Log on /=20 register
Feedback | = Support

Top
Abstract=20
Background<= /A>
Methods=
Results=
Discussion<= /A>
Conclusion<= /A>
Competing=20 interests
Authors'=20 contributions
References =
Pre-publica= tion=20 history

BMC=20 Cancer
Volume=20 8

Viewing options:

Abstract=20
Full text=20
PDF = (332KB)

Associated = material:

Readers'=20 comments
Pre-publicati= on=20 history
PubMed=20 record

Related literature:

Articles citing this article
on=20 Google Scholar

on ISI Web = of Science

on=20 PubMed Central
Other articles by authors

on=20 Google Scholar

= von=20 Hoff K
K= ieffer=20 V
= Habrand=20 JL
Ka= lifa=20 C
Dellatolas=20 G
Gri= ll=20 J

on=20 PubMed

von=20 Hoff K
Kieffer=20 V
Habrand=20 JL
Kalifa=20 C
Dellatolas=20 G
Grill=20 J
Related articles/pages

on=20 Google

on=20 Google Scholar

on=20 PubMed

Tools:

Download=20 references
Download=20 XML
Email to=20 a friend
Order=20 reprints
Post=20 a comment=20
Sign up for=20 article alerts

Post to:

= Citeulike=20
&= nbsp;Connotea=20
= Del.icio.us=20
= ;Digg=20
&= nbsp;Facebook=20

Research article

Impairment of intellectual functions after surgery and = posterior fossa=20 irradiation in children with ependymoma is related to age and = neurologic=20 complications

Katja von = Hoff¹^,2=20 ,=20 Virginie Kieffer¹^,3 ,=20 Jean-Louis Habrand⁴ ,=20 Chantal Kalifa¹ ,=20 Georges Dellatolas⁵ and = Jacques Grill¹

¹Department of paediatric and adolescent oncology, = Gustave=20 Roussy institute, 39 rue Camille Desmoulins, 94805 Villejuif = cedex,=20 France

²Childrens university hospital Wuerzburg, = Josef-Schneider=20 str.2, 97080 Wuerzburg, Germany

³Ressource center for patients with brain injuries, = national=20 rehabilitation hospital, 14 rue du val d'Osne, 94415 = Saint-Maurice,=20 France

⁴Department of radiotherapy, Gustave Roussy = institute, 39=20 rue Camille Desmoulins, 94805 Villejuif cedex, France

⁵Laboratory of psychology and cognitive = neurosciences,=20 CNRS-FRE 2987, 71 avenue Edouard Vaillant, 92774 = Boulogne-Billancourt,=20 France

=20 author email=20 corresponding author email

BMC Cancer 2008, = 8:15doi:10.1186/1471-2407-8-15

The electronic version of this article is the complete one and = can be=20 found online at: http://www.biomedcen= tral.com/1471-2407/8/15

Received:	23 March 2007
Accepted:	21 January 2008
Published:	21 January 2008

=C2=A9 2008 von Hoff et al; licensee BioMed Central = Ltd.
This is an Open=20 Access article distributed under the terms of the Creative Commons = Attribution License (http://creativecommon= s.org/licenses/by/2.0),=20 which permits unrestricted use, distribution, and reproduction in = any=20 medium, provided the original work is properly cited.

Abstract

Background

To investigate the neuropsychological outcome of children = treated with=20 surgery and posterior fossa irradiation for localized = infratentorial=20 ependymoma.

Methods

23 patients (age 0.3 =E2=80=93 14 years at diagnosis) who were = treated with=20 local posterior fossa irradiation (54 Gy) underwent one (4 = patients) or=20 sequential (19 patients) neuropsychologic evaluation. The last = evaluation=20 was performed at a median of 4.5 (1 to 15.5) years after RT.

Results

Mean last full scale IQ (FSIQ), verbal IQ (VIQ) and PIQ were = 89.1,=20 94.0, and 86.2 respectively. All patients had difficulties with = reading,=20 and individual patients showed deficits in visuospatial, memory = and=20 attentional tasks. There was no trend for deterioration of = intellectual=20 outcome over time. All 5 children with IQ scores =E2=89=A4 75 were = under the age=20 of four at diagnosis. There was a significant association between = the=20 presence of cerebellar deficits and impaired IQ (72.0 vs 95.2, p = <=20 0,001). The absence of hydrocephalus was an indicator of better=20 neuropsychologic outcome (mean FSIQ of 102.6 vs 83.9, p =3D = 0.025).

Conclusion

Within the evaluated cohort, intellectual functions were = moderately=20 impaired. Markedly reduced IQ scores were only seen with early = disease=20 manifestation and treatment, and postoperative neurological = deficits had a=20 strong impact on intellectual outcome.

Background

Within the posterior fossa, ependymoma is the second most = common=20 malignant tumour in children [1]. As = with=20 other paediatric central nervous system (CNS) tumours, finding the = balance=20 between effective treatment and preservation of psychomotor = development is=20 challenging. Modern approaches aim at maximizing surgical = resection while=20 reducing the volume of irradiation since complete tumour removal = is the=20 main prognostic factor [2-4].

As a consequences of brain damage caused by the tumor itself = and the=20 surgery, some children develop neurologic deficits such as = cerebellar=20 dysfunction and cranial nerve palsies [5]. = Indeed,=20 radiation therapy rarely causes neurologic damage in the absence = of=20 complications such as radionecrosis or stroke. More aggressive = surgery may=20 thus increase the risk of neurologic damage.

Progressive intellectual impairment is a serious side effect of = whole=20 brain irradiation [6-9], = the extent=20 to which intellectual capacities are also diminished due to local=20 radiation to the posterior fossa remains to be determined. = Intellectual=20 quotient (IQ) is preserved in patients with ependymoma after = posterior=20 fossa irradiation only, compared to children with medulloblastoma = who=20 received craniospinal irradiation (CSI) [10].=20 Furthermore preliminary data suggest that there may be only = limited=20 decline in neurocognitive functions after local posterior fossa=20 irradiation [3,11].

To determine the risk factors for intellectual impairment and = to define=20 the neuropsychological profile of long term survivors of localised = infratentorial ependymoma we analysed the long-term = neuropsychological=20 outcome of children who received posterior fossa radiotherapy in a = cohort=20 of patients treated between 1986 and 2003 either at diagnosis (in = children=20 over 5 years of age) or after first relapse following chemotherapy = in=20 younger children. Patients who were diagnosed 1998 and later were=20 evaluated prospectively.

All potential risk factors for intellectual impairment [12] = were=20 studied, including pre-operative complications such as = hydrocephalus,=20 surgical complications and persistent cerebellar deficits, age, = and=20 radiation volume (conformational versus whole posterior = fossa)

Methods

Patients

Patients were included in this study if they (i) were diagnosed = and=20 operated on a localised infratentorial ependymoma, (ii) received = local=20 posterior fossa irradiation at the Institute Gustave-Roussy in = Villejuif=20 between 1986 and 2003, as initial treatment or after chemotherapy=20 according to the BBSFOP protocol (Carboplatin/Procarbazin;=20 Etoposide/Cisplatin; Vincristine/Cyclophosphamide) [13], = (iii) had=20 at least one standardised neuropsychologic evaluation, and (iii) = had no=20 abnormal premorbid psychomotor development as reported by the = parents.

Twenty-three patients fulfilled these criteria. Informed = consent was=20 obtained from all patients. Patient characteristics are shown in = table 1. Age = at=20 diagnosis ranged between 0.3 and 14.2 years (median 7.2). Of ten = patients=20 who were under the age of five at diagnosis, eight were irradiated = under=20 the age of five, three of them were irradiated before the age of = three.=20 There was a male predominance with 17 boys within the group. 16 = patients=20 were presenting with signs of intracranial pressure at disease=20 manifestation. All patients had surgical resection with gross = total=20 resection achieved in 18 patients. Four patients received = postoperative=20 chemotherapy according to the French BBSFOP protocol [13] = and=20 commenced to radiotherapy due to progression of residual tumour or = relapse.

Tab= le=20 1. General characteristics of 23 patients included in = the=20 study.

Radiotherapy

All patients were treated using megavoltage equipments (4.5 to = 20 MV=20 photons of a linear accelerator). Total dose ranged between 50 and = 62 Gy,=20 administered in 5 weekly sessions of 1.8 Gy per day, with each = beam=20 treated every day. The highest doses correspond to patients with = gross=20 residual disease present at the time of radiations. A computerized = dose-distribution was made available in all patients using the=20 DOSIGRAY^=C2=AEsoftware. In early patients, it was = based on=20 radiographic simulation films with hand-drawn tailored shieldings, = based=20 on physician knowledge of the anatomical landmarks, and tumour=20 characteristics. More recently, a 3D high definition CT-scan based = representation of dose-distribution superimposed with the = posterior fossa=20 structures, and tumour contour was made available. Dose-volume = histograms=20 for structures of interest were also generated. As far as = technical=20 considerations, early patients were treated in a straightforward = approach=20 combining two opposed laterals; recent 3D simulation, allowed = conformation=20 to the target with optimal sparing of adjacent organs (mainly = pituitary,=20 cochleas, chiasm). The gross tumour volume (GTV) for the primary = site=20 boost included the postoperative tumor bed. The clinical target = volume=20 (CTV) included the GTV, with an anatomically confined margin of 2 = cm in=20 the adjacent brain, whereas the planned target volume (PTV) = expanded the=20 CTV with a geometric margin of 1 cm. Multiple beams arrangements = were=20 used, ie 2 to 4 wedge anterior and/or posterior obliques. The = early=20 approach induced full dose of radiations in the entire posterior = fossa,=20 along with occipital and posterior temporal lobes. Only pituitary = located=20 at anterior margin, was kept to an acceptable level. The recent = approach=20 allowed marked reduced maximal dose to most structures outside the = posterior fossa, including cochleas occipital and parietal lobes. = The=20 reverse side is that doses to the pituitary as well as integral = dose to=20 the temporal lobes were slightly increased due to beams exits.

Neuropsychologic evaluation

A battery of age adapted standard neuropsychological tests was = applied=20 to all patients. This included an IQ measure using Wechsler scales = WAIS-R=20 for adults, WISC-III for children =E2=89=A5 7 years and WPPSI-R = for children aged=20 < 7 years [14,15]. = WISC-III=20 consists of 10 obligatory and 3 optional subtests with a range of = test=20 scores between 1 and 19 (average: 10). Complementary tests were = used to=20 describe patients neurocognitive abilities as previously described = by our=20 group [16].=20 Additionally reading skills were measured by using the test of the = alouette [17]. = Executive=20 functions were evaluated using the Wisconsin card sorting test = (WCST). The=20 evaluation was completed by the judgment of line orientation [18], = facial=20 recognition [19], = a copy of=20 the Rey =E2=80=93 Osterrieth complex figure for children over 7 = and analysis of=20 fine motor skills with the Purdue pegboard test [20]. = This=20 latter test evaluates fine motor speed with the dominant and = non-dominant=20 hand both separately and together. The tests were timed, and a = period of=20 three hours was allowed for the entire evaluation. They were = always=20 performed in the same order. Information regarding school = placement, both=20 before disease onset and at the time of the neuropsychological = evaluation,=20 was also collected from parent's interview.

Tests were performed longitudinally in 19 patients. Of them 13 = patients=20 were evaluated prospectively and had baseline evaluation within = the first=20 year after the completion of radiotherapy. One of them was too = young for=20 WPPSI-R and received K-ABC [21]. = Six=20 patients were first tested >1 year after completion of = radiotherapy=20 (1.1=E2=80=9311.6 years, median 7.6). Four patients had only one=20 neuropsychological evaluation between 3.9 and 8.6 years after = completion=20 of RT (median 7).

Presence of cerebellar syndrome (Ataxia, Dysmetria, Nystagmus) = was=20 graduated as mild, moderate, or severe according to the impact on = daily=20 activities by an independent physician unaware of the = neuropsychological=20 performance using Riva's rating scale [22].

All patients were regularly screened for endocrinologic = deficits and=20 hearing impairment.

Statistical Analysis

Statistical analysis was conducted using SPSS software (12.0 = Version).=20 Test results of the neuropsychological test (except IQ measures = and=20 subgroups) were normalised and transferred into Z-scores where = score >=3D=20 2 corresponds to a probability of 95% to be outside of normal=20 distribution.

The neuropsychologic profile was analysed descriptively based = on the=20 results of Wechsler subtests and the above mentioned additional = test.

For analysis of risk factors for intellectual impairment, = patients were=20 divided into groups according to: age at radiotherapy (<5 y vs. = =E2=89=A5 5 y);=20 cerebellar syndrome, fine motor achievment; hydrocephalus at = presentation,=20 radiotherapy volume (conformal vs. posterior fossa). For each = patient the=20 result of the last FSIQ test was used. Comparison was done using=20 Mann-Whitney-U test for non-parametrical data.

The age limit of 5 years was chosen due to reasons of clinical=20 practice. Patients below 5 years of age were eligible for adjuvant = BBSFOP=20 chemotherapy [13]. = Patients=20 aged 5 or older would receive immediately adjuvant radiotherapy = according=20 to our institutional standard. Influence of age at RT was also = analysed=20 using linear regression.

Due to the small group size a multivariable analysis of risk = factors=20 was not reasonable and was therefore omitted.

Longitudinal data of achievment (FSIQ and reading) were = analysed=20 descriptively. Due to the small sample size and limited = reliability of=20 potential findings a random coefficient model was not used.

Results

The last neuropsychologic evaluation was done at a median of = 4.5 years=20 after the completion of radiotherapy (range 1 =E2=80=93 15.5 = years). At the last=20 testing mean full scale intelligence quotient (FSIQ), verbal IQ = (VIQ) and=20 performance IQ (PIQ) were 89.1 (standard deviation SD 14.6), 94.0 = (SD=20 12.4), and 86.2 (SD 16.1). Of the 23 evaluable patients FSIQ was = 90 or=20 above in 10 patients (43%), between 80 and 90 in eight patients = (35%), and=20 below 80 in five patients (22%).

Profile of neuropsychological evaluation

In most of the WISC III subtests, scores were within average = limits=20 with mean scores above or equal 8 in 10 of 13 subtests (norms are = 10 +/- 2=20 for each subtest). None of the patients showed a significant = (=E2=89=A4 2 SD)=20 impairment of VIQ. Of the VIQ subtests 2 children achieved low = test scores=20 in the "information" subtest testing general knowledge. Four = patients=20 showed significant impairments in the optional memory subtest. PIQ = was=20 below VIQ in 17 patients with a mean difference of 7.8 points and = a=20 significant impairment in 6 patients. A marked impairment was seen = in the=20 chessboard/coding subtest and limited impairments in object = assembly,=20 symbols and picture arrangement, subtests evaluating speed of = written=20 performance and the capacity of visuo-spatial observation and=20 organisation, respectively. Mean processing speed was also = reduced. Table=20 2 = shows=20 the IQ subtest scores.

Tab= le=20 2. Wechsler scale (WISC-III) subtest results of 23=20 patients.

All of the 12 tested patients showed impairment in their = reading=20 skills, with a lag of 1 to 5 years between "reading age" and = chronological=20 age (median 3.8 years). The discrepancy was growing with time in = all 8=20 patients with sequential testing (figure 1).=20 Concerning visuospatial capacities, 3 of 16 patients had severe=20 difficulties in reproducing the Rey-Osterrieth complex figure = (mean=20 Z-score of the whole group being -1.01 SD) while none of 11 = patients=20 tested had severe difficulties with the benton line orientation = test (mean=20 Z-score of the whole group being 0.56 SD). Short term memory = measured by=20 digit span was significantly diminished in 1 of 10 tested patients = (mean=20 Z-score of the whole group being-0.62 SD) and long term memory = measured by=20 word list was significantly diminished in 3 of 19 tested patients = (mean=20 Z-score of the whole group being -0.92 SD). Overall results of 12 = patients=20 who received the wisconsin card sorting test (WSCT) were within = average=20 limits, but 7 of these 12 showed attentional deficits with slow = adaption,=20 the tendency to keep one strategy, difficulties with reasoning, = and=20 problems to maintain the intentional thread. Difficulties within = the WSCT=20 were not correlated to the IQ scores.

Figure 1. Reading=20 performances. Differences between chronological age and = reading=20 age in years of 12 patients at different time points after therapy = (4 had=20 one test, 8 had sequential testing); Results of individual = patients are=20 connected with lines.

Except reading skills, none of the tests showed declining = results over=20 time after therapy.

Risk factor analysis (see table 3)

Tab= le=20 3. Medical history of the patients and full scale IQ = scores=20 at last evaluation.

Age

Low IQ results occurred mainly in the young age group. Figure = 2 = shows=20 the distribution of FSIQ values at last evaluation and age at = irradiation.=20 Comparing the IQ results of children younger than 5 years at = irradiation=20 with children who were older at the time of irradiation, the = difference=20 failed significance but there was a trend for poorer outcome in = younger=20 children. Mean FSIQ was 82.7 (n =3D 8, SD 17.2), and 92.5 (n =3D = 15, SD 12.8)=20 respectively (p =3D 0.1).

Figure 2. Of ten patients with = age < 5=20 years at diagnosis, eight were irradiated before the age of 5. = Regression=20 line is also indicated (r =3D 0.22; p =3D 0.3). Black circles =3D = patients >=20 5 y at diagnosis. Red circles =3D patients < 5 y at=20 diagnosis.

Interval since RT and FSIQ

Of 13 patients with a baseline evaluation, 10 patients were = tested=20 before the start of irradiation and three patients within the = first year=20 after completion of radiotherapy. One of these patients was below = 3 years=20 of age at diagnosis, therefore he received age adapted tests = without IQ=20 testing. Of the remaining 12 patients mean baseline FSIQ, VIQ and = PIQ (SD)=20 was 91.6 (10.6), 98.4 (8.9) and 85.8 (13.6). The only patient with = a=20 baseline FSIQ below 75 was diagnosed at 2.5 years and received = delayed RT=20 with neuropsychological evaluation and onset of irradiation at 5.9 = years.

In the longitudinal analysis there was no trend for loss of=20 intellectual capacity over time after completion of irradiation. = Figure 3 = shows=20 FSIQ scores of the sequential evaluations over time. There were = six=20 patients showing a decline in the measured IQ results, while five = were=20 gaining points. Of 11 patients who had baseline IQ testing and = evaluation=20 2=E2=80=935 years after radiotherapy (median 3.5 years) mean FSIQ = (SD) at baseline=20 and at last evaluation were 91.9 (11.0) and 91.3 (13.2), = respectively. The=20 difference between evaluations ranged between -10 to +12 points, = with a=20 median difference of 0.

Figure 3. FSIQ of the 23 = patients=20 at different time points after RT. Results of individual = patients=20 are connected with lines.

Cerebellar syndrome and other influencing factors

At the time of last neuropsychological evaluation 6 patients = had a=20 cerebellar syndrome. There was a strong correlation with decreased = IQ=20 scores. Mean FSIQ (SD) was 72.0 (6.3) within the group of children = with=20 persisting cerebellar syndrome compared to 95.2 (12.0) within the = group of=20 children showing no signs of cerebellar syndrome. This difference = was=20 highly significant (p < 0.001) (figure 4). = Hand=20 motor speed measured by Purdue pegboard evaluation was also highly = correlated with FSIQ results (p =3D 0.003). With only 5 patients = showing no=20 signs of elevated intraventricular pressure (IVP) at initial = presentation,=20 the negative influence of elevated IVP was however significant (p = =3D=20 0.025). Mean FSIQ with and without IVP at presentation was 83.9 = (SD12.5)=20 and 102.6 (SD14.4), respectively.

Figure 4. FSIQ with = absent (n =3D 16)=20 and present (n =3D 6) cerebellar syndrome (CS) at time of = neuropsychological=20 evaluation. Representation of the results is given as a = Tukey and=20 Cleveland's box-plot. The box represents the 5 principal centiles, = ie 50%=20 of the distribution. The line in the middle of the box represents = the=20 median. The line on top of the box joins the 90th centile. The = line below=20 the box joins the 10th percentile. Circles indicate the extreme=20 values.

Patients with opposite lateral beams radiotherapy showed a = lower mean=20 FSIQ (SD) of 86.2 (16.7) compared to patients with conformal = radiotherapy=20 with FSIQ (SD) of 92.4 (12.6). The difference did not reach = significance=20 (p =3D 0.21).

Schooling

Three patients attended a regular school but had a delay of 2 = or more=20 years. One patient visited a special institute. The other patients = were=20 attending regular schools with no more than one year delay.

Endocrine deficits

Four patients had endocrinologic deficits which needed = substitution=20 (precocious puberty 2, growth hormone deficit 2). None of the = patients had=20 severe hearing impairment.

Discussion

We conducted neuropsychological evaluations in patients with = localised=20 infratentorial ependymoma who received surgery and irradiation = limited to=20 the posterior fossa. Mean IQ scores evaluated with Wechsler IQ = tests=20 showed an overall moderate impairment but mean FSIQ remained in = the normal=20 range. While some patients had significant impairments in their = PIQ=20 scores, no significant impairment was seen in the VIQ score. = Compared to=20 previously published outcome scores of children who received whole = brain=20 irradiation these impairments were limited [16]. = In our=20 study IQ measures showed a high variability at all evaluated time = points.=20 Within the group which could be evaluated longitudinally (19/23), = there=20 was no trend for deterioration of intellectual functioning over = time. This=20 finding is in contrast to studies on medulloblastoma patients = receiving=20 CSI, who showed a deterioration of intellectual functions for = years after=20 the completion of therapy [7,23-25], and = it=20 is supporting the data by Merchant et al, who evaluated the = influence of=20 conformal RT for the treatment of ependymoma to intellectual = outcome. With=20 radiation limited to the tumour volume, they described stable = intellectual=20 functions with a median follow up of 3 years [3]. A = recent=20 report from Fouladi et al. also showed no significant longitudinal = decline=20 of IQ measures of patients with infratentorial tumours who = received local=20 RT compared to CSI [26].

With local posterior fossa irradiation, large parts of the=20 supratentorial hemispheres and white matter are spared from = irradiation,=20 which might explain that there is no gradual IQ drop as it is seen = after=20 whole brain irradiation. Our data support this hypothesis since = children=20 receiving conformal RT tended to show a better outcome than those = treated=20 with opposite lateral beams. Merchant et al analysed with = radiation=20 dosimetry models that volume and dose of irradiation of the = supratentorial=20 brain was predictive for IQ in localised infratentorial ependymoma = [27], = which=20 supports the above mentioned concept.

In our study very low IQ results were only observed in young = children,=20 but there was no statistical significant correlation between age = at=20 irradiation and intellectual outcome within our limited study = population.=20 While in different studies on patients who received CSI the = progressive=20 deterioration of neuropsychological functions was more pronounced = in=20 younger children [28,29], = in our=20 study there was no significant age dependent decrease of = intellectual=20 functions, and IQ results achieved at baseline evaluation and at = follow up=20 evaluations showed no difference. There was however a trend for = worse=20 outcome in younger children. But larger sample may be necessary to = show a=20 clear difference in outcome. Therefore we suppose that local = posterior=20 fossa RT is unlikely to be the only factor causing worse=20 neuropsychological outcome in young children. As in our study, = there were=20 only 3 children, who were treated with radiotherapy before the age = of 3,=20 we are not able to draw definite conclusion about the role of very = young=20 age in the intellectual deficit after posterior fossa RT.

The intellectual deficits reported in our study might reflect = also=20 damages accrued by the disease and surgical therapy. This concept = is=20 supported by studies showing that IQ is impaired in survivors of = posterior=20 fossa tumours even in the absence of radiotherapy [22,30,31] = suggesting=20 contributing factors of the disease itself and surgical therapy on = neuropsychological outcome.

Looking for other factors which could predict for low IQ = performance we=20 found a strong correlation of IQ and cerebellar damage, measured = by the=20 presence of cerebellar syndrome at the time of neuropsychological=20 evaluation. A pivotal role of cerebellar damage for the presence = of=20 intellectual deficits was described recently by our group in a = study=20 evaluating 76 children with posterior fossa tumours, where disease = factors=20 and surgical complications were exceeding the negative effects of = adjuvant=20 therapy. Interestingly persistent cerebellar syndrome was more = frequent in=20 the latter described study population (51%) which consisted mainly = of=20 medulloblastoma patients, compared to this study (26%) [32].

Another factor which showed a trend to negatively influence the = intellectual outcome in our study was hydrocephalus at = presentation.=20 Merchant et al analysed ventricular enlargement by MRI at = different time=20 points in patients with infratentorial ependymoma. They stated a = relevant=20 influence of hydrocephalus on intellectual achievement, while they = postulated that the negative influence of ventricular enlargement = was=20 reversible if ventricular size decreases over time [33]. = Since=20 there was no regular longitudinal measurement in our cohort, we = could not=20 evaluate the influence of change in ventricular size.

Concerning the neuropsychological profile, the subtest analysis = of the=20 Wechsler IQ test showed impairments concerning processing speed = and visual=20 motor skills. Individual patients had reduced scores in subtests=20 reflecting visual perceptive and memory problems, whereas the = overall=20 performance on these tasks was just slightly decreased. The = impaired=20 reading capacities may reflect problems with speed and possibly = also=20 visual problems. The battery of additional tests showed an = increase of the=20 lag between reading age and chronological age over time since = therapy in=20 all tested patients, which is likely due to a reduced rate of = skill=20 acquisition. This highlights that tests exploring reading skills = are=20 usefull read-outs for the monitoring of the outcome of these = children.=20 Furthermore there were individual deficits in visuospatial = capacities, in=20 attention and in memory functions. Similar deficits are described = in=20 patients suffering cerebellar astrocytoma [34-36] and=20 medulloblastomas [16]. = Although=20 there seems to be a common spectrum of deficits, we like to = emphasis, that=20 there was a wide variability and that we couldn't detect a clear = pattern=20 of impairment. Possibly the diversity of impairments reflects the=20 differing influence of perioperative and intraoperative damage = done to the=20 brain.

Conclusion

In conclusion, our data show that intellectual functions are = moderately=20 impaired in survivors of infratentorial ependymoma. Compared to = children=20 who received CSI, neuropsychological outcome was favourable in = children=20 who received only local posterior fossa radiotherapy.

There was a wide variability of the level of intellectual = achievements=20 and specific impairments. The high variability is likely to be = caused by=20 cerebellar and cerebral damage reflecting the influence of disease = and=20 surgery-related factors. Studies looking at therapy optimization = should=20 include neurological and cognitive evaluations to further describe = the=20 influencing factors and possible mechanisms of intellectual = impairment.=20 This report also indicates that further refinement of adjuvant = therapy for=20 ependymoma should include means to deliver radiation with limited = fields=20 and better chemotherapies to defer radiotherapy in the youngest = patients.=20 Children should also be always monitored for neurological and=20 neuropsychological outcome to ensure that they get the necessary = support=20 for rehabilitation.

Competing interests

The author(s) declare that they have no competing = interests.

Authors' contributions

KVH participated in the design of the study, collected the = data,=20 performed the statistical analysis and drafted the manuscript.

VK conceived the study, participated in its design and = coordination,=20 performed and evaluated the neurocognitive tests, and helped in = drafting=20 the manuscript.

JLH participated performed the irradiation and evaluated the=20 radiotherapy files.

CK conceived the study and followed the patients = clinically.

GD conceived the study, participated in its design and=20 coordination.

JG conceived the study, participated in its design and = coordiantion,=20 helped to collect the data, to perform the statistical analysis, = and to=20 draft the manuscript.

All authors read and approved the final manuscript.

References

Kaatsch P, Rickert CH, Kuhl J, Schuz J, = Michaelis J:=20 Population-based epidemiologic data on brain tumors in = German=20 children.
Cancer 2001,=20 92(12):3155-3164. PubMed=20 Abstract | Publisher Full Text
Perilongo G, Massimino M, Sotti G, = Belfontali T,=20 Masiero L, Rigobello L, Garre L, Carli M, Lombardi F, Solero C, = Sainati=20 L, Canale V, del Prever AB, Giangaspero F, Andreussi L, Mazza C, = Madon=20 E: Analyses of prognostic factors in a retrospective = review of=20 92 children with ependymoma: Italian Pediatric Neuro-oncology = Group.=20
Med Pediatr Oncol 1997,=20 29(2):79-85. PubMed=20 Abstract | Publisher Full Text
Merchant TE, Mulhern RK, Krasin MJ, Kun LE, = Williams=20 T, Li C, Xiong X, Khan RB, Lustig RH, Boop FA, Sanford RA:=20 Preliminary results from a phase II trial of conformal = radiation=20 therapy and evaluation of radiation-related CNS effects for = pediatric=20 patients with localized ependymoma.
J Clin Oncol 2004,=20 22(15):3156-3162. PubMed=20 Abstract | Publisher Full Text
Horn B, Heideman R, Geyer R, Pollack I, = Packer R,=20 Goldwein J, Tomita T, Schomberg P, Ater J, Luchtman-Jones L, = Rivlin K,=20 Lamborn K, Prados M, Bollen A, Berger M, Dahl G, McNeil E, = Patterson K,=20 Shaw D, Kubalik M, Russo C: A multi-institutional = retrospective=20 study of intracranial ependymoma in children: identification of = risk=20 factors.
J Pediatr Hematol Oncol 1999,=20 21(3):203-211. PubMed=20 Abstract | Publisher Full Text
van Veelen-Vincent ML, Pierre-Kahn A, Kalifa = C,=20 Sainte-Rose C, Zerah M, Thorne J, Renier D: Ependymoma = in=20 childhood: prognostic factors, extent of surgery, and adjuvant = therapy.=20
J Neurosurg 2002,=20 97(4):827-835. PubMed=20 Abstract | Publisher Full=20 Text
Mulhern RK, Merchant TE, Gajjar A, Reddick = WE, Kun LE:=20 Late neurocognitive sequelae in survivors of brain = tumours in=20 childhood.
Lancet Oncol 2004,=20 5(7):399-408. PubMed=20 Abstract | Publisher Full Text
Spiegler BJ, Bouffet E, Greenberg ML, Rutka = JT,=20 Mabbott DJ: Change in neurocognitive functioning after = treatment=20 with cranial radiation in childhood.
J Clin Oncol 2004,=20 22(4):706-713. PubMed=20 Abstract | Publisher Full Text
Duffner PK, Cohen ME, Thomas P: Late = effects=20 of treatment on the intelligence of children with posterior = fossa=20 tumors.
Cancer 1983,=20 51(2):233-237. PubMed=20 Abstract | Publisher Full Text
Palmer SL, Goloubeva O, Reddick WE, Glass = JO, Gajjar=20 A, Kun L, Merchant TE, Mulhern RK: Patterns of = intellectual=20 development among survivors of pediatric medulloblastoma: a = longitudinal=20 analysis.
J Clin Oncol 2001,=20 19(8):2302-2308. PubMed=20 Abstract | Publisher Full Text
Grill J, Renaux VK, Bulteau C, Viguier D,=20 Levy-Piebois C, Sainte-Rose C, Dellatolas G, Raquin MA, Jambaque = I,=20 Kalifa C: Long-term intellectual outcome in children = with=20 posterior fossa tumors according to radiation doses and volumes. =
Int J Radiat Oncol Biol Phys 1999,=20 45(1):137-145. PubMed=20 Abstract | Publisher Full Text
Kieffer-Renaux V, Viguier D, Raquin MA,=20 Laurent-Vannier A, Habrand JL, Dellatolas G, Kalifa C, Hartmann = O, Grill=20 J: Therapeutic schedules influence the pattern of = intellectual=20 decline after irradiation of posterior fossa tumors.
Pediatr Blood Cancer 2005,=20 45(6):814-819. PubMed=20 Abstract | Publisher Full Text
Grill J, Kieffer V, Kalifa C: = Measuring the=20 neuro-cognitive side-effects of irradiation in children with = brain=20 tumors.
Pediatr Blood Cancer 2004,=20 42(5):452-456. PubMed=20 Abstract | Publisher Full Text
Grill J, Le Deley MC, Gambarelli D, Raquin = MA,=20 Couanet D, Pierre-Kahn A, Habrand JL, Doz F, Frappaz D, Gentet = JC, Edan=20 C, Chastagner P, Kalifa C: Postoperative chemotherapy = without=20 irradiation for ependymoma in children under 5 years of age: a=20 multicenter trial of the French Society of Pediatric Oncology. =
J Clin Oncol 2001,=20 19(5):1288-1296. PubMed=20 Abstract | Publisher Full Text
Wechsler D: Manual for the Wechsler = preschool=20 and primary scale of intelligence - Revised.
The Psychological Corporation 1989.
Wechsler D: Wechsler intelligence = scale for=20 children - Third edition.
The Psychological Corporation 1991.
Kieffer-Renaux V, Bulteau C, Grill J, = Kalifa C,=20 Viguier D, Jambaque I: Patterns of neuropsychological = deficits=20 in children with medulloblastoma according to craniospatial = irradiation=20 doses.
Dev Med Child Neurol 2000,=20 42(11):741-745. PubMed=20 Abstract | Publisher Full Text
Lefavrais P: Test de = l=E2=80=99alouette.
Les editions du centre de psychologie = appliqu=C3=A9=C3=A9 1967.
Benton AL, Sivan AB, deS. Hamsher K, Varney = NR,=20 Spreen O: Judgment of line orientation, forme V: = Contribution to=20 neuropsychological assessment.
oxford university press, inc 1983.
Benton AL, Sivan AB, deS. Hamsher K, Varney = NR,=20 Spreen O: Facial recognition. Contribution to = neuropsychological=20 assessment.
oxford university press, inc 1978.
Tiffin J: Examiner manual for the = purdue=20 pegboard.
Purdue research foundation (version fran=C3=A7aise par = ECPA)=20 1961.
Kaufman AS, Kaufman NL: K.ABC. = Batterie pour=20 l'examen psychologique de l'enfant.
American Guidance Service, Inc 1983.
Riva D, Giorgi C: The cerebellum = contributes=20 to higher functions during development: evidence from a series = of=20 children surgically treated for posterior fossa tumours. =
Brain 2000, 123 ( Pt=20 5):1051-1061. PubMed=20 Abstract | Publisher Full Text
Ris MD, Packer R, Goldwein J, Jones-Wallace = D, Boyett=20 JM: Intellectual outcome after reduced-dose radiation = therapy=20 plus adjuvant chemotherapy for medulloblastoma: a Children's = Cancer=20 Group study.
J Clin Oncol 2001,=20 19(15):3470-3476. PubMed=20 Abstract | Publisher Full Text
Palmer SL, Gajjar A, Reddick WE, Glass JO, = Kun LE, Wu=20 S, Xiong X, Mulhern RK: Predicting intellectual outcome = among=20 children treated with 35-40 Gy craniospinal irradiation for=20 medulloblastoma.
Neuropsychology 2003,=20 17(4):548-555. PubMed=20 Abstract | Publisher Full Text
Merchant TE, Kiehna EN, Li C, Shukla H, = Sengupta S,=20 Xiong X, Gajjar A, Mulhern RK: Modeling radiation = dosimetry to=20 predict cognitive outcomes in pediatric patients with CNS = embryonal=20 tumors including medulloblastoma.
Int J Radiat Oncol Biol Phys 2006,=20 65(1):210-221. PubMed=20 Abstract | Publisher Full Text
Fouladi M, Gilger E, Kocak M, Wallace D, = Buchanan G,=20 Reeves C, Robbins N, Merchant T, Kun LE, Khan R, Gajjar A, = Mulhern R:=20 Intellectual and functional outcome of children 3 years = old or=20 younger who have CNS malignancies.
J Clin Oncol 2005,=20 23(28):7152-7160. PubMed=20 Abstract | Publisher Full Text
Merchant TE, Kiehna EN, Li C, Xiong X, = Mulhern RK:=20 Radiation dosimetry predicts IQ after conformal = radiation=20 therapy in pediatric patients with localized ependymoma. =
Int J Radiat Oncol Biol Phys 2005,=20 63(5):1546-1554. PubMed=20 Abstract | Publisher Full Text
Dennis M, Spiegler BJ, Hetherington CR, = Greenberg ML:=20 Neuropsychological sequelae of the treatment of children = with=20 medulloblastoma.
J Neurooncol 1996,=20 29(1):91-101. PubMed=20 Abstract | Publisher Full Text
Mulhern RK, Kepner JL, Thomas PR, Armstrong = FD,=20 Friedman HS, Kun LE: Neuropsychologic functioning of = survivors=20 of childhood medulloblastoma randomized to receive conventional = or=20 reduced-dose craniospinal irradiation: a Pediatric Oncology = Group study.=20
J Clin Oncol 1998,=20 16(5):1723-1728. PubMed=20 Abstract | Publisher Full Text
Levisohn L, Cronin-Golomb A, Schmahmann JD: = Neuropsychological consequences of cerebellar tumour = resection=20 in children: cerebellar cognitive affective syndrome in a = paediatric=20 population.
Brain 2000, 123 ( Pt=20 5):1041-1050. PubMed=20 Abstract | Publisher Full Text
Beebe DW, Ris MD, Armstrong FD, Fontanesi = J, Mulhern=20 R, Holmes E, Wisoff JH: Cognitive and adaptive outcome = in=20 low-grade pediatric cerebellar astrocytomas: evidence of = diminished=20 cognitive and adaptive functioning in National Collaborative = Research=20 Studies (CCG 9891/POG 9130).
J Clin Oncol 2005,=20 23(22):5198-5204. PubMed=20 Abstract | Publisher Full Text
Grill J, Viguier D, Kieffer V, Bulteau C, = Sainte-Rose=20 C, Hartmann O, Kalifa C, Dellatolas G: Critical risk = factors for=20 intellectual impairment in children with posterior fossa tumors: = the=20 role of cerebellar damage.
J Neurosurg 2004, 101(2=20 Suppl):152-158. PubMed=20 Abstract | Publisher Full=20 Text
Merchant TE, Lee H, Zhu J, Xiong X, Wheeler = G, Phipps=20 S, Boop FA, Sanford RA: The effects of hydrocephalus on=20 intelligence quotient in children with localized infratentorial=20 ependymoma before and after focal radiation therapy.
J Neurosurg 2004, 101(2=20 Suppl):159-168. PubMed=20 Abstract | Publisher Full=20 Text
Carpentieri SC, Waber DP, Pomeroy SL, Scott = RM,=20 Goumnerova LC, Kieran MW, Billett AL, Tarbell NJ:=20 Neuropsychological functioning after surgery in children = treated=20 for brain tumor.
Neurosurgery 2003,=20 52(6):1348-56; discussion = 1356-7. PubMed=20 Abstract | Publisher Full Text
Aarsen FK, Van Dongen HR, Paquier PF, Van = Mourik M,=20 Catsman-Berrevoets CE: Long-term sequelae in children = after=20 cerebellar astrocytoma surgery.
Neurology 2004,=20 62(8):1311-1316. PubMed=20 Abstract | Publisher Full Text
Ronning C, Sundet K, Due-Tonnessen B, = Lundar T,=20 Helseth E: Persistent cognitive dysfunction secondary to = cerebellar injury in patients treated for posterior fossa tumors = in=20 childhood.
Pediatr Neurosurg 2005,=20 41(1):15-21. PubMed=20 Abstract | Publisher Full Text

Pre-publication history

The pre-publication history for this paper can be accessed = here:

http://www.bi= omedcentral.com/1471-2407/8/15/prepub

Have=20 something to say? Post a comment on this article!

Terms and=20 Conditions Privacy=20 statement Information = for=20 advertisers Jobs at = BMC Contact us=20

=C2=A9 1999-2008 BioMed Central Ltd = unless otherwise=20 stated =