Search :     
1999-12-03-05 Answer of case of the week #14 © Jeanty

Answer to Case #14

December 3-17, 1999

Submitted and discussed by Philippe Jeanty, MD, PhD, Nashville, TN and Gianluigi Pilu, MD. Bologna, Italy

These are 9 images from 2 midtrimester exams (3 weeks apart) on a little baby boy. The spine was normal. The fetus was then referred elsewhere where he was considered normal. A CT at birth was also read as normal.


Clearly the images demonstrated a small degree of ventriculomegaly. As you know me already I rarely put the measurements on the images. I think it is much better to reason, estimate and, when needed, the graticules on the side are sufficient.

Another puzzling area is the posterior fossa.

1. The first six images do not demonstrate the vermis. Instead there is a great big cleft in between lateral lobes of the cerebellum.

2. Image 7 demonstrates a small amount of ventriculomegaly—which many found—but also notice the appearance of the cerebellum: the "buttock sign" (Well this one I will not explain). Actually to me it looks more like a dog’s head, with big eyes and the snout !

            3. Picture 8 was mild ventriculomegaly

            4. Picture 9 showed the cavum of the septi pellucidi, thus excluding dysgenesis of the corpus callosum.

We knew that the baby had a normal spine and I did not mentioned or showed any significant other anomaly in this baby.

Many thought that this fetus had dysgenesis of the corpus callosum, but the presence of the cavum of the septum pellucidum is against the diagnosis.

When I saw this fetus I could only come up with vermian agenesis. But many of you went further and recognized that this is a big clue to a rare diagnosis: Joubert syndrome.

An interesting report from the mother of this baby is also available. And it is important for those of us in the medical profession to have the opportunity to listen to what our patient say, their concern and unfortunately the lack of support that we provide.

As reported in the introduction a second opinion scan elsewhere was considered normal for the vermis. A CT done after birth was also read as normal. However, when the child was one year old, a geneticist put together the clinical findings of this baby and obtained a MRI:


Several axial views that demonstrate the absence of the vermis and the “molar tooth” sign (see below)


Coronal views with absence of the vermis and demonstration of the corpus callosum


Sagittal views with absence of the vermis and demonstration of the corpus callosum

Several also thought this could be Dandy-Walker or Dandy-Walker variant since these conditions indeed present with a vermian defect 


Dandy-Walker complex

The sonographic findings include:

  1. Mild enlargement of the cerebral lateral ventricles;
  2. A prominent fourth ventricle that seems to communicate with a slightly enlarged cisterna magna.

Although some  have suggested the possibility of agenesis of corpus callosum, this is ruled out (at least in its complete form) by the well documented presence of the cavum septi-pellucidi.

Dandy-Walker malformation (or Dandy-Walker syndrome) has been suggested by most participants.

The term Dandy-Walker syndrome was introduced to indicate the association of:

  1. ventriculomegaly of variable degree
  2. a large cisterna magna and
  3. a defect in the cerebellar vermis through which the cyst communicates with the fourth ventricle. [3]

In more recent years, different definitions have been suggested to indicate a group of posterior fossa abnormalities similar to the classic Dandy-Walker syndrome. Most of these definitions agree in including the following findings:

  1. cystic dilatation of the fourth ventricle;
  2. dysgenesis of the cerebellar vermis and
  3. a high position of the tentorium.

At present, the term Dandy-Walker complex is used to indicate a spectrum of anomalies of the posterior fossa that are classified by axial CT scans as it follows:

1.        classic Dandy-Walker malformation (enlarged posterior fossa, complete or partial agenesis of the cerebellar vermis, elevated tentorium);

2.        Dandy-Walker variant (variable hypoplasia of the cerebellar vermis with or without enlargement of the posterior fossa);

3.        mega-cisterna magna (enlarged cisterna magna with integrity of both cerebellar vermis and fourth ventricle). [4]

This classification has been challenged after the introduction of Magnetic Resonance. For a number of reasons, the axial scans traditionally employed in CT scans do not have the capability of assessing clearly the status of the cerebellar vermis and may both underestimate and overestimate the size of the defect. The excellent resolution of sagittal planes made possible by Magnetic Resonance has allowed to demonstrate that the classification based upon CT axial planes is inadequate to describe the anatomic derangement encountered in the Dandy-Walker complex. Some degree of vermian dysgenesis can be found in all cases, even with megacisterna magna, while classic Dandy Walker malformation and Dandy-Walker variant have so many similarities that a clear-cut distinction is often impossible. [5]

Although hydrocephalus has been classically considered to be an essential diagnostic element of this condition, more recent evidence suggests that even with classic Dandy-Walker ventriculomegaly it is not overtly present at birth in most patients, but it develops usually in the first months of life. [6]

Dandy-Walker malformation has an estimated prevalence of about 1:30,000 births, and is found in 4-12% of all cases of infantile hydrocephalus. [7]

The incidence of Dandy-Walker variant and mega-cisterna magna is unknown.

Genetic factors have a major role in the etiology of this condition. Dandy-Walker malformation may occur as a part of Mendelian disorders and chromosomal aberrations (Table 1).

Table 1. Abnormalities associated with Dandy-Walker malformation after Murray77






Warburg (AR)

Aase-Smith (AD)

Ruvalcaba syndrome (AD/X-linked)

Coffin-Siris (AR)

Oro-facio-digital syndrome type II (AR)

Meckel Gruber syndrome (AR)

Aicardi syndrome (X-linked dominant)

Ellis Van Creveld (AR)

Fraser cryptophtalmus (AR)




dup 5p

dup 8p

dup 8q

trisomy 9


dup 17q








congenital heart disease

neural tube defects

cleft lip/palate


Cornelia de Lange syndrome

Goldenhar syndrome

Kidney abnormalities

Facial hemangiomas

Klippel-Feil syndrome


AR, AD autosomal recessive, autosomal dominant

In the absence of a recognizable syndrome, a recurrence risk of 1-5% is suggested. In rare cases, the disease is inherited as an autosomal recessive trait. [8] Environmental factors, including viral infections, alcohol and diabetes, have also been suggested to play a role in the genesis of Dandy-Walker malformation, but the evidence in uncertain. Dandy-Walker malformation is frequently associated with other neural defects, mostly ventriculomegaly, other midline anomalies, such as agenesis of the corpus callosum and holoprosencephaly, cephaloceles. Other deformities include polycystic kidneys, cardiovascular defects and facial clefting. Postnatal studies indicate a frequency of associated malformation ranging between 50 and 70%.

The sonographic diagnosis of classic Dandy-Walker malformation is straightforward since midgestation, [9] and has been reported as early as 14 weeks by using vaginal sonography: [10] in the transcerebellar view, an enlarged cisterna magna is connected to the area of the fourth ventricle through a defect in the cerebellar vermis (Fig. 1). Borderline to overt ventriculomegaly and other neural and extra-neural malformations are frequently present.


Figure 1. Classic Dandy-Walker malformation demonstrated by postnatal MRI (left) and prenatal ultrasound. A large defect of the cerebellar vermis connects the fourth ventricle with a cystic cisterna magna. The sagittal view (right) confirms the connection between the cisterna magna and the area of the fourth ventricle. Only the superior portion of the cerebellar vermis can be demonstrated.

However, in cases with partial agenesis of the vermis, confined to the inferior portion, the traditional transcerebellar view can be unremarkable, and meticulous scanning may be required to identify the defect. Demonstration in a lower section that the inferior cerebellar vermis separates the fourth ventricle from the cisterna magna is required to rule out this condition (Figure 2).


Figure 2. Demonstration of the integrity of the cerebellar vermis with antenatal ultrasound. The transcerebellar view is obtained at the level of the superior vermis and does not rule the defects occurring in the inferior portion. A further scan, obtained caudally to the standard transcerebellar view is required to demonstrate the inferior vermis separating the fourth ventricle from the cisterna magna.

Criteria for a definitive diagnosis of mega-cisterna magna and Dandy-Walker variant in the fetus have not been firmly established thus far. The former condition should be suspected when the cisterna magna has a depth greater than 10 mm, [11] the latter when a thin communication is found between the  fourth ventricle and the cisterna magna (Figure 3). [12]


Figure 3. The Dandy-Walker complex. Left panel: a cisterna magna measuring 14 mm with an intact cerebellum: this is megacisterna magna. Right panel: a thin connection is found between the fourth ventricle and a large cisterna magna. This is Dandy-Walker variant.

Caution is warranted while making these diagnoses. In the early second trimester, the sonographic appearance of the normal cerebellar development can resemble pathology: the relatively large fourth ventricle and the incompletely formed inferior cerebellar vermis may give the false impression of a vermian defect. [13] , [14] It is therefore imprudent to make diagnosis of a defect in the vermis at this gestational age. A follow-up scan at 18 weeks or later is recommended. Even in the second and third trimester, however, a scanning angle too steep may create the impression of an excessive size of the cisterna magna and even of a vermian defect (Figure 4). [15] , [16]


Figure 4. Pseudo-Dandy-Walker variant. In this third trimester fetus, a scan obtained along plane 1 demonstrates a fourth ventricle and a cisterna magna of normal size, separated by a seemingly intact cerebellar vermis. A scan obtained with a steeper angulation and probably passing in-between the cerebellum and the brain stem creates the impression of a communication between the cisterna magna and the fourth ventricle, raising the suspicion of a Dandy-Walker variant. The infant was found to be normal by postnatal investigations.

The origin of this artifact is unclear. It has been speculated that it could be caused by fluid within the vallecula cerebelli that tends to expand slightly in the anterior aspect and has a very thin membranous roof, that is certainly impossible to visuzalize with the resolution granted even by current ultrasound. The juxtaposition of the vallecula with the adjacent cerebellar tonsils creates the impression of a connection between the fourth ventricle and the cisterna magna, similar to that described by CT studies of Dandy-Walker variant. Visualization of the posterior fossa in the median plane may be helpful in these cases, in that it allows the visualization of  the vermis in the sagittal plane. It is unlikely however that even this approach will allow an accurate identification of subtle dysgenesis of the cerebellum such as those described in postnatal MRI studies. Indeed, in our experience, the vast majority of fetuses with either an isolated enlargement of the cisterna magna or an image suggestive of a small vermian defect were found to be entirely normal at birth, both by neuroimaging and by clinical evaluation. On the other hand, despite meticolous multiplanar imaging, we have been unable to demonstrate any cerebellar defect in a fetus that had a borderline cisterna magna, that was found at birth to have Dandy-Walker variant and overt ventriculomegaly requiring a shunt.

On the basis of both the available evidence and our own experience, we believe that antenatal sonography allows a certain diagnosis only of the most severe anatomic varieties of the Dandy-Walker complex, those characterized by both an enlarged cisterna magna and a wide defect in the cerebellar vermis, and commonly referred to as classic Dandy-Walker malformation. The experience thus far is limited with less conspicuous varities. It is usually impossible to solve antenatally the doubt of either a large cisterna magna or a small inferior defect of the vermis.

Classic Dandy-Walker malformation is usually clinically manifest within the first year of life, with symptoms of hydrocephalus and/or other neurologic symptoms. Mortality rates as high as 24% have been reported in the first neurosurgical series, but because of advances in pediatric anesthesia and surgical techniques deaths have certainly become less common.  Intellectual development in survivors is controversial. Given the rarity of this condition, only limited series are available. Nevertheless, a subnormal intelligence is reported in 40 to 70% of cases.,75,76, [17]

The clinical significance of Dandy-Walker variant and mega-cisterna magna is uncertain. These conditions are frequently seen in association with neurologic compromise, but no clear-cut prognostic data exist. Prenatal series document a frequent association with other  malformations and/or chromosomal aberrations.80,81

When the classic form of Dandy-Walker malformation is found prior to viability, pregnancy termination can be offered to the parents. In continuing pregnancies, no modification of standard obstetric management exist. Cesarean delivery is indicated only if macrocrania is present. A careful search for associated malformation and postnatal follow-up seem indicated also in fetuses with either a suspicion of Dandy-Walker variant or with a cisterna magna depth greater than 10 mm.

In the present case, the diagnosis of a cerebellar abnormalities was favoured by the concomitance of a presumable cleft into the cerebellum with a prominent fourth ventricle and mild enlargement of the lateral ventricles. 


Joubert Syndrome


Marie Joubert described the syndrome in 1968 [18] -69 [19] . It is a rare developmental defect of the cerebellar vermis, with autosomal recessive inheritance. The phenotype is highly variable and may include episodic hyperpnea, abnormal eye movements, hypotonia, ataxia, developmental delay, and mental retardation



Prenatal diagnosis

The first ultrasonic diagnosis was made by Campbell in 1984 [20] . The following findings have been found in fetus with Joubert syndrome [21] , [22] :

  1. vermian agenesis
  2. abnormal cerebellar shape
  3. hydrocephalus
  4. renal anomalies
  5. nuchal lucency [23]

Interestingly there is another report of a fetus in which the vermis was considered abnormal, had a normal postnatal examination and the diagnosis was only made later22 .


The diagnosis is based on the clinical findings and supported by the pathognomonic “molar tooth” sign on MRI.

MRI findings:

The recently described “molar tooth” sign on axial sections results from a combination of midbrain, vermian, and superior cerebellar peduncle abnormalities [24] . Other findings include [25] , [26] :

  1. thinned optic tracts,
  2. enlarged temporal horns in the absence of hydrocephalus,
  3. high-signal of the cerebral periventricular white matter,
  4. abnormal signal in the decussation of the superior cerebellar peduncles,
  5. abnormal embryonic vessels associated with the dysplastic folia of the cerebellar hemispheres
  6. dilatation of the fourth ventricle with some appearing bat-wing shaped,
  7. elongation and stretching of the superior cerebellar peduncles,
  8. dysplasia of the vermis,
  9. widening of the foramen of Magendie and the posterior cistern

On postnatal diagnosis the following findings can be recognized [27] :

  1. molar tooth sign
  2. dysgenesis of the isthmic portion of the brain stem at the pontomesencephalic junction,
  3. abnormally thick superior cerebellar peduncles perpendicular to the brain stem,
  4. hypoplasia of the cerebellar vermis with enlargement of the 4th ventricle and rostral shift of the fastigium,
  5. sagittal vermis clefting
  6. mild prominence of the ventricles and subarachnoid spaces

Not all findings are present in all babies

Clinical findings:


  1. hypotonia
  2. truncal ataxia
  3. developmental delay
  4. tachy/hyperpnea/apnea
  5. abnormal eye movements
  6. tongue protrusion  

And less commonly [28] :

  1. seizures
  2. hemifacial spasms
  3. polydactyly 8%
  4. colobomas 4%
  5. renal cysts 2%
  6. soft tissue tumors of the tongue2%
  7. occipital meningocele [29]


Siblings from the same family and even monozygotic twins [30] with Joubert syndrome may present with phenotypes ranging from severely handicapped to minimally handicapped. The motor handicap varies from wheelchair bound to being able to walk and run and the mental handicap from severely retarded, nonverbal, and autistic to verbal.


A recent report (this week) suggests that the anomaly is genetically heterogeneous and that one locus maps to the telomeric region of chromosome 9q, close to the marker D9S158, with a multipoint LOD score of Z=+3.7 [31] .

Associated anomalies

The following cculomotor anomalies have been described [32] :

  1. decreased smooth pursuit gain
  2. hypometric volitional saccades
  3. optic nerve dysplasia
  4. severe visual loss
  5. oculomotor apraxia
  6. pendular nystagmus
  7. gaze-holding nystagmus
  8. pigmentary changes in the fundus
  9. decreased vestibulo-ocular reflexes

Cognitive functions:

A variety of deficits in cognition, verbal memory, visuomotor, motor, and language-related tasks are described as well as problems in temperament, hyperactivity, aggressiveness, and dependency [33] . The degree of developmental delay (DG: 30-85)41 in Joubert syndrome and the severity of gross central nervous system malformations appear independent [34] . In one case there was no mental retardation [35] and in another marked improvement appeared after some delay [36] .

Other anomalies like ectodermal dysplasia [37] , Gaucher disease [38] , multicystic kidney disease, hepatic fibrosis [39] , hypertension [40] have also been associated.


A decreased life span has been found in many patients [41] .

Autopsy findings

The following anomalies have been reported [42] , [43] :

  1. aplasia or agenesis of the cerebellar vermis
  2. fragmentation several brainstem nuclei (dentate nuclei, inferior olives, and basis pontis)
  3. dysplasia of structures at the pontomesencephalic junction and caudal medulla
  4. abnormal decussation of the superior cerebellar peduncles
  5. enlarged iter (rostral 4th ventricle)
  6. elongated tegmental nuclei (including the locus coeruleus)
  7. reduction of the neurons of the basis pontis and reticular formation
  8.  malformations of the medulla (hypoplasia of the inferior olivary nuclei, solitary nuclei and tracts, and the nucleus and spinal tracts of trigeminal nerve (cranial nerve V).
  9. dysplasia of the caudal medulla at the cervicomedullary junction (absence of a posterior median sulcus, neuronal swelling and axonal spheroids in the region of malformed nuclei gracilis and cuneatus, and absence of pyramidal decussation)

Thus aside from vermal agenesis, Joubert syndrome may be associated with malformation of multiple brainstem structures. This could explain the hyperpnea and oculomotor anomalies.

Differential diagnosis

Vermian cleft without Joubert [44] , Dandy-Walker, Dandy-Walker variant, Down’s syndrome [45]

Social support

There is a support group in the department of Marie Joubert and she can be contacted at : . Other resource include the Joubert Syndrome Resource and Joubert Syndrome at the National Institute of Neurological Disorders and Stroke Kris, the mom of this baby just started another list at .

In view of the few cases diagnosed there is not much support for parents of affected kids. It has been shown that the “parental burden depends more on the parents" coping skills and the level of family functioning rather than on the degree of the child"s impairment” [46] .


[1] Beware of the sonologist with a ruler: strict adhesion to “cut-off” measurements is rarely critical, and the overall appearance and judgment is more important

[2] Adamsbaum C, Moreau V, Bulteau C, Burstyn J, Lair Milan F, Kalifa G Vermian agenesis without posterior fossa cyst. Pediatr Radiol 1994;24(8):543-6

[3] Benda CE: The Dandy-Walker syndrome or the so-called atresia of the foramen Magendie. J Neuropathol Exp Neurol 13:14-27, 1954.

[4] Harwood Nash DC, Fitz CR: Neuroradiology in infants and children. Vol. 3 St Louis; Mosby, 1014-19, 1976.

[5] Barkovich AJ, Kjos BO, Normal D et al: Revised classification of the posterior fossa cysts and cystlike malformations based on the results of multiplanar MR imaging. AJNR 10:977-88, 1989.

[6] Hirsch JF, Pierre Kahn A, Reiner D, Sainte-Rose C, Hoppe-Hirsch E: The Dandy-Walker malformation. A review of 40 cases. J Neurosurg 61:515-22, 1984.

[7] Osenbach RK, Menezes AH: Diagnosis and management of the Dandy-Walker malformation: 30 years of experience. Pediatr Neurosurg 18:179-85, 1991

[8] Murray JC, Johnson JA, Bird TD: Dandy-Walker malformation: etiologic heterogeneity and empiric recurrence risk. Clin Genet 28:272, 1985

[9] Pilu G, Goldstein I, Reece EA, et al: Sonography of fetal Dandy-Walker malformation: a reappraisal. Ultrasound Obstet Gynecol 1992; 2:151

[10] Achiron R, Achiron A: Transvaginal ultrasonic assessment of the early fetal brain. Ultrasound Obstet Gynecol 1991; 1:336.

[11] Nyberg DA, Mahony BS, Hegge FN et al: Enlarged cisterna magna and the Dandy-Walker malformation: factors associated with chromosome abnormalities. Obstet Gynecol 77:436-42, 1991.

[12] Estroff JA, Scott MR. Benacerraf BR: Dandy-Walker variant: prenatal sonographic diagnosis and clinical outcome. Radiology 185:755-58, 1992.

[13] Bromley B, Nadel AS, Pauker S, Estroff JA, Benacerraf BR: Closure of the cerebellar vermis: evaluation with second trimester US. Radiology 193:761-3, 1994.

[14] Babcock CJ, Chong BW, Salamat MS, Ellis WG, Goldstein RB: Sonographic anatomy of the developing cerebellum: normal embryology can resemble pathology. AJR 166:427-33, 1996.

[15] Mahony BS, Callen PW, Filly RA, Hoddick WK: The fetal cisterna magna. Radiology 153:773-6, 1984.

[16] Laing FC, Frates MC, Brown DL, Benson CB, Di Salvo DN, Doubilet PM: Sonography of the fetal posterior fossa: false appearance of mega-cisterna magna and Dandy-Walker variant. Radiology 192:247-51, 1994.

[17] Sawaya R, McLaurin RL: Dandy-Walker syndrome: clinical analysis of 23 cases. J Neurosurg 1981, 55:89.

[18] Joubert M, Eisenring JJ, Andermann F Familial dysgenesis of the vermis: a syndrome of hyperventilation, abnormal eye movements and retardation. Neurology 1968 Mar;18(3):302-3

[19] Joubert M, Eisenring JJ, Robb JP, Andermann F Familial agenesis of the cerebellar vermis. A syndrome of episodic hyperpnea, abnormal eye movements, ataxia, and retardation. Neurology 1969 Sep;19(9):813-25

[20] Campbell S, Tsannatos C, Pearce JM The prenatal diagnosis of Joubert"s syndrome of familial agenesis of the cerebellar vermis. Prenat Diagn 1984 Sep-Oct;4(5):391-5

[21] Anderson JS, Gorey MT, Pasternak JF, Trommer BL Joubert"s syndrome and prenatal hydrocephalus. Pediatr Neurol 1999 May;20(5):403-5

[22] Ni Scanaill S, Crowley P, Hogan M, Stuart B Abnormal prenatal sonographic findings in the posterior cranial fossa: a case of Joubert"s syndrome. Ultrasound Obstet Gynecol 1999 Jan;13(1):71-4

[23] Reynders CS, Pauker SP, Benacerraf BR First trimester isolated fetal nuchal lucency: significance and outcome. J Ultrasound Med 1997 Feb;16(2):101-5

[24] Maria BL, Hoang KB, Tusa RJ, Mancuso AA, Hamed LM, Quisling RG, Hove MT, Fennell EB, Booth-Jones M, Ringdahl DM, Yachnis AT, Creel G, Frerking B "Joubert syndrome" revisited: key ocular motor signs with magnetic resonance imaging correlation. J Child Neurol 1997 Oct;12(7):423-30

[25] Sener RN MR imaging of Joubert"s syndrome. Comput Med Imaging Graph 1995 Nov-Dec;19(6):481-6

[26] Shen WC, Shian WJ, Chen CC, Chi CS, Lee SK, Lee KR MRI of Joubert"s syndrome. Eur J Radiol 1994 Feb;18(1):30-3

[27] Quisling RG, Barkovich AJ, Maria BL Magnetic resonance imaging features and classification of central nervous system malformations in Joubert syndrome. J Child Neurol 1999 Oct;14(10):628-35

[28] Pellegrino JE, Lensch MW, Muenke M, Chance PF Clinical and molecular analysis in Joubert syndrome. Am J Med Genet 1997 Oct 3;72(1):59-62

[29] Suzuki T, Hakozaki M, Kubo N, Kuroda K, Ogawa A A case of cranial meningocele associated with Joubert syndrome. Childs Nerv Syst 1996 May;12(5):280-2

Back to case
Help Support :