British Journal of Neurosurgery (1991), 5,475-484
ORIGINAL ARTICLE
The Dandy-Walker syndrome: diagnostic and surgical considerations M. MARINOV, S. GABROVSKY & S . UNDJIAN
Br J Neurosurg Downloaded from informahealthcare.com by UB Kiel on 11/08/14 For personal use only.
Department of Neurosurgery, Medical Academy, G. Sofijsky-Str. 1, 1431 Sofia, Bulgaria
Abstract Thirteen cases of the Dandy-Walker syndrome were seen over a period of 12 years. In this study their preoperative evaluation, surgical treatment and outcome is reviewed. The incidence of hydrocephalus in the series was 77%. Eight of the patients had associated central nervous system anomalies, and the need to identify these abnormalities is stressed. Because the two conditions require different surgical management, during the diagnostic work-up it is crucial to differentiate between the Dandy-Walker syndrome and extra axial cerebrospinal fluid (CSF) filled cysts, and to establish the patency of the aqueduct. The reliability of different diagnostic methods is discussed and the value of metrizamide computed tomographic ventriculography is emphasized. Unsatisfactory results following membrane excision in early cases, and observations of CSF dynamic disturbances confirm the futility of the direct approach for the management of the Dandy-Walker syndrome. The aqueduct was invariably patent and allowed either primary cyst or ventricular shunting, although simultaneous double shunting using a ‘Y’ connector may need to be carried out later. The mortality rate for the series was 18%. At follow-up 54.5% of the children were judged to have normal IQs.
Key words: Dandy- Walker syndrome, posterior fossa extra-axial cyst, metrizamide CT ventriculography, CSF pow disturbances, CSF shunting, outcome.
Introduction The first clinical description of the DandyWalker syndrome was published by Dandy and Blackfan in 1914.’ Further classical observations were made during the next decades in a series of publications by Dandy in 1921; Taggart and Walker in 1942: and Walker in 1944.4 In 1954, Bendas named the condition the ‘Dandy-Walker syndrome’. A number of features characterize the syndrome, the most important of these being cystic dilatation of the IVth ventricle, inconsistent patency of its outlet foramina, dysgenesis of the inferior vermis, and a high incidence of hydrocephalus. Atresia or occlusion of the foramina, which was formerly thought to be the cause of the cyst, is present in only half of
the cases.6 There is no correlation between the degree of hydrocephalus, the size of the expanded IVth ventricle, the extent of the vermian dysgenesis and the patency of the outlets of the IVth ventricle. As well as associated brain anomalies there are also systemic malformations which favour the view that this complex syndrome is the result of developmental failure of both the CNS and other organ^.^ Many theories have been published but none fully explain the whole phenomenon. Because of the relative rarity of the Dandy-Walker syndrome and lack of knowledge about how it develops, management is still debated. Evidence has accumulated that a direct posterior fossa operation is not effective.’-I3 The alternative is to carry out a
475
476
M. Marinov, S. Gabrovsky & S. Undjian
Br J Neurosurg Downloaded from informahealthcare.com by UB Kiel on 11/08/14 For personal use only.
shunting procedure, but some aspects of this remain controversial and require further investigation. In 1978 one of us (S.G.) reported the first cases of the Dandy-Walker syndrome diagnosed in our institute.I4 During the following years we saw an increasing number of children with Dandy-Walker cysts, and we present our experience based on this relatively large series followed over a period of 12 years. Clinical management and preoperative evaluation are emphasized and the principles of surgical treatment are discussed.
Clinical material and methods During the 12-year period of 1975-86, 13 patients with the Dandy-Walker syndrome were seen by the Paediatric Division of the Department of Neurosurgery of the Medical Academy, Sofia (Table I). Their records were reviewed to ascertain the presentation, diagnostic work-up, and treatment of each case. A standard preoperative intelligence quotient was performed by a clinical psychologist using the Stanford Binet Intelligence Scale or the Wechsler Intelligence Scale for children. Cerebral angiography was the most common investigation, being performed on 12 occasions. Ten C T scans were carried out, twice with ventriculography. Three patients had pneumoencephalography and two had direct cystography. Radionuclide dynamic CSF investigations (one "'InCaDTPA cysternogram and three ventriculograms with 9 9 n T pertech~ netate) were performed in four children. Up to 1977 the preferred method of investigation was a combination of angiography and air studies (three cases), between 1977 and 1983 C T scans were combined with angiography and radionuclide ventriculography (six cast.s) ,and after 1983 (Metrizamide C T ventriculography was the favoured method of investigation (four cases), especially if the ventricles were enlarged. Patient 13 was not operated on because the intracranial pressure was not raised, but the patient was kept under observation. Another patient only had a cranio-cervical meningocele
repaired. The operative procedures carried out in the remaining 11 children are outlined in Table 11. After 1980 the direct approach was abandoned, and only shunting procedures were performed. One patient died in the immediate postoperative period and another was lost to follow-up. The remaining children were observed for periods ranging from 7 months to 10 years (average 4 years and 5 months).
Results More than half the cases were seen before they were a year old. There was a trend towards earlier diagnosis in the later years of the study. There was a female to male predominance of 1.6:l. Table I shows that most patients presented with non-specific signs and symptoms of intracranial hypertension, and the striking absence of abnormal cerebellar and brain-stem signs. The IQs of the children on admission were subnormal or borderline. Two children, aged 2 and 3 months, had normal head circumferences. Patient 10 had a large head but a soft anterior fontanelle and no ventricular dilatation at diagnosis. There was questionable ventricular dilatation in patient 12. If the latter case is excluded, the incidence of hydrocephalus in the series is 77%. Occipital bossing, although considered to be a characteristic of a Dandy-Walker cyst, was only seen in half the cases. Associated anomalies, often multiple, were recorded in 61.5% of the patients. The only child with severe anomalies (patient 3) died the day after membrane excision. Two important questions need to be answered during the preoperative work-up: (1) What is the character of the posterior fossa cyst? Is it an intra- or extra-axial lesion? (2) What is the dynamic relationship between the fluid in the dilated IVth ventricle and the CSF pathways? Every effort should be made to obtain answers to these questions in order to plan the appropriate surgical procedure. Comparison of the angiographic findings of 12 children with Dandy-Walker cysts with those of 10 cases of posterior fossa arathnoid cysts, showed that this was a relatively unrelia-
~~
3 mo
4
3 mo F 4 mo F 2 mo M
~
~
Enlarged head, occipital bossing, gait disturbances Enlarged head, occipital bossing, delayed milestones, irritability, hypotonia Enlarged head, irritability, vomiting, bulging anterior fontanel Enlarged head, occipital bossing, febrile convulsions, visual impairment Enlarged head, vomiting bulging anterior fontanel Enlarged head, soft anterior fontanel, occipital bossing, delayed milestones Enlarged head, bulging fontanel, irritability, vomiting Enlarged head, soft fontanel, occipital bossing Sunken fontanel, seizures, delayed milestones Generalized hypotenia, IQ = 55 Hyperactive reflexes, IQ = 45
Increased ICP, IQ = 70
IQ = 22
Hyperactive reflexes,
Increased ICP, mild right hemiparesis, right amaurosis, IQ = 58 Increased ICP, IQ = 64
Increased ICP, IQ = 68
Increased ICP, mild spastic quadriparesis, ataxia, IQ = 70 Increased ICP, cerebellar signs, IQ = 44
IQ = 50
Suboccipital mass
Enlarged head, delayed milestones
Increased ICP, spastic quadriparesis; IQ = 37 Right spastic hemiparesis, right abducens paresis, severe psychomotor retardation Hyperactive reflexes, IQ = 48
Neurological findings; IQ score
Enlarged head, occipital bossing, delayed milestones Enlarged head, seizures, delayed milestones
Presenting symptoms
F, female; M, male; ICP, intracranial pressure; AVM, arteriovenous malformation.
13
12
11
F
3 mo
M
8 mo
9
10
1y3mo F
F
8 mo
M
3Y F ly2mo
8
7
6
5
F
1y4mo M
3
2
2y2mo F 3y9mo M
Age at diagnosis; sex
1
Case No.
TABLE I. Clinical summary in 13 patients with Dandy-Walker syndrome
Agenesis of the corpus callosum, plagiccephaly
-
Hemangioma faciei
Hypertelorism, bilateral syndactily
Hemangioma faciei, AVM a. vertebralis right/pars extracranialis), ganglioneuromas reg. colli
Spina bifida occulta C4-C5
-
Cardiac anomaly (ventricular septa1 defect), bilateral hydronephrosis and megaureter, Klippel-Feil syndrome, chromosome abnormalities ( G 2 ,trisomia), bilateral inquinal hernia Meningocele at the craniocervical junction, Klippel-Feil syndrome
Agenesis of the corpus callosum
Associated anomalies
Br J Neurosurg Downloaded from informahealthcare.com by UB Kiel on 11/08/14 For personal use only.
Proven
Present
Present Present Not present
Present Present Present Present Present
Not present Present ?
1
2 3
5 6
7
10 11 12
Proven Proven Proven
Proven Proven Proven Proven Not proven
-
CAA CAA CAA
EM EM Closure MC EM and VAA EM VAA VAA CAA
CAA
Initial treatment
3Y 1Y
-
-
VAA
2y7mo 7y6mo 4Y 7y 10 Y Lost to follow-up 2Y
2Y
-
16 mo
VPA VAA VA and CAA (‘Y’) CAA CA and VAA (‘Y’)
Revision CAA and EM -
Further treatment
Follow-up period
Disabled Normal Normal
No change Died 24 h postoperation Normal Normal Normal Normal Normal
Died (meningoencephalitis)
Neurological
Outcome
Severe PMR Normal Moderate PMR
Mild PMR Normal Normal Normal Normal
No change
Intellectual
VA(P)A, ventriculoatrio(peritoneo)nastomosis;CAA, cystoatrioa anastomosis; ‘Y’, Y-connector; EM, excision of membranes; PMR, psychomotor retardation.
9
8
4
Not proven Not proven
Aqueduct patency
No.
Ventricular dilatation on admission
Case
TABLE 2. Applied surgical procedures and outcome in 12 cases with Dandy-Walker syndrome
Br J Neurosurg Downloaded from informahealthcare.com by UB Kiel on 11/08/14 For personal use only.
rp 4 W
Br J Neurosurg Downloaded from informahealthcare.com by UB Kiel on 11/08/14 For personal use only.
The Dandy- Walker syndrome ble method of differentiating between the two conditions. In the children with DandyWalker cysts the posterior inferior cerebellar artery (PICA) was well defined in two cases, hypoplastic in three, and not visualized in seven. The anterior inferior cerebellar artery (AICA) was identified in all except one case where the extremely hypoplastic cerebellum was supplied by only the superior cerebellar arteries (SCAs). A small cerebellum was visualized by a capillary and venous ‘blush’ in seven cases, and a large cerebellum by a ‘blush’ in one case. The torcula Herophili was high in seven cases but normal in position in two. In the patients with arachnoid cysts the posterior inferior cerebellar artery (PICA) was not visualized in one case, and it was hypoplastic in another two. A ‘blush’ was absent in two cases in whom the cerebellum was deformed and compressed in a forward and upward direction. In five patients the torcula was high. Neither the courses of the basilar artery, the posterior cerebral arteries and the SCA, nor the type of displacement of the great vein of Galen and the straight sinus were helpful in differentiating between the Dandy-Walker syndrome and arachnoid cysts. The only reliable criterion was the type of displacement undergone by the PICA (if it was visualized); in retrocerebellar arachnoid cysts the PICA was always pushed
479
anteriorly and superiorly, whereas the large encysted IVth ventricles of the Dandy-Walker complex displaced the vessel in a posteroinferior direction. Cisternography was performed in four patients but was of little help. The investigation showed a communication between the encysted IVth ventricle and the basal subarachnoid spaces in two cases (a patent aqueduct was seen in one of them); in the other two cases there was no communication between the IVth ventricle and the subarachnoid spaces. Altogether the aqueduct was demonstrated to be patent by ventriculography or cisternography in eight cases; in two of them Metrizamide C T dynamic studies were also carried out. C T scanning is indispensable for assessing the structures in the superior and inferior tentorial compartments, the degree of ventricular enlargement and the presence of associated CNS anomalies. Mid-sagittal CT scan reconstructions can demonstrate the difference between a classical Dandy-Walker cyst and a retrocerebellar cyst (Fig. 1). In recent years dynamic Metrizamide C T ventriculography has been found to be very useful to demonstrate posterior fossa arachnoid cysts and patency of the aqueduct (Figs. 2a, b and 3a, b). Early in the series membranectomy was
FIG.1 . Midsagittal CT reconstruction: (a) in Dandy-Walker cyst; note almost total absence of cerebellar vermis and the pronounced cystic transformation of IVth ventricle (intra-axial cyst); (b) in a retrocerebellar (extra-axial) CSF cyst with normal sized IVth ventricle; vermis cerebelli is present.
Br J Neurosurg Downloaded from informahealthcare.com by UB Kiel on 11/08/14 For personal use only.
480
M.Marirtov, S. Gabrovsky & S. Undjian
FIG. 2. CT scan in a case with huge retrocerebellar arachnoid cyst. (a) IVth ventricle is not visible because of massive displacement, which makes differentiation from Dandy-Walker cyst difficult. (b) Metrizamide CT ventriculography demonstrates an isolated retrocerebellar cyst, normal IVth ventricle (arrowhead) with attenuated, but preserved vermis above it. (c) Native C T scan after posterior fossa cyst excision.
performed to try to establish a free flow of CSF, and to prevent recurrence of the cyst. In patient 6 the procedure failed and a ventricular shunt had to be inserted 10 days later to control the intracranial pressure. Patient 3 died on the first postoperative day, probably due to brain-stem dislocation following resection of the cyst membranes (this patient also had severe associated anomalies). In patient 2 the direct approach did not affect the clinical outcome, and in patient 5 simultaneous membranectomy and ventricular shunting was carried out successfully, although a shunt revision had to be performed 5 years later. Seven patients underwent primary shunting procedures. In five a cysto-atrial shunt was performed, three of them were well at the time of follow-up, but shunt failure occurred in the other two and shunt revisions had to be carried out 9 months after insertion. Fenestration of the cyst was attempted at the time of shunt revision in patient 1; this caused CSF leakage, prolonged intracranial infection, and finally death of the child. In patient 10 a cyst shunt was converted to a ventricular one because of progressive secondary ventricular enlargement. In two children who had primary ventricular shunting the outcome assessed
after 6 and 6.5 years, respectively, was excellent, but then there was a sudden dramatic deterioration in their condition with signs of brain-stem compression shortly afterwards. In one of the children, as well as shunt malfunction, there was a lack of communication between the supra- and sub-tentorial CSF compartments which was confirmed during operation and it was probable that a secondary obstruction of the aqueduct had occurred. The tubing in the cyst was joined to a ventricular catheter by a ‘Y-connector. In the other child, in spite proven patency of the aqueduct, it was decided to convert the drainage system to a cysto-atrial shunt rather than carry out a revision. Disappearance of the cyst on follow-up C T scans was only seen in patient 11 (Fig. 3); in the remaining cases incomplete re-expansion of the cerebellum occurred. In summary, the most favourable results were seen with cysto-atrial and cystoatrial combined with ventriculo-atrial shunting. Excluding the patient who was lost to follow-up, the mortality rate for the series was 18.2% (2/11); six children (54.5%) had normal motor and intellectual developmental outcomes.
Br J Neurosurg Downloaded from informahealthcare.com by UB Kiel on 11/08/14 For personal use only.
The Dandy- Walker syndrome
481
FIG.3. CT investigations in patient 11: (a) native C T scan of the Dandy-Walker cyst before treatment; (b) patency of aqueductus of Sylvii was confirmed by means of metrizamide CT ventriculography; (c) total cerebellar re-expansion one year after cysto-atrial shunting with re-establishment of posterior fossa architecture.
onuclides were also found to be unhelpful in contradistinction to some report^.^^-^^ In a Several points concerning the treatment of previous study2' we came to the conclusion patients with the Dandy-Walker syndrome are that the Dandy-Walker syndrome and posterworthy of comment. The diagnosis of symptoior fossa arachnoid cysts are associated with matic patients is relatively easy. Current methods of neuro-imaging provide the means of CSF-flow disturbances. In both cases commumaking an early and precise diagnosi~.'~-'~ With nicating and occlusive hydrocephalus are preincreasing frequency prenatal ultrasonography sent. There is documented evidence that at is detecting the Dandy-Walker syndrome and least 50-82% of patients with the DandyWalker syndrome have patent IVth ventricular associated anomalies in the fetus.18J9 . ~ ~ . ~supports ~. the earlier During this study a variety of diagnostic f ~ r a m i n e . ~ . ~ This belief that the rhombencephalic anomaly intermethods were used, but some of them proved feres with the normal development of the to be of little value in confirming the diagnosis subarachnoid spaces. The observations of Milor helping to plan appropriate treatment. Contrary to earlier reportsZ0J1and in agree- horat et al. in 197128have shown, however, ment with the conclusions reached by Rai- that the subarachnoid spaces in the Dandymondi et al. in 1984,22we found that angiogra- Walker syndrome are well developed but phy is not a reliable method of distinguishing compressed by the raised intracranial pressure. between the Dandy-Walker syndrome and Lack of patency of the subarachnoid spaces posterior fossa arachnoid cysts. C T scanning provides an explanation for the unsatisfactory Metrihas also been shown to be of little help in results of cyst e x c i s i ~ n . ~ ~ ~ , ' distinguishing between the two conditions. It zamide C T ventriculography greatly improved has been suggested that pneumoencephalogra- the preoperative assessment of the patients.'O phy or injecting air into the cyst via the cisterns Like Maria et ul.,16 this investigation was magna is the best method of investigation, and found to be superior to any other method of it did confirm the diagnosis and the patency of neuroimaging. As well as demonstrating pathe aqueduct in the two children who were tency of the aqueduct, it also ruled out the investigated in this way in this series, but both presence of a posterior fossa extra-axial cyst. deteriorated immediately afterwards. Because Whilst some authors have ignored the diagnosof this and the unreliability of lumbar pneuno- tic and surgical peculiarities of different postencephalography, air studies are no longer erior fossa developmental CSF cyst^^'-^^ it is performed. Cisternographic studies using radi- important to know if an extra-axial cyst is
Discussion
1~13~16~22723929
Br J Neurosurg Downloaded from informahealthcare.com by UB Kiel on 11/08/14 For personal use only.
482
M.Marinow, S. Gabrowsky & S. Undjian
present since arachnoid cysts may benefit from membrane excision rather than cyst shunting. During recent years shunting has become the treatment of choice for the Dandy-Walker syndrome. If hydrocephalus is not present at the time of diagnosis, a cysto-peritoneal shunt is advocated.22There is controversy concerning which shunt to use if hydrocephalus is present: lateral ventricular,8Js ~ y s t i c > ’ ~and J~ simultaneous ‘Y’-shunting of the lateral ventricles and the cyst have been used. ‘Y’shunting was proposed by Raimondi et al. in 196911and has been favoured by other^^^,^^,^' to avoid progressive upward herniation of the cyst which can cause brain-stem compression and secondary aqueduct ,stenosis. A more flexible approach towards treatment has been advocated depending on the presence and severity of hydrocephalus,’ the age of the ~ a t i e n t and, ~ ~ . most ~ ~ important of all, on whether the aqueduct is patent.16 In contradistinction to the often-cited concept of ‘functional’ or true aqueduct steno~is,6.~IJ~J~ like Sawaya and McLaurin,” we found the aqueduct to be patent in all our cases. Based on these findings, our policy has been to shunt the lateral ventricles or the cyst rather than performing a primary double-shunt procedure. In o w opinion double shunting does not insure hydrodynamic balance between the supra- and sub-tentorial compartments. Moreover, occlusion of one or other of the proximal catheters may accentuate the pressure differences even if a single regulating pressure device is used, as recommended by Serlo et al.34 Although double shunting is not advocated as a primary procedure, it is not excluded for later use if shunt revision is necessary. There was a high shunt failure rate in our series; 75% of the primary shunts required revision. Late secondary aqueduct stenosis is seen after ventricular shunting, as occurred in patient 7. Cyst shunting has the advantage of encouraging the passage of CSF through the aqueduct in the normal direction of flow. It also leads to a marked reduction in the size of the posterior fossa cyst and may allow the normal architecture of the posterior fossa to be re-established as occurred in patient 11 (Fig. 3).
Because of the relatively small number of cases in our series, we have reservations about recommending posterior fossa shunting alone for the treatment of the Dandy-Walker syndrome. All the different shunting procedures were found to effectively control the intracranial hypertension and lead to decompression of the posterior fossa structures. No operative mortality occurred in this subgroup of patients, and four out of six had a favourable outcome, although one was moderately retarded. The mortality rate for the whole series was 18.2% which is considerably less than the 40-5046 recorded in earlier series,7y8JoJ and was comparable to more recently reported rate^.^.'^.^^ The intellectual outcome of 54.5% of our patients was normal which is similar to mean values calculated for different series by Hirsch et ~ 1 but . ~less good than the recent results reported by Maria et all6and Golden et a1.37 Associated anomalies, although frequently seen in our patients, were not severe except in one patient. There was no correlation between the presence of anomalies and the intellectual outcome. As stressed by Sawaya and McLaurin,” the factor having the greatest influence on outcome is an early and aggressive treatment approach to the control of the intracranial hypertension. Careful evaluation of extracranial anomalies such as congenital heart disease,4O is important from the point of view of reducing operative risks and postoperative complications. 1939
References 1 Dandy WE, Blackfan KD. Internal hydrocephalus. An
experimental, clinical and pathological study. Am J Dis Child 1914; 8:406-82. 2 Dandy WE. The diagnosis and treatment of hydrocephalus due to occlusions of the foramina of Magendie and Luschka. Surgery Gynec Obstet 1921; 112-24. 3 Taggart JK, Walker AE. Congenital atresia of the foramens of Luschka and Magendie. Arch Neurol Psychiat 1942; 48583-612. 4 Walker AE. A case of congenital atresia of the foramina of Luschka and Magendie. J Neuropathol exp Neurol 1944; 3~368-73.
5 Benda CE. The Dandy-Walker syndrome or the socalled atresia of foramen Magendie. J Neuropath exp Neurol 1954; 13:14-29.
Br J Neurosurg Downloaded from informahealthcare.com by UB Kiel on 11/08/14 For personal use only.
The Dandy- Walker syndrome 6 Hart MN, Malamud N, Ellis WG. The Dandy-Walker syndrome. A clinicopathological study based on 28 cases. Neurology 1972; 771-80. 7 Carmel PW, Antunes JL, Hilal SK, Gold AP. DandyWalker syndrome:clinicopathological features and reevaluation of modes of treatment. Surg Neurol 1977; 8~132-8. 8 Fischer GE. Dandy-Walker syndrome: an evaluation of surgical treatment. J Neurosurg 1973; 39:615-21. 9 Hirsch J-F, Pierre-Kahn A, Renier D, Saint-Rose C, Hoppe-Hirsch E. The Dandy-Walker malformation. A review of 40 cases. J Neurosurg 1984; 61:515-22. 10 James HE, Kaiser G, Schut L, Bruce DA. Problems of diagnosis and treatment in the Dandy-Walker syndrome. Child’s Brain 1979; 5:24-30. 11 Raimondi AJ, Samuelson G, Yarzagaray L, Northon T. Atresia of the foramina of Luschka and Magendie. The Dandy-Walker cyst. J Neurosurg 1969; 31:202-16. 12 Raimondi AJ. Cystic transformation of the IV ventricle (the Dandy-Walker cyst). In: Hoffman HJ, Epstein F (eds). Disorders of the developing nervous system: diagnosis and treatment. Blackwell Sci Pub1 Boston; 1986; 235-46. 13 Sawaya R, McLaurin RL. Dandy-Walker syndrome. Clinical analysis of 23 cases. J Neurosurg 1981; 55939-98. 14 Gabrovsky S. Hydrocephalus in atresia of the foramina of Magendie and Luschka (Dandy-Walker syndrome). Neurol Psychiat i Neurochirurg (Sofia), 1978; XVII: 107-16. 15 Hough DRYMengel MB, Malik AH. Mega cisterna magna: diagnosis using metrizamide computed tomographic cisternography. Neurosurgery 1982; 11:260-2. 16 Maria BL, Zinreich SJ,Carson BC, Rosenbaum AE, Freeman J M Dandy-Walker syndrome revisited. Pediat Neurosci 1987; 1345-51. 17 Masdeu JC, Dobben GD, Azar-Kia B. Dandy-Walker syndrome studied by computed tomography and pneumoencephalography. Radiology 1983; 147:109-14. 18 Fileni A, Colosimo C, Mirk P, De Gaetano AM, di Rocco C. Dandy-Walker syndrome: diagnosis in utero by means of ultrasound and C T correlations. Neuroradiology 1983; 24233-5. 19 Newman GC, Buschi AI, Sugg NK, Kelly TE, Miller JQ. Dandy-Walker syndrome diagnosed in utero by ultrasonography. Neurology 1982; 32:180-4. 20 La Torre E, Formna A, Occhipinti E. Angriographic differentiation between Dandy-Walker cyst and arachnoid cyst of the posterior fossa in newborn infants and children. J Neurosurg 1973; 38:298-308. 21 Wolpen SM, Haller JS, Rabe EF. The value of angiography in the Dandy-Walker syndrome and posterior fossa extra-axial cysts. Am J Roentg 1970; 109:261-72. 22 Raimondi AJ, Sato K, Shimoji T. The Dandy-Walker syndrome, S. Karger, Basel, 1984. 23 French BN. The Dandy-Walker malformation. In: Youmans TR (ed) Neuroloeical Suraerv. vol 3. W.B. Saunders, Phiiadelphia, 198:; 1236-i380.
483
24 Glasauer FE. Isotope cisternography and ventriculography in congenital CNS anomalies. J Neurosurg 1975; 43:18-26. 25 Harbert JC, McCullough DC. Radionuclide studies in an unusual case of Dandy-Walker cyst. Radiology 1971; 101:363-6. 26 Palma A, Nazar N, Castro M y Olea E, Guzman G. Dandy-Walker malformation and the contribution of radioisotopic studies of cerebral spinal fluid dynamics to its diagnosis. Acta Neurochir (Wien) 1982; 61~319-24. 27 Marinov M, Undjian S, Wetzka P. An evaluation of the surgical treatment of intracranial arachnoid cysts in children. Child’s Nerv Syst 1989; 5:177-83. 28 Milhorat TH, Hammock MK, Chandra RS. The subarachnoid space in congenital obstructive hydrocephalus. 2. Microscopic findings. J Neurosurg 1971; 35~7-15. 29 Matson DD. Neurosurgery of infancy and childhood, ed. 2, Charles C Thomas, Springfield, Ill., 1969; 259-68. 30 Marinov M, Lesev M, Undjian S. Computer tomographic ventriculography in congenital intracranial cysts in childhood. Rentg i radio1 (Sofia) 1985; XXIV:3 1-37. 31 Cartery A, Gerosa M, Gaini S, Villani R. The dysraphic state of the posterior fossa: clinical review of the Dandy-Walker syndrome and the so-called arachnoid cyst. J Neurosurg Sci (Ita) 1979; 23:43-59. 32 Haller JS, Wolpert SM, Rabe EF, Hills JR. Cystic lesions of the posterior fossa in infants: a comparison of the clinical, radiological and pathological findings in Dandy-Walker syndrome and extraaxial cysts. Neurology 1971; 21~494-506. 33 Palma L, di Lorenzo N, Nicole S.Developmental CSF cysts of the posterior fossa. Analysis of 20 operated cases. Neurosurg Rev 1979; 2:159-69. 34 Serlo W, Wendt L, Heikkinen E, Saukkonen A-L, Nystrom St. Shunting procedures in the management of intracranial cerebro-spinal fluid cysts in infancy and childhood. Acta Neurochir (Wien), 1985; 76:lll-16. 35 Udvarhelyi GB, Epstein MH. The so-called DandyWalker syndrome: analysis of 12 operated cases. Child’s Brain 1975; 1:158-82. 36 Edwards MS, Raffel C. Discussion of articles by Maria et d. and Golden et d. Pediat Neurosci 1987; 13:50-5 1. 37 Golden JA, Rorke LB, Bruce DA. Dandy-Walker syndrome and associated anomalies. Pediat Neurosci 1987; 13:38-44. 38 D’Agostino AN, Kernohan JW, Brown JR. The Dandy-Walker syndrome. J Neuropath exp Neurol 1963; 22:450-70. 39 Tal Y, Freigang B, Dunn HG, Durity FA, Mayers PD. Dandy-Walker syndrome: analysis of 21 cases. Dev Med Child Neurol 1980; 22189-201. 40 Olson GS, Halpe DCE, Kaplan AM, Spataro J. Dandv-Walker malformation and associated cardiac anomalies. Child‘s Brain 1981; 8:173-80.
ORIGINAL ARTICLE
The Dandy-Walker syndrome: diagnostic and surgical considerations M. MARINOV, S. GABROVSKY & S . UNDJIAN
Br J Neurosurg Downloaded from informahealthcare.com by UB Kiel on 11/08/14 For personal use only.
Department of Neurosurgery, Medical Academy, G. Sofijsky-Str. 1, 1431 Sofia, Bulgaria
Abstract Thirteen cases of the Dandy-Walker syndrome were seen over a period of 12 years. In this study their preoperative evaluation, surgical treatment and outcome is reviewed. The incidence of hydrocephalus in the series was 77%. Eight of the patients had associated central nervous system anomalies, and the need to identify these abnormalities is stressed. Because the two conditions require different surgical management, during the diagnostic work-up it is crucial to differentiate between the Dandy-Walker syndrome and extra axial cerebrospinal fluid (CSF) filled cysts, and to establish the patency of the aqueduct. The reliability of different diagnostic methods is discussed and the value of metrizamide computed tomographic ventriculography is emphasized. Unsatisfactory results following membrane excision in early cases, and observations of CSF dynamic disturbances confirm the futility of the direct approach for the management of the Dandy-Walker syndrome. The aqueduct was invariably patent and allowed either primary cyst or ventricular shunting, although simultaneous double shunting using a ‘Y’ connector may need to be carried out later. The mortality rate for the series was 18%. At follow-up 54.5% of the children were judged to have normal IQs.
Key words: Dandy- Walker syndrome, posterior fossa extra-axial cyst, metrizamide CT ventriculography, CSF pow disturbances, CSF shunting, outcome.
Introduction The first clinical description of the DandyWalker syndrome was published by Dandy and Blackfan in 1914.’ Further classical observations were made during the next decades in a series of publications by Dandy in 1921; Taggart and Walker in 1942: and Walker in 1944.4 In 1954, Bendas named the condition the ‘Dandy-Walker syndrome’. A number of features characterize the syndrome, the most important of these being cystic dilatation of the IVth ventricle, inconsistent patency of its outlet foramina, dysgenesis of the inferior vermis, and a high incidence of hydrocephalus. Atresia or occlusion of the foramina, which was formerly thought to be the cause of the cyst, is present in only half of
the cases.6 There is no correlation between the degree of hydrocephalus, the size of the expanded IVth ventricle, the extent of the vermian dysgenesis and the patency of the outlets of the IVth ventricle. As well as associated brain anomalies there are also systemic malformations which favour the view that this complex syndrome is the result of developmental failure of both the CNS and other organ^.^ Many theories have been published but none fully explain the whole phenomenon. Because of the relative rarity of the Dandy-Walker syndrome and lack of knowledge about how it develops, management is still debated. Evidence has accumulated that a direct posterior fossa operation is not effective.’-I3 The alternative is to carry out a
475
476
M. Marinov, S. Gabrovsky & S. Undjian
Br J Neurosurg Downloaded from informahealthcare.com by UB Kiel on 11/08/14 For personal use only.
shunting procedure, but some aspects of this remain controversial and require further investigation. In 1978 one of us (S.G.) reported the first cases of the Dandy-Walker syndrome diagnosed in our institute.I4 During the following years we saw an increasing number of children with Dandy-Walker cysts, and we present our experience based on this relatively large series followed over a period of 12 years. Clinical management and preoperative evaluation are emphasized and the principles of surgical treatment are discussed.
Clinical material and methods During the 12-year period of 1975-86, 13 patients with the Dandy-Walker syndrome were seen by the Paediatric Division of the Department of Neurosurgery of the Medical Academy, Sofia (Table I). Their records were reviewed to ascertain the presentation, diagnostic work-up, and treatment of each case. A standard preoperative intelligence quotient was performed by a clinical psychologist using the Stanford Binet Intelligence Scale or the Wechsler Intelligence Scale for children. Cerebral angiography was the most common investigation, being performed on 12 occasions. Ten C T scans were carried out, twice with ventriculography. Three patients had pneumoencephalography and two had direct cystography. Radionuclide dynamic CSF investigations (one "'InCaDTPA cysternogram and three ventriculograms with 9 9 n T pertech~ netate) were performed in four children. Up to 1977 the preferred method of investigation was a combination of angiography and air studies (three cases), between 1977 and 1983 C T scans were combined with angiography and radionuclide ventriculography (six cast.s) ,and after 1983 (Metrizamide C T ventriculography was the favoured method of investigation (four cases), especially if the ventricles were enlarged. Patient 13 was not operated on because the intracranial pressure was not raised, but the patient was kept under observation. Another patient only had a cranio-cervical meningocele
repaired. The operative procedures carried out in the remaining 11 children are outlined in Table 11. After 1980 the direct approach was abandoned, and only shunting procedures were performed. One patient died in the immediate postoperative period and another was lost to follow-up. The remaining children were observed for periods ranging from 7 months to 10 years (average 4 years and 5 months).
Results More than half the cases were seen before they were a year old. There was a trend towards earlier diagnosis in the later years of the study. There was a female to male predominance of 1.6:l. Table I shows that most patients presented with non-specific signs and symptoms of intracranial hypertension, and the striking absence of abnormal cerebellar and brain-stem signs. The IQs of the children on admission were subnormal or borderline. Two children, aged 2 and 3 months, had normal head circumferences. Patient 10 had a large head but a soft anterior fontanelle and no ventricular dilatation at diagnosis. There was questionable ventricular dilatation in patient 12. If the latter case is excluded, the incidence of hydrocephalus in the series is 77%. Occipital bossing, although considered to be a characteristic of a Dandy-Walker cyst, was only seen in half the cases. Associated anomalies, often multiple, were recorded in 61.5% of the patients. The only child with severe anomalies (patient 3) died the day after membrane excision. Two important questions need to be answered during the preoperative work-up: (1) What is the character of the posterior fossa cyst? Is it an intra- or extra-axial lesion? (2) What is the dynamic relationship between the fluid in the dilated IVth ventricle and the CSF pathways? Every effort should be made to obtain answers to these questions in order to plan the appropriate surgical procedure. Comparison of the angiographic findings of 12 children with Dandy-Walker cysts with those of 10 cases of posterior fossa arathnoid cysts, showed that this was a relatively unrelia-
~~
3 mo
4
3 mo F 4 mo F 2 mo M
~
~
Enlarged head, occipital bossing, gait disturbances Enlarged head, occipital bossing, delayed milestones, irritability, hypotonia Enlarged head, irritability, vomiting, bulging anterior fontanel Enlarged head, occipital bossing, febrile convulsions, visual impairment Enlarged head, vomiting bulging anterior fontanel Enlarged head, soft anterior fontanel, occipital bossing, delayed milestones Enlarged head, bulging fontanel, irritability, vomiting Enlarged head, soft fontanel, occipital bossing Sunken fontanel, seizures, delayed milestones Generalized hypotenia, IQ = 55 Hyperactive reflexes, IQ = 45
Increased ICP, IQ = 70
IQ = 22
Hyperactive reflexes,
Increased ICP, mild right hemiparesis, right amaurosis, IQ = 58 Increased ICP, IQ = 64
Increased ICP, IQ = 68
Increased ICP, mild spastic quadriparesis, ataxia, IQ = 70 Increased ICP, cerebellar signs, IQ = 44
IQ = 50
Suboccipital mass
Enlarged head, delayed milestones
Increased ICP, spastic quadriparesis; IQ = 37 Right spastic hemiparesis, right abducens paresis, severe psychomotor retardation Hyperactive reflexes, IQ = 48
Neurological findings; IQ score
Enlarged head, occipital bossing, delayed milestones Enlarged head, seizures, delayed milestones
Presenting symptoms
F, female; M, male; ICP, intracranial pressure; AVM, arteriovenous malformation.
13
12
11
F
3 mo
M
8 mo
9
10
1y3mo F
F
8 mo
M
3Y F ly2mo
8
7
6
5
F
1y4mo M
3
2
2y2mo F 3y9mo M
Age at diagnosis; sex
1
Case No.
TABLE I. Clinical summary in 13 patients with Dandy-Walker syndrome
Agenesis of the corpus callosum, plagiccephaly
-
Hemangioma faciei
Hypertelorism, bilateral syndactily
Hemangioma faciei, AVM a. vertebralis right/pars extracranialis), ganglioneuromas reg. colli
Spina bifida occulta C4-C5
-
Cardiac anomaly (ventricular septa1 defect), bilateral hydronephrosis and megaureter, Klippel-Feil syndrome, chromosome abnormalities ( G 2 ,trisomia), bilateral inquinal hernia Meningocele at the craniocervical junction, Klippel-Feil syndrome
Agenesis of the corpus callosum
Associated anomalies
Br J Neurosurg Downloaded from informahealthcare.com by UB Kiel on 11/08/14 For personal use only.
Proven
Present
Present Present Not present
Present Present Present Present Present
Not present Present ?
1
2 3
5 6
7
10 11 12
Proven Proven Proven
Proven Proven Proven Proven Not proven
-
CAA CAA CAA
EM EM Closure MC EM and VAA EM VAA VAA CAA
CAA
Initial treatment
3Y 1Y
-
-
VAA
2y7mo 7y6mo 4Y 7y 10 Y Lost to follow-up 2Y
2Y
-
16 mo
VPA VAA VA and CAA (‘Y’) CAA CA and VAA (‘Y’)
Revision CAA and EM -
Further treatment
Follow-up period
Disabled Normal Normal
No change Died 24 h postoperation Normal Normal Normal Normal Normal
Died (meningoencephalitis)
Neurological
Outcome
Severe PMR Normal Moderate PMR
Mild PMR Normal Normal Normal Normal
No change
Intellectual
VA(P)A, ventriculoatrio(peritoneo)nastomosis;CAA, cystoatrioa anastomosis; ‘Y’, Y-connector; EM, excision of membranes; PMR, psychomotor retardation.
9
8
4
Not proven Not proven
Aqueduct patency
No.
Ventricular dilatation on admission
Case
TABLE 2. Applied surgical procedures and outcome in 12 cases with Dandy-Walker syndrome
Br J Neurosurg Downloaded from informahealthcare.com by UB Kiel on 11/08/14 For personal use only.
rp 4 W
Br J Neurosurg Downloaded from informahealthcare.com by UB Kiel on 11/08/14 For personal use only.
The Dandy- Walker syndrome ble method of differentiating between the two conditions. In the children with DandyWalker cysts the posterior inferior cerebellar artery (PICA) was well defined in two cases, hypoplastic in three, and not visualized in seven. The anterior inferior cerebellar artery (AICA) was identified in all except one case where the extremely hypoplastic cerebellum was supplied by only the superior cerebellar arteries (SCAs). A small cerebellum was visualized by a capillary and venous ‘blush’ in seven cases, and a large cerebellum by a ‘blush’ in one case. The torcula Herophili was high in seven cases but normal in position in two. In the patients with arachnoid cysts the posterior inferior cerebellar artery (PICA) was not visualized in one case, and it was hypoplastic in another two. A ‘blush’ was absent in two cases in whom the cerebellum was deformed and compressed in a forward and upward direction. In five patients the torcula was high. Neither the courses of the basilar artery, the posterior cerebral arteries and the SCA, nor the type of displacement of the great vein of Galen and the straight sinus were helpful in differentiating between the Dandy-Walker syndrome and arachnoid cysts. The only reliable criterion was the type of displacement undergone by the PICA (if it was visualized); in retrocerebellar arachnoid cysts the PICA was always pushed
479
anteriorly and superiorly, whereas the large encysted IVth ventricles of the Dandy-Walker complex displaced the vessel in a posteroinferior direction. Cisternography was performed in four patients but was of little help. The investigation showed a communication between the encysted IVth ventricle and the basal subarachnoid spaces in two cases (a patent aqueduct was seen in one of them); in the other two cases there was no communication between the IVth ventricle and the subarachnoid spaces. Altogether the aqueduct was demonstrated to be patent by ventriculography or cisternography in eight cases; in two of them Metrizamide C T dynamic studies were also carried out. C T scanning is indispensable for assessing the structures in the superior and inferior tentorial compartments, the degree of ventricular enlargement and the presence of associated CNS anomalies. Mid-sagittal CT scan reconstructions can demonstrate the difference between a classical Dandy-Walker cyst and a retrocerebellar cyst (Fig. 1). In recent years dynamic Metrizamide C T ventriculography has been found to be very useful to demonstrate posterior fossa arachnoid cysts and patency of the aqueduct (Figs. 2a, b and 3a, b). Early in the series membranectomy was
FIG.1 . Midsagittal CT reconstruction: (a) in Dandy-Walker cyst; note almost total absence of cerebellar vermis and the pronounced cystic transformation of IVth ventricle (intra-axial cyst); (b) in a retrocerebellar (extra-axial) CSF cyst with normal sized IVth ventricle; vermis cerebelli is present.
Br J Neurosurg Downloaded from informahealthcare.com by UB Kiel on 11/08/14 For personal use only.
480
M.Marirtov, S. Gabrovsky & S. Undjian
FIG. 2. CT scan in a case with huge retrocerebellar arachnoid cyst. (a) IVth ventricle is not visible because of massive displacement, which makes differentiation from Dandy-Walker cyst difficult. (b) Metrizamide CT ventriculography demonstrates an isolated retrocerebellar cyst, normal IVth ventricle (arrowhead) with attenuated, but preserved vermis above it. (c) Native C T scan after posterior fossa cyst excision.
performed to try to establish a free flow of CSF, and to prevent recurrence of the cyst. In patient 6 the procedure failed and a ventricular shunt had to be inserted 10 days later to control the intracranial pressure. Patient 3 died on the first postoperative day, probably due to brain-stem dislocation following resection of the cyst membranes (this patient also had severe associated anomalies). In patient 2 the direct approach did not affect the clinical outcome, and in patient 5 simultaneous membranectomy and ventricular shunting was carried out successfully, although a shunt revision had to be performed 5 years later. Seven patients underwent primary shunting procedures. In five a cysto-atrial shunt was performed, three of them were well at the time of follow-up, but shunt failure occurred in the other two and shunt revisions had to be carried out 9 months after insertion. Fenestration of the cyst was attempted at the time of shunt revision in patient 1; this caused CSF leakage, prolonged intracranial infection, and finally death of the child. In patient 10 a cyst shunt was converted to a ventricular one because of progressive secondary ventricular enlargement. In two children who had primary ventricular shunting the outcome assessed
after 6 and 6.5 years, respectively, was excellent, but then there was a sudden dramatic deterioration in their condition with signs of brain-stem compression shortly afterwards. In one of the children, as well as shunt malfunction, there was a lack of communication between the supra- and sub-tentorial CSF compartments which was confirmed during operation and it was probable that a secondary obstruction of the aqueduct had occurred. The tubing in the cyst was joined to a ventricular catheter by a ‘Y-connector. In the other child, in spite proven patency of the aqueduct, it was decided to convert the drainage system to a cysto-atrial shunt rather than carry out a revision. Disappearance of the cyst on follow-up C T scans was only seen in patient 11 (Fig. 3); in the remaining cases incomplete re-expansion of the cerebellum occurred. In summary, the most favourable results were seen with cysto-atrial and cystoatrial combined with ventriculo-atrial shunting. Excluding the patient who was lost to follow-up, the mortality rate for the series was 18.2% (2/11); six children (54.5%) had normal motor and intellectual developmental outcomes.
Br J Neurosurg Downloaded from informahealthcare.com by UB Kiel on 11/08/14 For personal use only.
The Dandy- Walker syndrome
481
FIG.3. CT investigations in patient 11: (a) native C T scan of the Dandy-Walker cyst before treatment; (b) patency of aqueductus of Sylvii was confirmed by means of metrizamide CT ventriculography; (c) total cerebellar re-expansion one year after cysto-atrial shunting with re-establishment of posterior fossa architecture.
onuclides were also found to be unhelpful in contradistinction to some report^.^^-^^ In a Several points concerning the treatment of previous study2' we came to the conclusion patients with the Dandy-Walker syndrome are that the Dandy-Walker syndrome and posterworthy of comment. The diagnosis of symptoior fossa arachnoid cysts are associated with matic patients is relatively easy. Current methods of neuro-imaging provide the means of CSF-flow disturbances. In both cases commumaking an early and precise diagnosi~.'~-'~ With nicating and occlusive hydrocephalus are preincreasing frequency prenatal ultrasonography sent. There is documented evidence that at is detecting the Dandy-Walker syndrome and least 50-82% of patients with the DandyWalker syndrome have patent IVth ventricular associated anomalies in the fetus.18J9 . ~ ~ . ~supports ~. the earlier During this study a variety of diagnostic f ~ r a m i n e . ~ . ~ This belief that the rhombencephalic anomaly intermethods were used, but some of them proved feres with the normal development of the to be of little value in confirming the diagnosis subarachnoid spaces. The observations of Milor helping to plan appropriate treatment. Contrary to earlier reportsZ0J1and in agree- horat et al. in 197128have shown, however, ment with the conclusions reached by Rai- that the subarachnoid spaces in the Dandymondi et al. in 1984,22we found that angiogra- Walker syndrome are well developed but phy is not a reliable method of distinguishing compressed by the raised intracranial pressure. between the Dandy-Walker syndrome and Lack of patency of the subarachnoid spaces posterior fossa arachnoid cysts. C T scanning provides an explanation for the unsatisfactory Metrihas also been shown to be of little help in results of cyst e x c i s i ~ n . ~ ~ ~ , ' distinguishing between the two conditions. It zamide C T ventriculography greatly improved has been suggested that pneumoencephalogra- the preoperative assessment of the patients.'O phy or injecting air into the cyst via the cisterns Like Maria et ul.,16 this investigation was magna is the best method of investigation, and found to be superior to any other method of it did confirm the diagnosis and the patency of neuroimaging. As well as demonstrating pathe aqueduct in the two children who were tency of the aqueduct, it also ruled out the investigated in this way in this series, but both presence of a posterior fossa extra-axial cyst. deteriorated immediately afterwards. Because Whilst some authors have ignored the diagnosof this and the unreliability of lumbar pneuno- tic and surgical peculiarities of different postencephalography, air studies are no longer erior fossa developmental CSF cyst^^'-^^ it is performed. Cisternographic studies using radi- important to know if an extra-axial cyst is
Discussion
1~13~16~22723929
Br J Neurosurg Downloaded from informahealthcare.com by UB Kiel on 11/08/14 For personal use only.
482
M.Marinow, S. Gabrowsky & S. Undjian
present since arachnoid cysts may benefit from membrane excision rather than cyst shunting. During recent years shunting has become the treatment of choice for the Dandy-Walker syndrome. If hydrocephalus is not present at the time of diagnosis, a cysto-peritoneal shunt is advocated.22There is controversy concerning which shunt to use if hydrocephalus is present: lateral ventricular,8Js ~ y s t i c > ’ ~and J~ simultaneous ‘Y’-shunting of the lateral ventricles and the cyst have been used. ‘Y’shunting was proposed by Raimondi et al. in 196911and has been favoured by other^^^,^^,^' to avoid progressive upward herniation of the cyst which can cause brain-stem compression and secondary aqueduct ,stenosis. A more flexible approach towards treatment has been advocated depending on the presence and severity of hydrocephalus,’ the age of the ~ a t i e n t and, ~ ~ . most ~ ~ important of all, on whether the aqueduct is patent.16 In contradistinction to the often-cited concept of ‘functional’ or true aqueduct steno~is,6.~IJ~J~ like Sawaya and McLaurin,” we found the aqueduct to be patent in all our cases. Based on these findings, our policy has been to shunt the lateral ventricles or the cyst rather than performing a primary double-shunt procedure. In o w opinion double shunting does not insure hydrodynamic balance between the supra- and sub-tentorial compartments. Moreover, occlusion of one or other of the proximal catheters may accentuate the pressure differences even if a single regulating pressure device is used, as recommended by Serlo et al.34 Although double shunting is not advocated as a primary procedure, it is not excluded for later use if shunt revision is necessary. There was a high shunt failure rate in our series; 75% of the primary shunts required revision. Late secondary aqueduct stenosis is seen after ventricular shunting, as occurred in patient 7. Cyst shunting has the advantage of encouraging the passage of CSF through the aqueduct in the normal direction of flow. It also leads to a marked reduction in the size of the posterior fossa cyst and may allow the normal architecture of the posterior fossa to be re-established as occurred in patient 11 (Fig. 3).
Because of the relatively small number of cases in our series, we have reservations about recommending posterior fossa shunting alone for the treatment of the Dandy-Walker syndrome. All the different shunting procedures were found to effectively control the intracranial hypertension and lead to decompression of the posterior fossa structures. No operative mortality occurred in this subgroup of patients, and four out of six had a favourable outcome, although one was moderately retarded. The mortality rate for the whole series was 18.2% which is considerably less than the 40-5046 recorded in earlier series,7y8JoJ and was comparable to more recently reported rate^.^.'^.^^ The intellectual outcome of 54.5% of our patients was normal which is similar to mean values calculated for different series by Hirsch et ~ 1 but . ~less good than the recent results reported by Maria et all6and Golden et a1.37 Associated anomalies, although frequently seen in our patients, were not severe except in one patient. There was no correlation between the presence of anomalies and the intellectual outcome. As stressed by Sawaya and McLaurin,” the factor having the greatest influence on outcome is an early and aggressive treatment approach to the control of the intracranial hypertension. Careful evaluation of extracranial anomalies such as congenital heart disease,4O is important from the point of view of reducing operative risks and postoperative complications. 1939
References 1 Dandy WE, Blackfan KD. Internal hydrocephalus. An
experimental, clinical and pathological study. Am J Dis Child 1914; 8:406-82. 2 Dandy WE. The diagnosis and treatment of hydrocephalus due to occlusions of the foramina of Magendie and Luschka. Surgery Gynec Obstet 1921; 112-24. 3 Taggart JK, Walker AE. Congenital atresia of the foramens of Luschka and Magendie. Arch Neurol Psychiat 1942; 48583-612. 4 Walker AE. A case of congenital atresia of the foramina of Luschka and Magendie. J Neuropathol exp Neurol 1944; 3~368-73.
5 Benda CE. The Dandy-Walker syndrome or the socalled atresia of foramen Magendie. J Neuropath exp Neurol 1954; 13:14-29.
Br J Neurosurg Downloaded from informahealthcare.com by UB Kiel on 11/08/14 For personal use only.
The Dandy- Walker syndrome 6 Hart MN, Malamud N, Ellis WG. The Dandy-Walker syndrome. A clinicopathological study based on 28 cases. Neurology 1972; 771-80. 7 Carmel PW, Antunes JL, Hilal SK, Gold AP. DandyWalker syndrome:clinicopathological features and reevaluation of modes of treatment. Surg Neurol 1977; 8~132-8. 8 Fischer GE. Dandy-Walker syndrome: an evaluation of surgical treatment. J Neurosurg 1973; 39:615-21. 9 Hirsch J-F, Pierre-Kahn A, Renier D, Saint-Rose C, Hoppe-Hirsch E. The Dandy-Walker malformation. A review of 40 cases. J Neurosurg 1984; 61:515-22. 10 James HE, Kaiser G, Schut L, Bruce DA. Problems of diagnosis and treatment in the Dandy-Walker syndrome. Child’s Brain 1979; 5:24-30. 11 Raimondi AJ, Samuelson G, Yarzagaray L, Northon T. Atresia of the foramina of Luschka and Magendie. The Dandy-Walker cyst. J Neurosurg 1969; 31:202-16. 12 Raimondi AJ. Cystic transformation of the IV ventricle (the Dandy-Walker cyst). In: Hoffman HJ, Epstein F (eds). Disorders of the developing nervous system: diagnosis and treatment. Blackwell Sci Pub1 Boston; 1986; 235-46. 13 Sawaya R, McLaurin RL. Dandy-Walker syndrome. Clinical analysis of 23 cases. J Neurosurg 1981; 55939-98. 14 Gabrovsky S. Hydrocephalus in atresia of the foramina of Magendie and Luschka (Dandy-Walker syndrome). Neurol Psychiat i Neurochirurg (Sofia), 1978; XVII: 107-16. 15 Hough DRYMengel MB, Malik AH. Mega cisterna magna: diagnosis using metrizamide computed tomographic cisternography. Neurosurgery 1982; 11:260-2. 16 Maria BL, Zinreich SJ,Carson BC, Rosenbaum AE, Freeman J M Dandy-Walker syndrome revisited. Pediat Neurosci 1987; 1345-51. 17 Masdeu JC, Dobben GD, Azar-Kia B. Dandy-Walker syndrome studied by computed tomography and pneumoencephalography. Radiology 1983; 147:109-14. 18 Fileni A, Colosimo C, Mirk P, De Gaetano AM, di Rocco C. Dandy-Walker syndrome: diagnosis in utero by means of ultrasound and C T correlations. Neuroradiology 1983; 24233-5. 19 Newman GC, Buschi AI, Sugg NK, Kelly TE, Miller JQ. Dandy-Walker syndrome diagnosed in utero by ultrasonography. Neurology 1982; 32:180-4. 20 La Torre E, Formna A, Occhipinti E. Angriographic differentiation between Dandy-Walker cyst and arachnoid cyst of the posterior fossa in newborn infants and children. J Neurosurg 1973; 38:298-308. 21 Wolpen SM, Haller JS, Rabe EF. The value of angiography in the Dandy-Walker syndrome and posterior fossa extra-axial cysts. Am J Roentg 1970; 109:261-72. 22 Raimondi AJ, Sato K, Shimoji T. The Dandy-Walker syndrome, S. Karger, Basel, 1984. 23 French BN. The Dandy-Walker malformation. In: Youmans TR (ed) Neuroloeical Suraerv. vol 3. W.B. Saunders, Phiiadelphia, 198:; 1236-i380.
483
24 Glasauer FE. Isotope cisternography and ventriculography in congenital CNS anomalies. J Neurosurg 1975; 43:18-26. 25 Harbert JC, McCullough DC. Radionuclide studies in an unusual case of Dandy-Walker cyst. Radiology 1971; 101:363-6. 26 Palma A, Nazar N, Castro M y Olea E, Guzman G. Dandy-Walker malformation and the contribution of radioisotopic studies of cerebral spinal fluid dynamics to its diagnosis. Acta Neurochir (Wien) 1982; 61~319-24. 27 Marinov M, Undjian S, Wetzka P. An evaluation of the surgical treatment of intracranial arachnoid cysts in children. Child’s Nerv Syst 1989; 5:177-83. 28 Milhorat TH, Hammock MK, Chandra RS. The subarachnoid space in congenital obstructive hydrocephalus. 2. Microscopic findings. J Neurosurg 1971; 35~7-15. 29 Matson DD. Neurosurgery of infancy and childhood, ed. 2, Charles C Thomas, Springfield, Ill., 1969; 259-68. 30 Marinov M, Lesev M, Undjian S. Computer tomographic ventriculography in congenital intracranial cysts in childhood. Rentg i radio1 (Sofia) 1985; XXIV:3 1-37. 31 Cartery A, Gerosa M, Gaini S, Villani R. The dysraphic state of the posterior fossa: clinical review of the Dandy-Walker syndrome and the so-called arachnoid cyst. J Neurosurg Sci (Ita) 1979; 23:43-59. 32 Haller JS, Wolpert SM, Rabe EF, Hills JR. Cystic lesions of the posterior fossa in infants: a comparison of the clinical, radiological and pathological findings in Dandy-Walker syndrome and extraaxial cysts. Neurology 1971; 21~494-506. 33 Palma L, di Lorenzo N, Nicole S.Developmental CSF cysts of the posterior fossa. Analysis of 20 operated cases. Neurosurg Rev 1979; 2:159-69. 34 Serlo W, Wendt L, Heikkinen E, Saukkonen A-L, Nystrom St. Shunting procedures in the management of intracranial cerebro-spinal fluid cysts in infancy and childhood. Acta Neurochir (Wien), 1985; 76:lll-16. 35 Udvarhelyi GB, Epstein MH. The so-called DandyWalker syndrome: analysis of 12 operated cases. Child’s Brain 1975; 1:158-82. 36 Edwards MS, Raffel C. Discussion of articles by Maria et d. and Golden et d. Pediat Neurosci 1987; 13:50-5 1. 37 Golden JA, Rorke LB, Bruce DA. Dandy-Walker syndrome and associated anomalies. Pediat Neurosci 1987; 13:38-44. 38 D’Agostino AN, Kernohan JW, Brown JR. The Dandy-Walker syndrome. J Neuropath exp Neurol 1963; 22:450-70. 39 Tal Y, Freigang B, Dunn HG, Durity FA, Mayers PD. Dandy-Walker syndrome: analysis of 21 cases. Dev Med Child Neurol 1980; 22189-201. 40 Olson GS, Halpe DCE, Kaplan AM, Spataro J. Dandv-Walker malformation and associated cardiac anomalies. Child‘s Brain 1981; 8:173-80.
