QUARTERLY REVIEW

Therapeutic Hypothermia for Treatment of Neonatal Encephalopathy

Neonatal encephalopathy, a condition resulting from perinatal asphyxia, occurs in 2.0–6.0 of every 1000 live births. Without treatment, prognosis is poor and resulting complications such as intellectual delay and cerebral palsy are often severe. Therapeutic hypothermia has emerged as an effective treatment for neonatal encephalopathy. Now, research is aimed at determining prognosis after encephalopathy and therapeutic hypothermia. Additionally, nurses play a large role in the identification and care of infants receiving therapeutic hypothermia and their families.

Introduction

Neonatal encephalopathy (NE) is a condition that often results in serious health consequences including death, cerebral palsy, developmental delay, and seizure disorder.[1] The incidence ranges from 2.0 to 6.0 per 1000 live births, with higher incidence in poorer countries.[1] Therapeutic hypothermia (TH) has become the gold standard in treatment of NE due to its effectiveness in preventing death and major disability during the neonatal period. This paper aims to discuss the pathophysiology of NE, review current literature, and discuss nursing care of the cooled infant.

---

Pathophysiology

NE is a condition characterized by abnormal neurological function in the newborn period. Infants with NE may display an abnormal level of consciousness, altered muscle tone or reflexes, apnea or altered respirations, and sometimes seizures.[2,3] Hypoxic-ischemic encephalopathy (HIE) occurs when NE is the product of hypoxic-ischemic brain injury. HIE is the result of any event that causes decreased blood supply to the brain. Disturbed uteroplacental blood flow, placental abruption, tight or knotted nuchal cord, umbilical cord prolapse, and uterine rupture are risk factors.^[3]

 

Figure 1. Key pathophysiological changes after hypoxic insult to neonatal brain.

 

 

The disruption of blood flow and oxygen delivery that occurs in HIE causes a two phase reaction within the brain tissue that eventually causes brain injury.[3,4] Phase one, also known as primary energy failure, occurs when blood flow to the brain is disturbed and delivery of oxygen and other required substrates is impaired causing the brain to enter anaerobic metabolism.[4] Subsequently, brain tissue acidosis due to lactic acid buildup disrupts neuronal ability to maintain ionic balance, synthesize proteins, and regulate neurotransmitter release and reuptake.[5] One neurotransmitter in particular, glutamate, is a prominent excitatory neurotransmitter.[5] Buildup of excessive extracellular glutamate leads to a process called excitotoxicity where channels activated by glutamate remain open, causing persistent neuronal depolarization, allowing excess extracellular calcium into the cell resulting in cell death.[3,5,6]

Primary energy failure is relieved by the resolution of hypoxia-ischemia, leading to restoration of aerobic metabolism and reduced acidosis. However, reperfusion is followed by a secondary energy failure that usually occurs within 8–16 hours of primary failure.[5–7] Secondary energy failure is thought to be a product of mitochondrial damage, inflammation, the continued existence of surplus extracellular excitatory neurotransmitters, free radical damage, and oxidative injury culminating in premature neuronal apoptosis.[4–6] The timing and severity of secondary energy failure correlate with the severity of primary energy failure. Quicker onset and increased severity of secondary energy failure follow more severe primary failure.[5]

---

Therapeutic Hypothermia

Treatment of NE is aimed at disruption of the cascade of events leading to secondary energy failure and neuronal death.[8] TH is thought to diminish the severity of secondary energy failure by reducing the brain's energy utilization, decreasing free radical production and release of extracellular excitatory neurotransmitters, and normalizing protein synthesis.[9,10] Combined, these effects lead to reduced neuronal apoptosis.[9,10]

Westin first reported the use of TH as treatment for asphyxiated infants in 1959.[4] Infants had improved neurological outcomes after being immersed in cold water to achieve core body temperatures of 23–25 °C for 20 minutes. However, clinical trials in the neonatal setting did not start in earnest until the late 1990s and early 2000s when pilot studies showed TH was safe for infants greater than 36 weeks gestation.[11,12] Then, in 2005, the results of the first multicenter phase III trial showed TH was both safe and effective for the treatment of mild and moderate NE.^[12]

Since then, multicenter randomized controlled trials (RCTs) including the U.K.'s Total Body Hypothermia Trial (TOBY), Europe's "neo.nEuro.network" trial of whole body cooling, the National Institute of Child Health and Human Development's (NICHD) trial, and Australia's Infant Cooling Evaluation (ICE) encompassing over 1000 infants have commenced.[13–15] The results from these trials were published in the early to mid 2000s and since then several meta-analyses of these and other studies have been published that help to summarize the outcomes.[16–18] Edwards, et al. (2010) performed a meta-analysis of 10 clinical trials encompassing 1320 infants.[16] Their analysis showed lower rates of mortality, cerebral palsy, hearing and visual impairment, and neurodevelopmental delay in cooled infants. They concluded that the overall effect of TH is significant for preventing the primary outcomes of death and major disability as well as secondary outcomes such as cerebral palsy caused by NE. Shah (2010) reviewed 19 reports of 14 clinical trials of 1440 patients and found significant reductions in the risk of mortality or moderate to severe developmental delay.[18] Tagin, et al. (2011) reviewed 7 trials encompassing 1214 newborns and also found overall reduced risk of death and major disability along with cerebral palsy, developmental delay, and blindness.[19] Though there is some overlap in the studies included in these meta-analyses, they highlight the growing pool of evidence supporting the efficacy of TH.

Since the initial results of the large RCTs were published, subsequent follow-up studies were done on the surviving participants. Researchers were interested in whether TH provided long-lasting neuroprotection. The 18-month to 8-year follow-up results are summarized in . In general, TH was shown to reduce the incidence of death and major disability into childhood.[13–15,17,20]

TH = Therapeutic hypothermia group; C = Control Group; NR = Not reported.

Of particular interest are outcome measures of treatment subjects who are now reaching school age. Shankaran, et al. performed follow-up evaluations on 122 surviving subjects of the NICHD RCT.[20] They evaluated survivors on IQ and motor function, including classifying the level of cerebral palsy if present. At 6–7 years post treatment, they found no statistically significant difference in the rates of the composite outcome of death or IQ below 70 between the hypothermia and control groups. However, reduced mortality continued through age 6–7 years in the hypothermia group. Additionally, survivors had no significant increase risk of negative neurodevelopmental outcomes. These results suggest that TH safely provides long-term neuroprotection.[20]

---

New Topics in Therapeutic Hypothermia Research

Early research has shown that TH can be beneficial in the treatment of NE. However, researchers are now attempting to identify methods for real-time assessment of effectiveness of treatment and ways to predict outcomes. Knowledge pertaining to these topics could aid in counseling parents as well as finding complimentary treatments.

Imaging

Magnetic resonance imaging (MRI) is a useful tool for defining the extent of brain injury in infants following perinatal asphyxia. A main focus in current literature is research aimed at expanding the utility of MRI beyond diagnosis. Studies are now reporting evaluations of MRI in measuring the effectiveness of TH as well as predicting neurodevelopmental outcomes. Additional information is being sought to determine appropriate timing of scans and whether TH interferes with the reliability of MRI readings.

Shankaran, et al. (2011) compared MRIs performed at 44 weeks gestational age and categorized by pattern of injury to neurodevelopmental assessments at 18–22 months.[21] Because some patterns of injury did not fit into earlier defined systems of categorization, a method called the NICHD Neonatal Research Network (NRN) pattern of injury was created that classified injuries into six groups. The most common pattern of injury included damage to the posterior limb of the internal capsule (PLIC), the anterior limb of the internal capsule (ALIC), or the basal ganglia and thalami (BGT). Infants in the hypothermia group tended to have more normal appearing MRIs, normal PLIC and ALIC, and significantly fewer areas of watershed infarction. They found excellent correlation between the NICHD NRN pattern of injury categorization system and the primary outcome of death or disability at 18–22 months. Additionally, there was no statistically significant effect of age at time of scan on the prognostic utility of the NICHD NRN pattern of injury system.[21]

Rutherford, et al. (2010) and Cheong, et al. (2012) analyzed T1 and T2 weighted MRIs and performed sub-studies of infants in the TOBY and ICE trials, respectively.[22,23] Lesions were categorized into 1 of 5 patterns of injury as defined by Okeraefor, et al.[24] Cheong, et al. found moderate to severe injuries in certain regions to be prognostic of poor outcomes at 2 years of age.[23] In both studies, cooled infants were more likely to have a normal scan and had reduced incidence and severity of MRI abnormalities in the basal ganglia and thalami as well as the white and gray matter.[22,23] They also inferred that MRIs performed during the neonatal period were predictive of outcomes at 18 months and 2 years and were not affected by TH.[22,23] Both groups suggest that timing of MRI may affect the reliability of the reading. Rutherford, et al. discovered more major abnormalities on infants scanned at less than 8 days of age when compared to those scanned at greater than 8 days of age.[22] The variation may be due to spontaneous resolution of lesions or increased severity of disease for those scanned earlier. Because the reason for differences in timing of obtaining MRI was unknown, generalizations about optimal timing were unable to be ascertained.[22]

Tusor, et al. (2012) aimed to identify an early biomarker that could help assess the efficacy of a treatment.[25] They also hypothesized that MRI results could be used to predict neurodevelopmental outcomes in infants with NE. The MRIs were evaluated using tract-based spatial statistical (TBSS) analysis of diffusion tensor imaging (DTI). DTI can be used to make presumptions about the underlying microstructure of tissue using the random motion of water molecules within that tissue. TBSS is an observer-independent, multi-subject analysis tool that can be used to assess DTI data for white matter degenerative changes. This study compared measures derived from DTI with results of Griffiths Mental Development Scales (Revised) (GMDS-R) performed at 24–28 months of age. They found TBSS to be more accurate at predicting outcomes than HIE staging. Their findings also suggest that TBSS can be used as an early biomarker to study additional neuroprotective interventions.[25]

HIE Staging

Shankaran, et al. (2012) hypothesized that the clinical course of NE, as measured by serial neurological exams, could have prognostic utility.[26] Modified Sarnat exams were performed at greater than 6, 24, 48, and 72 hours during the intervention and once again at discharge. Follow-up with 204 surviving infants was performed at 18–22 months when they were assessed for severe disability. There was no significant difference in stage of encephalopathy at greater than 6 and at 72 hours as well as at discharge. However, the cooled group had earlier improvement in stage of encephalopathy, with reduced staging at 24 and 48 hours. Persistent severe encephalopathy through the 72-hour mark was strongly correlated with increased risk of death or severe disability at 18 months of age. They suggest that NE stage at the end of cooling intervention can be used as a good predictor of the risk of death or severe disability.[26]

---

Gaps in Literature

Though long-term study results are beginning to be published and appear promising, there is still limited information regarding the childhood outcomes of encephalopathic infants treated with TH. Additionally, more research aimed at real-time evaluation of the effectiveness of TH could be useful for identifying adjunct treatments. Studies focused on these topics are in progress and the results will be helpful in developing the TH treatment process and improving outcomes for infants.

---

Nursing Considerations

What does all this mean for nursing? To start with, we know that TH is an effective treatment for NE and more nurseries are implementing its use. Therefore, education is needed regarding care of the cooled infant. Knowledge regarding management of TH from the identification of potential candidates to the management of treatment itself can allow for seamless transition from one phase to the next.

Cooling Methods

There are two main methods for implementing TH – whole body cooling, which is the most common, and selective head cooling. The type of cooling varies by institution; however, both methods have been shown to be safe and effective at bringing down core body temperature to the target level.[10,27] Selective head cooling involves a cap placed on the infant's head that circulates cold water to lower the infant's core temperature.[12]With this method, the infant's head and brain structures reach a cooler temperature than the body.[10] Target rectal or esophageal temperature during selective head cooling is 34–35 °C.[19] Whole body cooling is typically implemented via a cooling blanket placed under the baby that circulates cold water to achieve homogenous cooling of the entire body.^[13] Target rectal or esophageal temperature during whole body hypothermia is 33–34 °C.[13,14] Both cooling devices have automatic control modes where the device monitors the baby's temperature with an attached temperature probe and maintains the desired target temperature, programmed by the user, by changing the temperature of circulating water. Security features, such as alarms and screen prompts notify users of unexpected changes in temperature.^[27,28]

It is important to note that not all hospital facilities are outfitted with the equipment required to implement cooling. Several studies have evaluated passive cooling as a technique to diminish heat retention prior to transport and active cooling.[29–31] Passive cooling was found to be a simple and efficient way to initiate TH as long as appropriate temperature monitoring was used concurrently.[29–31] Additionally, passive cooling allowed for therapy to begin nearly 5 hours earlier than if therapy was started after transport.[29] To initiate passive cooling, turn off external heating devices and remove hats, clothing, and blankets.[29] Serial monitoring of rectal temperatures every 15 minutes during passive cooling should be done to prevent the temperature from getting too low.^[29]

---

Who Should Be Cooled?

A gestational age of 36 weeks is usually the minimum age required for TH.[11,12] Cooling of younger infants has not been studied on a large scale; however, some institutions have expanded the eligibility down to 34 weeks gestational age.[32,33] Additionally, the therapeutic window for hypothermia ends at 6 hours of age and it is believed that implementation after that time does not yield the same level of protection.[4] Therefore, the infant must be less than 6 hours old. General exclusions include presence of major congenital anomalies or weight less than 1800 grams.[34] Then, a combination of clinical findings, lab results, and neurological examination is used to diagnose NE. Some institutions may also use amplitude-integrated electroencephalography (aEEG) to identify cooling candidates.^[35]

Clinical findings that indicate NE include history of an acute perinatal event such as cord prolapse or placental abruption, Apgar ≤ 5 at 5 or 10 minutes, and continued need for resuscitation or ventilator support after delivery.[34] Lab results indicative of increased risk for NE include cord blood or postnatal blood gas pH of ≤ 7.0 at ≤ 1 hour of life or base deficit ≥ 16 mEq/L at ≤ 1 hour of life.[34] Neurological status is assessed using the modified Sarnat score and infants are categorized as mild, moderate, or severe if they present with 3 of the 6 indicators for a stage.[34] Generally, only infants with moderate to severe NE will be cooled.[34] A decision tree adapted from the NICHD NRN Whole Body Hypothermia trial combining all eligibility criteria for cooling is provided in .^[36]

 

 

Information adapted from NICHD NRN Whole Body Hypothermia study.³⁵

Treatment Implementation and Monitoring

Once it has been determined that the infant will receive TH and informed consent has been obtained, many steps must occur to implement treatment during the therapeutic window. Preparation includes setting up the cooling device, assisting with placement of central or peripheral access lines, obtaining blood samples for baseline laboratory tests, and initiating passive cooling if possible.[27–29] Before cooling begins, a rectal or esophageal temperature probe will be placed to monitor core temperatures. Esophageal placement will be verified by chest x-ray. Additionally, an axillary temperature probe attached to the warmer bed may be used though the heating element should be off.^[27–29]

After preparation is complete, initiate cooling and begin monitoring of temperatures, vital signs, laboratory results, and aEEG if applicable. Monitoring schedule will vary by institution but often starts with core, axillary, and water temperature every 15 minutes for 2–4 hours and then spaces out to every 30–60 minutes.[27,28] Blood pressure, heart rate, respiratory rate, and pre-ductal and post-ductal oxygen saturations are often documented on the same schedule as temperatures.[27,28] Expect for the cooled infant's heart rate to be lower than usual, often below 100 beats per minute.[34] Lower heart rates are tolerated as long as there is sinus heart rhythm and perfusion and blood pressure remain adequate.[34] Additionally, serial monitoring of serum electrolytes, BUN and creatinine, blood glucose, blood gases, prothrombin time (PT), partial prothrombin time (PPT), and international normalized ration (INR) may be done.^[27,28]

Because infants with NE are at risk for developing seizures, monitoring for seizure activity is imperative to prompt treatment. Stiffening, rhythmic movements of one or more extremity, and repetitive sucking or extension of the tongue are signs of seizure activity. Amplitude-integrated EEG (aEEG) is a newer, continuous brain monitoring technique that is useful for detecting seizures, especially in sedated infants.[37] Since aEEG is a bedside device that may be more widely available than continuous full channel array video EEG, it allows nurses to play a critical role in brain monitoring.[37,38] Routine use in asphyxiated and cooled infants has become more common because training is simple and users are able to easily detect abnormal activity.[37] Subsequent early detection and notification of the provider allow for early treatment of seizures, which can lead to improved outcomes.^[38]

Continuous assessment of skin and monitoring for pain are also important during TH. Skin complications such as subcutaneous fat necrosis and cold panniculitis may occur.[39] Each present with painful, reddened or bluish-purple indurated areas over the skin that was in contact with the cooling device. Periodic mobilization and turning, if tolerated by the infant, can prevent these complications.[39] Additional treatment may be required if skin complications occur.

The infant's level of stress and agitation is greatly affected by his or her comfort level. Additionally, excess agitation or activity due to pain can lead to increased temperatures and difficulty reaching the target temperature.[28] Pain scores should be performed routinely so that appropriate interventions can be implemented. However, infants undergoing TH are frequently treated with anti-epileptics, which are sedating, or are otherwise sedated and ventilated, so pain scoring may not be a reliable measure. In this case, monitoring vital signs closely for increases in heart rate and blood pressure can help in assessment of pain.[40] Neither the NICHD Whole Body Hypothermia trial, nor the TOBY, ICE, or CoolCap trials included sedation or pain control as part of the study intervention. There is no standard or recommended clinical approach to use of sedatives or pain control for infants with NE undergoing cooling.^[36]

Rewarming

Rewarming begins after 72 hours of cooling. The automatic mode of the cooling device will be changed to allow a gradual increase in temperature of 0.5 °C per hour until a core temperature of 36.5 °C is reached.[27,28,34] The device will then be used to hold the infant's temperature at 36.5 °C for 24 hours.[32,33] Serial monitoring of core temperature during this time is imperative to prevent overheating. Hyperthermia is associated with poorer outcomes in infants that experience NE.^[41]

Parental Support

Throughout the cooling process, nurses also have the responsibility of supporting parents through a scary and unpredictable situation. Parents may have feelings of helplessness, fear, and uncertainty when their newborn is critically ill.[27,28,42] They also feel a loss of control since the care of the infant is transferred to the nurses and healthcare providers. Nursing staff can support parents by offering education, opportunities to participate in care, and limiting separation when possible.^[42]

Conclusion

NE is a condition that has severe consequences when left untreated. Over the past few decades, TH has emerged as a useful option for treating NE. Research is still needed to increase understanding of the NE disease process, identify adjunct treatments, and develop more precise ways of predicting outcomes. As the process of TH is perfected, the use of this treatment will spread more rapidly and more infants will benefit from its use.

References

1. Kurinczuk J, et al. Early Hum Dev. 2010;86:329–38,http://dx.doi.org/10.1016/j.earlyhumdev.2010.05.010.

2. Ferriero DM. N Engl J Med. 2004;351:1985–95.

3. Tan S, Wu Y. Retrieved from http://www.uptodate.com/contents/etiology-and-pathogenesis-of-neonatal-encephalopathy].

4. Cotton M, Shankaran S. Expert Rev Obstet Gynecol. 2010;5:227–39, http://dx.doi.org/10.1586/eog.10.7.

5. Johnston MV, et al. Pediatr Res. 2001;49:735–41.

6. Volpe J. Ment Retard DevDisabil Res Rev. 2001;7:56–64.

7. Lai M, Yang S. J Biomed Biotech. 2011:1–6, http://dx.doi.org/10.1155/2011/609813.

8. Lust WD, et al. Metab Brain Dis. 2002;17:113–21.

9. Badr LK, Purdy I. J Perinat Neonatal Nurs. 2006;2:163–75.

10. Higgis RD, et al. J Pediatr. 2011;159:851–8, http://dx.doi.org/10.1016/j.jpeds.2011.08.004.

11. Gunn AJ, Gluckman PD, Gunn TR.Pediatrics. 1998;102:885–92.

12. Gluckman PD, et al. Lancet. 2005;365:663–70.

13. Azzopardi DV, et al. N Engl J Med. 2009;361:1349–58.

14. Simbrunner G, et al. Pediatrics. 2010;126:771–8,http://dx.doi.org/10.1542/peds.2009-2441.

15. Jacobs SE, et al. Arch PediatrAdolesc Med.2011;165:692–700.

16. Edwards AD, et al. Br J Med. 2010;340:1–7, http://dx.doi.org/10.1136/bmj.c363.

17. Guillet R, et al. Pediatr Res. 2012;71:205–9, http://dx.doi.org/10.1038/pr.2011.30.

18. Shah PS. Semin Fetal Neonatal Med. 2010;15:238–46, http://dx.doi.org/10.1016/j.siny.2010.02.003.

19. Tagin MA, et al. Arch PediatrAdolesc Med. 2012;166:558–66,http://dx.doi.org/10.1001/archpediatrics.2011.1772.

20. Shankaran S, et al. N Engl J Med. 2012;366:2085–92, http://dx.doi.org/10.1056/NEJMoa1112066.

21. Shankaran S, et al. Arch Dis Child. 2011;97:398–404, http://dx.doi.org/10.1036/archdischild-2011-301524.

22. Rutherford M, et al. Lancet Neurol. 2010;9:39–45, http://dx.doi.org/10.1016/S1474-422(09)70295-9.

23. Cheong JL, et al.Arch PediatrAdolesc Med. 2012;166:634–40,http://dx.doi.org/10.1001/archpediatrics.2012.284.

24. Okeraefor A, et al. Pediatrics. 2008;121:906–14, http://dx.doi.org/10.1452/peds.2007-0770.

25. Tusor N, et al. Pediatr Res. 2012;72:63–9, http://dx.doi.org/10.1542/peds.2007-0770.

26. Shankaran S, et al. J Pediatr. 2012;160:567–72,http://dx.doi.org/10.1016/j.peds.2011.09.018.

27. Merill L. NursWomens Health. 2012;16:126–34, http://dx.doi.org/10.1111/j.1751-486X.2012.01718.x.

28. Selway LD. Adv Neonatal Care. 2010;10:60–6, http://dx.doi.org/10.1097/ANC.0b013e3181d54b30

29. Kendall GS, et al. Arch Dis Child Fetal Neonatal Ed. 2010;95:408–12,http://dx.doi.org/10.1136/adc.2010.187211.

30. Anderson ME, et al. J Perinatol. 2007;27:592–3, http://dx.doi.org/10.1038/sj.jp.7211781.

31. Hallberg B, et al. ActaPaediatr(Stockholm). 2009;98:942–6.

32. Duke ICN Clinical Practice Council. Hypothermia/controlled core temperature: care of the patient. 2013.

33. Vanderbilt University Medical Center. [Retrieved from http://clinicaltrials.gov/show/NCT00620711].

34. Higgins RD, Shankaran S. Early Hum Dev. 2009;85:49–52, http://dx.doi.org/10.1016/j.earlyhumdev.2009.08.015.

35. SpitzmillerRE, et al. J Child Neurol. 2009;22:1069–78, http://dx.doi.org/10.1177/0883073807306258.

36. Shankaran S, et al. N Engl J Med. 2005;253:1574–84, http://dx.doi.org/10.1056/NEJMcps050929.

37. Tao JD, Mathur AM. J Perinatol. 2010;30:73–81, http://dx.doi.org/10.1038/jp.2010.93.

38. Bonifaciao SL, et al. Nat Rev Neurol. 2011;7:485–94, http://dx.doi.org/10.1038/nrneurol.2011.119.

39. Filippi L, et al. J Matern Fetal Neonatal Med. 2012;25:2115–21,http://dx.doi.org/10.3109/14767058.2012.683898.

40. Chirinian N, Mann N. Crit Care Nurs. 2011;31:e1, http://dx.doi.org/10.4037/ccn2011873.

41. Laptook A, et al. Pediatrics. 2008;122:491–9.

42. Nassef SK, et al. J ObstetGynecol Neonatal Nurs. 2012;42:38–47, http://dx.doi.org/10.1111/j.1552-6909.2012.01429.x.y"