Hypoxic-Ischemic Encephalopathy Treatment & Management

Updated: Jul 18, 2018
  • Author: Santina A Zanelli, MD; Chief Editor: Dharmendra J Nimavat, MD, FAAP  more...
  • Print

Medical Care

Therapeutic hypothermia is indicated for infants with moderate-to-severe hypoxic-ischemic encephalopathy (HIE). Supportive management is also critical to prevent additional injury from seizure activity, poor perfusion, electrolyte imbalance, and abnormal glycemic control.

Following initial resuscitation and stabilization, treatment of HIE includes hypothermia therapy for moderate to severe encephalopathy as well as supportive measures focusing on adequate oxygenation, ventilation and perfusion, careful fluid management, avoidance of hypoglycemia and hyperglycemia, and treatment of seizures. [4, 5] Intervention strategies aim to avoid any further brain injury in these infants. [66]

In cases of posterior cranial fossa hematoma, surgical drainage may be lifesaving if no additional pathologies are present.

In patients with HIE and suspected neonatal sepsis receiving gentamicin and hypothermia treatment, modified gentamicin dosing regimens are required owing to the reduced clearance of this agent potentially leading to toxicity in these infants from higher gentamicin concentrations during hypothermia therapy. [129]


Infants who present in a level I or II center may require transfer to a tertiary neonatal intensive care unit (NICU) for definitive neurodiagnostic studies (electroencephalography [EEG] and neuroimaging), consultation with a pediatric neurologist, and evaluation for therapeutic hypothermia. Based on the current recommendations, therapeutic hypothermia must be initiated within 6 hours after birth. [127] Timely referral is essential to provide therapeutic hypothermia. If that window (of 6 hours) has passed, infants will still benefit from the expertise of level III and higher centers.

Discharge considerations

Physical therapy and developmental evaluations are needed prior to discharge of patients with HIE. Even after discharge, close monitoring and regular follow-ups are essential for better outcomes. Referring to early intervention is a must at the time of discharge.

Continuation of seizure medications should depend on evolving central nervous system (CNS) symptoms and EEG findings. In most cases, antiseizure medications can be discontinued prior to NICU discharge. Follow-up by a pediatric neurologist is recommended.


Initial Resuscitation and Stabilization

Delivery room management follows standard Neonatal Resuscitation Program (NRP) guidelines. Close attention should be paid to appropriate oxygen delivery, perfusion status, avoidance of hypoglycemia and hyperglycemia, as well as avoidance of hyperthermia.

A lot of attention has been focused on resuscitation with room air versus 100% oxygen in the delivery room. Several clinical trials indicate that room air resuscitation for infants with perinatal asphyxia is as effective as resuscitation with 100% oxygen. In addition, infants resuscitated with room air have a lower level of circulating markers of oxidative stress. However, studies indicating that time to return to spontaneous circulation is equivalent with room air resuscitation are lacking. Based on this evidence, International Liaison Committee on Resuscitation (ILCOR) and NRP guidelines were updated and are now recommending the use of 21% oxygen for the initial resuscitation of term infants. If despite effective ventilation, the infant does not improve, higher concentrations of oxygen should be used and should be guided by the use of pulse oxymetry. [67, 68]


Supportive Care in Patients with Hypoxic-ischemic Encephalopathy

Most infants with severe hypoxic-ischemic encephalopathy (HIE) need ventilatory support during the first few days after birth. Although animal data suggest that permissive hypercapnia may be neuroprotective, no such evidence is available in newborn. Therefore, the role of mechanical ventilation is to maintain the blood gases and acid-base status in the physiologic ranges and prevent hypoxia, hyperoxia, hypercapnia, and hypocapnia. Hypocapnia in particular may lead to severe brain hypoperfusion and cellular alkalosis and has been associated with worse neurodevelopmental outcomes. Of note, evidence indicates that increased FiO2 in the first 6 hours of life is a significant risk factor for adverse outcomes in infants with hypoxic-ischemic encephalopathy treated with hypothermia therapy. This association is independent of underlying respiratory pathology and further emphasizes the benefit of resuscitation and stabilization with room air in this patient population. [69]

Infants with HIE are also at risk for pulmonary hypertension and should be monitored. Inhaled nitric oxide (iNO) may be used according to published guidelines if pulmonary hypertension is suspected. [70]


Perfusion and Blood Pressure Management

Studies indicate that a mean blood pressure (BP) above 35-40 mm Hg is necessary to avoid decreased cerebral perfusion. Hypotension is common in infants with severe hypoxic-ischemic encephalopathy (HIE) and is due to myocardial dysfunction, capillary leak syndrome, and hypovolemia; hypotension should be promptly treated. Dopamine or dobutamine can be used to achieve adequate cardiac output in these patients. If a cardiac injury is suspected, then administration of dobutamine or milrinone may be beneficial to support the injured heart.


Fluid and Electrolytes Management

Because of the concern for acute tubular necrosis (ATN) and syndrome of inappropriate antidiuretic hormone (SIADH) secretion, fluid restriction is typically recommended for these infants until renal function and urine output can be evaluated. However, this recommendation is not based on evidence from randomized controlled trials. [71] Therefore, fluid and electrolyte management must be individualized on the basis of clinical course, changes in weight, urine output, and the results of serum electrolyte and renal function studies.

The role of prophylactic theophylline, given early after birth, in reducing renal dysfunction after hypoxic-ischemic encephalopathy (HIE) has been evaluated in 3 small randomized controlled trials. [72, 73, 74] In these studies, a single dose of theophylline (5-8 mg/kg) given within 1 hour of birth resulted in (1) decreased severe renal dysfunction (defined as creatinine level >1.5 mg/dL for 2 consecutive days); (2) increased creatine clearance; (3) increased glomerular filtration rate (GFR); and (4) decreased b2 microglobulin excretion. The clinical significance of these findings remains unclear. Larger studies are warranted to confirm the safety of adenosine inhibitor use following HIE.

Fluid and glucose homeostasis should be achieved. Avoid hypoglycemia and hyperglycemia because both may accentuate brain damage. Hypoglycemia in particular should be avoided. In a retrospective study, Salhab et al showed that initial hypoglycemia (< 40 mg/dL) is significantly associated with adverse neurologic outsomes. [75]


Hyperthermia Avoidance

Hyperthermia has been shown to be associated with increased risk of adverse outcomes in neonates with moderate-to-severe hypoxic-ischemic encephalopathy (HIE). [6] In this observational secondary study, the risk of death or moderate-to-severe disability was increased 3.6-fold to 4-fold for every 1°C increase in the mean of the highest quartile of skin or esophageal temperature.


Treatment of Seizures

Hypoxic-ischemic encephalopathy (HIE) is the most common cause of seizures in the neonatal period. Seizures are generally self-limited to the first days after birth but may significantly compromise other body functions, such as maintenance of ventilation, oxygenation, and blood pressure. Additionally, studies suggest that seizures, including asymptomatic electrographic seizures, may contribute to brain injury and increase the risk of subsequent epilepsy. [76, 77, 78]

Current therapies available to treat neonates with seizures have limited efficacy, and safety concerns remain specifically for infants undergoing therapeutic hypothermia. Antiseizure drugs used in this population include phenobarbital, levetiracetam, phenytoin, lidocaine, and benzodiazepines. However, phenobarbital has been shown to be effective in only 29-50% of cases, [79, 80, 81]  and phenytoin only offers an additional 15% efficacy. Benzodiazepines, particularly lorazepam, may offer some additional efficacy. [82, 83]  Newer antiseizure medications such as levetiracetam are increasingly used in infants with HIE and seizures despite the lack of strong evidence regarding safety or efficacy in this population.


Hypothermia Therapy

Extensive experimental data suggest that mild hypothermia (3-4°C below baseline temperature) applied within a few hours (no later than 6 h) of injury is neuroprotective. The neuroprotective mechanisms are not completely understood. Possible mechanisms include (1) reduced metabolic rate and energy depletion; (2) decreased excitatory transmitter release; (3) reduced alterations in ion flux; (4) reduced apoptosis due to hypoxic-ischemic encephalopathy; and (4) reduced vascular permeability, edema, and disruptions of blood-brain barrier functions. [84, 85]

Randomized clinical trials

The clinical efficacy of therapeutic hypothermia in neonates with moderate-to-severe hypoxic-ischemic encephalopathy (HIE) has been evaluated in multiple randomized controlled trials. [27, 28, 86, 87, 88, 89, 90, 91] for a total of greater than 1500 infants enrolled. Inclusion criteria varied slightly between studies and are summarized as follows:

  • Near-term infants born at 36 weeks' gestation or more with birth weight of 1800-2000 g or more, younger than 6 hours at admission. Some trials have enrolled infants as young as 35 weeks' gestation (Eicher and ICE trial)

  • Evidence of acute event around the time of birth – Apgar score of 5 or less at 10 minutes after birth (In the study by Shankaran et al, this needed to be in conjunction with either evidence of acute perinatal event or need for assisted ventilation for at least 10 min. [28] ), severe acidosis, defined as pH level of less than 7 or base deficit of 16 mmol/L or less (cord blood or any blood gas obtained within 1 h of birth), continued need for resuscitation at 10 minutes after birth

  • Evidence of moderate to severe encephalopathy at birth – Clinically determined (at least 2 of the following: lethargy, stupor, or coma; abnormal tone or posture; abnormal reflexes [suck, grasp, Moro, gag, stretch reflexes]; decreased or absent spontaneous activity; autonomic dysfunction [including bradycardia, abnormal pupils, apneas]; and clinical evidence of seizures), moderately or severely abnormal amplitude-integrated electroencephalography (aEEG) background or seizures (CoolCap, TOBY and Neo-Neuro, Neonatal Network trial)

All of these studies have shown benefits, and 9 independent meta-analyses have confirmed a consistent and robust beneficial effect of therapeutic hypothermia for moderate-to-severe encephalopathy with a number needed to treat between 5 and 9.

The 2013 cochrane review included 11 randomized controlled trials and 1505 infants and found that therapeutic hypothermia resulted in the following:

  • A decrease in the combined outcomes of mortality/major neurodevelopmental disability at 18 months (8 studies, 1344 infants): relative risk [RR] 0.75 (0.68-0.83); number needed to benefit (NNTB) 7 (5-10)

  • A reduction in mortality (11 studies, 1468 infants): RR 0.75 (0.64-0.88); NNTB 11 (8-25)

  • A reduction in neurodevelopmental disability in survivors (8 studies, 917 infants): RR 0.77 (0.63-0.94); NNTB 8 (5-14)

Adverse effects

Many theoretical concerns surround hypothermia and its side effects, which include coagulation defects, leukocyte malfunctions, pulmonary hypertension, worsening of metabolic acidosis, and abnormalities of cardiac rhythm, especially during rewarming.

Randomized trials have been reassuring thus far regarding the safety and applicability of therapeutic hypothermia. [92] In a 2013 Cochrane review, significant adverse effects were limited to sinus bradycardia (RR 11.59 [4.94-27.17]; number need to harm [NNTH] 11 [9-14]), and thrombocytopenia (RR 1.21 [1.05-1.40]; NNTH 17 [10-50]).

Long-term outcomes

School-age outcomes of infants in the NICHD trial were published in 2012. [93] At age of follow-up (6-7 years, 91% follow-up), the combined outcome of death or IQ score below 70 occurred in 62% of infants in the control group versus 47% of infants in the hypothermia group (P = 0.06). More infants in the control group died (44%) compared to 28% in the hypothermia group (P = 0.04). Reassuringly, this finding was not associated to increased risk of neurodevelopmental disability in survivors with a risk of death or severe disability in 60% of controls versus 41% in the hypothermia group (P = 0.03). [93]

In 2014, more follow-up results from the TOBY trial were published of children aged 6-7 years who had asphyxia encephalopathy as infants and were treated with hypothermia. [94] There were 75 of 145 survivors (52%) in the hypothermia group relative to 52 of 132 children (39%) in the control group. Children who received hypothermia shortly after birth were significantly more likely to have an IQ of 85 or higher at age 6-7 years, and they were less likely to moderate-to-severe disability compared to the control group. The study was insufficiently powered to determine whether hypothermia treatment led to positive neurocognitive effects at older ages, although the investigators were able to establish that early assessment at ages 18-21 months reliably predicted good functional outcomes at school age. [94]

Therapeutic hypothermia when applied within 6 hours of birth and maintained for 72 hours is the only therapy currently available that improves the outcomes of infants with moderate-to-severe HIE. [95, 96]

Remaining questions

What is the optimal timing of initiation of hypothermia therapy?

Cooling must begin early, within 6 hours of injury. However, experimental evidence strongly suggest that the earlier the better.

Reports on the feasibility and safety of cooling on transport indicate that initiation of hypothermia therapy at referring centers is possible, provided that ongoing education is in place. [97] The ICE trial confirmed that a simplified method using widely available icepacks is an effective way to provide hypothermia therapy in referring centers while awaiting transfer to a tertiary neonatal intensive care unit (NICU). [88]

However, a favorable outcome may be possible if the cooling begins beyond 6 hours after injury. A current National Institute of Child Health and Human Development (NICHD) study is evaluating the efficacy of delayed hypothermia therapy for infants presenting at referral centers beyond 6 hours of life or with evolving encephalopathy.

What is the optimal duration of hypothermia therapy?

The greater the severity of the initial injury, the longer the duration of hypothermia needed for optimal neuroprotection. The optimal duration of brain cooling in the human newborn has not been established. A 2014 NICHD trial indicated that longer and deeper cooling does not provide additional benefits over current protocols. [98]

What is the best method?

Two methods have been used in clinical trials: selective head cooling and whole body cooling.

In selective head cooling, a cap (CoolCap) with channels for circulating cold water is placed over the infant's head, and a pumping device facilitates continuous circulation of cold water. Nasopharyngeal or rectal temperature is then maintained at 34º-35°C for 72 hours.

In whole body hypothermia, the infant is placed on a commercially available cooling blanket, through which circulating cold water flows, so that the desired level of hypothermia is reached quickly and maintained for 72 hours.

The relative merits and limitations of these 2 methods have not been established; however, whole body hypothermia is most widely used modality to provide therapeutic hypothermia.

What is the optimal rewarming method?

Rewarming is a critical period. In clinical trials, rewarming was carried out gradually, over 6-8 hours.

Can the use of aEEG improve candidates selection?

Predefined subgroup analysis in the CoolCap trial suggested that head cooling had no effect in infants with the most severe aEEG changes.

The findings were beneficial only in infants with less severe aEEG changes.

Hypothermia therapy has been recommended as the standard of care since 2010 by the AHA and ILCOR: "During the postresuscitation period in greater than or equal to 36-week gestation neonates with evolving moderate or severe encephalopathy, hypothermia should be offered in the context of clearly defined protocols similar to published trials.”

Hypothermia therapy should be conducted under strict protocols and reserved to regional referral centers offering comprehensive multidisciplinary care and planning to conduct long-term neurodevelopmental follow-up. Implementation requires thorough and ongoing education to avoid complications such as overcooling. [99] Ideally, all infants should be registered in national registry whenever possible.


Future Neuroprotective Strategies

Several groups are investigating other neuroprotective strategies whether alone or in combination with hypothermia therapy (summarized in the image below). [100]

Summary of potential neuroprotective strategies. Summary of potential neuroprotective strategies.

Promising avenues include the following:

  • Prophylactic barbiturates: In a small randomized trial, high-dose phenobarbital (40 mg/kg) was given over 1 hour to infants with severe hypoxic-ischemic encephalopathy. Treated infants had fewer seizures (9 of 15) than untreated control infants (14 of 16). Treated infants also had fewer neurologic deficits at age 3 years (4 of 15) than untreated infants (13 of 16). [101] In another small study, thiopental given within 2 hours and over 24 hours, did not result in improved rate of seizures or neurodevelopmental outcomes at 12 months. [102] Hypotension was more common in infants who received thiopental. Thus, the role of prophylactic barbiturate remains unclear. Further studies are needed. [103]

  • Erythropoietin: In one study, low-dose erythropoietin (300-500 U/kg) administered for 2 weeks starting in the first 48 hours of life decreased the incidence of death or moderate and severe disability at age 18 months (43.8% vs 24.6%; P < 0.05) in infants with moderate-to-severe hypoxic-ischemic encephalopathy. Subgroup analysis indicated that only infants with moderate disability benefited from this therapy. [104] Currently being evaluated in NCT 01913340.

  • Allopurinol: Slight improvements in survival and cerebral blood flow (CBF) were noted in a small group of infants tested with this free-radical scavenger in one clinical trial. [105]

  • Excitatory amino acid (EAA) antagonists: MK-801, an EAA antagonist, has shown promising results in experimental animals and in a limited number of adult trials. However, this drug has serious cardiovascular adverse effects.

  • Stem cell therapy: The use of mesenchymal stem cells and autologous stem cells to treat infants with hypoxic-ischemic encephalopathy (HIE) is under extensive study. Early evidence suggest this may be an effective therapeutic avenue. More work is required to determine the type of cells, dose, timing, and duration. In addition, more studies are also required to understand the underlying protective mechanisms. [106, 107, 108]

  • Other adjuvant therapies under investigation include Xexon (NCT0271394 and NCT01545271), topiramate (NCT01765218), and MgSO4 (NCT01646619).



A pediatric neurologist should help assist in the management of seizures, interpretation of electroencephalograms (EEGs), and overall care of the infant with hypoxic-ischemic encephalopathy (HIE). The neurologist should also work with the primary care physician to address long-term disabilities.

Follow-up by a developmental pediatrician is also recommended to assist with planning for the infant's long-term assessments of neurodevelopment and care.



In most cases (particularly in severe hypoxic-ischemic encephalopathy [HIE]), the infant is restricted to nothing by mouth (NPO) until the general level of alertness and consciousness improves and the hemodynamic status stabilizes. In addition, most infants undergoing therapeutic hypothermia should remain NPO until rewarmed. A study of 51 neonates with HIE indicated that minimal enteral nutrition (1-2 mL/kg boluses every 3h) may be safe in hemodynamically stable infants undergoing therapeutic hypothermia. [109]

Enteral feeds should be carefully initiated, and the use of trophic feeds is recommended for 24-48 hours (2 mL/kg every 3 h). Infants should be monitored carefully for signs and symptoms of necrotizing enterocolitis, for which infants with perinatal asphyxia are at high risk. Individualize increments in feeding volume and composition.



The use of intrapartum markers such as fetal heart rate monitoring are poor predictors of neonatal outcomes and long-term risk of cerebral palsy. [110]

Most treatments under investigation have discussed earlier and remain experimental. With the exception of therapeautic hypothermia, none has consistently shown efficacy in human infants.


Long-Term Monitoring

The goal of follow-up is to detect impairments and promote early intervention for those infants who require it. [111]

Growth parameters including head circumference should be closely monitored in all infants with hypoxic-ischemic encephalopathy (HIE).

Infants with moderate-to-severe HIE should be followed closely after neonatal intensive care unit (NICU) discharge by a developmental pediatrician and, in some cases, a pediatric neurologist (if there is a history of seizure and/or abnormal neurologic examination). Additionally, evaluation by a pediatric ophthalmologist is recommended during the first year of life, because damage to the posterovisual cortex can occur. Standard hearing test screening should occur prior to NICU discharge. A repeat hearing screen is also recommended in the first 2 years of life. 

If therapeutic hypothermia was used in the neonatal period, follow-up is recommended for the continued evaluation of the long-term efficacy of this therapy. Data should be entered into the available registries, local databases, or both, whenever possible.

Infants with mild HIE generally do well and do not require specialized follow-up.