Article Text

Download PDFPDF

Images In...
Ataxia, hypoplastic vermis and keyhole connection sign: is it the variant form of Dandy-Walker complex?
  1. Osama S M Amin
  1. Department of Neurology, Sulaimaniya General Teaching Hospital, Sulaimaniya City, Iraq
  1. Correspondence to Dr Osama S M Amin, dr.osama.amin{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


A 12-year-old girl has a progressive instability of stance and gait since the age of 2 years. She demonstrates mild mental subnormality and signs of cerebellar incoordination. A diagnosis of an ataxic form of cerebral palsy was made at the age of 3 years. When she was 9 years old, she started to develop occasional vomiting and headache which were ascribed to ‘hydrocephalus’. Accordingly, shunting surgery was performed (figure 1). Both headaches and vomiting remarkably improved; however, the cerebellar dysfunction remained. Her up-to-date brain MRI is shown (figure 2).

Figure 1

Axial non-contrast CT brain scan of the patient, which was done 3 years ago. Note the site of the shunt (yellow arrow), which was placed into the posterior fossa's cisterna magna (red arrow). The patient developed a few bouts of headache and vomiting at that time and her paediatrician referred her to a neurosurgeon, who suggested that hydrocephalus is the cause of these complaints and he did this shunt. We think that the paediatrician and the neurosurgeon had missed the diagnosis of Dandy-Walker complex-variant form in this patient and the dilated cisterna magna prompted them to think of hydrocephalus.

Figure 2

Axial T1-weighted brain MRI film at the level of the posterior fossa. Note the mildly enlarged posterior fossa, absence of the lower midline cerebellar vermis (white arrow), keyhole connection sign between the 4th ventricle and cisterna magna (marked by yellow dots) and the enlarged pre-pontine cistern (red arrow). The bony defect in the right occipital bone represents the site of the shunt (blue arrow). The brainstem per se looks normal. The patient has a variant form of Dandy-Walker complex. In general, her cerebellar signs and mild mental subnormality have not changed since early childhood; her gait and stance instability had been somewhat stable during the past few years.

The symptomatology and brain imaging do not categorise the girl as having Dandy-Walker malformation or Dandy-Walker mega cisterna magna; she fits the ‘variant form’ of Dandy-Walker complex.1 This form is very heterogeneous and results in a multitude of phenotypes.2 The brainstem per se and the lateral ventricles are normal and the posterior fossa is normal or typical mildly enlarged. The midline vermis is hypoplastic and the cisterna magna communicates, through an enlarged vallecula, with the posterior/inferior aspect of the fourth ventricle forming the so-called keyhole connection sign. Therefore, posterior fossa cystic spaces appear to be enlarged.3

Supratentorial malformations are uncommon and hydrocephalus is found in 25% of patients only. The typical ‘torcular-lambdoid inversion sign’ of Dandy-Walker malformation is not encountered in the variant form.

Patients with less severe symptoms and mild anomalies usually hold a good prognosis. The presence of severe malformations (including multiple intracranial and extracranial anomalies) portends a gloomy outlook.

Learning points

  • Dandy-Walker complex-variant form defines patients who do not fit the fully-fledged Dandy-Walker malformation or Dandy-Walker mega cisterna magna categories and represents a less severe form of this ciliopathy with a relatively good prognosis.

  • Characteristically, torcular-lambdoid inversion sign of Dandy-Walker malformation is not encountered in the variant form. The posterior fossa may be mildly enlarged and this, together with enlarged posterior fossa cystic spaces, may prompt the neurologist to think of hydrocephalus and unnecessary shunting surgery may ensue.

  • Coexistent intracranial and extracranial anomalies are uncommon but their presence portends a poor prognosis.


View Abstract


  • Competing interests None.

  • Patient consent Obtained.

  • Provenance and peer review Not commissioned; externally peer reviewed.