to 23% of these patients exhibit electrographic status epilepticus in the first days of life.29,31 On the other hand, even experienced clinicians have difficulty distinguishing convulsive seizures from non-epileptic motor phenomena based on observation alone, so there is a risk of overdiagnosis of neonatal seizures. In consequence, the International League Against Epilepsy (ILAE) has underscored the importance of EEG monitoring for the accurate diagnosis of neonatal seizures.32

The seizure burden in newborns with moderate to severe HIE managed with or without TH is high, is dependent on the severity of HIE and peaks around 24h post birth.29,30 The seizure burden is also an important predictor of lesions in neuroimaging and of neurodevelopmental disorders, and its association with adverse outcomes is independent of the severity of HIE and the use (or lack thereof) of TH.33,34 Therefore, prompt and effective treatment of neonatal seizures is of the essence (Fig. 2, Table 4, Appendix A S4). Treatment algorithms must call for caution in the approach to refractory seizures, broadening the differential diagnosis to include aetiologies different from or concurrent with hypoxia-ischaemia (Fig. 2).

Figure 2.

Algorithm for the treatment of seizures in resource-limited settings (without therapeutic hypothermia).

(0.85MB).

Since many neonatal units in RL settings have limited access to continuous EEG monitoring or aEEG, treatment of neonatal seizures depends on their identification based on clinical features, on which health care professionals must therefore be trained. The Brighton Collaboration has proposed a has proposed a scheme with five levels of diagnostic certainty that can guide treatment decisions and help avoid overtreatment if EEG is not available35 (Appendix A Table S5).

On the other hand, continued exposure to antiepileptic drugs can have noxious effects on the developing brain in the long term and, due to the sedative effects, delay the establishment of oral feeding and hinder bonding between the newborn and the parents. Therefore, antiepileptic treatment should only be given for as long as they are needed.

There is limited consensus regarding the duration of antiepileptic treatment once the seizures have ceased.32,36 However, there is growing evidence that neonatal seizures in infants with HIE are rare past day 4 or 5 post birth and that recurrence is also infrequent once they are controlled.29,31

Delivery of TH for 72h using passive cooling or manual (not servo-controlled) active cooling is not recommended. These interventions should be restricted to very short periods with close monitoring of body temperature, for instance, for induction of hypothermia prior to transport to a facility where servo-controlled cooling is available.7,8

With the exception of TH, the evince on neuroprotective interventions is scarce, and most of the published studies address their feasibility rather than their efficacy. Since the introduction of TH, most of these interventions have been assessed as adjuvants of TH, which limits our knowledge of their performance as monotherapy in RL settings. The only intervention that has proven beneficial in RL settings is the administration of phenobarbital for prophylaxis (Appendix A Table S6).

Multiple organ dysfunction is frequent and its extension and severity are associated with those of HIE and with neonatal mortality.37 It can also cause significant morbidity and exacerbate the brain injury resulting from the hypoxic-ischaemic insult. The monitoring and treatment methods proposed in Table 5 are not based on evidence, as few studies have compared diagnostic or therapeutic interventions. Notwithstanding, systemic support cannot be neglected. The evidence on the close association between systemic and cerebral haemodynamics has motivated the widespread implementation of noninvasive regional cerebral oxygen saturation monitoring and functional echocardiography. The availability of an ultrasound machine can guide the identification and finetuning of the most appropriate treatment.

Fluid administration must be carefully controlled. Since infants with HIE can have delayed micturition, acute renal failure and, in some cases, syndrome of inappropriate antidiuretic hormone secretion (SIADH), they are at risk of fluid overload and hyponatraemia. Fifteen to forty percent of non-cooled infants with HIE have abnormal renal function test results and more than 30% have hyponatraemia.

More than 50% of newborns with HIE exhibit additional electrolyte abnormalities.20,22,38 The most frequent ones are, in decreasing order, hypocalcaemia, hyperkalaemia, hypokalaemia, hyponatraemia e hypomagnesemia.22 We recommend monitoring of serum electrolytes every 12–14h (Table 6 and Appendix A S7).

As regards nutrition in newborns with HIE, the RCTs that assessed the efficacy of TH had patients under nil per os for the first 72h due to the risk of intestinal ischaemic-reperfusion injury.38 However, the incidence of severe intestinal damage is very low, and trophic and progressive feedings do not seem to be associated with an increase in intestinal complications (Table 6 and Appendix AS7).

The routine use of antibiotherapy in patients with neonatal asphyxia or HIE is not justified, although a high index of suspicion must be maintained along with close monitoring of markers of infection in blood and cerebrospinal fluid (CSF), especially in patients with a history of chorioamnionitis, fever, elevation of acute phase reactants or in whom the course of neurologic impairment differs from what has been expected.

Two widely available methods, the conventional EEG and the cranial ultrasound (CUS) can be particularly helpful in establishing the prognosis in RL settings (Appendix A Table S8).

Before the introduction of continuous EEG monitoring, severe background pattern disturbances in the EEG in the first days of life had been found to be strongly correlated to the severity of HIE, MRI abnormalities and a poor neurodevelopmental outcome.39–41 This traditional neurophysiological assessment method requires a minimum of 15min; it can be implemented at the bedside, using a limited number of electrodes, and it is relatively easy to interpret.42

Cranial ultrasound is a noninvasive imaging test that is performed at the bedside and can be repeated to observe the evolution of detected abnormalities. Modern equipment with grayscale imaging and probes with variable frequencies have improved the sensitivity and specificity of the diagnosis of lesions by means of CUS. In addition, the widespread availability and affordability of portable ultrasound machines has made CUS increasingly feasible in RL settings. The performance and interpretation of CUS by trained neonatologists has improved access and reduced the cost of this technique.43

In infants with HIE, CUS abnormalities between 24 and 72h have been found to correlate to post-mortem findings44 as well as MRI findings in the first week, and are predictive of a poor outcome.45 Repeated scans allow monitoring of the changes in the identified lesions and defining the pattern and severity of brain injury. In infants with HIE that do not receive TH, the resistive index measured in the anterior cerebral artery is a good prognostic factor.

This cornerstone of the comprehensive care of infants with HIE has received little attention until TH was introduced in clinical practice. The birth and hospitalization of infants with HIE is a stressful and traumatic experience for the family. Factors like the technology surrounding the infant, the fear that the infant may die and the uncertainty about the prognosis give rise to feelings of guilt, sadness and hopelessness in parents. In addition, the separation from the infant, the inability or difficulty to hold the infant or to be involved in the infant’s care are factors that hinder parent-child bonding and the development of parental roles, and have a negative impact on the emotional wellbeing of the family (Appendix A Table S9).