Outcome after The prevalence of residual physical, behavioural, and/or cog-

nitive impairments after ischaemic stroke in children is clear-

ly of importance when considering the risk-to-benefit ratio
ischaemic stroke in for acute interventions, such as anticoagulation or thrombol-
ysis. Anecdotally, children who have had ischaemic strokes
childhood are perceived to have a favourable outcome; this view, how-
ever, conflicted with our experience and that reported by
their parents. The literature also provides little support for
this position, reporting that a significant number of children
V Ganesan* MD, Neurosciences Unit; have residual impairments (Schoenberg et al. 1978, Giroud
A Hogan BA, Cognitive Neuroscience Unit, Institute of Child et al. 1995).
Health, University College London; We undertook a preliminary investigation using a parental
N Shack MCSP SRP; questionnaire to evaluate outcome in children who had had
A Gordon SROT, Great Ormond Street Hospital for Children an ischaemic stroke. We elected to investigate simple func-
NHS Trust; tional parameters reflecting residual impairments experi-
E Isaacs PhD, Cognitive Neuroscience Unit; enced by this group of children in daily life. To validate
F J Kirkham FRCP, Neurosciences Unit, Institute of Child parental report as a measure of outcome, we compared,
Health, University College London, London, UK. where possible, the results of the questionnaire with qualita-
tive ratings of motor and behavioural function as judged by a
*Correspondence to first author at Wolfson Centre, paediatric occupational therapist and a paediatric physio-
Mecklenburgh Square, London WC1N 2AP, UK. therapist, and with quantitative measures of cognitive func-
E-mail: [Link]@[Link] tion as evaluated by a neuropsychologist. Within the context
of an ongoing longitudinal study, we were also able to exam-
ine the influence of clinical factors, such as the age of the
child at the time of the stroke, the occurrence of seizures,
and the location of the lesion, on this measure of outcome.
A parental questionnaire was used to investigate the outcome
for children who had had ischaemic stroke, who were seen at Method
Great Ormond Street Hospital, London between 1990 and PARTICIPANTS
1996. The results of functional assessments carried out by a Children with ischaemic stroke who were seen at Great
physiotherapist and an occupational therapist, and of Ormond Street Hospital, London between 1990 and 1996
quantitative evaluations carried out by a neuropsychologist were eligible for inclusion in the study. The study group
were used for validation where possible. The relationship included patients who had been referred with acute
between clinical and radiological factors and outcome were ischaemic stroke as well as those who were referred to the
examined. The children were aged between 3 months and 15 paediatric stroke clinic. Patients who had had ischaemic
years at the time of stroke (median age 5 years) and the stroke within the first month of life were excluded from the
period of follow-up ranged from 3 months to 13 years study. Aetiological conditions investigated were cerebrovas-
(median duration 3 years). Of the 90 children for whom data cular abnormalities, cardiac disease, and the presence of any
were obtained, 13 (14%) had no residual impairments. haematological or metabolic risk factors for stroke according
Outcome was good in 37 children (40%) and poor in 53 to a previously published protocol (Kirkham 1994).
(60%) (defined according to whether impairments interfered The study group comprised 128 children with ischaemic
with daily life). Agreement, as measured by Cohen’s kappa, stroke. Information on outcome was available for 105 chil-
was good or very good between the parents’ responses and the dren (i.e. 82%); data were not available for 23 children, 14 of
qualitative measures provided by the medical professionals whom had moved and could not be contacted by post or tele-
and the therapists, but only fair to moderate for the phone. Fifteen children were known to have died. Of the six
quantitative measures provided by the neuropsychologists. children who died soon after the stroke, death was directly
This may reflect different parental perceptions of the physical related to the stroke in three. Of the nine children who died
and cognitive aspects of outcome. Younger age at time of the later, seven died secondary to cardiac disease; the cause of
stroke was the only significant predictor of adverse outcome. death was unknown to us in the remaining two.

PROCEDURE
To evaluate outcome, a questionnaire was designed (see
Appendix) and, with approval from the local ethics commit-
tee, distributed to the parents of all children who had sur-
vived the stroke. For non-responders, the questionnaire was
sent again. If the child was under active follow-up, the par-
ents were given an opportunity to complete the question-
naire in the clinic. The questionnaire was intentionally
simple in design as it has previously been established in
adults that the use of simple questions to evaluate outcome
after stroke corresponds well with the results of formal
assessments (Lindley et al. 1994). Most questions required

Developmental Medicine & Child Neurology 2000, 42: 455–461 455

Downloaded from [Link] University of Rochester, on 18 Nov 2019 at [Link], subject to the Cambridge Core terms of use, available at [Link]
. [Link]
 ‘yes’ or ‘no’ responses. The only items which required quan- terms of arterial territories and more specifically whether
tification related to the severity of the motor impairments in cortical tissue was involved in the lesion.
the upper and lower limbs (questions 5 and 6). In these
instances, structured guidelines were given for the ratings STATISTICAL ANALYSIS
(see Appendix). To examine the agreement between parental and therapists’
Some questions addressed global function, such as reports, and between parental and neuropsychological eval-
whether the child required assistance with activities of daily uations, Cohen’s kappa (KC) was calculated. The accepted
living and whether this was evident in the home or school guidelines for the interpretation of kappa statistics are that 0
environment. Others were specific to certain domains of to 0.2 implies poor agreement, 0.21 to 0.4 fair, 0.41 to 0.6
function (such as motor, speech, language, behaviour, and moderate, 0.61 to 0.8 good, and 0.81 or more very good
education). Parents were also asked to report if the child had (Altman 1995).
epilepsy and had been prescribed anticonvulsant therapy. A logistic regression analysis was used to examine the
To gain an overall indication of the cumulative burden of independent effects of age at time of stroke, time since
residual impairments experienced by the child, the parental stroke, previously recognised risk factor for stroke (present
responses were scored and a composite outcome score was or absent), seizures during the acute presentation (present
calculated for each child. The composite scores ranged from or absent), and infarct location (cortical or subcortical and
0 to 13, with a high score indicating the greatest extent of unilateral or bilateral) on outcome (good or poor). The
residual impairments. The patients were subsequently divid- analysis was performed using both forward and backward
ed into ‘good’ or ‘poor’ outcome subgroups depending on stepwise methods, with p<0.1 as a criterion for removal of
whether their composite score was below or above the arbi- variables from the model. The backward method yielded the
trary cut-off point of 4. Children in the ‘good’ outcome better fit and therefore the results reported here relate to
group were felt to have impairments which were unlikely to this method of analysis. The required two-tailed significance
interfere with daily life, whereas those in the ‘poor’ outcome value was set at 0.05.
group were felt to have impairments which resulted in signif-
icant disability. Although the composite score was a rating, it Results
nevertheless allowed the separation of patients into sub- The clinical characteristics of the patients are summarised in
groups on the basis of the overall severity of residual impair- Tables I and II. They were aged between 3 months and 16
ments, thus facilitating the evaluation of the effects of the years (median age 5 years) at the time of stroke. The interval
clinical factors discussed previously on reported outcome. between stroke and collection of outcome data was between
The parents of 90 children responded to the question- 3 months and 13 years (median 3 years).
naire. The questionnaire was also completed by a paediatric The ischaemic lesions involved the territory of the anteri-
occupational therapist (AG) and a paediatric physiothera- or cerebral circulation in 66 children (73%), the territory of
pist (NS), allowing a comparison of their ratings with those the posterior circulation in 13 children (14%), both distrib-
of the parents. utions in six children (7%), and were confined to the inter-
Neuropsychological evaluations were performed by AH. A nal border zone in two children (2%). Three children (3%)
measure of overall cognitive ability was obtained with the had cerebral infarction secondary to cerebral venous throm-
Bayley Scales of Infant Development (BSID-II; Bayley 1969) bosis. Thirty children (34%) had left-hemisphere lesions, 31
for children under the age of 4 years, while the appropriate
Wechsler scale (Wechsler 1986, 1990, 1992) was used for
older children. Although the BSID-II is a developmental test
rather than an intelligence test, both types of assessment
Table I: Risk factors for stroke
establish a level of current functioning. The tests provide
quantitative scores, but the emphasis in this paper was for Risk factors for stroke Nr of children (n = 90)
the results to describe outcome in a way which could be cor-
related with the parental assessment. In addition, the Moyamoya syndrome 17
Clinical Evaluation of Language Fundamentals, either in Cardiac abnormality 15
revised (CELF-R; Semel et al. 1987) or preschool form (CELF- Sickle cell anaemia 8
CNS infection 5
Preschool; Semel et al. 1992), was used to establish measures
Malignancy 6
of receptive and expressive language. These tests formed Other 6
part of a much wider neuropsychological battery but only
these results will be presented here.
Neuropsychological assessments were carried out in 22
children, 15 of whom had also been examined by the thera-
pists. Four children were assessed using the BSID-II. IQ was Table II: Clinical presentation of stroke
measured with the Wechsler Preschool and Primary Scale of Clinical presentation Nr of children (n = 90)
Intelligence – Revised in five children, the Wechsler Intelli-
gence Scale for Children in 10 children, and the Wechsler Hemiparesis (right:left) 75 (36:39)
Adult Intelligence Scale – Revised in three children. Thirteen Cerebellar syndrome 4
children were assessed with the CELF-R, and two with the Coma 2
CELF-Preschool. Bilateral motor signs 6
Other 3
The reports of MRI scans performed at the time of the
Seizures during presentation 30
stroke were reviewed to find the location of the lesions in

456 Developmental Medicine & Child Neurology 2000, 42: 455–461

Downloaded from [Link] University of Rochester, on 18 Nov 2019 at [Link], subject to the Cambridge Core terms of use, available at [Link]
. [Link]
 (34%) had right-hemisphere lesions, and 28 (31%) had clearly developed handedness before the stroke; 23 of them
bilateral lesions. Fifty-three children (59%) had lesions (26%) had changed handedness after the stroke. Four chil-
which involved the cortex (superficial grey matter) on CT dren (4%) were not able to walk at the time of follow-up; all of
or MRI scans. them had had a second neurological insult after the stroke.
Thirty-eight children (42%) were reported to have speech
OUTCOME REPORTED BY PARENTS and language difficulties; of these, 14 had left-hemisphere
Table III summarises the results obtained from the parental lesions, eight had right-hemisphere lesions, and 16 had bilat-
questionnaires. The overlap in reported impairments is eral lesions.
shown in Figure 1. Fifty-nine percent of the children were reported by their
Seventy-eight of the children (87%) were reported to have parents to need help relative to their peers in the school envi-
ronment. The proportion who had had a statement of educa-
tional needs or who were having educational support is
shown in Figure 2.
Table III: Outcome reported by parents Overall, only 13 children (14%) were reported to have no
residual impairments. Thirty-seven children (41%) had a
Area investigated Nr of children (%) ‘good’ outcome and 53 (59%) had a ‘poor’ outcome.
Child needs additional help in some areas 58 (64)
(more than other children of comparable age) THERAPISTS ’ ASSESSMENTS
Child needs help at home 54 (60) Twenty-two children (24%) had been seen by the paediatric
Child needs help at school 53 (59) therapists who subsequently completed the questionnaire
Child has motor impairment 66 (74) for these children.
Child has difficulty using hemiparetic hand 59 (66) Table IV summarises the agreement between parental
Child has difficulty using hemiparetic leg 59 (66) report and therapists’ assessment. Agreement was good or
Child has difficulty with speech and language 38 (43) very good for the questions which required a qualitative
Parents concerned about child’s behaviour 33 (37)
assessment, with the exception of the assessment of speech
Child is receiving anticonvulsants for epilepsy 13 (15)
Child attends mainstream school 73 (81)
and language where agreement was moderate.
[with help] [30 (33)] Agreement was moderate for the two questions which
Child attends special school 17 (19) required quantitation (questions 5 and 6), although agree-
Child has had statement of educational needs 43 (48) ment was very good as to whether a motor impairment was
No residual impairment 13 (14) present in either the upper or the lower limb. This is not
Good outcome 37 (41)
Poor outcome 53 (59)

50

60 Residual impairment

40 S/L
Motor
50 SEN
Language
Extra support
Nr of children

Educational 30
40
Behavioural
Nr of children

Epilepsy
20
30

10
20

10 0
Needed Not needed
Help at school (parental report)

0
Needed Not needed Figure 2: Educational impairment/support relative to
Recovery: need for additional help parental report of whether child needs help at school.
S/L, children reported to have speech/language
Figure 1: Overlap in residual impairments. ‘Educational’ impairment; SEN, children who have had a statement of
impairment refers to those children who were receiving educational needs; extra support, children receiving
additional educational input. Motor impairment was additional educational support (including help in
usually hemiparesis. mainstream education or special educational provision).

Outcome after Stroke V Ganesan et al. 457

Downloaded from [Link] University of Rochester, on 18 Nov 2019 at [Link], subject to the Cambridge Core terms of use, available at [Link]
. [Link]
 surprising as it has previously been shown that increasing is obtained by considering level of function within each
the number of possible responses tends to diminish the domain – Verbal and Performance.
interobserver reproducibility (Newton et al. 1995). Where We asked parents to report their child’s educational place-
there was a discrepancy in the scoring of the motor impair- ment rather than their general cognitive ability. These data
ments, the parents rated limb function as being worse than are also shown in Table V.
the therapists for all except one child. Information about speech and language function was
available for 19 children. Although the four children who
NEUROPSYCHOLOGICAL EVALUATIONS were tested with the BSID-II did not have additional lan-
Four children under the age of 4 years were assessed with the guage assessment, a language age equivalent facet score can
BSID-II. While there is no universally accepted definition of be determined with the BSID-II language item. Two of the
developmental delay, one criterion which is used frequently children showed an impairment of more than 12 months
is the demonstration of a 20% delay in functioning compared when language-equivalent age was compared with chrono-
with same-age peers (Bayley 1969). By this standard, two of logical age, while two were age appropriate.
the four children demonstrated developmental delay on The CELF gives a total language score as well as expressive
both of the overall developmental indices, mental and psy- and receptive scores, based on a set of subtests measuring
chomotor (MDI and PDI), obtained from this test. One child function in each domain. Eight of the 15 children receiving
had an impaired PDI but average MDI and one child showed the CELF had total language scores which were average or
the reverse pattern. above, four had a mild impairment (defined as performance
The results of the IQ assessments are presented in Table V, between 1 and 2 SDs below the mean), while three had a
which shows the number of children in each of Wechsler’s severe impairment (more than 2 SDs below the mean).
ranges, based on their Full-Scale IQ scores. Language scores were within 5 points of the Verbal IQ in five
Although there is no definitive cut-off point below which of the seven children with an impairment, but were lower by
children are considered to have low cognitive ability, the 15 points or more in two children.
frequencies shown in Table V indicate that more children The receptive and expressive scores were examined sepa-
fall in the bottom half of the distribution than would be rately. In the receptive domain, 11 children showed average
expected in a normal population (χ2, p<0.01). As premorbid or above performance, three showed a mild impairment, and
IQ information was not available, we do not know whether one a severe impairment. Expressive language was more
some of these children would have had low scores before
stroke, but the probability is that stroke has resulted in a fall
in IQ in at least some of the children in the present study.
Twelve of the 18 children who had IQ assessment Table V: IQ results
showed a significant difference between their Verbal and
Wechsler category for IQ Nr of Nr who had Nr receiving
Performance IQ scores according to the criteria set by children an assessment additional
Wechsler (Wechsler 1986, 1990, 1992). In seven of the chil- of educational help at
dren, the Verbal IQ was significantly higher than the needs school
Performance IQ (five had a right-hemisphere lesion, one
was left-sided, and one bilateral), while the opposite was High average (110–119) 3 1 1
true in five children (three with right-sided lesion and two Average (90–109) 6 2 3
Low average (80–89) 4 2 2
with left). In view of these discrepancies, Full-Scale IQ
Low (70–79) 3 1 3
scores may not adequately represent the ability level of the
Exceptionally low (≤ 69) 2 2 2
children, and a more accurate reflection of cognitive ability

Table IV: Agreement between parental report and therapists’/neuropsychological assessments

Question Nr of children who Nr in whom both Agreement – KC

had parental and assessments (95% confidence
professional assessment agreed (%) intervals)

Does the child need additional help in any areas? 18 17 (94) 0.82 (0.49, 1.15)
(more than you would expect for their age)
Does the child need help at home? 14 13(93) 0.76 (0.31, 1.21)
Does the child need help at school? 13 13 (100) 1.00
Does the child have difficulties with speech and language? 18 12 (67) 0.60 (0.15, 1.05)
(parental response compared with therapists response)
Does the child have difficulties with speech and language? 21 13 (62) 0.22 (–0.17, 0.61)
(parental response compared with neuropsychological assessment)
Does the child have a residual motor impairment in upper limb? 10 10 (100) 1.00
How well can the child use their weaker hand? (score 0 to 3) 18 11 (61) 0.47 (0.18, 0.76)
Does the child have a residual motor impairment in the lower limb? 20 19 (95) 0.83 (0.5, 1.16)
How well can the child use his/her weaker leg? (score 0 to 3) 20 11 (55) 0.42 (0.17, 0.67)

KC, Cohen’s kappa.

458 Developmental Medicine & Child Neurology 2000, 42: 455–461

Downloaded from [Link] University of Rochester, on 18 Nov 2019 at [Link], subject to the Cambridge Core terms of use, available at [Link]
. [Link]
 severely affected, with seven children showing no impair- in children (Hariman et al. 1991), these do not touch on
ment, five showing a mild impairment, and three a severe many of the activities of daily living relevant to a child’s life,
impairment. e.g. education, behaviour, and play. Such scales have been
Agreement was only fair between parental perception and shown to correspond well to the responses to simple ques-
neuropsychological evaluation as to whether a child had a tions about recovery and dependence in the investigation of
speech and language impairment. Most disagreements were outcome after stroke in adults (Lindley et al. 1994). The
in children who had evidence of language impairment on design of our questionnaire attempted to modify such sim-
psychometric evaluation which was not reported in the ple outcome questions to explore areas of function more
parental questionnaire. specifically relevant to children.
The immaturity of both the brain and neuromuscular sys-
EFFECT OF CLINICAL FACTORS ON OUTCOME tem in children means that there is significant potential for
A logistic regression model was used to examine the effect of physical, cognitive, and behavioural impairments as well as
clinical factors on outcome (Table VI). R2 for the model was epilepsy after acquired brain injury (Gold et al. 1973).
0.15. Age at the time of stroke emerged as a significant pre- Overall, previous studies have reported that around 75% of
dictor of outcome: the age of the patients with poor outcome children have residual sequelae after ischaemic stroke (Eeg-
was significantly lower. None of the other variables exam- Olofsson and Ringheim 1983, Andrew et al. 1997), similar to
ined (time since stroke, presence of a previous risk factor for the rate reported here. Previous studies of children with
stroke, infarct location – cortical versus subcortical and uni- acute hemiplegia indicated that outcome was particularly
lateral versus bilateral – and seizures at acute presentation) poor in children presenting before the age of 2 years, who
was significantly predictive of outcome. had seizures at onset (Aicardi et al. 1969, Solomon et al.
Despite the wide variation in the duration of follow-up, 1970, Vargha-Khadem et al. 1992). It has previously been
this factor was not a significant determinant of outcome. noted that the cognitive sequelae of hemiplegia (as mea-
sured by IQ) are particularly severe if brain lesions are
Discussion acquired between 1 and 5 years of age (Goodman and Yude
These data suggest that although some children with 1996, Goodman 1997). The data presented here would sug-
ischaemic stroke are left with few sequelae, most have resid- gest that global outcome is also related to the age of lesion
ual difficulties which encompass a wide range of functions, acquisition and that children who experience an insult at a
leading to difficulties both in the home and in the education- young age are especially vulnerable. However, although it
al environment. Younger children appear to have a worse has been suggested that other factors, such as the presence
prognosis than older children; however, other clinical vari- of an underlying risk factor for stroke (Dusser et al. 1986,
ables do not influence outcome. Wanifuchi et al. 1988, Mancini et al. 1997) or lesion location,
This study is hampered by problems common to any mea- particularly involvement of cortical tissue (Dusser et al.
surement of outcome. Although the parental report is not an 1986, Abram et al. 1996) also have an influence on outcome,
objective measure, overall there was broad agreement this was not evident in our group of patients. Possible rea-
between the parents and objective observers. Interestingly, sons for this difference include improved management of
where there were discrepancies, these did not occur in a con- associated conditions, such as cardiac disease, and that our
sistent direction. The relative overestimation of the severity measure of outcome was related to functional impairments.
of motor impairments by parents probably relates to differ- Characterisation of the outcome for children who have
ences in perception of severity between a parent and a thera- had ischaemic stroke is of obvious value to those involved in
pist. The relative underrecognition of language impairment the planning and provision of rehabilitation. Overall, resid-
is likely to reflect the relative insensitivity of informal evalua- ual motor impairments, usually hemiparesis, are common
tion compared with standardised tests of language function. after ischaemic stroke in childhood and may affect up to 90%
The main drawback of this study is clearly the lack of vali- of children (Solomon et al. 1970, Schoenberg et al. 1978,
dation of the questionnaire used here with a standardised Dusser et al. 1986, Giroud et al. 1995). Children with striato-
outcome measure. Although formal measures of indepen- capsular infarcts may experience significant residual dysto-
dence used to assess adult stroke patients such as the nia (Dusser et al. 1986); this appears to be a sequel of
Barthel scale (Barthel and Mahoney 1965) have been used basal-ganglia injury which is unique to childhood (Demierre

Table VI: Results of logistic regression analysis examining effect of clinical factors on outcome

Variable Regression SE (b) Odds ratio 95% CI for p

coefficient (b) (Exp b) odds ratio

Age at time of stroke –0.12 0.06 0.89 0.79–0.99 0.04

Time since stroke –0.02 0.07 0.98 0.85–1.12 0.76
Lesion location
Cortical versus subcortical –0.73 0.24 0.48 0.59 0.48
Unilateral versus bilateral 1.28 –1.23 0.40–4.07 0.13 0.68
Previous risk factor for stroke –0.73 0.53 0.48 0.17–1.36 0.17
(‘symptomatic’ versus ‘idiopathic’ stroke)
Seizures at time of stroke (present or absent) –0.12 0.51 0.89 0.79–1.00 0.82

Outcome after Stroke V Ganesan et al. 459

Downloaded from [Link] University of Rochester, on 18 Nov 2019 at [Link], subject to the Cambridge Core terms of use, available at [Link]
. [Link]
 and Rondot 1983, Jaap Kapelle et al. 1989, Giroud et al. – Parental Report Form (Achenbach 1991) showed that
1995). Our clinical impression in all the patients was that patients tended to have more behavioural and emotional
motor recovery occurred most slowly and incompletely in problems than 12 sibling control children. Although all T
the area of hand function and it may be useful to target this scores were in the average range for both groups, the
area specifically in rehabilitation therapy. As was observed in patients scored significantly higher (indicating greater
this group, it is usual for children to regain ambulation after abnormality) in three of the eight domains examined; ‘with-
ischaemic stroke, unless they have secondary insults (Lanska drawn’, ‘social’, and ‘attention’ (personal communication –
et al. 1991, Hurvitz 1999). AH). These preliminary findings suggest that the data report-
The incidence of epilepsy after ischaemic stroke varies ed in this report are likely to be representative. While behav-
widely between 25 and 50% but appears to be higher than in ioural sequelae may be a direct consequence of lesions of the
adults (Gold et al. 1973, Isler 1984, Dusser et al. 1986). frontal lobes or basal ganglia (Lou et al. 1990), secondary
Cortical lesions and persistence of seizures beyond 2 weeks behavioural problems may also arise as a result of functional
of the acute insult have previously been identified as risk fac- disability. Hyperkinesis, impulsivity, reduced attention, and
tors for residual epilepsy (Yang et al. 1995, Abram et al. unrecognised cognitive impairments may impede function
1996). The incidence of secondary epilepsy is relatively low within a mainstream school environment (Gold et al. 1973);
in recent studies of ischaemic stroke compared with the early difficulties with peer group relations may compound this.
accounts of ‘acute infantile hemiplegia’. Although this is Anxiety may also be an important factor underlying behav-
probably due to exclusion of patients with hemiplegic syn- ioural difficulties (Goodman 1997).
dromes such as HHE (hemiplegia, hemiconvulsions, epilep- One of the frequent comments made by parents surveyed
sy) from the later groups, a short duration of follow-up in in this study concerned the lack of awareness and informa-
some studies may also underestimate the prevalence of even- tion about the spectrum of difficulties experienced by chil-
tual secondary epilepsy. dren after stroke; this may reflect differences between the
The effect of brain injury on language function is com- perception of clinicians and of caregivers about potential
plex; although early clinical observations suggested that lan- areas of difficulty. The lack of information and rehabilitative
guage function is usually spared after early brain injury, this facilities for children with stroke has also received some
has been challenged by subsequent investigations (Vargha- recent attention in the media (Dobson 1996). The personal
Khadem et al. 1994). However, relative preservation of ver- and economic sequelae of childhood stroke in adult life
bal skills may occur at the expense of visuospatial skills or remain unexplored. Although more detailed studies will be
reduction in overall IQ (Vargha-Khadem et al. 1994, required to better characterise these impairments, this pre-
Goodman and Yude 1996). Children functioning within the liminary survey has identified some of the areas in which chil-
‘normal range’ may have qualitative impairments in speech dren who have had ischaemic stroke experience difficulties
and language function (Lees and Neville 1990). Although our and may serve as a guide for professionals involved in the
data suggest that significant speech and language problems, rehabilitation of such children.
recognisable by the child’s parents, are relatively common,
the prevalence of impairments in higher language function Accepted for publication 22nd December 1999.
may be higher than reported here. It was apparent from the
small subgroup of children who underwent formal evalua- Acknowledgements
VG, AH, and FJK were supported by the Wellcome Trust. EI is
tion of language that significant language impairment can go supported by the Medical Research Council. We are grateful to the
unrecognised by lay observers and this area should be specif- parents and children who participated in this study and to Dr
ically targeted for formal evaluation as there may be potential Charles Newton for his comments.
for intervention or educational support.
In a previous study, Abram (1996) found that most chil-
dren who had had a stroke could be educated in a main- Appendix
stream school environment. Although most of the children
in this study were in mainstream education, 56% of our pop- Name: Age at time of stroke:
ulation were in primary school at the time of data collection;
Date of birth: Time since stroke:
secondary education may prove more challenging to this
group of children. Detailed psychometry may reveal subtle 1. Have there been any further episodes similar to the initial stroke?
cognitive impairments in this population, as evident in the No ■
high frequency of children with significant discrepancies Yes, but they resolved fully ■
Yes, my child has had more than one stroke ■
between Verbal and Performance IQs. In particular, the cog-
nitive sequelae of striatocapsular infarction deserve further 2. What type of school does your child go to?
investigation in the paediatric stroke population where such Mainstream school Score 0 ■
lesions are common (Donnan et al. 1991, Nicolai et al. 1995). Mainstream school with help Score 1 ■
Other (please give details) Score 2 if not in mainstream
We were surprised by the high rate of behaviour difficul-
education
ties in our population as this has not been previously high-
lighted as a sequel of ischaemic stroke in children, although 3. Does your child need help in any areas (more than you would
it is similar to that previously noted in a study of children expect at his/her age)? Score 1 if “yes” Yes ■ No ■
with hemiplegia (Goodman 1997). Although our findings If yes, is this at school Score 1 if “yes” Yes ■ No ■
at home Score 1 if “yes” Yes ■ No ■
will need to be confirmed with a standardised behavioural 4. In your view, does your child have any difficulties with speech or
scale, data from a pilot study investigating behaviour out- language? Score 1 if “yes” Yes ■ No ■
come in this population using the Child Behaviour Checklist

460 Developmental Medicine & Child Neurology 2000, 42: 455–461

Downloaded from [Link] University of Rochester, on 18 Nov 2019 at [Link], subject to the Cambridge Core terms of use, available at [Link]
. [Link]
 5. How well is your child able to use his/her weaker hand? Hurvitz E, Beale L, Ried S, Nelson VS. (1999) Functional outcome of
Without any difficulty Score 0 ■ paediatric stroke survivors. Paediatric Rehabilitation 3: 43–51.
Has some difficulties but could use it to hold a Score 1 ■ Isler W. (1984) Stroke in childhood and adolescence. European
jar while taking the lid off with the other hand Neurology 23: 421–4.
Has significant difficulties but can use it as a Score 2 ■ Jaap Kapelle L, Willemse J, Ramos LMP, van Gijn J. (1989) Ischaemic
prop or support e.g. to hold down paper stroke in the basal ganglia and internal capsule in childhood.
when writing Brain Development 11: 283–92.
Is unable to use it at all Score 3 ■ Kirkham FJ. (1994) Stroke in childhood. Current Paediatrics
4: 208–15.
6. How well is your child able to use his/her weaker leg? Lanska MJ, Lanska DJ, Horwitz SJ, Aram DM. (1991) Presentation,
Without any difficulty Score 0 ■ clinical course, and outcome of childhood stroke. Pediatric
Minimal difficulty – e.g. has difficulty running Score 1 ■ Neurology 7: 333–41.
and may drag leg when tired Lees JA, Neville BGR. (1990) Acquired aphasia in childhood: case
Has significant difficulties e.g. can only walk Score 2 ■ studies of five children. Aphasiology 4: 463–78.
short distances Lindley RI, Waddell F, Livingstone M, Sandercoul P, Dennis MS,
Unable to walk Score 3 ■ Slattery J, Smith B, Warlow C. (1994) Can simple questions assess
outcome after stroke? Cerebrovascular Disease 4: 314–24.
7. Do you have any concerns about your child’s behaviour Lou HC, Hentiksen L, Bruhn P. (1990) Focal cerebral dysfunction in
Score 1 if “yes” Yes ■ No ■ developmental learning disabilities. Lancet 335: 8–11.
If so, what are they? Mancini J, Girard N, Chabrol B, Lamoureux S, Livet MO, Thuret I,
Pinsard N. (1997) Ischaemic cerebrovascular disease in children:
8. Has your child ever had a seizure (fit)? Yes ■ No ■ retrospective study of 35 patients. Journal of Child Neurology
If so, when was this? 12: 193–9.
Does your child still Score 1 if “yes” Yes ■ No ■ Newton CRJC, Kirkham FJ, Johnston B, Marsh K. (1995) Inter-
have seizures? observer agreement of the assessment of coma scales and
brainstem signs in non-traumatic coma. Developmental
Please feel free to make any additional comments: Medicine & Child Neurology 37: 807–13.
Nicolai A, Lazzarino LG, Biasutti E. (1995) Large striatocapsular
infarcts: clinical features and risk factors. Journal of Neurology
References 243: 44–50.
Abram H, Knepper E, Warty VS, Painter MJ. (1996) Natural history, — — — (1992)The CELF-Preschool. London: Psychological
prognosis and lipid abnormalities of idiopathic ischemic Corporation.
childhood stroke. Journal of Child Neurology 11: 276–82. Schoenberg BS, Mellinger JF, Schoenberg DG. (1978)
Achenbach TM. (1991) Manual for the Child Behaviour Checklist/ Cerebrovascular disease in infants and children: a study of
4-18 and 1991 profile. Burlington, VT: University of Vermont incidence, clinical features and survival. Neurology 28: 763–8.
Department of Psychiatry. Semel E, Wiig EH, Secord W. (1987) The CELF-R. London:
Aicardi J, Amsili J, Chevrie JJ. (1969) Acute hemiplegia in infancy Psychological Corporation.
and childhood. Developmental Medicine & Child Neurology Solomon GE, Hilal SK, Gold AP, Carter S. (1970) Natural history of
11: 162–73. acute hemiplegia of childhood. Brain 93: 107–20.
Altman DG. (1995) Practical Statistics for Medical Research. Vargha-Khadem F, Isaacs E, Van Der Werf S, Robb S, Wilson J. (1992)
London: Chapman & Hall. p 404. Development of intelligence and memory in children with
Andrew M, David M, deVeber G, Brookner LA. (1997) Arterial hemiplegic cerebral palsy. The deleterious consequences of early
thromboembolic complications in pediatric patients. seizures. Brain 115: 315–29.
Thrombosis and Haemostasis 78: 715–25. — — Muter V. (1994) A review of cognitive outcome after unilateral
Barthel D, Mahoney F. (1965) Functional evaluation. State Medical lesions sustained during childhood. Journal of Child Neurology
Journal 2: 61–4. 9: 2S67–2S73.
Bayley N. (1969) Bayley Scales of Infant Development Manual. Wanifuchi H, Kagawa M, Takeshita M, Izawa M, Kitamura K. (1988)
London: Psychological Corporation. Ischemic stroke in infancy, childhood and adolescence. Child’s
Demierre B, Rondot P. (1983) Dystonia caused by putamino- Nervous System 4: 361–4.
capsulo-caudate vascular lesions. Journal of Neurology, Wechsler D. (1986) The Wechsler Adult Intelligence Scale – Revised
Neurosurgery and Psychiatry 46: 404–9. UK. London: Psychological Corporation.
Dobson R. (1996) A stroke. It can hit you when you’re 65 ...or 12. — (1990) The Wechsler Preschool and Primary Scale of Intelligence
Independent Section Two. p 6–7. – Revised UK. London: Psychological Corporation.
Donnan GA, Bladin PF, Berkovic SF, Longley WA, Saling MM. (1991) — (1992) The Wechsler Intelligence Scale for Children – III UK.
The stroke syndrome of striatocapsular infarction. Brain 114: 51–70. London: Psychological Corporation.
Dusser A, Goutieres F, Aicardi J. (1986) Ischaemic strokes in Yang JS, Park YD, Hartlage PL. (1995) Seizures associated with
children. Journal of Child Neurology 1: 131–6. stroke in childhood. Pediatric Neurology 12: 136–8.
Eeg-Olofsson O, Ringheim Y. (1983) Stroke in children. Clinical
characteristics and prognosis. Acta Paediatrica Scandinavica
72: 391–5.
Giroud M, Lemesle M, Gouyon J, Nivelon J, Milan C, Dumas R.
(1995) Cerebrovascular disease in children under 16 years in
the city of Dijon, France: a study of incidence and clinical
features from 1985 to 1993. Journal of Clinical Epidemiology
48: 1343–8.
Gold AP, Challenor YB, Gilles FH, Hilial SP, Leviton A, Rollins EI,
Solomon GE, Stein BM. (1973) Report of the joint committee for
stroke facilities. IX. Strokes in children. 1. Diagnosis and medical
treatment of strokes in children. Stroke 4: 871–94.
Goodman R. (1997) Psychological aspects of hemiplegia. Archives
of Disease in Childhood 76: 177–81.
— Yude C. (1996) IQ and its predictors in childhood hemiplegia.
Developmental Medicine & Child Neurology 38: 881–9.
Hariman LM, Griffith ER, Hurtig AL, Keehn MT. (1991) Functional
outcomes of children with sickle-cell disease affected by stroke.
Archives of Physical Medicine and Rehabilitation 72: 498–502.

Outcome after Stroke V Ganesan et al. 461

Downloaded from [Link] University of Rochester, on 18 Nov 2019 at [Link], subject to the Cambridge Core terms of use, available at [Link]
. [Link]