Outcomes of Hypoxic-Ischemic E n c e p h a l o p a t h y in N e o n a t e s Treated with Hypothermia Seetha Shankaran,
MD
KEYWORDS Encephalopathy Neonatal Hypothermia Outcome Predictors KEY POINTS Hypothermia for neonatal hypoxic-ischemic encephalopathy is safe and effective in reducing death and disability at 18 months of age. This neuroprotection continues into childhood, as demonstrated by a reduction in mortality and major disability at 6 to 7 years of age. Outcome at 18 months of age is a good predictor of childhood outcome.
OUTCOMES OF CHILDREN WITH BIRTH DEPRESSION/ENCEPHALOPATHY BEFORE HYPOTHERMIA THERAPY
Before the advent of neuroprotective therapy with hypothermia, studies evaluating outcomes of children born at term with birth depression or encephalopathy were generally cohort studies, each having unique inclusion criteria, evaluation methods, and duration of follow-up. Among these studies, the outcome of children with acute perinatal asphyxia and/or neonatal encephalopathy was a disability rate of 6% to 21% in children with moderate encephalopathy and 42% to 100% in those with severe encephalopathy.1–5 Among nondisabled children who were able to undergo psychometric and behavioral testing, lower executive function and delays in school readiness (reading, spelling, and arithmetic) and lower scores in language, memory, and sensorimotor perception have been noted.1,3,6 The possibility of an increase in the rate of symptoms associated with attention-deficit/hyperactivity disorder, including anxiety, depression, attention regulation, time perception, and thought problems, as well as an increased risk of autism has been noted in children with encephalopathy or depression at birth.6–9 A recent systematic review of long-term neurodevelopmental outcomes after intrauterine and neonatal insults, especially in low- and middle-income countries (LMIC), noted that the 27 studies evaluating 2708 infants reported sequelae
The author has no financial disclosures. Division of Neonatal/Perinatal Medicine, Children’s Hospital of Michigan, 3901 Beaubien, Detroit, MI, USA E-mail address: [email protected] Clin Perinatol 41 (2014) 149–159 http://dx.doi.org/10.1016/j.clp.2013.10.008 perinatology.theclinics.com 0095-5108/14/$ – see front matter Ó 2014 Elsevier Inc. All rights reserved.
150
Shankaran
in 1002 (37%) children. Cognitive, general developmental delay or learning difficulties were noted in 45%, cerebral palsy (CP) in 29%, deafness/hearing loss in 9%, impaired vision/blindness in 26%, gross motor coordination problems in 17%, epilepsy in 12%, and behavioral problems in 1% of children. Multiple impairments were reported in 20.5% of children while at least 1 impairment was noted in 44%, and severe sequelae in at least 1 domain in 27%.10 OUTCOMES OF CHILDREN WITH MODERATE OR SEVERE ENCEPHALOPATHY AT 18 MONTHS OF AGE FOLLOWING HYPOTHERMIA THERAPY
All of the randomized controlled trials (RCTs) of neuroprotection with hypothermia for neonatal hypoxic-ischemic encephalopathy (HIE) have had very similar inclusion criteria (36 weeks gestational age, severe acidosis or birth depression, moderate or severe encephalopathy with or without an abnormal amplitude-integrated electroencephalogram [aEEG]) and cooling initiated within 6 hours of age. The exclusion criteria were infants of age greater than 6 hours, those with major congenital or chromosomal abnormalities, or those with severe intrauterine growth restriction. The neurodevelopmental assessment tools evaluating 18-month outcome were similar and comparable between trials. Details of the results of the primary outcomes, components of the primary outcome, and predefined secondary outcomes are summarized herein. Predictors of 18-month outcomes are also presented, as this information may be helpful in discussions with families of infants with HIE. CoolCap Trial
The CoolCap Trial was the first large RCT of hypothermia to be published.11 Selective head cooling to a target temperature of 34 to 35 for 72 hours, or intensive care alone following moderate or severe encephalopathy and an abnormal aEEG or seizures, was compared among 234 neonates, 116 infants assigned to cooling and 118 to conventional care. Eight infants were lost to follow-up in each group. The primary outcome was mortality or severe neurodevelopmental disability at 18 months. Severe disability was defined as gross motor function level (GMF) of 3 to 5, Bayley Scales of Infant Development (BSID II) with either Mental Development Index (MDI) or a Psychomotor Development Index (PDI) less than 70 (2 standard deviations [SD] below normal), or bilateral cortical visual impairment. Nine survivors did not have MDI scores and 4 of 9 were missing visual outcome data, but all had GMF scores of 3 or higher. Death or severe disability was noted in 59 of 108 (55%) cooled infants versus 73 of 110 (66%) control-group infants, odds ratio (OR) 0.61 (95% confidence interval [CI] 0.34–1.09); with mortality rates of 33% versus 38%, OR 0.81 (0.47–1.41) and severe disability in 14 of 72 (19%) versus 21 of 68 (31%), OR 0.54 (0.25–1.17) in the cooled and control groups, respectively. A Bayley MDI score of lower than 70 occurred in 21 of 70 (30%) versus 24 of 61 (39%), and visual impairment in 7 of 72 (10%) versus 11 of 64 (17%) cooled and control infants (P not significant). Predictors of 18-month outcome in the CoolCap RCT12: 1. The primary outcome was lower among infants who received hypothermia. 2. Primary outcome was also lower among infants with a less severe aEEG pattern at study intervention. 3. A worse outcome was noted among infants with severe HIE in comparison with infants with moderate HIE. 4. The absence of seizures by aEEG was associated with better outcome. 5. Elevated temperature in control-group infants was associated with worse outcomes.
Outcomes of Hypoxic-Ischemic Encephalopathy
6. Age at randomization of study infants within 6 hours did not influence the outcome. 7. Hypothermia altered the accuracy of the neuroexamination after cooling.13 The National Institute of Child Health and Human Development (NICHD) RCT
Whole-body hypothermia to a target temperature of 33 to 34 C or usual care was assigned randomly following moderate or severe HIE.14 There were 102 infants in the hypothermia group and 106 in the control group. The primary outcome was death or moderate or severe disability at 18 months. Severe disability was defined as Bayley II MDI less than 70, Gross Motor Function Classification System (GMFCS) level 3 to 5, or hearing impairment requiring aids or blindness. Moderate disability was defined as MDI 70 to 84 and GMFCS level 2, hearing impairment without amplification, or a persistent seizure disorder. Data were unavailable for 3 infants. The primary outcome was seen in 45 of 102 (44%) hypothermia and 64 of 103 (62%) control infants. Risk ratio (RR) adjusted for center was 0.72 (95% CI 0.54–0.95). Death occurred in 24 of 102 (24%) and 38 of 106 (37%) infants, RR 0.68 (0.44–1.05), in the hypothermia and control groups, respectively. Moderate/severe CP occurred in 15 of 78 (19%) and 19 of 68 (30%), blindness in 7% and 14%, and hearing impairment in 4% and 6% of infants in the hypothermia and control groups, respectively. Moderate disability occurred in 3 infants, 2 in the hypothermia group and 1 in the control group.15 The treatment was reported to be safe, and was associated with a consistent trend for decreasing frequency in each of the components of disability.15 Predictors of 18-month outcome in the NICHD whole-body hypothermia trial with secondary analyses of data from the NICHD trial data have found the following: 1. An elevated temperature (in the control group of infants) was associated with an increase in the risk of death or disability at 18 months.16 2. An elevated urinary lactate to creatinine ratio was associated with poor outcome.17 3. Classification and regression-tree analysis revealed that cord pH, spontaneous activity of the infants, base deficit of the first postnatal blood gas, and the level of partial pressure of carbon dioxide (PCO2) in the cord gas was useful in predicting outcome.18 4. Death or moderate severe HIE was noted to be common among infants with a persistently low 10-minute Apgar score; however, not all infants with an Apgar score lower than 3 at 10 minutes had a uniformly poor outcome.19 5. Persistence of severe encephalopathy at 72 hours increased the risk of death or disability, as did the presence of an abnormal neurologic examination at neonatal hospital discharge.20 6. Clinical seizures were not associated with outcome when adjusted for confounding variables.21 7. Location of birth (inborn vs outborn) did not influence outcome.22 8. Smaller, sicker infants had more decreases in temperature, both during induction and maintenance phase of cooling.23 9. Time to initiation of cooling did not influence outcome.15 10. In a subset of trial infants and prospectively cooled infants (after trial accrual closed), the aEEG was noted to add minimally to predictive value of stage of encephalopathy.24 11. The administration of phenobarbital for seizures before the onset of initiation of cooling was associated with a lower temperature during the induction phase of cooling.25
151
152
Shankaran
12. Both minimum PCO2 and cumulative PCO2 lower than 35 mm Hg from birth to 12 hours of age were associated with poor outcome.26 13. The 18- to 22-month outcome was not associated with the APOE genotype among survivors.27 TOBY Trial
The TOBY Trial inclusion criteria involved both the presence of moderate or severe HIE and an abnormal aEEG.28 Whole-body hypothermia to a target of 33.5 C for 72 hours or intensive care alone was compared between the 2 study arms. Primary outcome was death or severe disability, defined as BSID II MDI lower than 70, GMFCS level 3 to 5, or bilateral cortical visual impairment. The study is the largest to date, with 325 (163 hypothermia and 162 control) infants enrolled. In both groups, 1 infant was lost to follow-up. The primary outcome was noted in 74 of 163 (45%) cooled versus 86 of 162 (53%) control, RR 0.86 (95% CI 0.68–1.07). Results were unchanged when adjusted for severity of aEEG background, sex, or age at randomization. Survivors with severe disability were 32 of 120 (27%) hypothermia and 42 of 117 (36%) control, RR 0.74 (0.51–1.09). The predefined secondary outcome of survival free of neurologic abnormalities was 71 of 163 (44%) hypothermia and 45 of 162 (28%) control, RR 1.57 (1.16–2.12). Among hypothermia and control-group infants, higher rates of BSID MDI of 85 or higher (70% vs 55%) and PDI of 85 or higher (68% vs 53%), normal neurologic examination (71% vs 54%), and lower rates of CP (28 vs 41%) and multiple disabilities (19% and 30%) were noted in the hypothermia group (all P
MD
KEYWORDS Encephalopathy Neonatal Hypothermia Outcome Predictors KEY POINTS Hypothermia for neonatal hypoxic-ischemic encephalopathy is safe and effective in reducing death and disability at 18 months of age. This neuroprotection continues into childhood, as demonstrated by a reduction in mortality and major disability at 6 to 7 years of age. Outcome at 18 months of age is a good predictor of childhood outcome.
OUTCOMES OF CHILDREN WITH BIRTH DEPRESSION/ENCEPHALOPATHY BEFORE HYPOTHERMIA THERAPY
Before the advent of neuroprotective therapy with hypothermia, studies evaluating outcomes of children born at term with birth depression or encephalopathy were generally cohort studies, each having unique inclusion criteria, evaluation methods, and duration of follow-up. Among these studies, the outcome of children with acute perinatal asphyxia and/or neonatal encephalopathy was a disability rate of 6% to 21% in children with moderate encephalopathy and 42% to 100% in those with severe encephalopathy.1–5 Among nondisabled children who were able to undergo psychometric and behavioral testing, lower executive function and delays in school readiness (reading, spelling, and arithmetic) and lower scores in language, memory, and sensorimotor perception have been noted.1,3,6 The possibility of an increase in the rate of symptoms associated with attention-deficit/hyperactivity disorder, including anxiety, depression, attention regulation, time perception, and thought problems, as well as an increased risk of autism has been noted in children with encephalopathy or depression at birth.6–9 A recent systematic review of long-term neurodevelopmental outcomes after intrauterine and neonatal insults, especially in low- and middle-income countries (LMIC), noted that the 27 studies evaluating 2708 infants reported sequelae
The author has no financial disclosures. Division of Neonatal/Perinatal Medicine, Children’s Hospital of Michigan, 3901 Beaubien, Detroit, MI, USA E-mail address: [email protected] Clin Perinatol 41 (2014) 149–159 http://dx.doi.org/10.1016/j.clp.2013.10.008 perinatology.theclinics.com 0095-5108/14/$ – see front matter Ó 2014 Elsevier Inc. All rights reserved.
150
Shankaran
in 1002 (37%) children. Cognitive, general developmental delay or learning difficulties were noted in 45%, cerebral palsy (CP) in 29%, deafness/hearing loss in 9%, impaired vision/blindness in 26%, gross motor coordination problems in 17%, epilepsy in 12%, and behavioral problems in 1% of children. Multiple impairments were reported in 20.5% of children while at least 1 impairment was noted in 44%, and severe sequelae in at least 1 domain in 27%.10 OUTCOMES OF CHILDREN WITH MODERATE OR SEVERE ENCEPHALOPATHY AT 18 MONTHS OF AGE FOLLOWING HYPOTHERMIA THERAPY
All of the randomized controlled trials (RCTs) of neuroprotection with hypothermia for neonatal hypoxic-ischemic encephalopathy (HIE) have had very similar inclusion criteria (36 weeks gestational age, severe acidosis or birth depression, moderate or severe encephalopathy with or without an abnormal amplitude-integrated electroencephalogram [aEEG]) and cooling initiated within 6 hours of age. The exclusion criteria were infants of age greater than 6 hours, those with major congenital or chromosomal abnormalities, or those with severe intrauterine growth restriction. The neurodevelopmental assessment tools evaluating 18-month outcome were similar and comparable between trials. Details of the results of the primary outcomes, components of the primary outcome, and predefined secondary outcomes are summarized herein. Predictors of 18-month outcomes are also presented, as this information may be helpful in discussions with families of infants with HIE. CoolCap Trial
The CoolCap Trial was the first large RCT of hypothermia to be published.11 Selective head cooling to a target temperature of 34 to 35 for 72 hours, or intensive care alone following moderate or severe encephalopathy and an abnormal aEEG or seizures, was compared among 234 neonates, 116 infants assigned to cooling and 118 to conventional care. Eight infants were lost to follow-up in each group. The primary outcome was mortality or severe neurodevelopmental disability at 18 months. Severe disability was defined as gross motor function level (GMF) of 3 to 5, Bayley Scales of Infant Development (BSID II) with either Mental Development Index (MDI) or a Psychomotor Development Index (PDI) less than 70 (2 standard deviations [SD] below normal), or bilateral cortical visual impairment. Nine survivors did not have MDI scores and 4 of 9 were missing visual outcome data, but all had GMF scores of 3 or higher. Death or severe disability was noted in 59 of 108 (55%) cooled infants versus 73 of 110 (66%) control-group infants, odds ratio (OR) 0.61 (95% confidence interval [CI] 0.34–1.09); with mortality rates of 33% versus 38%, OR 0.81 (0.47–1.41) and severe disability in 14 of 72 (19%) versus 21 of 68 (31%), OR 0.54 (0.25–1.17) in the cooled and control groups, respectively. A Bayley MDI score of lower than 70 occurred in 21 of 70 (30%) versus 24 of 61 (39%), and visual impairment in 7 of 72 (10%) versus 11 of 64 (17%) cooled and control infants (P not significant). Predictors of 18-month outcome in the CoolCap RCT12: 1. The primary outcome was lower among infants who received hypothermia. 2. Primary outcome was also lower among infants with a less severe aEEG pattern at study intervention. 3. A worse outcome was noted among infants with severe HIE in comparison with infants with moderate HIE. 4. The absence of seizures by aEEG was associated with better outcome. 5. Elevated temperature in control-group infants was associated with worse outcomes.
Outcomes of Hypoxic-Ischemic Encephalopathy
6. Age at randomization of study infants within 6 hours did not influence the outcome. 7. Hypothermia altered the accuracy of the neuroexamination after cooling.13 The National Institute of Child Health and Human Development (NICHD) RCT
Whole-body hypothermia to a target temperature of 33 to 34 C or usual care was assigned randomly following moderate or severe HIE.14 There were 102 infants in the hypothermia group and 106 in the control group. The primary outcome was death or moderate or severe disability at 18 months. Severe disability was defined as Bayley II MDI less than 70, Gross Motor Function Classification System (GMFCS) level 3 to 5, or hearing impairment requiring aids or blindness. Moderate disability was defined as MDI 70 to 84 and GMFCS level 2, hearing impairment without amplification, or a persistent seizure disorder. Data were unavailable for 3 infants. The primary outcome was seen in 45 of 102 (44%) hypothermia and 64 of 103 (62%) control infants. Risk ratio (RR) adjusted for center was 0.72 (95% CI 0.54–0.95). Death occurred in 24 of 102 (24%) and 38 of 106 (37%) infants, RR 0.68 (0.44–1.05), in the hypothermia and control groups, respectively. Moderate/severe CP occurred in 15 of 78 (19%) and 19 of 68 (30%), blindness in 7% and 14%, and hearing impairment in 4% and 6% of infants in the hypothermia and control groups, respectively. Moderate disability occurred in 3 infants, 2 in the hypothermia group and 1 in the control group.15 The treatment was reported to be safe, and was associated with a consistent trend for decreasing frequency in each of the components of disability.15 Predictors of 18-month outcome in the NICHD whole-body hypothermia trial with secondary analyses of data from the NICHD trial data have found the following: 1. An elevated temperature (in the control group of infants) was associated with an increase in the risk of death or disability at 18 months.16 2. An elevated urinary lactate to creatinine ratio was associated with poor outcome.17 3. Classification and regression-tree analysis revealed that cord pH, spontaneous activity of the infants, base deficit of the first postnatal blood gas, and the level of partial pressure of carbon dioxide (PCO2) in the cord gas was useful in predicting outcome.18 4. Death or moderate severe HIE was noted to be common among infants with a persistently low 10-minute Apgar score; however, not all infants with an Apgar score lower than 3 at 10 minutes had a uniformly poor outcome.19 5. Persistence of severe encephalopathy at 72 hours increased the risk of death or disability, as did the presence of an abnormal neurologic examination at neonatal hospital discharge.20 6. Clinical seizures were not associated with outcome when adjusted for confounding variables.21 7. Location of birth (inborn vs outborn) did not influence outcome.22 8. Smaller, sicker infants had more decreases in temperature, both during induction and maintenance phase of cooling.23 9. Time to initiation of cooling did not influence outcome.15 10. In a subset of trial infants and prospectively cooled infants (after trial accrual closed), the aEEG was noted to add minimally to predictive value of stage of encephalopathy.24 11. The administration of phenobarbital for seizures before the onset of initiation of cooling was associated with a lower temperature during the induction phase of cooling.25
151
152
Shankaran
12. Both minimum PCO2 and cumulative PCO2 lower than 35 mm Hg from birth to 12 hours of age were associated with poor outcome.26 13. The 18- to 22-month outcome was not associated with the APOE genotype among survivors.27 TOBY Trial
The TOBY Trial inclusion criteria involved both the presence of moderate or severe HIE and an abnormal aEEG.28 Whole-body hypothermia to a target of 33.5 C for 72 hours or intensive care alone was compared between the 2 study arms. Primary outcome was death or severe disability, defined as BSID II MDI lower than 70, GMFCS level 3 to 5, or bilateral cortical visual impairment. The study is the largest to date, with 325 (163 hypothermia and 162 control) infants enrolled. In both groups, 1 infant was lost to follow-up. The primary outcome was noted in 74 of 163 (45%) cooled versus 86 of 162 (53%) control, RR 0.86 (95% CI 0.68–1.07). Results were unchanged when adjusted for severity of aEEG background, sex, or age at randomization. Survivors with severe disability were 32 of 120 (27%) hypothermia and 42 of 117 (36%) control, RR 0.74 (0.51–1.09). The predefined secondary outcome of survival free of neurologic abnormalities was 71 of 163 (44%) hypothermia and 45 of 162 (28%) control, RR 1.57 (1.16–2.12). Among hypothermia and control-group infants, higher rates of BSID MDI of 85 or higher (70% vs 55%) and PDI of 85 or higher (68% vs 53%), normal neurologic examination (71% vs 54%), and lower rates of CP (28 vs 41%) and multiple disabilities (19% and 30%) were noted in the hypothermia group (all P
