7/23/2019 Pi is 0022347615001559

http://slidepdf.com/reader/full/pi-is-0022347615001559 1/7

Cerebral Palsy after Neonatal Encephalopathy: How Much Is Preventable?Jarred Garnkle, MDCM1,2 , Pia Wintermark, MD3 ,4, Michael I. Shevell, MDCM1,2 ,3 , Robert W. Platt, PhD3,5,

and Maryam Oskoui, MDCM, MSc1,2 ,3, on behalf of the Canadian Cerebral Palsy Registry*

Objectives To determine the expected proportion of term cerebral palsy (CP) after neonatal encephalopathy (NE)that could theoretically be prevented by hypothermia and elucidate the perinatal factors associated with CP afterNE in those who do not meet currently used clinical criteria required to qualify for hypothermia (“cooling criteria”).Study design Using the Canadian CP Registry, we categorized children born at $ 36 weeks with birth weight$ 1800 g with CP after moderate or severe NE according to the presence or absence of cooling criteria. Maternal,perinatal, postnatal, and placental factors were compared between the 2 groups. A number needed to treat of 8(95% CI 6-17) to prevent one case of CP was used for calculations.Results Among the 543 term-born children with CP, 155 (29%) had moderate or severe NE. Sixty-four of 155(41%) met cooling criteria and 91 of 155 (59%) did not. Shoulder dystocia was more common in those who didnot meet cooling criteria (OR 8.8; 95% CI 1.1-71.4). Low birth weights (20% of all singletons), small placentas(42%), and chorioamnionitis (13%) were common in both groups.Conclusions The majority of children with CP after NE did not meet cooling criteria. An estimated 5.1% (95% CI

2.4%-6.9%) of term CP after NE may be theoretically prevented with hypothermia. Considering shoulder dystociaas an additional criterion may help recognize more neonates who could potentially benet from cooling. In all cases,a better understanding of the antenatal processes underlying NE is essential in reducing the burdenof CP. (J Pediatr 2015;167:58-63) .

See editorial, p 8 andrelated article, p 25

T herapeutic hypothermia has proven to be an effective neuroprotective intervention and has been applied increasingly toneonates with specic clinical criteria and moderate or severe neonatal encephalopathy (NE) within the rst 6 hours of life.1-3 The number of such neonates needed to treat (number nee ded to treat; NNT) to prevent 1 subsequent case of

cerebral palsy (CP) is 8 (95% CI 6-17) according to recent meta-analyses. 2,3 Some pediatric neurologists are hoping that hy-pothermia will reduce the frequency of CP in term-born neonates with NE, but the overall proportion of term CP that could beprevented by hypothermia in this population has not yet been established. 4

Furthermo re, there is no neuroprotective strategy for term neonates with NE who do not meet current clinical criteria forhypothermia. 5 More specic information co ncerning the pathways to CP in this population is needed before potentially effec-tive therapies can be developed and applied. 6,7 Using the Canadian Cerebral Palsy Registry (CCPR), we aimed to investigateterm-born children with later CP after moderate or severe NE. The objectives were: (1) to determine the expected proportionof term CP after NE that could theoretically be prevented by hypothermia; and (2) to elucidate the perinatal factors associatedwith CP after NE in those who do not meet the clinical criteria required to qualify for hypothermia.

Methods

The study was conducted using data extracted from the CCPR, which builds onthe development and implementation of the Cerebral Palsy Register of Qu ebec.The CCPR captures cases of CP identied through both pediatric rehabilitationcenters and university hospitals at which provincial pediatric neurology anddevelopmental pediatric services are located, from the birth year of 1999 until2011. The registry covers most of Quebec, the greater Toronto area of Ontario,and the entire provinces of British Columbia, Alberta, Nova Scotia, and

From the 1 Department of Neurology/Neurosurgery,McGill University; 2 Division of Pediatric Neurology,McGill University Health Centre; 3 Department of Pediatrics, McGill University; 4 Division of Neonatology,McGill University Health Center; and 5 Department of Epidemiology, Biostatistics, and Occupational Health,McGill University, Montreal, Quebec, Canada

*Listof membersof theCanadianCerebralPalsy Registryis available at www.jpeds.com ( Appendix ).

The Cerebral Palsy Register of Qu ebec has been fundedby the R eseau de recherche sur le d eveloppement, lasant e et le bien-

^

etre de l’enfant (RSDE) des Fonds deRecherche en Sant e du Qu ebec and NeuroDevNet Na-tional Centre of Excellence. The authors declare noconicts of interest.

0022-3476/$ - see front matter. Copyright ª 2015 Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.jpeds.2015.02.035

CCPR Canadian Cerebral Palsy RegistryCP Cerebral palsyHIE Hypoxic-ischemic encephalopathyNE Neonatal encephalopathyNNT Number needed to treat

58

7/23/2019 Pi is 0022347615001559

http://slidepdf.com/reader/full/pi-is-0022347615001559 2/7

Newfoundland, and includes more than one-half (approxi-mately 18 million individuals) of the Canadian population.Once cases are identied, parental consent is obtained, andthe maternal and child’s records are reviewed. These dataare supplemented by a standardized parental interview andphysical examination of the child by a pediatric neurologist,developmental pediatrician, or child physiatrist. Local ethicsboard approval was obtained from each participating institu-tion. The Montreal Children’s Hospital-McGill University Health Center research ethics board provided centralapproval for data storage, analysis, and overall operations.

To be enrolled in the CCPR, a child must be at least 2 yearsof age and meet current diagnostic consensus criteria for CP,which include a clinical diagnosis of a nonprogressive motorimpairment result ing from a presumably early insult to thedeveloping brain. 8 The diagnosis is conrmed wheneverpossible at 5 years of age, and the children without CP wereremoved from the registry. 9 Patients within the CCPR included for analysis in this study: (1) were bornat $ 36 weeks; (2) had a birth weight of $ 1800 g; and (3) ful-lled criteria for moderate or severe NE during the rst week of life. We then categorized the neonates according to thepresence or absence of currently used clinical criteriarequired to qualify for hypothermia (“cooling criteria”).

Criteria for NE were derived from the Sarnat score. 10

Neonates with moderate NE were either lethargic, hypo-tonic, manifested seizures, or had decreased reexes. Thosewith severe NE manifested either accid coma, brainstemndings, or difcult-to-control seizures. Clinical coolingcriteria were derived from the National Institute of ChildHealth and Human Development randomized controlledtrial of therapeutic hypothermia and the Position State-ment of the Canadian Paediatric Society. 11,12 Neonateswere required to have evidence of both “fetal” and“neonatal” distress. “Fetal distress” was evident by at leastone of the following: (1) sentinel event (abruptio placenta,cord accident, uterine rupture, or shoulder dystocia); (2)pH # 7.0 or base decit $ 16 mEq/L from the umbilicalcord or blood collected within the rst hour of life; or(3) severe fetal heart rate abnormalities. “Neonataldistress” was evident by at least one of the following: (1)Apgar score # 5 at 10 minutes; or (2) delivery room resus-citation (intubation, cardiac massage, or administration of epinephrine or normal saline bolus). For the purposes of

cooling group assignment, when the 10-minute Apgarscore or pH was not documented in the records, they were assumed to have been normal, and when sentinelevents, severe fetal heart rate abnormalities, and delivery room resuscitation were not documented in the records,they were assumed to have been absent.

Maternal, paternal, perinatal, postnatal, and placental fac-tors were compared in an attempt to identify variables thatpotentially distinguish those who met cooling criteria fromthose who did not.

Birth weight,placental weight, andplacental-to-birth weightpercentiles of singletons were basedon sexand ges tatio nbythe

use of contemporary Canadian percentile curves.13,14

Seven randomized controlled trials have evaluated hypo-thermia and each assessed the incidence of CP. Two subse-quent meta-analyses established that the NN T to preventone case of CP was 8 (95% CI 6-17 and 6-16).2,3 This NNTis applicable to our cohort of children with CP who: (1)had moderate or severe NE; and (2) met clinical coolingcriteria.

Statistical analysis was performed with SPSS 20.0 (SPSSInc, Chicago, Illinois). For comparisons between the twogroups, the c

2 test or 2-sided Fisher exact test were usedfor univariate analysis of categorical variables as appropriateand the Student t test for continuous variables. ORs and their95% CIs were calculated where appropriate. A P of <.05 was apriori considered signicant. We applied unconditional lo-gistic regression analysis selectively when 2 variables thatact along the same causal pathway were signicant.

Results

Between 1999 and June 2011, a total of 1001 patients wereenrolled into the CCPR. Of these, 543 were $ 36 weeks’ gesta-tion and had a birth weight $ 1800 g. Of them, 155 of 543(29%) had a history of moderate or severe NE (it is this groupthat comprises the study cohort) and 388 of 543 (71%) didnot (361 did not have moderate or severe NE and 27 didnot have documentation of a neurologic examination duringthe neonatal period). Among those with NE, 64 of 155 (41%)met currently used cooling criteria, but 91 of 155 (59%) didnot. Cord or rst hour of life blood pH was available in 62 of 64 and 65 of 91 neonates, and 10-minute Apgar score wasavailable in 55 of 64 and 53 of 91 neonates in either group(Table I ). In 41 of 47 cases (87%) in which the 10-minute

Apgar score was unavailable, the 5-minute Apgar score was>5. Eleven neonates with NE who met cooling criteria(17%) and 2 who did not meet them (2%) were actually cooled, all of whom were born in 2009 or later. Amongthose without moderate or severe NE, 9 of 388 (2%)otherwise met cooling criteria (2 with mild NE and 7without NE), and 379 of 388 (98%) did not meet coolingcriteria (5 with mild NE and 374 without NE) ( Figure ;available at www.jpeds.com).

The features and degree of NE in the 155 children withmoderate or severe NE are summarized in Table II .Hypotonia as a feature of NE was signicantly more

common in those who met cooling criteria ( P = .020). Theclinical characteristics of children with CP after NE arelisted in Table I . There were no signicant differencesbetween the 2 groups with respect to antenatal maternaland paternal variables. Thirty of 151 (20%) singletons were<10th percentile for weight (similar in both groups).

Signs of perinatal distress, such as low pH, cesarean deliv-ery, low Apgar scores, and resuscitation at birth requiringintubation, cardiac massage, or epinephrine, were signi-cantly more common in those who met cooling criteria(P < .0001), as expected. Overall, 22 of 64 (34%) of thosewho met cooling criteria vs 8 of 91 (9%) of those who did

not had an abruptio placenta, uterine rupture, or cord

Vol. 167, No. 1 July 2015

59

7/23/2019 Pi is 0022347615001559

http://slidepdf.com/reader/full/pi-is-0022347615001559 3/7

accident ( P < .0001). Multisystem involvement, dened asNE and an objectively documented impairment in at leastone other system, was almost uniformly documented in thosewho met cooling criteria (60/64 [94%]).

Shoulder dystocia, in contrast to the other sentinelevents, was 8 times more common in those with NEwho did not meet cooling criteria (OR 8.8; 95% CI 1.1-71.4, P = .015). Gestational diabetes was 4 times morecommon in those who did not meet cooling criteria (OR 4.0; 95% CI 1.1-14.5, P = .039). Only 4 of the 11 (36%)neonates with shoulder dystocia who did not meet cooling

criteria were born to a mother with gestational diabetes.Their mean pH, available in 9 of 11 of them, was 7.17(SD 0.10). Shoulder dystocia was still signicantly more common in those who did not meet cooling criteriaafter adjusting for gestational diabetes in an unconditionallogistic regression analysis (OR 1.8; 95% CI 1.0-3.2;P = .042). Chorioamnionitis, which was diagnosed histo-logically or when there was a maternal fever of undeter-mined origin, was present in 9 of 63 (14%) and 10 of 88(11%) of singletons in either group.

Placentas were available foranalysis in 93 of 151 (62%) sin-gletons. Interestingly, 39/93 (42%) placentas were <10th

percentile for weight, and 38 of 93 (41%) were <10th percen-tile for placental weight-to-birth weight ratio; neither factordiffered signicantly between the 2 groups.

Discussion

Approximately 1 in 4 children with CP in this registry born at$ 36 weeks’ gestation and with a birth weight $ 1800 g had ahistory of moderate or severe NE, less than one-half of whommet clinical cooling criteria. In the Western Australian CPRegister between 1980 and 1995, one-third of term childrenhad antecedent moderate or severe NE. 6 In the Oxford

Table I. Clinical characteristics of children with CP andNE (n = 155)

Clinical factors

Clinical criteria forhypothermia

P valueYes (n = 64) No (n = 91)

Maternal and paternalMaternal age, years, mean SD 30.6 5.2 29.5 5.8 .56Maternal university studies, n (%) 26 (41) 38 (42) 1.00Paternal university studies, n (%) 23 (36) 27 (30) .49Maternal e thnicity white , n (%) 46 (72) 69 (76) .58Paternal e thnicity white , n (%) 43 (67) 68 (75) .37Gravida $ 2, n (%) 34 (53) 49 (54) 1.00Parity $ 2, n (%) 44 (69) 55 (60) .31 Aborta $ 1, n (%) 26 (41) 32 (35) .51Hypertension, n (%) 7 (11) 10 (10) .97Gestational diabetes, n (%) 3 (5) 15 (17) .039*Insulin-dependent gestational

diabetes, n (%)1 (2) 6 (7) .241

Trauma during the pregnancy, n(%)

8 (13) 8 (9) .49

Bleeding during rst trimester, n(%)

7 (11) 13 (14) .79

Bleeding during second trimester,

n (%)

2 (5) 5 (6) .75

Bleeding during third trimester, n(%)

4 (6) 6 (7) .96

Alcohol use, n (%) 11 (17) 18 (20) .46Tobacco use, n (%) 16 (25) 13 (14) .099Maternal illness, n (%) 13 (20) 21 (23) .46Twin pregnancy, n (%) 1 (2) 3 (3) .64Pre-eclampsia, n (%) 2 (3) 4 (4) .89TORCH infection, n (%) 8 (13) 6 (7) .29GBS positive, n (%) 12 (19) 15 (17) .58

PerinatalGestational age, wk 39.5 1.2 39.3 1.5 .46Gestational age 36-38 + 6 wk,

n (%)12 (19) 29 (32) .032*

Gestational age 41-43 wk,n (%)

12 (19) 24 (26) .11

Male sex, n (%) 34 (53) 54 (59) .51Congenital malformation, n (%) 8 (13) 15 (17) .65Birth weight, g† 3284 575 3390 558 .255Birth weight <10th percentile,

n (%)†15 (24) 15 (17) .31

Birth weight $ 90th percentile,n (%)†

4 (6) 10 (11) .40

Cesarean delivery, n (%) 44 (69) 24 (26) <.0001*Maternal fever during labor or

delivery, n (%)6 (9) 13 (14) .27

Prolonged rupture of membranes,n (%)

6 (9) 13 (14) .37

Chorioamnionitis, n (%)† 9 (14) 10 (11) .76 Abruptio placenta, n (%) 11 (17) 0 (0) <.0001*Shoulder dystocia, n (%) 1 (2) 11 (12) .015*Uterine rupture, n (%) 3 (5) 0 (0) .068Cord accident, n (%) 12 (19) 7 (8) .048*pH, mean SDz 6.98 0.20 7.21 0.10 <.0001* Apgar at 1, mean SD 1.6 1.6 5.1 3 .0 <.0001 * Apgar at 5, mean SDx 3.9 2.2 7.1 2 .5 <.0001 * Apgar at 10, mean SD{ 4.9 2.5 7.1 2 .4 <.0001 *Resuscita tion a t b irth, n (%) 64 (100) 41 (45) <.0001*

PostpartumSepsis, n (%) 4 (6) 3 (3) .69 Administration of antibiotics for

10 days, n (%)8 (13) 6 (7) .30

Seizures during the rst 24 hours,n (%)

49 (77) 56 (62) .10

Seizures during the rst 72 hours,n (%)

55 (86) 68 (76) .28

Multisystem involvement, n (%) 60 (94) 57 (63) <.0001*(continued )

Table I. Continued

Clinical factors

Clinical criteria forhypothermia

P valueYes (n = 64) No (n = 91)

Icterus, n (%) 8 (13) 18 (20) .44Placental†,**

Placental weight, g 547 169 601 159 .114

Placental weight <10thpercentile, n (%) 23 (51) 16 (33) .096

Placental weight $ 90thpercentile, n (%)

1 (2) 1 (2) 1.00

Placental weight/birth weight ratio<10th percentile, n (%)

22 (49) 16 (33) .14

Placental weight/birth weight ratio$ 90th percentile, n (%)

1 (2) 4 (8) .36

GBS , group B streptococcus; TORCH , toxoplasmosis, other (syphilis, varicella-zoster, parvo-virus B19), rubella, cytomegalovirus, and herpes.*P < 0.05†Singletons only (n = 63 and 88, respectively).zpH available in 62/64 and 65/91 neonates.x5-minute Apgar scores available in 62/64 and 88/91.{ 10-minute Apgar scores available in 55/64 and 53/91.**Placentas available in 45/63 and 48/88 singletons.

THE JOURNAL OF PEDIATRICS www.jpeds.com Vol. 167, No. 1

60 Garnkle et al

7/23/2019 Pi is 0022347615001559

http://slidepdf.com/reader/full/pi-is-0022347615001559 4/7

Regional Health Authority Registry between 19 84 and 1987,one-fth of term children had evidence of NE. 15 However,neither of these registry-based studies, conducted in thepre-cooling era, evaluated the proportion of children withCP after NE that met cooling criteria. The Western AustralianCP Register estimated that 63% of term neonates with CP af-ter NE had “hypoxic-ischemic encephalopathy” (HIE) deter-mined on thebasis clinical diagnoses entered into themedicalrecord, but sentinel events were documented in only one-fth of those labeled as HIE. These authors’ use of HIEmay hav e been less specic than the clinical cooling criteriawe used.6,7

On the basis of an NNT of 8 (95% CI 6-17), theoreti-cally, if all 64 neonates who had NE and met coolingcriteria had been cooled, an estimated 8 (64/8; 95% CI64/17-64/6 = 3.8-10.7) would not have developed CP.Therefore, of all 155 children born at term with CP afterNE, an estimated 5.1% (8/155 100%; 95% CI 3.8/155-10.7/155 100% = 2.4%-6.9%) could have theoretically been spared CP with the application of hypothermia. Thereare limitations inherent in this estimate, which we addressin the sections to follow.

The majority of children with CP after NE did not meetthe current cooling criteria (59%), and many of them likely had NE due to causes other than hypoxic-ischemic injury.Shoulder dystocia may be an exception whereby neonateswith NE who did not meet clinical cooling criteria still suf-

fered from hypoxic-ischemic injury. Shoulder dystoc ia isknown to complicate approximately 1% of deliveries. 16 Inour study, it was present in 2% of those who met coolingcriteria and 12% of those who did not, and remainedmore frequent in those who did not meet cooling criteria af-ter adjusting for gestational diabetes, suggesting that neo-nates with NE and shoulder dystocia and later CP do notnecessarily manifest clinical signs of both “fetal” and“neonatal” distress. In one study, the mean umbilical artery pH of shoulder dystocia cases wa s only 0.04 lower than themean pH of all vaginal deliveries.17 In our study, the meancord or rst hour of life blood pH in neonates with shoulder

dystocia who did not meet cooling criteria was 7.17, which is

within the normal range. This can be explained by the factthat compression of the fetal carotid vessels by maternal tis-sue at the level of the pelvic outlet may cause isolated cere-bral hypoxia rather than systemic acidosis. The cord bloodpH, which reects the systemic circulation, may fail toreect the actual degree of underlying cerebral pathology.On the basis of our data, caregivers should not be reassuredby a cord pH over 7 in neonates with shoulder dystocia, asthese neonates may indeed be at risk of developing CP.Additional prospective studies are needed to properly assessthe neurodevelopmental outcomes of children with NE aftershoulder dystocia. At the very least, when the cord pH iswithin normal limits, a blood gas could be performed inthese neonates within the rst hour of life, as it may betterreect possible cerebral injury.

Current recommendations that only neonates with mod-erate or severe NE be treated with hypothermia may be toorestrictive. 12 This recommendation is based on the random-ized controlled trials, which were designed to exclude infantswho did not demonstrate moderate or severe NE. 2,3,11,18

However, in our registry, 2% of children with CP born at$ 36 weeks’ gestation with a birth weight $ 1800 g met cool-ing criteria in the absence of moderate or severe NE. Only 2of them had mild NE, but we suspect that mild NE wasunderreported. Our study is not the rst to suggest that clin-ical signs of fetal and neonatal distress in the absence of moderate or severe NE are not uniformly benign. Forinstance, one randomized controlled trial, in violation of its protocol, enrolled 42 neonates classied as having mildNE at less than 6 hours of lif e; 28.5% of them later diedor developed major disability. 18 In a recent retrospectivecohort study, 20% of neonates with perinatal ac idemiaand mild NE had abnormal short-term outcomes. 19 Ourstudy looking at children with CP thus further emphasizesthe need to collect prospective data on the neurologic out-comes of neonates with mild NE to evaluate whether they should qualify for hypothermia.

Both groups included a large proportion of singletons withsmall placentas, low birth weights (less than the 10th percen-tile for gestation), and chorioamnionitis, but the incidence of these 3 variables did not differ between the 2 groups.

Similarly, single-center cohort studies of: (1) neonateswith NE subsequent to a broad denition of hypoxic-ischemic injury; and (2) neonates meeting criteria for hypo-

thermia found high frequencies of these variables.20,21

Ameta-analysis found that both low birth weights and cho-rioamnionitis are established risk factors for CP in term-born children. 22 Even neonates with signs of perinataldepression and subs equent CP have a high frequency of ante-natal complications. 23 The fact that these risk factors for NEand later CP frequently are present in both groups of neo-nates suggests that the pathway to CP frequently involvesantenatal processes in both groups. A case-control study of CP in term and late preterm singletons examining the poten-tial contribution of sentinel events, inammation, fetalgrowth restriction, and birth defects estimated that growth

restriction and inammation contributed to a total of 20%

Table II. Features and degree of encephalopathy of children with CP and moderate or severe NE (n = 155)

Degree and features

Met clinical criteria forhypothermia

P valueYes (n = 64) No (n = 91)

Degree .17Moderate 44 (69) 70 (77)Severe 20 (31) 21 (23)

FeaturesLethargy 15 (23) 21 (23) 1.00Hypotonia 45 (70) 46 (51) .020*Seizures 55 (86) 68 (76) .28Decreased reexes 16 (25) 15 (17) .22Flaccid coma 7 (11) 4 (4) .20Difcult-to-control seizures 20 (31) 21 (23) .27

*P < .05.

July 2015 ORIGINAL ARTICLES

Cerebral Palsy after Neonatal Encephalopathy: How Much Is Preventable? 61

7/23/2019 Pi is 0022347615001559

http://slidepdf.com/reader/full/pi-is-0022347615001559 5/7

of CP in infants with NE with HIE and 34% in those with NEnot otherwise specied. 6 Antenatal and placental processesmay predispose neonates to later CP, regardless of whetherthere are signs of fetal and/or neonatal distress. A better un-derstanding of these other mechanisms will be paramount inreducing the burden of CP following NE, especially becausehypothermia addresses only a minority of these patients.

The strengths of this study include its prospective design ina geographically dened population, the large number of cases of CP in term-born children based on well-deneddiagnostic criteria, and the precise categorization of childreninto well-dened groups. Moreover, placental data wereavailable for almost two-thirds of the cohort. We attemptedto identify etiologically more specic pathways by focusingour analysis on those with moderate or severe NE anddichot-omizing them according to the presence or absence of cool-ing criteria. By requiring signs of both fetal and neonataldistress, we applied current clinical criteria used to evaluateneonates for hypothermia. Assuming, for cooling groupassignment purposes, that undocumented pH were in allcases normal represents a limitation but provides a moreconservative estimate. In addition, our registry does notinclude information on the onset of NE, but rather specieswhether the n eonate had NE in the rst week of life as orig-inally dened. 10 In calculating the proportion of term CP af-ter NE that cooling could theoretically prevent, we rstassumed that the onset of NE was within 6 hours of life (asused in the randomized controlled trials of hypothermia),but it is probable that the onset was beyond 6 hours of lifein some.

Neonatal seizures, which were very common in the rst24 hours of life in this study, would mostly have occurredafter 6 hours of life.24 As such, our estimate may overem-phasize the proportion of term CP after NE that coolingcould theoretically prevent. Fortunately, the degree of NEusually improves over time, so we should not have greatly overestimated the number of neonates w ith moderate or se-vere NE during the rst 6 hours of life. 25 We made a secondassumption in applying the NNT derived from the meta-analyses to our cohort: that the neonates with NE did notreceive hypothermia. We acknowledge that 11 of the 64 ne-onates with NE who met cooling criteria actually werecooled.

Moreover, the NNT from the meta-analyses was based on a

more heterogeneous group including neonates who died,which may introduce some bias into our calculations. Finally,cerebral magnetic resonance imaging was not routinely per-formed in this era. Therefore, we were not permitted tofurther characterize the neonates with NE according to theirprecise neuroanatomic injury type.

Unfortunately, cooling addresses only a minority of CP af-ter NE. Future studies should evaluate the neurodevelop-mental outcomes of neonates with shoulder dystocia andNE in the absence of acidemia or neonatal distress and of ne-onates meeting clinical cooling criteria in the absence of moderate or severe NE. Despite the limitations inherent in

applying the NNT derived from the meta-analyses, our esti-

mate of the proportion of term CP after NEthat could be pre-vented by hypothermia should encourage further research onthe antenatal processes underlying NE to develop other pre-ventative strategies to reduce the burden of CP. n

Submitted for publication Jul 4, 2014; last revision received Dec 29, 2014;accepted Feb 9, 2015.

Reprint requests: Maryam Oskoui, MDCM, MSc, Division of Pediatric

Neurology Montreal Children’s Hospital, 2300 Tupper Street, A-512 Montreal,Quebec, H3H 1P3. E-mail: maryam.oskoui@mcgill.ca

References

1. Azzopardi D, Strohm B, Linsell L, Hobson A, Juszczak E, Kurinczuk JJ,et al. Implementation and Conduct of Therapeutic Hypothermia forPerinatal Asphyxial Encephalopathy in the UK—Analysis of NationalData. PLoS One 2012;7:e38504.

2. Jacobs SE, Berg M, Hunt R, Tarnow-Mordi WO, Inder TE, Davis PG.Cooling for newborns with hypoxic ischaemic encephalopathy. Co-chrane Database Syst Rev 2013;1:CD003311 .

3. Tagin MA, Woolcott CG, Vincer MJ, Whyte RK, Stinson DA. Hypo-

thermia for neonatal hypoxic ischemic encephalopathy: an updatedsystematic review and meta-analysis. Arch Pediatr Adolesc Med 2012;166:558-66.

4. Himmelmann K, Hagberg G, Uvebrant P. The changing panorama of ce-rebral palsy in Sweden. X. Prevalence and origin in the birth-year period1999-2002. Acta Paediatr 2010;99:1337-43 .

5. Badawi N, Felix JF, Kurinczuk JJ, Dixon G, Watson L, Keogh JM, et al.Cerebral palsy following term newborn encephalopathy: a population-based study. Dev Med Child Neurol 2005;47:293-8 .

6. McIntyre S, Blair E, Badawi N, Keogh J, Nelson KB. Antecedents of ce-rebral palsy and perinatal death in term and late preterm singletons. Ob-stet Gynecol 2013;122:869-77.

7. Ellenberg JH, Nelson KB. The association of cerebral palsy with birthasphyxia: a denitional quagmire. Dev Med Child Neurol 2013;55:210-6.

8. Rosenbaum P, Paneth N, Leviton A, Goldstein M, Bax M, Damiano D,et al. A report: the denition and classication of cerebral palsy April2006. Dev Med Child Neurol Suppl 2007;109:8-14 .

9. Oskoui M, Joseph L, Dagenais L, Shevell M. Prevalence of cerebral palsy in Quebec: alternative approaches. Neuroepidemiology 2013;40:264-8 .

10. Sarnat HB, Sarnat MS. Neonatal encephalopathy following fetal distress.A clinical and electroencephalographic study. Arch Neurol 1976;33:696-705.

11. Shankaran S, Laptook AR, Ehrenkranz RA, Tyson JE, McDonald SA,Donovan EF, et al. Whole-body hypothermia for neonates withhypoxic-ischemic encephalopathy. N Engl J Med 2005;353:1574-84 .

12. Peliowski-Davidovich A. Hypothermia for newborns with hypoxicischemic encephalopathy. Paediatr Child Health 2012;17:41-6 .

13. Kramer MS, Platt RW, Wen SW, Joseph KS, Allen A, Abrahamowicz M,et al. A new and improved population-based Canadian reference forbirth weight for gestational age. Pediatrics 2001;108:E35 .

14. Almog B, Shehata F, Aljabri S, Levin I, Shalom-Paz E, Shrim A. Placentaweight percentile curves for singleton and twins deliveries. Placenta2011;32:58-62.

15. Gaffney G, Flavell V, Johnson A, Squier M, Sellers S. Cerebral palsy andneonatal encephalopathy. Arch Dis Child Fetal Neonatal Ed 1994;70:F195-200.

16. Overland EA, Vatten LJ, Eskild A. Pregnancy week at delivery and therisk of shoulder dystocia: a population study of 2,014,956 deliveries.BJOG 2014;121:34-41. discussion 2.

17. Stallings SP, Edwards RK, Johnson JW. Correlation of head-to-body de-livery intervals in shoulder dystocia and umbilical artery acidosis. Am JObstet Gynecol 2001;185:268-74 .

18. Jacobs SE, Morley CJ, Inder TE, Stewart MJ, Smith KR, McNamara PJ,

et al. Whole-body hypothermia for term and near-term newborns

THE JOURNAL OF PEDIATRICS www.jpeds.com Vol. 167, No. 1

62 Garnkle et al

7/23/2019 Pi is 0022347615001559

http://slidepdf.com/reader/full/pi-is-0022347615001559 6/7

with hypoxic-ischemic encephalopathy: a randomized controlled trial.Arch Pediatr Adolesc Med 2011;165:692-700 .

19. DuPont TL, Chalak LF, Morriss MC, Burcheld PJ, Christie L,Sanchez PJ. Short-term outcomes of newborns with perinatal acidemiawho are not eligible for systemic hypothermia therapy. J Pediatr 2013;162:35-41.

20. Harteman JC, Nikkels PG, Benders MJ, Kwee A, Groenendaal F, deVries LS. Placental pathology in full-term infants with hypoxic-ischemic neonatal encephalopathy and association with magnetic reso-

nance imaging pattern of brain injury. J Pediatr 2013;163:968-95.e2 .21. Wintermark P, Boyd T, Gregas MC, Labrecque M, Hansen A. Placental

pathology in asphyxiated newborns meeting the criteria for therapeutichypothermia. Am J Obstet Gynecol 2010;203:579.e1-9 .

22. McIntyre S, Taitz D, Keogh J, Goldsmith S, Badawi N, Blair E. A system-atic review of risk factors for cerebral palsy in children born at term indeveloped countries. Dev Med Child Neurol 2013;55:499-508 .

23. Nelson KB, Grether JK. Selection of neonates for neuroprotective thera-pies: one set of criteria applied to a population. Arch Pediatr AdolescMed 1999;153:393-8 .

24. Glass HC, Nash KB, Bonifacio SL, Barkovich AJ, Ferriero DM,Sullivan JE, et al. Seizures and magnetic resonance imaging-detectedbrain injury in newborns cooled for hypoxic-ischemic encephalopathy.

J Pediatr 2011;159:731-5.e1 .25. Shankaran S, Laptook AR, Tyson JE, Ehrenkranz RA, Bann CM, Das A,

et al. Evolution of encephalopathy during whole body hypothermia forneonatalhypoxic-ischemicencephalopathy. J Pediatr2012;160:567-72.e3 .

50 Years Ago in T HE J OURNAL OF P EDIATRICS

Primary Liver Cell Carcinoma Associated with Biliary Cirrhosis Due toCongenital Bile Duct AtresiaOkuyama K. J Pediatr 1965;67:89-93

O kuyama’s 1965report on the development of liver cell carcinoma in a child with biliary atresia demonstrates whatwas once the inevitable natural history of this disorder. Of interest, reection on this case underscores the ad-

vances that have been made in the management of this devastating disease.Biliary atresia is a rapidly progressive obstructive cholangiopathy of infants. Untreated, 90% of affected children die

by 36 months of age from complications related to biliary cirrhosis. Malignant transformation, usually hepatocellularcarcinoma, is a well-recognized complication of end-stage liver disease and , although rare, has been documented inpatients with biliary atresia, with an estimated prevalence of 0.73%-2.44%. 1

Fortunately, surgical, medical, and radiographic advancements have dramatically improved the outcomes of chil-dren diagnosed with biliary atresia. Dr Morio Kasai’s development of the hepatoportoenterostomy procedure enabledthe placement of a surgical conduit to allow bile drainage and arrest the progression of biliary atresia. Although theoutcomes following hepatoportoenterostomy are variable and likely related to multiple factors, the overall improvedsurvival following portoenterostomy ensured long-term survival in affected patients for the rst time.

Augmenting the surgical innovations were medical and radiographic advancements to improve the care for childrenwith biliary atresia. Prevention of cholangitis, stimulation of choleresis, and aggressive nutritional support have collec-tively improved the medical care provided to children following portoenterostomy. Additionally, as it relates toOkuyama’s original report, biomarkers such as alpha-fetoprotein and improved imaging using gadolinium-ethoxybenzyl-diethylenetriamine penta-acetic acid (Gd-EOB-DTPA) enhanced magnetic resonance imaging haveenabled early detection of malignant transformation with resultant timely intervention.

Finally, although Kasai portoenterostomy and optimal medical management allowed survival beyond 3 years of life,disease progression occurred in the majority of patients. However, with the advent of liver transplant, there has been aclear reversal of fortune for children diagnosed with bilia r y atresia. Although survival rates in 1965 hovered around10%, families today can expect survival rates above 90%. 2

James E. Squires, MD, MSDivision of Gastroenterology

Cincinnati Children’s Hospital Medical CenterCincinnati, Ohio

http://dx.doi.org/10.1016/j.jpeds.2014.12.073References

1. Kim JM, Lee SK, Kwon CH, Joh JW, Choe YH, Park CK. Hepatocellular carcinoma in an infant with biliary atresia younger than 1 year. J PediatrSurg 2012;47:819-21.

2. Ryckman FC, Fisher RA, Pedersen SH, Balistreri WF. Liver transplantation in children. Semin Pediatr Surg 1992;1:162-72 .

July 2015 ORIGINAL ARTICLES

Cerebral Palsy after Neonatal Encephalopathy: How Much Is Preventable? 63

7/23/2019 Pi is 0022347615001559

http://slidepdf.com/reader/full/pi-is-0022347615001559 7/7

Appendix

Additional members of the Canadian Cerebral Palsy Registry include:

John Andersen, MD (Glenrose Rehabilitation Hospital,Edmonton, AB), David Buckley, MBChB (Janeway Chil-dren’s Hospital, St. John’s, NL), Darcy Fehlings, MD (Bloor-view Kids Rehab, Toronto, ON), Adam Kirton, MD, MSc(Alberta Children’s Hospital, Calgary, AB), Alison Moore,MD (Alberta Children’s Hospital, Calgary, AB), Esias van Re-nsburg, MD (BC Children’s Hospital, Vancouver, BC), andEllen Wood, MD (IWK Health Centre, Halifax, NS).

Cerebral Palsy,≥ 36 weeks and

BW ≥ 1800g543

No moderate tosevere NE388 (71%)

Moderate tosevere NE155 (29%)

Cooling Criteria64 (41%)

No Cooling Criteria91 (59%)

Cooling Criteria9 (2%)

No Cooling Criteria379 (98%)

Cerebral Palsy1001 Excluded: n = 458

< 36 weeks: 350Unknown GA: 93BW < 1800g: 1Unknown BW: 14

Included inanalysis

Figure. Flow chart. BW , body weight; GA, gestational age.

THE JOURNAL OF PEDIATRICS www.jpeds.com Vol. 167, No. 1

63.e1 Garnkle et al