Cerebral palsy after neonatal encephalopathy: do neonates with suspected asphyxia have worse outcomes?

Transcription

1 DEVELOPMENTAL MEDICINE & CHILD NEUROLOGY ORIGINAL ARTICLE Cerebral palsy after neonatal encephalopathy: do neonates with suspected have worse outcomes? JARRED GARFINKLE 1 PIA WINTERMARK 1,2 MICHAEL I SHEVELL 1,3,4 MARYAM OSKOUI 1,3,4 ON BEHALF OF THE CANADIAN CEREBRAL PALSY REGISTRY* 1 Department of Pediatrics, Montreal Children s Hospital-McGill University Health Center, Montreal, QC; 2 Division of Neonatology, Montreal Children s Hospital- McGill University Health Center, Montreal, QC; 3 Department of Neurology/Neurosurgery, Montreal Children s Hospital-McGill University Health Center, Montreal, QC; 4 Division of Pediatric Neurology, Montreal Children s Hospital-McGill University Health Center, Montreal, QC, Canada. Correspondence to Maryam Oskoui at Division of Pediatric Neurology, Montreal Children s Hospital, 2300 Tupper Street, A-512, Montreal, QC H3H 1P3, Canada. maryam.oskoui@mcgill.ca *Canadian Cerebral Palsy Registry members are listed in the Acknowledgements. This article is commented on by Marret on pages of this issue. PUBLICATION DATA Accepted for publication 15th September Published online 11th November ABBREVIATIONS CCPR Canadian Cerebral Palsy Registry REPACQ Registre de la Paralysie Cerebrale au Quebec AIM We sought to investigate how brain injury and severity, and neurological subtype of cerebral palsy (CP) differed in term-born children with CP after neonatal encephalopathy, between those with suspected birth and those without. METHOD Using the Canadian CP Registry, which included 1001 children, those with CP born at 36wks after moderate or severe neonatal encephalopathy, were dichotomized according to the presence or absence of suspected birth. Gross Motor Function Classification System (GMFCS) scores, neurological subtypes, comorbidities, and magnetic resonance imaging findings were compared. RESULTS Of the 147 term-born children with CP (82 males, 65 females; median age 37 months, interquartile range [IQR] ) who after moderate or severe neonatal encephalopathy had the required outcome data, 61 (41%) met criteria for suspected birth. They had a higher frequency of non-ambulatory GMFCS status (odds ratio [OR] 3.4, 95% confidence interval [CI] ), spastic quadriplegia (OR 2.8, 95% CI ), nonverbal communication skills impairment (OR 4.2, 95% CI ), isolated deep grey matter injury (OR 4.1, 95% CI ), a lower frequency of spastic hemiplegia (OR 0.17, 95% CI ), focal injury (OR 0.20; 95% CI ), and more comorbidities (p=0.017) than those who did not meet criteria. INTERPRETATION Term-born children who develop CP after neonatal encephalopathy with suspected birth have a greater burden of disability than those without suspected birth. Cerebral palsy (CP) is a disorder acquired early in life and represents the most common physical disability of childhood. 1 The most widely accepted consensus definition explicitly recognizes the diversity of CP. 2 Medical disorders (e.g. epilepsy), sensory deficits (e.g. visual), and other non-motor developmental deficits (e.g. intellectual disability) are frequently associated with CP. 3 Unfortunately, the prevalence of CP in term-born children has not declined in recent decades and there has been a trend towards the more devastating spastic quadriplegic and dyskinetic subtypes. 4 One-quarter of CP in term-born children is estimated to occur after moderate or severe neonatal encephalopathy. 5,6 Moreover, term-born children with CP after neonatal encephalopathy have more debilitating features than those without preceding neonatal encephalopathy. 6,7 It is now well established that therapeutic hypothermia initiated within the first 6 hours of life can prevent brain injury in neonates with neonatal encephalopathy in the context of suspected birth. 8 The current clinical criteria required to qualify for hypothermia are designed to exclude neonates with neonatal encephalopathy (and later CP) of likely non-l origin. 8 In addition, neonates with suspected birth are known to be particularly susceptible to injury to areas of the brain that are associated with later neuromotor development. 9 We thus hypothesize that term-born children with CP after neonatal encephalopathy with suspected birth have a greater proportion of deep grey matter injury and non-ambulatory and spastic quadriplegic CP, as well as a greater numerical burden of comorbidities than those without suspected birth. The differences in 2015 Mac Keith Press DOI: /dmcn

2 severity and neurological subtype of CP, as well as the associated disabilities between these two subgroups have not yet been examined and doing so will assist us in such domains as counseling, prognostication, and selective interventional program emphasis. In term-born children with CP after neonatal encephalopathy, we therefore aimed to investigate how the brain injury and severity, and neurological subtype of CP differ between those with suspected birth and those without. METHOD The study was conducted using data extracted from the Canadian Cerebral Palsy Registry (CCPR), which builds on the earlier development and implementation of the Registre de la Paralysie Cerebrale au Quebec (REPACQ). The CCPR captures cases of CP identified through both pediatric rehabilitation centers and university hospitals where provincial pediatric neurology and developmental pediatric services are located, from the birth year of It covers most of Quebec, the greater Toronto area of Ontario, and the entire provinces of British Columbia, Alberta, Nova Scotia, and Newfoundland. This includes more than half (approximately 18 million individuals) of the Canadian population. Once cases are identified, parental consent is obtained and the maternal medical and obstetric records, as well as the child s neonatal, medical, and rehabilitation records are reviewed. These data are supplemented by a standardized parental interview and documented physical examination of the child by a pediatric neurologist, developmental pediatrician, or pediatric physiatrist. Local Ethics Board approval was obtained from each participating institution. The Montreal Children s Hospital-McGill University Health Center Research Ethics Board provided central approval for data storage, analysis, and overall operations. To be enrolled in the CCPR, a child must be at least 2 years of age and meet current diagnostic consensus criteria for CP, which include a clinical diagnosis of a non-progressive motor impairment resulting from a presumably early insult to the developing brain. 2 The diagnosis was confirmed when possible at 5 years of age and the children without CP were removed from the registry. 10 Patients within the CCPR included for analysis in this study were born at 36 weeks or more; had a birthweight of 1800g or more; and fulfilled criteria for moderate or severe neonatal encephalopathy during the first week of life. We then categorized the neonates according to the presence or absence of suspected birth. Criteria for neonatal encephalopathy were derived from the Sarnat score. 11 Neonates with moderate neonatal encephalopathy either manifested seizures, were lethargic, were hypotonic, or had depressed peripheral reflexes. Those with severe neonatal encephalopathy manifested either a flaccid coma, brainstem findings, or difficult to control seizures. Suspected birth was defined according to specific signs of both fetal and neonatal distress that were What this paper adds Children with cerebral palsy after neonatal encephalopathy with suspected birth were compared to those without suspected birth. They are three times more likely to have a non-ambulatory Gross Motor Function Classification System status and/or spastic quadriplegia. They are four times more likely to have non-verbal communication skills impairment. They more frequently have isolated deep grey matter injury. derived from the National Institute of Child Health and Human Development randomized controlled trial of therapeutic hypothermia and the Position Statement of the Canadian Paediatric Society, as previously described. 8,12,13 Fetal distress was evident by at least one of the following: (1) sentinel event (placenta abruptio, cord accident, uterine rupture, or shoulder dystocia); (2) ph 7.0 or base deficit 16mEq/L from the umbilical cord or blood collected within the first hour of life; or (3) severe fetal heart rate abnormalities (e.g. late decelerations, persistent bradycardia). Neonatal distress was evident by at least one of the following: Apgar score 5 at 10 minutes, or caseroom resuscitation (intubation, cardiac massage, or administration of epinephrine or normal saline bolus). For the purpose of group assignment, when the 10-minute Apgar score or ph were not documented in the records, they were assumed to have been normal, and when sentinel events, severe fetal heart rate abnormalities, and caseroom resuscitation were not documented in the records, they were assumed to have been absent. The subtype of CP was classified according to the quality and topographic pattern of motor impairment last documented by an objective assessment employing the following clinical scheme: (1) spastic quadriplegic, spastic hemiplegic, spastic diplegic, or spastic triplegic; (2) dyskinetic (athetosis, chorea, or dystonia); and (3) ataxic-hypotonic. Those with a mixed profile of dyskinetic and spastic features were categorized as dyskinetic. Further details are provided elsewhere. 14 The last objective assessment was used to assign functional mobility according to the Gross Motor Function Classification System (GMFCS). 15 Independent ambulation, without (levels I and II) and with assistance (level III), was distinguished from non-ambulation (levels IV and V). We excluded from data analysis children who lacked information on either CP subtype or GMFCS level. Information on several comorbidities was systematically collected in the registry: severe cortical visual impairment, substantial auditory impairment, non-verbal communication skills impairment, gavage feeding status, and epilepsy. The age of the children (between 2y and 5y) precluded reliable assessment of possible cognitive impairment. Lack of access to psychiatric information precluded data collection regarding associated behavioral disorders. Cortical visual impairment required diagnosis by an ophthalmologist and included homonymous hemianopsia. Substantial auditory impairment was defined as a 70dB or greater hearing loss (bilateral) on audiometric testing. Non-verbal communication skills impairment referred to the absence 190 Developmental Medicine & Child Neurology 2016, 58:

3 of specific words or recognizable vocabulary regardless of possible etiology (i.e. motor or cognitive limitations). Gavage feeding referred to the use of a temporary or permanent artificial tube to administer the bulk of a child s nutrition. Epilepsy was defined as the occurrence of afebrile seizures in the 12 months preceding the last Registry inscription. Clinical descriptions of the magnetic resonance imaging (MRI) studies were extracted from the data set. If more than one MRI was performed on a particular child, only the last performed study was registered in the data set. MRIs were originally interpreted by university-based pediatric radiologists. MRI findings were classified into one of seven categories by two of the authors (JG and PW), who were blinded to clinical information. The classification system was adapted from one previously developed. 16 Each mutually exclusive MRI finding is described in Table SI (online supporting information). Differences in classification were discussed and resolved by consensus, and when there was discordance, the more severe category was assigned. Statistical analysis was performed with SPSS 20.0 (Chicago, IL). CP severity and neurological subtype and comorbidities were compared between those with and without suspected birth using the v 2 test or twosided Fisher s exact test for univariable analysis of categorical variables as appropriate and the Student s t-test or Wilcoxon Mann Whitney U test for continuous variables as appropriate. A p-value of <0.05 was chosen a priori to denote significance. RESULTS As of July 2013, a total of 1001 patients were enrolled into the CCPR, all of whom were born between 1999 and June Of them, 543 were born at 36 weeks or more and had a birthweight of 1800g or more. Of these children, 155 (29%) had a documented diagnosis of moderate or severe neonatal encephalopathy and 388 (71%) did not (seven had mild neonatal encephalopathy, 354 had no neonatal encephalopathy, and 27 did not have written documentation of a neurological examination during the neonatal period). Concurrent information on CP subtype and GMFCS score was available in 147 (95%) of these children (82 males and 65 females) and it is this group that comprises the study cohort. Sixty-one (41%) of those with CP after neonatal encephalopathy met criteria for suspected birth as outlined above while 86 (59%) did not (Fig. 1 and Table I). Cord or first hour of life ph was available in 59 out of 61 with suspected birth and 60 out of 86 neonates in the group without. Additionally, the 10-minute Apgar score was available in 53 out of 61 with suspected birth and 49 out of 86 neonates in the group without. In 38 out of 45 cases (84%) in which the 10-minute Apgar score was unavailable, the 5-minute Apgar score was >5. Ten neonates with neonatal encephalopathy and suspected birth (16%) and two without suspected birth (2%) received therapeutic hypothermia, all of whom were born in 2009 or CP n=1001 Excluded n=458 <36wks: 350 Unknown GA: 93 CP 36wks, and BW 1800g n=543 Moderate or severe NE n=155 (29%) Suspected birth n=61 (41%) 36wks & BW <1800g: 1 36wks & unknown BW: 14 No moderate or severe NE n=388 (71%) Excluded No GMFCS or no CP subtype data: n=8 No suspected birth n=86 (59%) Figure 1: Flow chart. BW, birthweight; CP, cerebral palsy; GA gestational age; NE, neonatal encephalopathy; GMFCS, Gross Motor Function Classification System. Table I: Demographic characteristics of children with cerebral palsy after neonatal encephalopathy (n=147) Suspected birth (n=61) No suspected birth (n=86) Maternal age (y:mo), 30:6 (5:2) 39:5 (5:11) 0.48 mean (SD) Maternal university 26 (43) 33 (38) 0.61 studies, n (%) Maternal ethnicity 44 (72) 66 (77) 0.57 (White), n (%) Gestational diabetes, 3 (5) 15 (17) n (%) Trauma during the 8 (13) 8 (9) 0.74 pregnancy, n (%) Alcohol use, n (%) 10 (16) 17 (20) 0.67 Twin pregnancy, n (%) 1 (2) 3 (4) 0.64 Preeclampsia, n (%) 2 (3) 3 (4) 0.98 Gestational age, wks 39.5 (1.2) 39.4 (1.5) (SD) Male sex, n (%) 31 (51) 51 (59) 0.32 Congenital malformation, 7 (12) 14 (17) 0.48 n (%) Birthweight a (g), mean (SD) 3296 (555) 3423 (554) 0.56 a Singletons only (n=60 and 83 respectively). later. The median age of the children was 37 months (interquartile range [IQR] ) at the time of initial registry inscription. Two children died after they were enrolled, both of whom were included in the cohort. Eighty-two (56%) were re-evaluated at 5 years of age. The distribution of neurological severity and subtype according to the presence or absence of suspected birth p Outcomes in CP after Neonatal Encephalopathy Jarred Garfinkle et al. 191

4 Percentage Percentage Suspected birth (n=61) Dyskinetic Spastic hemiplegia Suspected birth (n=61) GMFCS I GMFCS IV GMFCS II GMFCS V is outlined in Figure 2. Spastic triplegia was classified under spastic quadriplegia because only five children had the former. Those with suspected birth had a higher frequency of non-ambulatory GMFCS status (levels IV and V) (OR 3.4, 95% CI ), spastic quadriplegia/triplegia subtype (OR 3.1, 95% CI ), and a lower frequency of spastic hemiplegia subtype (OR 0.17, 95% CI ). The individual comorbidity frequency distribution according to the presence or absence of suspected birth is shown in Table II. Non-verbal communication skills impairment was the most frequent comorbidity in those with suspected birth (32/61, 52%), while epilepsy was the most common comorbidity in those without suspected birth (18/86, 21%). Of the five comorbidities evaluated, non-verbal communication skills impairment was the only one that was significantly more No suspected birth (n=86) Spastic diplegia Spastic quadriplegia/triplegia No suspected birth (n=86) GMFCS III Figure 2: Distribution of cerebral palsy severities and subtypes. GMFCS, Gross Motor Function Classification System. Table II: Comorbidities and magnetic resonance imaging findings in children with cerebral palsy after neonatal encephalopathy (n=147 and 112 respectively) Suspected birth Comorbidities (%) (n=61) (n=86) No suspected birth Non-verbal communication 32 (52) 18 (21) <0.001 skills impairment Severe visual impairment 17 (28) 18 (21) 0.34 Severe auditory 4 (7) 4 (5) 0.72 impairment Gavage feeding 9 (15) 11 (13) 0.81 Epilepsy 20 (33) 19 (22) 0.19 MRI findings (%) (n=40) (n=72) Normal 3 (8) 4 (6) 0.70 WMI 8 (20) 12 (17) 0.80 WMI and cortical 4 (10) 14 (19) 0.28 injury Deep GMI 7 (18) 3 (4) Near-total injury 15 (38) 18 (25) 0.20 Focal injury 2 (5) 15 (21) Malformation 1 (3) 6 (8) 0.42 MRI, magnetic resonance imaging; WMI, white matter injury; GMI, grey matter injury. frequent in those with suspected birth (OR 4.2, 95% CI ). Those with suspected birth experienced more comorbidities than those without, with this difference being significant (median 1, IQR 0 2; compared to median 0, IQR 0 1; p=0.017). Children with suspected birth comprised 41% of the children with moderate or severe neonatal encephalopathy. However, they contributed 58% of the non-ambulatory children (GMFCS levels IV and V) and 55% of the children with spastic quadriplegia. In addition, of the 152 separate comorbidities documented among the 147 children, 82 (54%) occurred among those with suspected birth. MRI descriptions were available in 40 out of 61 (66%) and 72 out of 86 (84%) children in each group. Almost half of the MRIs were performed in the first month of life (Table SII, online supporting information). The distribution of MRI findings in the 112 children with MRIs is shown in Table II. The initial agreement between the two imaging reviewers was excellent (j=0.86). Isolated deep grey matter injury was more common in those with suspected birth (18% vs. 4%, OR 4.9, 95% CI ), whereas focal injury was less common in that group (5% vs. 21%, OR 0.20, 95% CI ). Focal injury was more common in children with spastic hemiplegia than those with other CP subtypes (43% vs. 3%, OR 28.1, 95% CI ). Injury involving the deep grey matter (near total brain injury and deep grey matter injury patterns) was more common in children with non-verbal communication skills impairment (51% vs. 20%, OR 4.1, 95% CI ), but was not significantly more or less common in children with any of the other comorbidities. p 192 Developmental Medicine & Child Neurology 2016, 58:

5 DISCUSSION In our cohort of term-born children with CP after to moderate or severe neonatal encephalopathy, drawn from a national CP registry, children with suspected birth were more likely to have a later non-ambulatory GMFCS status, a more severe neurological subtype (spastic quadriplegia), and non-verbal communication skills impairment, and have a greater numerical burden of comorbidities compared to those without suspected birth. On a frequency basis, those with suspected birth were three times more likely to be non-ambulatory, three times more likely to have spastic quadriplegia, and four times more likely to have non-verbal communication skills impairment. The other comorbidities (gavage feeding, seizure disorder, and severe visual or auditory impairments) were not significantly different between the two groups. Fortunately, therapeutic hypothermia specifically targets these neonates who are at risk for a more severe form of CP. Previous CP registries have shown that the overall burden of disability was greater in children who developed CP after neonatal encephalopathy than in children with CP without preceding neonatal encephalopathy. 6,7,17 From the Western Australian Cerebral Palsy Register, CP after neonatal encephalopathy was more likely to be severe and spastic quadriplegic or dyskinetic than CP without preceding neonatal encephalopathy. It was also more frequently associated with cognitive impairment, non-verbal communication skills impairment, and epilepsy. 6 Our study evaluated this high-risk group with antecedent neonatal encephalopathy and dichotomized it according to the presence or absence of suspected birth. Based on our results, those with suspected birth constitute a minority (41%) of children with CP after neonatal encephalopathy; 13 however, they have a greater burden of later morbidity. We found that 62% of children with suspected birth had non-ambulatory CP (in GMFCS levels IV V) compared to 33% of those without suspected birth. This increased burden of impairment in children with CP after moderate or severe neonatal encephalopathy with suspected birth has important implications on their quality of life and the need for, and response to, therapeutic interventions. It also provides a secure yardstick to evaluate the potential benefits of therapeutic hypothermia. For many individuals with CP and their families, the non-neuromotor comorbidities may be the major locus of challenges and burdens of care that prompt the utilization of health services. 3,18 Physicians should be aware that the CP that manifests in survivors of neonatal encephalopathy with suspected birth is often, but not always, nonambulatory and associated with multiple comorbidities. The increased frequency of non-verbal communication skills impairment in children with CP after suspected birth may be a consequence of their brain injury, which more frequently involved the deep grey nuclei. A retrospective study of children with neonatal encephalopathy and suspected birth found that non-verbal communication skills impairment was almost uniquely found in those with later CP and that only the severity of basal ganglia and thalamic injury was independently associated with non-verbal communication skills impairment. 19 Clinical and functional-imaging observations indicate that the motor execution of speech production (i.e. the coordination of vocal tract gestures) depends upon several structures including the thalami and the basal ganglia. 20 The overall frequency of spastic hemiplegia in our cohort was 30%. This proportion is slightly higher than the 17% to 25% reported in other registry- and cohortbased studies that assessed children with CP after neonatal encephalopathy. 6,7,17,21 Many of the children with spastic hemiplegia had focal injuries. Spastic hemiplegia was four times more frequent in the group without suspected birth, at least in part because that group had a high proportion of focal injuries. Neonatal encephalopathy be it manifested by seizures, abnormal tone, or altered level of consciousness is known to be a common presenting feature of neonatal stroke. 22,23 Moreover, neonates with stroke do not usually require major caseroom resuscitation, a feature that is more common in neonates who have suspected birth. 22,23 An Executive Summary by the American College of Obstetricians and Gynecologists Task Force on Neonatal Encephalopathy stated that in the context of acute intrapartum hypoxic ischemic events, [the] developmental outcome is spastic quadriplegia or dyskinetic cerebral palsy and that other subtypes of cerebral palsy are less likely. 24 Our results support this element of the Executive Summary, since the majority (79%) of children with neonatal encephalopathy with suspected birth had either a spastic quadriplegia or dyskinetic neurological subtype. The Executive Summary goes on to state that other developmental abnormalities may occur. Our results also support this element since the majority of children with CP after neonatal encephalopathy with suspected birth had at least one associated disability. The strengths of this study include its prospective design in a geographically defined population, the large number of cases of CP in term-born children based on well-defined diagnostic criteria, standardized classification of motor function with the GMFCS, 15 systematic variable ascertainment, and the precise categorization of children into welldefined groups. Our study represents the first report from a CP registry regarding the outcomes of term-born children with CP after moderate or severe neonatal encephalopathy dichotomized according to the presence or absence of suspected birth. Assuming, for group assignment purposes, that undocumented ph were in all cases normal is a limitation but provides a more conservative estimate of the proportion of children who had suspected birth. A substantial limitation was our reliance on MRI descriptions rather than a systematic Outcomes in CP after Neonatal Encephalopathy Jarred Garfinkle et al. 193

6 review of imaging studies by blinded neuroradiologists. We attempted to mitigate this limitation by having two authors blinded to the clinical information (PW and JG) classify the MRI reports based on an adaptation of a previously described classification system. Furthermore, imaging of the brain was not uniformly attained. We dichotomized children according to suspected birth rather than birth per se because the latter is difficult to accurately identify clinically. 25 Other exposures such as inflammation or placental vasculopathy can result in a phenotype indistinguishable from that of birth. In conclusion, term-born children with CP after neonatal encephalopathy with suspected birth are more likely to have a non-ambulatory GMFCS status, spastic quadriplegia, and more comorbidities. Their particular brain injury, in comparison to that of the child with CP after neonatal encephalopathy without suspected birth, may account for the increased severity of their CP. Our results can complement existing prognostic information for clinicians caring for neonates with neonatal encephalopathy and provide a benchmark for evaluating the potential benefits of regional therapeutic hypothermia programs. ACKNOWLEDGEMENTS The Canadian Cerebral Palsy Registry includes the following contributors who assisted with recruitment of participants: John Andersen, Glenrose Rehabilitation Hospital, Edmonton, AB; David Buckley, Janeway Children s Hospital, St. John s, NL; Darcy Fehlings, Bloorview Kids Rehab, Toronto, ON; Adam Kirton, Alberta Children s Hospital, Calgary, AB; Alison Moore, Alberta Children s Hospital, Calgary, AB; Esias van Rensburg, BC Children s Hospital, Vancouver, BC; Ellen Wood, IWK Health Centre, Halifax, NS, Canada. REPACQ has been funded by the Reseau de recherche sur le developpement, la sante et le bien-^etre de l enfant des Fonds de Recherche en Sante du Quebec, and NeuroDevNet National Centre of Excellence. No funding sources were involved in the design, analysis, manuscript preparation, or manuscript submission decisions of this study. The authors have stated that they had no interests that might be perceived as posing a conflict or bias. SUPPORTING INFORMATION The following additional material may be found online: Table SI: Definitions for magnetic resonance imaging findings. Table SII: Timing of magnetic resonance imaging. REFERENCES 1. Rosenbaum P. Cerebral palsy: what parents and doctors want to know. BMJ 2003; 326: Rosenbaum P, Paneth N, Leviton A, et al. A report: the definition and classification of cerebral palsy April Dev Med Child Neurol Suppl 2007; 109: Shevell M, Dagenais L, Oskoui M. The epidemiology of cerebral palsy: new perspectives from a Canadian registry. Semin Pediatr Neurol 2013; 20: Himmelmann K, Hagberg G, Uvebrant P. The changing panorama of cerebral palsy in Sweden. X. Prevalence and origin in the birth-year period Acta Paediatr 2010; 99: McIntyre S, Blair E, Badawi N, Keogh J, Nelson KB. Antecedents of cerebral palsy and perinatal death in term and late preterm singletons. Obstet Gynecol 2013; 122: Badawi N, Felix JF, Kurinczuk JJ, et al. Cerebral palsy following term newborn encephalopathy: a populationbased study. Dev Med Child Neurol 2005; 47: Kyriakopoulos P, Oskoui M, Dagenais L, Shevell MI. Term neonatal encephalopathy antecedent cerebral palsy: a retrospective population-based study. Eur J Paediatr Neurol 2013; 17: Shankaran S, Laptook AR, Ehrenkranz RA, et al. Whole-body hypothermia for neonates with hypoxic-ischemic encephalopathy. N Engl J Med 2005; 353: Rutherford M, Ramenghi LA, Edwards AD, et al. Assessment of brain tissue injury after moderate hypothermia in neonates with hypoxic-ischaemic encephalopathy: a nested substudy of a randomised controlled trial. Lancet Neurol 2010; 9: Oskoui M, Joseph L, Dagenais L, Shevell M. Prevalence of cerebral palsy in Quebec: alternative approaches. Neuroepidemiology 2013; 40: Sarnat HB, Sarnat MS. Neonatal encephalopathy following fetal distress. A clinical and electroencephalographic study. Arch Neurol 1976; 33: Peliowski-Davidovich A. Hypothermia for newborns with hypoxic ischemic encephalopathy. Paediatr Child Health 2012; 17: Garfinkle J, Wintermark P, Shevell MI, Platt RW, Oskoui M. Cerebral palsy following neonatal encephalopathy: how much is preventable? J Pediatr 2015; 167: Shevell MI, Dagenais L, Hall N. Comorbidities in cerebral palsy and their relationship to neurologic subtype and GMFCS level. Neurology 2009; 72: Palisano R, Rosenbaum P, Walter S, Russell D, Wood E, Galuppi B. Development and reliability of a system to classify gross motor function in children with cerebral palsy. Dev Med Child Neurol 1997; 39: Towsley K, Shevell MI, Dagenais L. Population-based study of neuroimaging findings in children with cerebral palsy. Eur J Paediatr Neurol 2011; 15: Gaffney G, Flavell V, Johnson A, Squier M, Sellers S. Cerebral palsy and neonatal encephalopathy. Arch Dis Child Fetal Neonatal Ed 1994; 70: F Majnemer A, Shevell M, Law M, Poulin C, Rosenbaum P. Indicators of distress in families of children with cerebral palsy. Disabil Rehabil 2012; 34: Martinez-Biarge M, Diez-Sebastian J, Wusthoff CJ, et al. Feeding and communication impairments in infants with central grey matter lesions following perinatal hypoxic-ischaemic injury. Eur J Paediatr Neurol 2012; 16: Ackermann H, Hage SR, Ziegler W. Brain mechanisms of acoustic communication in humans and nonhuman primates: an evolutionary perspective. Behav Brain Sci 2014; 37: Evans K, Rigby AS, Hamilton P, Titchiner N, Hall DM. The relationships between neonatal encephalopathy and cerebral palsy: a cohort study. J Obstet Gynaecol 2001; 21: Kirton A, Armstrong-Wells J, Chang T, et al. Symptomatic neonatal arterial ischemic stroke: the International Pediatric Stroke Study. Pediatrics 2011; 128: e Rutherford MA, Ramenghi LA, Cowan FM. Neonatal stroke. Arch Dis Child Fetal Neonatal Ed 2012; 97: F Executive summary: neonatal encephalopathy and neurologic outcome, second edition. Report of the American College of Obstetricians and Gynecologists Task Force on Neonatal Encephalopathy. Obstet Gynecol 2014; 123: Ellenberg JH, Nelson KB. The association of cerebral palsy with birth : a definitional quagmire. Dev Med Child Neurol 2013; 55: Developmental Medicine & Child Neurology 2016, 58: