Hydromyelia and Chiari malformation in children and adolescents

Hydromyelia and Chiari malformation in children and adolescents

R. R C. Jones FRACSFRCS(ENG) J. G. J. A y e r BSC(MED) MBBS W. A. S t e n i n g FRACSFRCS(ENG) Department of Neurosurgery, Prince of Wales Children's Hospital, Randwick, NSW 2031, Australia

A series of 24 patients with Chiari malformation and hydromyelia, treated at the Prince of Wales Children's Hospital between 1975 and 1991, is reviewed. The age range of these patients was 3 to 19 years. Eleven had a Chiari 1 malformation and 13 had a Chiari 2 malformation. The follow-up period ranged from 6 months to 16 years with a mean of 5 years. Twenty-one patients had posterior fossa decompression procedures as the primary treatment, with or without plugging of the central canal at the obex. Forty-five per cent of the Chiari 1 patients and 62% of the Chiari 2 patients showed sustained neurological improvement. A further 36% of the Chiari 1 patients and 15% of the Chiari 2 patients had their neurological deficits stabilised. Following operation, sustained radiological collapse of the cavity occurred in 9 out of the 11 cases of Chiari 1 malformation. The follow-up radiology in the Chiari 2 patients has been incomplete. Complications following posterior fossa procedures with plugging of the central canal were transient and there has not been long-term morbidity. Posterior fossa decompression, fourth ventriculostomy, obex plugging and dural grafting is recommended in those cases where the surgical anatomy allows dissection of the tonsils from the bralnstem. Plugging of the central canal at the obex is of value if the muscle plug is firmly secured. Journal of Clinical Neuroscience 1996 3 (1):34-45

© Pearson Professional 1996

Keywords: Chiari malformation, Hydromyelia

Introduction Spinal cord cavities have been classified into communicating and non-communicating types? Communicating syrinxes are continous with the fourth ventricle whereas non-communicating syrinxes are separated from the fourth ventricle by a syrinx free segment of the spinal cord. The term syringomyelia has been reserved for those spinal cord cavities that originate in and extend into the substance of the spinal cord, whereas hydromyelia has been used to describe accumulation of fluid within a dilated central canal. This distinction is hard to make since the majority of cavities are eccentrically located and lined by both glia and ependyma. 2,3 In this paper the term hydromyeliarefers to spinal cord cavities occurring in" association with developmental or acquired abnormalities of the craniocervical junction, most commonly with Chiari malformation? ,4,5These cavities may or may not communicate with the fourth ventricle at the time of examination. The term does not apply to cases associated with trauma, t u m o u r or degenerative diseases of the spinal cord. The pathogenesis of hydromyelia remains controversial. Gardner has proposed a hydrodynamic theory. 6,7 This theory relates to persistent dilatation of the neural

34

tube during embryological life, p r o d u c e d by obstruction of CSF drainage from the ventricles. According to Gardner, hydromyelia develops because of failure of the rhombic roof to perforate normally during prenatal life. Pulsatile CSF pressure, which is generated by arterial pulsation in the choroid plexus, decompresses down the patent central canal of the cord creating progressive spinal cord cavitation. On the basis of this pathophysiological proposal, Gardner has developed a surgical procedure which aims to open up the adhesions blocking the foraminal exits from the fourth ventricle and plug the obex region to close off the communication between the fourth ventricle and the central canal. Williams has proposed modifications to the hydrodynamic theory while maintaining the importance of a patent central canal. 8 He has emphasised the importance of pressure differences between the intracranial and intraspinal fluid compartments p r o d u c e d by episodic venous pressure changes, such as occurs with a cough or Valsalva's manoeuvre. O t h e r theories have been suggested to explain the presence of hydromyelia. For example, Ball and Dayan

J. Clin. Neuroscience Volume 3 Number 1 January 1996

w

Oh

LD

c~J

0-

3

Z e-

LM

e-

o<

8

Z tD e-

t~

Age at operation

Sex

15 12

13 11 9

7 8

9 10 11

F F M

M M

M

PFD+FV+PO+ DG PFD+FV+PO+ DG PFD+FV+PO

PFD+FV+PO+ DG PFD+FV+PO+ DG

PFD+FV+PO

PFD+FV+PO+ DG PFD+FV+PO

11 4 19

22 23 24

M M M

F

F

F

F

M F

F F

F

PFD+FV+PO+ DG PFD+FV+PO+ DG PFD+DG

PFD+FV+PO

PFD+FV+PO

PFD+FV+PO

SSS

PFD+FV+PO PFD+FV+PO+ DG

SSS PFD+FV+DG

23/11/87 23/6/89 18/12/91

14/9/87

6/7/87

25/5/87

19/12/86

26/5/86 18/7/86

11/1/85 4/3/85

11/11/83

21/7/75

10/2/90 23/3/90 12/4/91

9/6/89 30/10/89

17/4/89

15/9/86 25/7/88

8/8/85

2/6/85

17/11/83

Date

NIL NIL NIL

NIL

NIL

NIL

NIL

NIL NIL

NIL pyrexia

NIL

CSF leakage from wound

NIL transient post operative fever NIL

transient post operative nausea and vomiting NIL NIL

intraoperative bradycardia NIL

intraoperative bradycardia and postop, clonus intraoperative bradycardia

NIL

Surgical complications

sustained improvement sustained improvement sustained improvement

stabilization

sustained improvement

sustained improvement

sustained improvement slow sustained improvement in gait stabilization

sustained improvement progression after initial improvement

progressive weakness and parasthesiae

progression after initial improvement

sustained improvement stabilisation stabilisation

stabilisation sustained improvement

sustained improvement

progression after initial improvement. sustained improvement sustained improvement

progression after intial improvement sustained improvement

Clinical outcome

MRI: marked collapse of syrinx MRI: marked shhrinkage of the syrinx MRI: moderate reduction in the syrinx MRI: marked reduction in syrinx MRI: moderate reduction MRI: marked reduction •

• MRI: collapsed spinal cord

marked shrinkage •

•

metrizamide myelogram: no change

MRI: marked shrinkage MRI: marked shrinkage (Fig. 3) MRI: moderate shrinkage MRI: moderate shrinkage MRI: marked shrinkage

MRI: marked shrinkage (Fig.2) MRI: marked shrinkage (Fig. 1) MRI: marked reduction MRI : recurrence after initial collapse(Fig.4) MRI: marked shrinkage

CT scan: no change

Radiological outcome

3 3 1.5

0.8 3

3

6 4

4

NIL NIL NIL

NIL

NIL

NIL

NIL

5 3.5 1.5

5

5.5

5.5

0.5

SSS resulting in 16 stabilization of symptoms SSS, PFD+FV+ PO 9 with continued progression NIL 7 PFD+FV+PO+ DG, 5 PFD with continued progression. NIL 6 NIL 6

NIL NIL NIL

NIL NIL

NIL

NIL NIL

NIL

7

6

Follow-up (yrs)

SSS, PFD+FV+DG, laminectomy NIL

Secondary procedures

. . . . . • . . . PFD=pstenor fossa decompression; FV=tburth ventnculostomy; PO=pluggmg ot the central canal at the obex; DG=dural graft 1. the postoperatwe films m 4. of the patmnts with myelomeningocoele were not able to be reviewed..~MM=myelomeningocoele;#H/C=hydrocephalus; VP=ventriculoperitoneal; R/O=removal of syringo-subarachnoid shunt.

10

21

*

3

11

20

19

6

6 7

16 17

18

8 12

14 15

14

PFD+DG

13

6

F M

PFD+FV+PO+ DG

13

15 6

4 5

F

PFD+FV+PO+ DG

PFD+FV+PO

14

3

M

SSS

Operation

Patients with MM § 12 3 F

14

2

Patients without MM § 1 6 F

Case

Table 1 Summary of the cases w i t h h y d r o m y e l i a treated at the Prince of Wales Childrens Hospital b e t w e e n 1975 and 1991

5"

e~

e-

Y.

Da

m.

8"

0a m

3

t'3

-I

3

-i"

Clinical studies

have suggested that spinal cord cavities may be p r o d u c e d by cerebrospinal fluid tracking u n d e r pressure into the spinal cord from the subarachnoid space along the perivascular (Virchow-Robin) spaces? Aboulker has proposed that obstruction at the cisterna magna, in association with high venous pressure, results in transmedullary passage of CSF which produces progressive cavitation. 1° In 1892 Abbe and Coley made the first attempt at surgical treatment of spinal cord cavities when they performed a laminectomy and aspiration, n Since this time numerous surgical interventions have been advocated. These procedures can be grouped into four categories : a) posterior fossa decompression; b) syrinx shunting; c) terminal venticulostomy; and d) percutaneous aspiration. In this report we review the results of surgery in 24 patients. Twenty-one patients had posterior fossa decompression as the primary procedure.

Materials and methods This series includes 24 patients younger than the age of 20 who u n d e r w e n t surgical treatment at the Prince of Wales Children's Hospital between 1975 and 1991. Eleven of these patients were diagnosed with hydromyelia and Chiari 1 malformation. None of these patients had spinal dysraphism. Thirteen patients were treated for hydromyelia associated with myelomeningocele and Chiari 2 malformation. Table 1 summarises these cases.

Chiari 1 malformation Eleven patients were treated between 1983 and 1991. The age range was 6-15 with a mean of 11 years 7 months. Six were male and 5 were female. All patients had a diagnosis of hydromyelia and Chiari malformation made prior to surgery by CT metrizamide* myelography or MRI. One patient had prior surgery at another institution for removal of a suspected spinal cord t u m o u r (case 7), however, histological examination failed to reveal a tumour. Another patient had a primary procedure at another centre for treatment of hydromyelia (case 6). Two patients had hydrocephalus treated by ventriculoperitoneal shunting at 3 years and 8 years of age (cases 5, 8).

Hydromyelia and Chiari malformations

Results Presenting

symptoms

Table 2 summarises the presenting symptoms. For the Chiari 1 patients the most c o m m o n presenting symptom was scoliosis which was f o u n d in 9 patients (82%). Four patients complained of persistent pain (36%). In one patient pain was felt between the scapulae, in another patient in the back and anterior chest wall, and two other patients in the chest wall alone. U p p e r limb weakness was complained of by 3 patients (27%) and lower limb motor disturbance in 3 patients (27%). Bladder dysfunction occurred in 4 patients (36%) at presentation. No patients complained of parasthesiae or numbness at presentation. Nine patients with Chiari 2 malformation complained of deteriorating gait and increasing lower limb weakness (69%). Upper limb weakness was a presenting symptom in 3 patients (23%) and cranial nerve dysfunction in one (8%). Table 2 Presenting symptoms in children with hydromyelia Symptom

Patients without MM* n=11

Patients with MM* n=13

Scoliosis Upper limb weakness/ wasting Gait disturbance/ progressive lower limb weakness Bladder dysfunction Pain Sensory disturbance Pes cavus/hemiatrophy Headache

9 3

1 3

3

9

4 4 1 2 1

3 3 0 0 1

*MM=myelomeningocele

Presenting signs

Table 3 summarises the clinical findings on admission. In the Chiari 1 patients the most c o m m o n finding was scoliosis, f o u n d in nine patients (82%). A dissociated sensory disturbance o c c u r r e d in 4 patients (36%). T h r e e of these patients had diminished sensation of pain over the trunk. One patient had dissociated sensory loss in both Table 3 Signs at admission in children with hydromyelia

Chiari 2 m a l f o r m a t i o n

Thirteen patients were treated between 1975 and 1991. The age range was 3-19 with a mean of 8 years 9 months. Four were male and 9 were female. In all patients the myelomeningocele had been repaired in infancy. Twelve children had prior surgery for hydrocephalus. In all cases the ventriculoperitoneal shunts were functioning at the time of operation.

*Winthrop Laboratories, Division of Sterling Pharmaceuticals, 82 Hughes Avenue, Ermington NSW 2115, Australia.

36

Sign

Patients without MM* n=11

Patients with MM* n=13

Scoliosis Upper limb weakness/ atrophy of intrinsic hand muscles Gait abnormality Dissociatedsensory loss Recent superficial reflex changes Recent onset of Babinski Nystagmus Syringomyelicsyndrome

9 2

4 5

2 4 8

12 3 0

4 0 1

4 2 0

* M M=myelomeningocele

J. Clin. Neuroscience V o l u m e 3 N u m b e r 1 January 1996

Hydromyelia and Chiari malformations

Clinical studies

hands. Two patients (18%) had unilateral atrophy of the intrinsic muscles of the hand.Loss of superficial reflexes occurred in 8 patients (72%), lower limb hyperreflexia in 3 patients (27%) and a Babinski response in 4. Only one patient presented with the 'syringomyelic syndrome' of both brachial amyotrophy and dissociated sensory loss. T h e most c o m m o n clinical finding in the Chiari 2 patients was progressive lower limb weakness, f o u n d in 12 patients (92%). O t h e r c o m m o n features included: h a n d atrophy, 2 u p p e r limb weakness, ~ cranial nerve abnormalit,/,2 dissociated sensory loss, 3 scoliosis 4 and nystagmus3 Neuroradiological

findings

All patients had a diagnosis of Chiari malfomation and hydromyelia prior to operation.

made on MRI when a low signal intensity area within the spinal cord was observed on the T1 weighted images. The diagnosis of Chiari malformation was made if the tonsils were below the level of the foramen magnum. The diagnosis of hydromyelia was made on CT metrizamide myelography when there was widening of the spinal cord with lucency at the centre of the cord. A delayed CT scan was not p e r f o r m e d routinely. The syrinx was holocord in 2 patients, u p p e r cervical to lower thoracic in 4 patients and u p p e r cervical to u p p e r thoracic in 2 patients. In 3 patients the extent of the cavitation was not defined. In all the patients who had preoperative CT metrizamide myelography there was minimal blockage of the passage of contrast up through the foramen m a g n u m despite the presence of a Chiari malformation (Fig. 1).

Patients w i t h Chiari 2 m a l f o r m a t i o n Patients with

Chiari 2 malformation

A preoperative MRI scan was p e r f o r m e d in 8 out of the 11 Chiari 1 patients (75%). One of these patients also had a CT metrizamide myelogram. T h r e e patients treated before the end of 1986 had CT metrizamide myelography alone. T1 and T2 weighted axial and sagittal MRI examinations were obtained. The diagnosis of hydromyelia was

A preoperative MRI scan was p e r f o r m e d in 2 patients. One of these patients also had CT metrizamide myelography. Four patients had a diagnosis made on plain CT and 3 patients had CT metrizamide myelography alone. One patient had a metrizamide myelogram alone. In 3 patients the preoperative studies were unable to be reviewed.

A

B

c D Fig 1. (Case 3) (A) Pre operative CT metrizamide myelogram showing the cerebellar tonsils clearly outlined at the level of C1 in a 14 year old girl(arrow). (B) Preoperative CT metrizamide myelogram at the level of the body of the C3 vertebrae demonstrating a large central hydromyelic cavity(arrow). (C) Two months following posterior fossa decompression, fourth ventriculostomy, plugging of the central canal and dural grafting collapse of the cavity had occurred as demonstrated by CT metrizamide myelography. (D) A T1 weighted MRI at the level of the body of C3 vertebrae 2 years following operation showed maintained collapse of the cavity despite recurrence of symptoms.

J. Clin. Neuroscience

Volume 3 Number 1 January 1996

37

Clinical studies

Hydromyelia and Chiari malformations

Table 4 Clinical result in patients w i t h hydromyelia treated at the Princeof Wales Childrens Hospital between 1975 and 1991 NUMBER Procedure t

PFD+FV+PO+DG PFD+FV+PO PFD+FV+DG PFD+PO+DG PFD+DG SSS TOTAL

WITHOUT MM*

Sustained improvement

Stabilisation

Progression after initial improvement

2 2

2 2

1 0

0 0

1

0

0

0

0 5

0 4

1 2

0 0

NUMBER

Progressive

Sustained improvement 2 3 0 0 1 2 8

Stabilisation

WITH Progression after initial improvement

1 1 0 0 0 0 2

0 1 1 0 0 0 2

MM* Progression

Total

0 0 0 0 1 0 1

8 9 1 1 2 3 24

tPFD=posterior fossa decompression;FV=fourth ventriculostomy;PO=plugging of the central canal at the obex;DG=dural graft;SSS=syringosubarachnoid shunting *MM=myelomeningocele

Procedures Two types of p r o c e d u r e s were used in this g r o u p of patients as p r i m a r y operative m a n a g e m e n t . Table 4 summarises the procedures p e r f o r m e d . Preoperative radiological studies and assessment of the surgical anatomy were used to tailor the operations.

Posterior fossa d e c o m p r e s s i o n T h e patients were in the p r o n e position and u n d e r assisted ventilation. A midline skin incision with an inverted Y muscle incision at the craniocervicaljunction was p e r f o r m e d . A suboccipital craniectomy and an u p p e r cervical laminectomy were p e r f o r m e d exposing the lowest level of the Chiari malformation. A Y shaped incision was m a d e in the dura. T h e goals of surgery were to decompress the Chiari malformation, to re-establish n o r m a l CSF flow f r o m the fourth ventricle and to prevent filling of the spinal cord cavity. D e p e n d i n g on the surgical anatomy, the operative steps included some or all of the following: lysis of adhesions between the cerebellar tonsils and the brainstem, fenestration of the f o r a m e n of Magendie (fourth ventriculostomy), plugging of the central canal at the obex and duroplasty. Plugging of the obex was p e r f o r m e d if the o p e n i n g of the central canal was patent at the obex. A small piece of muscle was placed into the dilated central canal. In some patients this was sutured into position and then b l o o d clot or a n o t h e r piece of muscle placed over the top. T h e muscle plug was sutured to a fine rim of fibrous tissue dorsal to the central canal by 10/0 suture. In other cases 2 to 3 pieces of muscle were placed in the obex without suturing. A graft was used to patch the dura if the area of the cisterna m a g n a a p p e a r e d crowded.The graft was m a d e of fascia, or rarely of dacron or lyophilised bovine dura. Dacron is no longer r e c o m m e n d e d due to p r o b l e m s with fibrosis and subarachnoid h a e m o r r h a g e . 12

A midline myelotomy was p e r f o r m e d or an incision was m a d e in the thinnest portion of the syrinx to minimize the possibility of neurological damage. A silastic* catheter was inserted into the hydromyelic cavity in a cephalad direction. T h e catheter was immobilised by a suture to the d u r a at the myelotomy site and the distal e n d of the catheter was inserted 2-3 cm into the subarachnoid space. T h e dura was closed with absorbable 4-5/0 Vicryl* sutures.

Surgical c o m p l i c a t i o n s T h e r e was no operative mortality. During o b e x plugging operations on the Chiari 1 patients, three patients suffered transient intraoperative bradycardia, one patient e x p e r i e n c e d transient postoperative nausea for 48 h and one patient e x p e r i e n c e d ankle clonus for a period of 48 h after o p e r a t i o n . O n e patient had transient postoperative pyrexia for which no specific cause could be found. Following operation on the patients with myelomeningocele one patient had a transient postoperative fever of u n k n o w n cause and one patient h a d a leak of CSF f r o m the w o u n d which settled following insertion of a ventriculoperitoneal shunt to reduce CSF pressure. O n e of the patients with m y e l o m e n i n g o c e l e died at the age of 15, 4 years after p r i m a r y operation, due to respiratory depression.

Results o f surgical t r e a t m e n t Tables 4 and 5 summarise the results of surgical therapy. Results were analysed in terms of clinical o u t c o m e and radiological resolution.

Clinical results T h e range of follow up has b e e n 6 m o n t h s to 16 years with a m e a n of 5 years. Patients were classified into 4 groups according to

kaminectomy and syringosubarachnoid shunting T h e patient was placed in a h e a d rest and a laminectomy p e r f o r m e d , the level of which was d e t e r m i n e d by preoperative radiological studies. Fine dissection of the arachnoid was p e r f o r m e d in o r d e r to preserve this m e m b r a n e .

38

*Heyer Schulte-American Hospital Supply Corporation, Chicago Illinois, USA. tEthicon,Johnson and Johnson, 1-5 Khartoum Road, North Ryde, NSW 2113, Australia.

J. Clin. Neuroscience V o l u m e 3 N u m b e r 1 J a n u a r y 1 9 9 6

Hydromyelia and Chiari malformations Table 5

Clinical studies

Radiological result in patients with hydromyelia treated at the Prince of Wales Childrens Hospital between 1975 and 1991 NUMBER

Proceduret

PFD+FV+PO+DG PFD+FV+PO PFD+FV+DG PFD+PO+DG PFD+DG SSS

Marked shrinkage

Moderate shrinkage

WITHOUT

MM*

NUMBER

Recurrence No change or Marked after p r o g r e s s i o n shrinkage shrinkage

Moderate shrinkage

WITH

MM *§

Recurrence after shrinkage

No change

2 3

3 0

0 1

0 0

2 2

1 1

0 0

0 1

1

0

0

0

0

0

0

0

0

0

1

0

2

0

0

0

No radiological review§ 0 1 1 0 2 0

tPFD=posteriorfossa decompression; FV=fourth ventriculostomy; PO=plugging of the central canal at the obex; DG=dural graft; SSS=syringosubarachnoid shunting *MM=myelomeningocele 5postoperativeradiological result for 4 patients with myelomeningocele was not able to be reviewed.

Overall, 45% of the Chiari 1 patients and 62% of the Chiari 2 patients showed sustained neurological improvement. A further 36% of the Chiari 1 patients and 15% of the Chiari 2 patients had their neurological deficits stabilised.

clinical result: sustained improvement; stabilisation; progression after initial improvement; and progression. 'Improvement' was defined as improvement in neurological signs and symptoms. 'Progression' was defined as worsening of existing signs and symptoms or the development of new symptoms. 'Stabilisation' was defined to have occurred when signs and symptoms remained unchanged without progression.

Five patients underwent posterior fossa decompression, fourth ventriculostomy, obex plugging and a dural graft.

A

B

Patients w i t h Chiari 1 m a l f o r m a t i o n

Fig 2. (Case 2) (A) Preoperative saggital T1 weighted MRI demonstrated hydromyelia and herniation of the cerebellar tonsils. (B) A

postoperative sagittal T1 weighted MRI demonstrating collapse of the cavity.

J. Clin. Neuroscience Volume 3 Number 1 January 1996

39

Clinical studies

Hydromyelia and Chiari malformations

A Fig 3. (Case8) Twelve year old boy who presented with scoliosis and pain, (A) Preoperativesagittal T1 weighted MRI scan demonstrating Chiari 1 malformation. (B) Preoperative sagittal T1 weighted MRI scan demonstrating hydromyelia (C) Postoperative T1 weighted MRI showing a marked reduction in the size of the cavity.

Two have had sustained improvement, 2 have been stabilised and one has deteriorated after initial improvement. MR/ has shown that the syrinx has remained collapsed and further treatment has not been required (Case 3, Fig 1). In addition, one patient has had posterior fossa decompression, plugging of the obex and dural grafting without fourth ventriculostomy as the ventricular outlets were patent. This patient has had sustained improvement. Four patients have undergone posterior fossa decompression, fourth ventriculostomy and obex plugging without a dural graft. Two have had sustained improvement and 2 have been stabilized. One patient underwent laminectomy, myelotomy and syringosubarachnoid shunting as a primary procedure (Case 1). This patient progressed after initial improvement and has subsequently undergone three further procedures: repeat syringosubarachnoid shunting, posterior fossa decompression and finally laminectomy alone. At laminectomy the syrinx was found to be collapsed and sustained improvement has followed this procedure. Patients with Chiari 2 m a l f o r m a t i o n

Three patients have had posterior fossa decompression, fourth ventriculostomy, obex plugging and dural grafting. Two have had sustained improvement and 1 has been stabilised. Five patients have had posterior fossa decompression,

40

fourth ventriculostomy and obex plugging. Three of these have exhibited sustained improvement, one has been stabilised and one patient has progressed after initial improvement. This patient went on to have as

J. Clin. Neuroscience Volume 3 Number 1 January 1996

Hydromyelia and Chiari malformations

Clinical studies

A

B

Fig 4. (Case 5) (A) T1 weighted MRI scan at 4 months following posteriordecompression,fourth ventriculostomyand obex plugging in a six year old boy, demonstrating a collapsedsyrinx.

a secondary p r o c e d u r e laminectomy and syringosubarachnoid shunting which has stabilised the neurological deficit. One patient has u n d e r g o n e posterior fossa decompression, fourth ventriculostomy and dural grafting. This patient progressed after initial improvement. Further decompression procedures have b e e n p e r f o r m e d without long-term relief. This patient presented for operation at an advanced clinical state. Further operative m a n a g e m e n t has been declined. Two patients have had a posterior fossa decompression and dural grafting alone. One has had sustained improvement while the other patient progressed. This patient has had two further procedures (syringosubarachnoid shunting and decompression with grafting) but has continued to progress. This patient also presented at a more advanced stage and has now refused further treatment (Case 13). Two patients have u n d e r g o n e laminectomy, myelotomy and syringosubarachnoid shunting as a primary procedure. Both patients have had sustained improvement.

Radiological result Following operation all patients have had serial structural studies (Mill or CT metrizamide myelography) to determine the size of the syrinx (Table 5).

Patients with Chiari 1 malformation O f the Chiari 1 patients six (55%) have shown a marked reduction in the size of the cavity. A futher three (27%)

have shown a moderate reduction. Two patients have (17%) have had a reaccumulation of the syrinx. In one patient this has resulted in a syrinx much smaller in size than preoperatively and this patient remains asymptomatic at 5 year follow up (Case 5, Fig. 4). No further p r o c e d u r e has been required.The other patient required three further procedures before the cavity collapsed (Case 1).

Patients with myelomeningocele Postoperative radiological results have been reviewed in only 9 out of 13 patients. Six have shown marked reduction (66%) and 2 have shown moderate reduction of the syrinx size (22%). One patient had no change in cavity size.

Discussion T h e traditional concept of hydromyelia is that it affects individuals in the second to fourth decades of life. An initial presentation in childhood was thought to be rare. Recently, hydromyelia has been identified with increasing frequency in patients younger than 20 years of age. 13,14Earlier detection of hydromyelia is now possible with the use of MRI in patients with few clinical symptoms or signs.The clinical features of hydromyelia in children differ from those in adults. Isu et al f o u n d that 65% of patients younger than 20 had scoliosis as an initial symptom. 1~ Only 15% of the adults in their series displayed scoliosis. Similarly, in our series 82% of the patients with Chiari 1 malformation had scoliosis on admission. It is r e c o m m e n d e d that MRI scans be p e r f o r m e d on all chil-

J. Clin. Neuroscience Volume 3 Number 1 January 1996

41

Clinical studies dren and adolescents with scoliosis and any evidence of neurological deficit. MRI is the investigation of choice in suspected hydromyelia and Chiari m a l f o r m a t i o n ? 5,I6 MRI has provided a non-invasive means of studying the anatomy before and after surgery and it has allowed for objective evaluation of the surgical results. However, it is clear that r e c u r r e n c e or persistence of clinical symptoms following operation is not always associated with r e a p p e a r a n c e of the syrinx (Fig. 1). Residual atrophy of the spinal cord may remain following successful collapse of the cavity. Similarly, r e c u r r e n c e of the syrinx as detected by routine postoperative MRI is not always associated with recurrence or worsening of symptoms (Fig.4).

Pathogenesis T h e hydrodynamic theories of Williams and G a r d n e r rely on the presence of a patent c o m m u n i c a t i o n between the fourth ventricle and the syrinx in o r d e r to explain initiation of hydromyelia. 6, 7. s Criticism of both hydrodynamic theories has focused on the frequent inability to detect a c o m m u n i c a t i o n between the fourth ventricle and the syrinx in pathological and radiological studies. At operation Foster and H u d g s o n were able to find a patent communication in only 26 out of 47 cases (55%). O f 4 autopsy cases, only one displayed a patent c o m m u n i c a tion. 17 Rice-Edwards has described a pathological study of 20 cases of syringomyelia, 9 of these having evidence of Chiari m a l f o r m a t i o n ? s In 17 cases a narrow c o m m u n i c a tion existed between the fourth ventricle a n d the syrinx but these were felt to be too narrow to act as a conduit for syrinx development. Recent studies have utilised MRI to look for a communication between the fourth ventricle and the syrinx.16,19,20,21,22, 23 Park et al, 23 c o m b i n i n g the MRI findings of a n u m b e r of reports, f o u n d that a c o m m u n i c a t i o n was present in only 4 of 127 cases (3%). Moreover, they f o u n d that there often exists a long segment of n o r m a l spinal cord between the fourth ventricle and the syrinx. In an attempt to explain these pathological and radiological findings, p r o p o n e n t s of a hydrodynamic theory have argued that the c o m m u n i c a t i o n partly or completely closes off at later stage of d e v e l o p m e n t of the cavity. 8'24'25 A n u m b e r of mechanisms have b e e n put forward to explain this closure. With c o n t i n u e d impaction, the Chiari m a l f o r m a t i o n may compress the central canal below the obex so that it can no longer dilate and thus progressively disappears. Alternatively, the c o m m u n i c a tion may b e c o m e compressed by a syringomyelic cavity that has f o r m e d in the paracentral white matter. In o u r series a noticeable c o m m u n i c a t i o n between the fourth ventricle and the spinal cord existed in all patients who had plugging of the obex as part of their procedure. This c o m m u n i c a t i o n was not necessarily seen on preoperative imaging.Early radiological techniques (myelography and CT myelography) were too limited to delineate a communication. 26 It may be that even the resolution of MRI is not sufficient to pick up narrow communications especially if such c o m m u n i c a t i o n s are

42

Hydromyelia and Chiari malformations collapsed. Mso when a c o m m u n i c a t i o n does exist it may lie away f r o m the midline or obliquely making it difficult to pick u p on MRI. A n u m b e r of researchers have failed to find obstruction of the fourth ventricular outlets and have used this to criticise G a r d n e r ' s t h e o r y . 27, 28 However, in o u r operative cases on Type 1 Chiari patients the f o r a m e n of Magendie was occluded in 10 out of 11 cases. This finding tends to favour G a r d n e r ' s hypothesis. Moreover, in three cases a distinct m e m b r a n e overlying the fourth ventricle was f o u n d to obstruct CSF flow. S u p p o r t for the presence of fourth ventricular outlet obstruction comes f r o m o t h e r sources. Levy et al r e p o r t e d 55 cases where the f o r a m e n of Magendie was described as being occluded and 10 where it was described as being open. 29 H a n k i n s o n described operative cases where there a p p e a r e d to be no drainage of CSF f r o m the fourth ventricle before the tonsils were opened, s° S u p p o r t for the theories of Aboulker a n d Ball and Dayan centre on the delayed filling of the spinal cord cavity in CT metrizamide myelography? 1 Ellertson and Greitz and Hall et al have f o u n d that intrasyrinx pressure exceeds subarachnoid pressure? ~'33To pass transneurally fluid would have to move against a pressure gradient which makes Ball a n d Dayans' theory improbable. T h e venous hypertension required u n d e r Aboulker's theory has not b e e n verified. It is likely that the migration of water soluble dyes into the syrinx represents an epiphenomenon.

Treatment N u m e r o u s surgical p r o c e d u r e s have b e e n advocated for the t r e a t m e n t of hydromyelia and Chiari malformation. Posterior fossa decompression has b e e n the most comm o n surgical p r o c e d u r e employed. Various additional steps have b e e n advocated including division of adhesions between the brain stem a n d the cerebellar tonsils, a m p u t a t i o n of the cerebellar tonsils, fenestration of the fourth ventricular outlets, plugging of the central canal with a muscle plug and insertion of a fourth ventriculosubarachnoid shunt. T h e efficacy of posterior fossa decompression procedures has v a r i e d . 6,14'29'34'35,36,37,~8 In 1965, G a r d n e r r e p o r t e d on the surgical t r e a t m e n t of 74 patients with Chiari malformation, 62 of w h o m h a d associated hydromyelia. 6 All these patients u n d e r w e n t posterior fossa decompression and plugging of the central canal at the obex. Seventy p e r c e n t showed i m p r o v e m e n t and 15% were stabilised. In a retrospective study, Levy et al r e p o r t e d on 127 patients. 29 Twenty-five patients had decompression and o p e n i n g of the fourth ventricle with 44% showing i m p r o v e m e n t a n d 24% having their neurological deficits stabilised. Sixty patients had decompression and a plugging p r o c e d u r e with 48% showing neurological i m p r o v e m e n t and 27% stabilising. The authors concluded that overall there was no benefit in plugging the central canal. They did show some benefit in plugging of the central canal in patients with progressive symptoms. Logue and Rice-Edwards, in a study of 68

J. Clin. Neuroscience Volume 3 Number 1 January 1996

Hydromyelia and Chiari malformations

Clinical studies

patients with Chiari malformation and syringomyelia, f o u n d improvement in 27% of those u n d e r g o i n g Gardner's p r o c e d u r e and 29% of those undergoing decompression alone, s6 They r e c o m m e n d e d simple decompression over Gardner's p r o c e d u r e as the latter was associated with greater risks. Schlesinger et al reported on 16 patients who underwent posterior fossa decompression. 28 I m p r o v e m e n t was seen in 8 patients, 5 were unchanged, 2 became worse and one died of respiratory failure. In a recent r e p o r t comparing the outcome of posterior fossa decompression and syringosubarachnoid shunting in patients with syringomyelia and Chiari malformation, Vaquero et al f o u n d neurological improvem e n t in 27% and stabilisation in 40% of cases following posterior fossa decompressionY Pillay et al p e r f o r m i n g Gardner's p r o c e d u r e in 24 patients with Chiari malformation and hydromyelia r e p o r t e d favourable results in 80% of patients. ~ H o f f m a n et al have reviewed the results of posterior fossa decompression in 36 children with Chiari malformation and hydromyelia. 39 Thirty-one patients underwent Gardner's p r o c e d u r e with 25 showing sustained improvement and a further 6 showing stabilisation. Five patients u n d e r w e n t posterior fossa decompression alone with 2 improving and 2 progressing. In our series a total of 21 patients u n d e r w e n t posterior fossa decompression. Sustained clinical improvem e n t o c c u r r e d in 11 patients (52%) and stabilisation of clinical deficit was achieved in a further 9 (29%). These results confirm the overall long-term benefits of posterior fossa surgery in improving CSF circulation and causing syrinx collapse. T h e benefits of plugging of the central canal at the obex has b e e n controversial. Levy et al and Logue and Rice-Edwards have f o u n d similar improvement following both Gardner's p r o c e d u r e and simple decompression. 29,~6The muscle plug in these cases was not sutured or glued into place. On several occasions in our series it was n o t e d at operation that the muscle plug would float away from the area of the obex if it were not secured. This p h e n e m e n o n has been n o t e d by others. 4° In our series 56% of patients have had sustained clinical improvement and a further 33% have had stabilisation following obex plugging procedures. This compares favourably with reports of treatment of hydromyelia by decompression alone. 29,36 An objection raised against plugging of the central canal is the increased risk of operative complications. Protracted nausea and vomiting, persistent bradycardia and respiratory depression have been associated with obex plugging. In our series obex plugging was associated with transient intraoperative bradycardia in 3 cases, transient postoperative nausea in one case and temporary ankle clonus in one case. T h e r e was no long-term morbidity. Our series shows that plugging of the obex can be a safe procedure. If the obex is accessible at the time of operation we would r e c o m m e n d plugging of the obex as an effective additional step. O t h e r complications have been r e p o r t e d with poste-

rior fossa surgery.In early reports operative mortality ranged from 0% to 15%. 6,41,42 The cause of death was often respiratory depression, postoperative haemorrhage, acute hydrocephalus and brainstem infarct. Recent reports have confirmed a m u c h lower mortality rate. la,3s,~9Morbidity from posterior fossa procedures can take the form of increased neurologic deficit, hydrocephalus, cardiac or respiratory irregularities and aseptic meningitis. The risk of the latter is significantly reduced by dural grafting. Our series confirms the low overall morbidity of m o d e r n posterior fossa decompression surgery for the treatment of hydromyelia. The development of microsurgical techniques and improved inert shunt materials has revived interest in laminectomy, myelotomy and syringostomy. Shunting of the syrinx to the subarachnoid space or to the peritoneal and pleural cavities has b e e n advocated. It is claimed that these procedures are relatively free of mortality and morbidity and that the shunt can provide continuous drainage of the syrinx. The early results r e p o r t e d on syringosubarachnoid shunting were disappointing. 43,44 Results in recent reports have been more encouraging. For example, Tator et al have reported on 20 patients with syringomyelia treated with a syringosubarachnoid shunt. 45 Tonsillar ectopia was absent or minimal in all cases. I m p r o v e m e n t occurred in 11 patients (55%) and cessation of progression occurred in 4 (20%).Vaquero et al r e p o r t e d on 30 patients with hydromyeliaY Fifteen were treated with syringosubarachnoid shunting and 15 with posterior fossa decompression. They concluded that syringosubarachnoid shunting was equally as effective as posterior fossa decmpression in inducing syrinx collapse. Isu et al treated 16 children with hydromyelia by insertion of a syringosubarachnoid shunt. 13 I m p r o v e m e n t occurred in 15 cases (94%). This improvement was largely subjective. In our series only three patients had syringosubarachnoid shunting as a primary p r o c e d u r e with two having favourable results. Some authors have advocated shunting to a low pressure cavity such as the pleura or peritoneum. Barbaro et al have r e p o r t e d on 39 cases of syringomyelia of various causes. 46 Fifteen patients were treated with a syringoperitoneal shunt with 53% showing a good result (improvement) and a further 27% a fair result (stabilisation). Nineteen patients had a syringosubarachnoid shunt with 32% having a good result and a further 27% a fair result. They concluded that syringoperitoneal shunting had the highest improvement rate but also the highest complication rate. Six patients had shunt related complications. Suzuki et al have r e p o r t e d their results of syringoperitoneal shunting in 29 patients with syringomyelia. 47 Twenty-two patients (76%) improved and 5 (17%) were stabilised. The authors recognised that the shunts were low flow systems and required periodic revision in m u c h the same way as ventriculoperitoneal shunts used in the treatment of hydrocephalus. Williams has advocated syringopleural shunting. 4s In a r e p o r t of 21 patients treated this way 6 (29%) showed long-term improvement and 6 (29%) were stabilised.

J. Clin. Neuroscience Volume 3 Number 1 January 1996

43

Clinical studies Williams r e c o m m e n d e d the use of syringopleural shunting as a salvage operation after failure of posterior fossa surgery. He also r e c o m m e n d e d it as a primary procedure in cases of hydromyelia with marked hindbrain arachnoiditis. Wisoff and Epstein have advocated the use of syringopleural shunting as a primary procedure in dysraphic children with large thoracic hydromyelic cavities. 14 Criticism of shunt procedures has centred on the shunt related complications. Shunts will need periodic revision. Moreover for the treatment of hydromyelia associated with Chiari malformation shunts do not address the problem of brainstem compression n o r do they disarm the filling mechanism of the syrinx. As a result shunting procedures should be restricted to those cases where hydromyelia recurs or fails to respond to posterior fossa decompression.

Hydromyelia and Chiari malformations 3 4 5 6 7 8 9 10 11

Conclusions Hydromyelia is capable of a wide variety of clinical presentations. A c o m m o n clinical presentation in children is progressive scoliosis with minimal neurological features. Increasing utilisation of MRI allows for early diagnosis of hydromyelia in such patients. MRI also allows for objective assessment of surgical outcome. Posterior fossa decompression, fourth ventriculostomy, dural grafting and obex plugging is associated with favourable surgical results.Fourth ventriculostomy improves CSF drainage by opening obstructed fourth ventricular outlets. Closure of the dura with graft lessens the risk of aseptic meningitis and therefore allows fenestration of the fourth ventrcle outlets to be functional. If it is practicable, depending u p o n accessibility to the obex at operation, a plug is an effective additional step that in our series has not been associated with significant complications. We favour reserving shunting for the minority of patients in whom this treatment has been unsuccessful.

Acknowledgments D e p a r t m e n t of Medical Illustration, University of New South Wales Teaching Hospitals for the illustrations. Received21 December 1993 Accepted 3 May 1995

12 13 14 15 16 17

18 19 20 21

22 23

Correspondence and offprint requests: R.F.C.Jones, Department of Neurosurgery, Prince of Wales Children's Hospital, Randwick, NSW 2031, Australia. References 1 Milhorat TH, Johnson WD, Miller JI, Bergland RM, Hollenberg-SherJ. Surgical treatment of syringomyelia based on magnetic resonance imaging criteria. Neurosurgery 1992; 31 (2):231-245 2 Foster JB, Hudgson T. The pathology of communicating syringomyelia. In: Barnett HJM, FosterJB, Hudgson T eds. Syringomyelia, Volume 1 in Major Problems in Neurology. Philidelphia: WB Saunders, 1973:79-103

44

24 25

26 27

J. Clin. Neuroscience Volume 3 Number 1 January 1996

Oakes WJ. Chiari malformations, hydromyelia, syringomyelia. In: Wilkins RH, Rengachary SS eds. Neurosurgery. New York: McGraw Hill, 1985:2102-2124 Park TS, Cail WS, Maggio WM, Mitchell DC. Progressive spasticity and scoliosis in children with myelomeningocoele.J Neurosurg 1985; 62:36%375 Cameron AH. The Arnold-Chiari and other neuroanatomical malformations associated with spina bifida. J Pathol 1957; 73:195-211 Gardner WJ. Hydrodynamic mechanism of syringomyelia: its relationship to myelocoele.J Neurol Neurosurg Psychiatry 1965; 28:247-259 Gardner WJ. The dysraphic states from syringomyelia to anencephaly. Amsterdam: Excerpta Medica, 1973 Williams B. On the pathogenesis of syringomyelia: a review. J R Soc Med 1980; 73:798-806 Ball MJ, Dayan AD. Pathogenesis of syringomyelia. Lancet 1972; 2:799-801 AboulkerJ. La syringomyelia et les liquides intrarachidiens. Neurochirugie(suppl 1) 1979; 25:1-144 Abbe R, Coley WB. Syringomyelia, operation-exploration of cord, withdrawal of fluid, exhibition of patient. J Nerv Ment Dis 1892; 17:512-520 Simpson DA, Robson A. Recurrent subarachnoid bleeding in association with dural substitute. Report of three cases.J Neurosurg 1984; 60:408-409 Isu T, lwaski Y, Akino M. Hydrosyringomyelia associated with a Chiari 1 malformation in children and adolescents. Neurosurgery 1990; 26:591-597 WisoffJH, Epstein E Management of hydromyelia. Neurosurgery 1989; 25:562-571 DoMing RJ, Tress BM. MRI-The investigation of choice in syringomyelia?. Australas Radiol 1989; 33:337-343 Kokmen E, Marsh WR, Baker HL. Magnetic resonance imaging in syringomyelia. Neurosurgery 1985; 17:267-270 FosterJB, Hudgson T. The surgical treatment of communicating syringomyelia. In: Barnett HJM, FosterJB, Hudgson Teds. Syringomyelia, in Volume 1, Major Problems in Neurology. Philidelphia: WB Saunders, 1973:65-77 Rice EdwardsJM. A pathological study of syringomyelia. J Neurol Neurosurg Psychiatry 1977; 40:198-199 Lee BCP, Zimmerman RD, Manning~J, Peck MDE MR imaging of syringomyelia and hydromyelia. AJR 1985; 144: 273-284 Pojunas K, Williams AL, Daniels DL, Haughton VM. Syringomyelia and hydromyelia: Magnetic resonance evaluation. Radiology 1984; 153:679-683 Samuelsson L, Bergstrom K, Thomas KA, Itemmingsson A, Wallenstem R. MR imaging of syringomyelia and Chiari malformation in myelomeningocoele patients with scoliosis. AJNR 1987; 8:539-546 ShermanJL, Barkovich AJ, Citrin CM. The MR appearance of syringomyelia: new observations.AJNR 1986; 7:985-995 Parks TS, Call WS, Broaddus WC, Walker MG. Lumboperitoneal shunt combined with myelotomy for treatment of syringhydromyelia.J Neurosurg 1989; 70: 721-727 Newman PK, Terenty TR, FosterJB. Some observations on the pathogenesis of syringomyelia. J Neurol Neurosurg Psychiatry 1981; 44:964-969 FosterJB, Hudgson R The pathogenesis of communicating syringomyelia. In: Barnett HJM, Foster JB, Hudgson T eds. Syringomyelia, in Volume 1, Major Problems in Neurology. Philidelphia: WB Saunders, 1973: 104-123 Gardner WJ, McMurry FG. 'Non-communicating' syringomyelia: a non-existent entity. Surg Neurol 1976; 6: 251-256 Williams B. Hypothesis: the distending force in the production of 'communicating syringomyelia'. Lancet 1969; 2:189

Hydromyelia and Chiari malformations

Clinical studies

28 Schlesinger EB, AntunesJL, Michelsen WJ, Louis KM. Hydromyelia: Clinical presentation and comparison of modalities of treatment. Neurosurgery 1981; 9:356-365 29 Levy WJ, Mason L, HahnJE Chiari malformation presenting in adults: A surgical experience in 127 cases. Neurosurgery 1983; 12:377-390 30 HankinsonJ. Syringomyelia and the surgeon. In: Williams D ed. Modern trends in neurology: series 5. London: Butterworth, 1970:127 31 Li KC, Chui MC. Conventional and CT metrizamide myelography in Arnold-Chiari 1 malformation and syringomyelia. AJNR 1987; 8:11-17 32 Ellertson AB, Greitz T. The distending force in the production of communicating syringomyelia (letter). Lancet 1970; 1:1234 33 Hall P, Turner M, Aichinger S, Bendick P, Campbell R. Experimental syringomyelia. The relationship between intraventricular and intrasyrinx pressures. J Neurosurg 1980; 52:812-817 34 Dyste GN, Menezes AH, VanGilderJC. Symptomatic Chiari malformations. An analysis of presentation, management, and long-term outcome. J Neurosurg 1989; 71:159-168 35 Cahan LD, BentsonJR. Considerations in the diagnosis and treatment of syringomyelia and the Chiari malformation.J Neurosurg 1982; 57:24-31 36 Logue V, Rice Edwards M. Syringomyelia and its surgical treatment-an analysis of 75 patients. J Neurol Neurosurg Psychiatry 1981; 44:273-284 37 Vaquero J, Martinez R, Arias A. Syringomyelia-Chiari complex: magnetic resonance imaging and clinical

evaluation of surgical treatment. J Neurosurg 1990; 73:64-68 38 Pillay PK, Awad IA, Little JR, HahnJE Surgical management of syringomyelia: a five year experience in the era of magnetic resonance imaging. Neurol Res 1991; 13:3-9 39 Hoffman HJ, NeillJ, Crone KR, Hendrick EB, Humphreys RP. Hydrosyringomyelia and its management in childhood. Neurosurgery 1987; 21:347-351 40 Peerless SJ, Durward QJ. Management of syringomyelia: a pathophysiological approach. Clin Neurosurg 1983; 30: 531-576 41 Williams B. A critical appraisal of posterior fossa surgery for communicating syringomyelia. Brain 1978; 101:223-250 42 Banerji NK, MillarJHD.Chiari malformation presenting in adult life-its relationship to syringomyelia. Brain 1974; 97:157-168 43 LoveJG, Olafson RA. Syringomyelia: Alook at surgical therapy.J Neurosurg 1966; 24:714-718 44 Wetzel N, Davis L. Surgical treatment of syringomyelia. Arch Surg 1954; 68:570-573 45 Tator CH, Meguro K, Rowed DW. Favourable results with syringosubarachnoid shunts for treatment of syringomyelia. J Neurosurg 1982; 56:517-523 46 Barbaro NM, Wilson CB, Gutin PH, Edwards MSB. Surgical treatment of syringomyelia. Favourable results with syringoperitoneal shunting. J Neurosurg 1984; 61: 531-538 47 Suzuki M, Davis C, Symon L, Gentili E Syringoperitoneal shunt treatment of cord cavitation. J Neurol Neurosurg Psychiatry 1985; 48:620-627 48 Williams B, Page N. Surgical treatment of syringomyelia with syringopleural shunting. BrJ Neurosurg 1987; 1:63-80

J. Clin. Neuroscience Volume 3 Number 1 January 1996

45