Pontocerebellar hypoplasia type 2: Variability in clinical and imaging findings

ARTICLE IN PRESS E U R O P E A N J O U R N A L O F PA E D I AT R I C N E U R O L O G Y

11 (2007) 146 – 152

Official Journal of the European Paediatric Neurology Society

Original article

Pontocerebellar hypoplasia type 2: Variability in clinical and imaging findings Maja Steinlina,, Andrea Kleinb, Karin Haas-Ludec, Dimitrios Zafeirioud, Susi Strozzia, Thomas Mu¨llere, Danielle Gubser-Mercatif, Thomas Schmitt Mechelkeg, Ingeborg Kra¨geloh-Mannc, Eugen Boltshauserb a

Division of Neuropaediatrics, University Children’s Hospital Berne, Iselspital, 3010 Bern, Switzerland Division of Neuropaediatrics, University Children’s Hospital Zu¨rich, Switzerland c Department of Neuropaediatrics, University Children’s Hospital Tu¨bingen, Germany d Division of Neuropaediatrics, University Children’s Hospital Thessaloniki, Greece e Division of Neuropaediatrics, Children’s Hospital Feldkirch, Austria f Division of Neuropaediatrics, Children’s Hospital Neuenburg, Switzerland g Division of Neuropaediatrics, Children’s Hospital Lucerne, Switzerland b

ar t ic l e i n f o

abs tra ct

Article history:

We report 24 children (14 girls) who presented with the typical neuroimaging findings of

Received 6 September 2006

pontocerebellar hypoplasia (PCH) to describe the clinical spectrum of type 2. Twenty-one

Received in revised form

presented with the classical form described by Barth; characteristic features (15/21) were

29 November 2006

breathing and/or sucking problems during neonatal period and early onset hyperkinetic

Accepted 29 November 2006

movement disorder. Eighteen were normocephalic at birth, but all developed microcephaly during infancy. Development was severely affected with none of the children being capable

Keywords:

of sitting, walking, or talking. Social contact and visual fixation were persistently poor.

Pontocerebellar hypoplasia

Dyskinetic movement disorder was present in all, in some together with mild spasticity.

Microcephaly

Seizures occurred in 14 (in 7 as neonates). Eight children died (age 1 day–6 years).

Cerebellar malformation

Neuroimaging showed an absent or severely flattened pons, different degrees of vermian

Dykinesia

hypoplasia, with cerebellar hemispheres (wing-like structures) being equally or more affected. Three (all girls) were less severely affected clinically and did not develop the dyskinetic movement disorder, motor and cognitive development were somewhat better. Microcephaly was also a prominent sign. Severity of pontocerebellar neuroimaging findings did not differentiate between the typical and atypical clinical group and did not correlate with clinical outcome. & 2007 European Paediatric Neurology Society. Published by Elsevier Ltd. All rights reserved.

Corresponding author. Tel.: +41 31 632 9424; fax: +41 31 632 9500.

E-mail address: [email protected] (M. Steinlin). 1090-3798/$ - see front matter & 2007 European Paediatric Neurology Society. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ejpn.2006.11.012

ARTICLE IN PRESS E U R O P E A N J O U R N A L O F PA E D I AT R I C N E U R O L O G Y

1.

Introduction

Pontocerebellar hypoplasia (PCH) is a descriptive structural abnormality with hypoplasia not only of the cerebellum but also of the ventral pons. Barth1 differentiates two major forms: PCH type 1, where PCH is associated with spinal muscular atrophy and usually death of the children before 1 year of age and PCH type 2. Barth et al.2 were the first to define the clinical syndrome of PCH type 2 in a larger cohort. Only few case reports have been added.3–5 PCH type 2 is characterized clinically by the findings of accompanying progressive microcephaly and dyskinetic movement disorder. In the literature, there is discussion about a severe fatal variant of PCH type 2 (also called type 4), leading to early death within the first few weeks of life.6–9 Most recently Patel et al.10 even reported three cases with an intrauterine variant (type 5?). Rajab et al.11 reported a family with PCH type 3, where optic atrophy was an additional mandatory feature. A linkage to chromosome 7 was possible in this family from Oman. There are other conditions, known to mimic the neuroimaging features of PCH, the most frequent certainly is pontocerebellar disruption as a consequence of brain damage after extreme prematurity.12,13 PCH has to be distinguished from early onset pontocerebellar atrophy as observed for example in some children with congenital glycosylation disorders14 and mitochondrial disorders.15 We would like to report a cohort of children with PCH type 2, defined on the basis of neuroimaging, to further characterize the clinical spectrum and neuroimaging findings.

2.

Patients and methods

We collected clinical findings in children with the neuroimaging findings of PCH, who were seen in consultation or for a second opinion at the University Hospitals of Berne (Switzerland), Thessaloniki (Greece), Tu¨bingen (Germany) and Zu¨rich (Switzerland). Inclusion criteria were the imaging findings of PCH, e.g. a markedly hypoplastic pons and cerebellum. Exclusion criteria were PCH type 1 (combined with spinal muscular atrophy), PCH with optic atrophy (type 3), severe prematurity (borno32 weeks of gestation) and children with progressive disorders. Metabolic problems were ruled out by exclusion of congenital disorders of glycosylation, normal lactate, urinary amino acids and organic acids. We reviewed the charts of the children for family history, history of pregnancy, birth, neonatal course, clinical findings during infancy and at last presentation, details of microcephaly, seizures, cognitive and motor development and dyskinetic movements such as chorea and dystonia during infancy. Neuroimaging findings were reviewed in axial, coronal and sagittal T1 and/or T2 weighted MR images. Gradings of imaging are illustrated in Figs. 1 and 2. Pontine prominence was graded as: flat (grade 1) or hypoplastic (grade 2). Vermis was graded as dysplastic streak (grade 1), hypoplastic (grade 2), or small shape, but shape preserved (grade 3). Cerebellar hemispheres were graded as streaky remnants (grade 1), thin

11 (200 7) 146 – 1 52

147

wing like hemispheres (grade 2), or thick wing like hemispheres (grade 3). Supratentorial abnormalities such as enlarged ventricles resulting from reduced white matter, and gyration abnormalities were looked for.

3.

Results

In 24 children (14 girls) the neuroimaging findings of PCH could be confirmed, but only 21 fulfilled also the clinical characteristics of PCH 2 as decribed by Barth et al.2 Parents originated from Germany, Switzerland, Croatia, Greece and Turkey.

3.1.

Typical PCH 2

The most important clinical findings are summarized in Table 1. From these 21 children (11 girls) pregnancy was uneventful for the majority, there was polyhydramnios in two pregnancies (patient 3 and 14). Fourteen women had an ultrasound during the second half of their pregnancy, being reported as normal. Two children were born prematurely, during the 35th and 36th week of gestation, respectively. None of the children had breech presentation. Three were born by caesarian section. At birth all except 3 showed head circumferences in the low normal range (none 450%). During the neonatal period 11 of these children had respiratory problems. One child was ventilated and a subsequently born affected sibling died the first day of life as it was decided not to provide ventilatory support. Thirteen children had sucking and/or feeding problems, which improved only in some. Four children were found to be hypotonic and three hypertonic during the neonatal period. In all these seven children dyskinetic movements were less prominent at birth, than in the other children. In all the other children abnormalities of tone were not reported, but they all manifested significant jitteriness and dyskinetic movements (jitteriness and myocloni) already in the first few weeks. Seizure activity was difficult to differentiate from these movements in some cases. However, only three children had neonatal seizures. During infancy the clinical course of these children was similar across the whole group. All became microcephalic. None showed dysmorphic features. Development was very limited, none learned to sit or crawl, and there was no speech development (vocalization of sounds only in one child). Eight children were able to fixate/follow with their eyes for social contact, for the others this was not clearly observed. All except two children developed a marked extrapyramidal dyskinetic movement disorder, with predominance of dystonia. Characteristically the movement disorder was unresponsive to medical treatment, and many of these children appeared distressed from it. In all twelve children who developed a seizure disorder antiepileptic treatment was effective. Mean age at last follow-up was 49 months (range 2 years), 7 children died (range 1 day–6 years). 4 months–1112 Two children died during an infection, in one of them high fever was combined with dehydration, in the other a bronchopneumonia without fever increased respiratory problems. In three children respiratory insufficiency was the

ARTICLE IN PRESS 148

E U R O P E A N J O U R N A L O F PA E D I AT R I C N E U R O L O G Y

11 (2007) 146 – 152

Fig. 1 – Grading of pons and vermis: (a) sagittal T1 weighted imaging with flat pons without pontine prominence and dysplastic streak of vermis; (b) sagittal T2 weighted imaging with minimal pontine prominence and hypoplastic vermis; (c) sagittal T2 weighted imaging with flat pons and small, but preserved vermis.

cause of death. Two children died at home and details are not known. MRI findings are summarized in Table 2, examples for the grading are given in Figs. 1 and 2. Supratentorial findings were ventricular dilatation, predominantly of the frontal horns/ reduced white matter in 11/20. The two severely affected brothers (patients 14 and 15) showed a querry simplified gyral pattern of frontal lobes. In three children a follow-up MRI 1 7 –512 years) was performed, in two of them (interval range 112

mild progressive supratentorial atrophy (with enlargement of all ventricles and also frontal lobe atrophy) was seen. Clinical findings, severity of movement disorder and developmental delay did not correlate with the degree of pontine or cerebellar hypoplasia on MRI. Autopsies were performed in two severely affected brothers (patients 14 and 15), who died at ages 110 12 years and 1 day respectively. In both there was evidence of hypoxic-ischaemic events before death. The vermis was considered to be normal

ARTICLE IN PRESS E U R O P E A N J O U R N A L O F PA E D I AT R I C N E U R O L O G Y

11 (200 7) 146 – 1 52

149

Fig. 2 – Grading of hemispheres: (a) axial T2 weighted imaging with small streaky remnants of cerebellar hemispheres; (b) coronal T2 weighted imaging with thin, wing like hemispheres; (c) coronal T1 weighted imaging with thick wing like hemispheres.

in one (on MRI findings were of a small but preserved shape), but there was significant hypoplasia of the pons and cerebellar hemispheres. There was almost complete lack of Purkinje and granular cells in the cerebellar hemispheres and the dentate nucleus was almost undetectable. In the base of the pons, inferior olivary nuclei and to a lesser extent in the thalamus and basal ganglia there was a general reduction of neurons demonstrated. There was no evidence of anterior horn cell degeneration. Cellular degeneration with astrogliosis was found mainly in the telencephalon and was considered to be secondary to hypoxia-ischaemia. The suspicion of simplified gyral pattern was not obvious at postmortem. Overall, the findings were consistent with PCH 2 as neuropathologically described earlier.2,3,6

3.2.

Atypical PCH

Three children (all girls) were considered not to have typical PCH type 2 clinically, despite characteristic infratentorial imaging findings. They were less severely affected than the

typical cases. In one of them there was bleeding in early pregnancy from a possible twin pregnancy, the other two had normal pregnancies. One girl showed jittery hand movements during the neonatal period, but none developed dyskinetic movement problems later. The degree of spasticity was variable, two of them had mild ataxia. All three showed better development than the group of typical PCH type 2, one of them was walking a few steps and talking a few words, the other two were able to stand and sit respectively, and both had good social non-verbal communication but no expressive speech. All three had seizures. There were no dysmorphic features. The infratentorial images were consistent with PCH 2. For one girl there was severe hypoplasia of the pons, vermis and hemispheres with almost complete lack of cerebellum. However, the other two showed grade 2 and 3 for pons, vermis and cerebellum The supratentorial findings were characterized by reduced white matter leading to a marked ventricular dilatation and enlarged extracerebral CSF spaces, clearly much more prominent than seen in the children with typical PCH 2.

ARTICLE IN PRESS 150

E U R O P E A N J O U R N A L O F PA E D I AT R I C N E U R O L O G Y

11 (2007) 146 – 152

Table 1 – Clinical findings in children with typical pontocerebellar hypoplasia type 2 Pat

Sex

CS sib

HC birth (cm)

Neo neuro

Last FU (in years)

Feeding problems

Social contact

HC

Movement disorder

Seizures

+

Death

1

F

!

33

Dysk

3 412

PEG

Fixation

oP3

Dyst spast

2

M

c

34

Dysk

9 212

+

Fixation

oP3

Dyst spast

+

3

F

e

32.5

Dysk

3

FTT

oP3

Dyst



4

F

c

33

Dysk

11/12

FTT

oP3

Dyst



5 6

M M

36 34

Hyper Dysk

 +

Fixation smile Fixation smile None None

oP3 oP3

Spast Dyst spast

+ +

7

M

32

Dysk

2 m 112



Fixation

oP3

Dyst choreatic



8 9 10 11

F M F M

normal 30 (oP3) 30 (oP3) 31 (oP3)

Dysk Dysk Dysk Hypo

8/12 5 7

+ FTT FTT Normal

None Fixation Fixation Fixation

oP3 oP3 oP3 oP3

Dyst spastic Dyst Dyst Dyst spast

 + + FS

12

F

32

Dysk

5 512

PEG

oP3

Dyst spast

+

6y

6 212

6 312

4/12 3y

13

F

+

33.5

Dysk

8 312

FTT

Fixation smile None

oP3

Dyst spast

+

8 3 y12

14

M

+e

34

Dysk

1 112

PEG

None

oP3

Dyst spast

+

1y10 12

15 16

M F

+c

33 33

Hyper Dysk

610 12

PEG

Fixation

oP3

Dyst spast

+

17 18

F F

32.5 32.8

Hyper Dysk

9

FTT PEG

None None

oP3 oP3

Spast Dyst spast

FS +

19

M

34

Dysk

2 1112

FTT

Vocals

oP3

Dyst spast

+

20 21

F M

32.5 normal

Dysk Dysk

4/12 4

+ +

Fixation Fixation smile

oP3 oP3

Dyst spast Dyst spast

 

**

1d

2 512

6y

Abbreviations: F ¼ female M ¼ male; CS+ ¼ consanguinity; Sib ¼ siblings e c; ** by history one affected sibling; HC ¼ head circumference; FU ¼ follow-up. Neo neuro: neonatal neurological findings: hypo ¼ hopotone, hyper ¼ hypertone; dyskinetic movements. FTT ¼ failure to thrive Dyst ¼ dystonic, Spast ¼ spastic; FS ¼ febrile seizures; PEG ¼ percutaneous endoscopis gastrostomy.

Table 2 – Neuroimaging findings in children with typical PCH 2 ( n ¼ 21) Pons Flat Hypoplastic

6 15

Vermis Dysplastic streak Hypoplastic Small, shape preserved

1 15 5

Hemispheres Streaky remnants Thin wing-like Thick wing-like

2 8 11

 Symmetrical in 19/21.

4.

Discussion

This is the largest reported series of PCH 2, confirming its clinical entity for the majority of cases but also suggesting a

spectrum of mild to severe involvement. Infratentorial findings on neuroimaging using MRI techniques does not predict the degree of clinical involvement. All 21 children with the typical form of PCH 2 had a characteristic history of normal pregnancy and delivery, but symptoms manifesting during the first few weeks of life. As already pointed out by Barth et al.2 these children have respiratory and/or feeding problems during the neonatal period. The most characteristic feature however is the extrapyramidal dyskinesia, which manifests in the majority already during the neonatal period or early months. It seems important to point out that distinguishing this from seizures is clinically difficult in many cases. Although only few had seizures during the first few weeks of life, half of the children developed epilepsy subsequently. Another striking feature was the minimal or even lack of development. All our children were severely handicapped, They developed next to no voluntary motor function, had most severe cognitive and language impairment without verbal or non-verbal communication, and approximately half of them only could fix and follow with their eyes. However, despite the severe developmental impairment, survival of these children into childhood was remarkable. The highest 2 years, only one boy age attained at last follow-up was 1112 died during the neonatal period.

ARTICLE IN PRESS E U R O P E A N J O U R N A L O F PA E D I AT R I C N E U R O L O G Y

The fact that in our group there were four pairs of affected siblings supports an autosomal recessive inheritance, as already illustrated by Barth et al.2 Consanguinity was only present in two of our families and ethnicity was not specific for these children. Whether one gene or different genes are responsible for PCH type 2 remains to be clarified. An important observation was that prenatal ultrasound did not detect the pathology in any affected child. In two families an attempt was made for a prenatal diagnosis after a first affected child, by ultrasound and by magnetic resonance imaging, respectively. In both subsequently affected children, the pontocerebellar hypoplasia could not be definitely detected during the second trimester in the 22nd gestational week. This difficulty of prenatal diagnosis was also experienced by Peter Barth (personal communication). In the literature there are several reports of severe neonatal variants of PCH, also categorized as type 4.6–9 There is suggestion, that these children differ by the presence of polyhydramnios, spasticity rather than hypotonia exhibited soon after birth and a severe neonatal course with respiratory and sucking insufficiency leading to early death. In our group, there were two children with polyhydramnios, both of them had hypertonia at birth. One died at 4 months, the other had long-term ventilatory support and died at 22 months. The second one had a sibling dying at 1 day of age, there was no polyhydramnios during pregnancy, but he was reported to have increased tone after birth. However, our children with hypertonia in the neonatal period and severe course did not show extensive dyskinetic movements, as reported by Chaves-Vischer et al.7 Interestingly, these two siblings with severe course showed also some supratentorial abnormalities with a querry simplified gyral pattern. In the literature the combination of PCH with supratentorial migration abnormalities is also linked to severe neonatal forms.16 Our data suggest, that there might be a continuum of severe neonatal and infantile types, rather than clearly defined subgroups. The three children with less severe clinical symptoms, ‘atypical PCH2’ could represent the other end of the severity spectrum. This, however, remains to be shown, once the gene for PCH type 2 is detected. In these three girls developmental problems only became evident during the first 6 months. Thus, lack of hyperkinetic movements during the neonatal period and also lack of dyskinetic movement problems during the first few months probably suggests an atypical course with better developmental outcome. There is a milder form of PCH reported in the literature.17 However, our three girls did not show the associated dysmorphic features reported in these two siblings. Also, the differential diagnosis of infantile onset spinocerebellar ataxia (IOSCA) seems excluded, as all three children did not show the clinical characteristics of normal development during the first years of life, progressive ataxia and cerebellar atrophy combined with axonal neuropathy and hearing impairment. Thus, genetic testing for IOSCA in these three children was not considered. We revaluated the neuroimaging findings and scored the degree of pontine and cerebellar hypoplasia relying on visual analysis. There was no correlation between the degree of hypoplasia and the degree of clinical involvement, nor was neuroimaging helpful in discriminating the atypical cases from the typical ones. Whether enlarged ventricular size

11 (200 7) 146 – 1 52

151

might be an indicator for the atypical form remains to be proven. In conclusion, most of the children with the neuroimaging criteria of PCH—excluding the cases with SMA (PCH type 1) and preterm born children with pontocerebellar disruption, which is thought to be a ‘lesional mimicry’—show very early a typical clinical picture, corresponding to ‘classical PCH type 2’ as described by Barth. These children are severely affected with a nearly untreatable dyskinetic movement disorder and most severe developmental delay. Death during early childhood is observed in about one third, however survival into the second decade is possible. In addition to this severe group we identified a smaller group with better developmental outcome, and without the dyskinetic movement disorder. Infratentorial neuroimaging findings did not differentiate between the two forms. Autosomal recessive inheritance is supported by four pairs of siblings affected in the classical form. Prenatal diagnosis will depend on the detection of the genetic markers as prenatal ultrasound appears to lack sensitivity in the detection of pathology.

Acknowledgements We thank the parents of the affected children for helping with medical records and anamnestic data. We thank Avril Castagna Sloane, MD, for help in preparing the manuscript. We thank the paediatricians of the Kinderklinik VillingenSchwenningen, Germany, for sharing patients data with us. We thank Dr. Antje Bornemann, neuropathology University Hospital Tu¨bingen, for allowing us to cite from the neuropathology records. R E F E R E N C E S

1. Barth PG. Pontocerebellar hypoplasia—how many types? Eur J Paediatr Neurol 2000;4(4):161–2. 2. Barth PG, Blennow G, Lenard HG, et al. The syndrome of autosomal recessive pontocerebellar hypoplasia, microcephaly, and extrapyramidal dyskinesia (pontocerebellar hypoplasia type 2): compiled data from 10 pedigrees. Neurology 1995;45(2):311–7. 3. Barbot C, Carneiro G, Melo J. Pontocerebellar hypoplasia with microcephaly and dyskinesia: report of two cases. Dev Med Child Neurol 1997;39(8):554–7. 4. Simonati A, Dalla Bernardina B, Colombari R, Rizzuto N. Ponto-cerebellar hypoplasia with dystonia: clinico-pathological findings in a sporadic case. Childs Nerv Syst 1997; 13(11-12):642–7. 5. Coppola G, Muras I, Pascotto A. Pontocerebellar hypoplasia type 2 (PCH2): report of two siblings. Brain Dev 2000;22(3):188–92. 6. Albrecht S, Schneider MC, Belmont J, Armstrong DL. Fatal infantile encephalopathy with olivopontocerebellar hypoplasia and micrencephaly. Acta Neuropathol 1993;85:394–9. 7. Chaves-Vischer V, Pizzolato GP, et al. Early fatal pontocerebellar hypoplasia in premature twin sisters. Eur J Paediatr Neurol 2000;4(4):171–6. 8. Hashimoto K, Takeuchi Y, Kida Y, et al. Three siblings of fatal infantile encephalopathy with olivopontocerebellar hypoplasia and microcephaly. Brain Dev 1998;20:169–74.

ARTICLE IN PRESS 152

E U R O P E A N J O U R N A L O F PA E D I AT R I C N E U R O L O G Y

9. Grosso S, Mosardini R, Cioni M, Galuzi G, Balestri P. Pontocerebellar hypoplasia type 2: further clinical characterization and evidence of positive response of dyskinesia to levodopa. J Neurol 2002;249(5):596–600. 10. Patel MS, Becker JE, Toi A, Armstrong DL, Chitayat D. Severe, fetal-onset form of olivopontocerebellar hypoplasia in three sibs: PCH type 5? Am J Med Genet A 2006;140(6):594–603. 11. Rajab A, Mochida GH, et al. A novel form of pontocerebellar hypoplasia maps to chromosome 7q11-21. Neurology 2003;60(10):1664–7. 12. Johnsen SD, Bodensteiner JB, Lotze TE. Frequency and nature of cerebellar injury in the extremely premature survivor with cerebral palsy. J Child Neurol 2005;20:60–4. 13. Messerschmidt A, Brugger PC, Boltshauser E, et al. Disruption of cerebellar development: potential complication of extreme prematurity. AJNR Am J Neuroradiol 2005;26:1659–67.

11 (2007) 146 – 152

14. Barone R, Pavone L, Fiumara A, Bianchini R, Jaeken J. Developmental patterns and neuropsychological assessment in patients with carbohydrate-deficient glycoconjugate syndrome type IA (phosphomannomutase deficiency). Brain Dev 1999;21(4):260–3. 15. De Koenig TJ, deVries LS, Groenendaal F, et al. Pontocerebellar hypoplasia associated with respiratory-chain defects. Neuropediatrics 1999;30(2):93–5. 16. Sztriha L, Johansen JG, Al-Gazali LI. Extreme microcephaly with agyria-pachygyria, partial agenesis of the corpus callosum, and pontocerebellar dysplasia. J Child Neurol 2005;20(2):170–2. 17. Dilber E, Aynaci FM, Ahmetoglu A. Pontocerebellar hypoplasia in two siblings with dysmorphic features. J Child Neurol 2002;17(1):64–6.