Colloid cyst of the third ventricle:

Dr.Ramamurthi Ravi,
Neurosurgeon, V.H.S.  Medical  Centre,  Chennai , India.

Approximately 0.5-1% of all primary brain tumors and 15-20% of all intraventricular masses are colloid cysts.

They may cause sudden death or longstanding symptoms from obstructive hydrocephalus.They are still associated with considerable morbidity and mortality. 

In 1858, Wallman first reported colloid cyst. Dandy accomplished the first successful resection of a colloid cyst in 1921.


Although these tumors are considered congenital, their presentation in childhood is rare. They usually present in the middle age.  

The origin of these cysts continues to be uncertain. Remnants of paraphysis, diencephalic ependyma, invagination of neuroepithelium of the ventricle, or the respiratory epithelium of endodermal origin are other etiologic possibilities.  

Colloid cysts are thought to enlarge through increases in their contents. This process is postulated to occur in several ways. The epithelial lining of the cell wall secretes a mucinous fluid. In addition, cyst cavities filled with blood degradation products and cholesterol crystals have been reported.  

Colloid cysts are slowly growing lesions, probably of endodermal origin, and usually founding the anterior third ventricle close to the foramen of Monro, but other locations, such as, the roof of the third ventricle, the columns of the fornix, or the choroid plexus, are possible.

Their fibrous walls are lined with simple or pseudostratified epithelial cells. Their shape is either flattened cuboidal or low columnar, and they rest on a thin capsule of collagen and fibroblasts. The cysts are mucin secreting and ciliated.


Colloid cyst (H&E):  cyst wall is lined by cuboidal to columnar epithelium(arrow), supported by delicate collagenous stroma

The Cells are periodic acid-Schiff (PAS) positive and stain positively for S100 and negatively for glial fibrillary acidic protein (GFAP), vimentin, and neurofilament. The stromal wall stains positively for vimentin. Contents of the cyst are usually greenish and of variable viscosity.


Clinical features:


Symptoms are usually caused by constant or intermittent hydrocephalus.  Headache related to position of the head and sudden drop attacks are typical.  Colloid cysts have been the cause of sudden death due to obstruction of CSF flow or hypothalamic disturbance of cardiovascular control. 

The majority of colloid cysts are detected in the work-up of milder symptoms: Headaches (68%), Gait disturbances (47%), Disturbed mention (37%), Nausea (37%), Blurred vision (24%), Incontinence (13%), Tinnitus (13%), Seizures (10%), Acute deterioration (10%), Diplopia (8%) 

Signs detected include: papilloedema (47%), gait disturbance, hyperreflexia, positive babinski sign (21-32%), 24% have a normal examination.

With the advent of MR and CT imaging an increasing number of colloid cysts will be incidental findings.


CT scan demonstrates a usually hyperdense (iso-and hypodense are also possible), and may enhance with contrast.

MRI enables visualization the same typical features, But with better anatomical detail and with delineation of venous anatomy. 

Differential diagnosis includes other tumors in the region such as hamartomas, astrocytomas and ependymomas, and benign cysts of the choroid plexus.  The latter are adjacent to the plexus, usually in the lateral ventricles, hypodense and non-contrast enhancing on CT. 


  Colloid cyst-CT


Common modes of treatments are stereotactic aspiration, microsurgical extirpation and endoscopic fenestration of the cysts. 

For small (<0.5), asymptomatic cysts that do not cause hydro-cephalus can probably be followed without treatment, although data on possible deterioration are lacking. This is, however. controversial.  This approach does restrict the patient to life long follow up. 

Ventricular shunting:

Traditionally bilateral shunts have been advocated.  The rationale was an assumption that the cyst occluded both foramina of Monro.  It is, however, possible that CSF flow is actually interfered with in the posterior part of the ventricle.  A single shunt would thus suffice.  Shunting carries a risk of sudden deterioration if dysfunction occurs, and does not alleviate symptoms caused by pressure on the fornices or the hypothalamus.  Shunt revisions for infection or malfunction are common. 

Microsurgical excision:

Transcallosal approaches are feasible with normal ventricles.  The main risks are venous infarction from interference with bridging veins, and damage to the pericallosal arteries.  Traction on the gyrus cinguli may produce (usually transient) mutism.  Disconnection syndromes are not detectable following a small (<2.5 cm) callosotomy, which is made starting 1-2 cm behind the genu.  An operation can be carried out through a 1.5 cm callosotomy.  Special tests may, however, demonstrate minor deficits. 

The other approaches apart from the transcallosal that are available for lesions in this area are the transcortical, transventricular and the subfrontal.  The subfrontal approach is used for tumors that arise inferiorly and compress the third ventricle from below.  The transcortical approach involves doing a frontal craniotomy usually on the right side,  cortical incision down to the ventricle and then locating the foramen of Monroe.  The transcortical route cannot be used when the ventricles are normal or narrow and requires a lot of brain incision and retraction.  The foramen of Monroe on both sides cannot be visualized if necessary.

The transcallosal approach which was first employed by Dandy and later on by others is a direct midline approach to either or both lateral ventricles and the third ventricle.  There is no cortical incision and only retraction.  The small opening made in the anterior part of the corpus callosum (1.5 to 2.5 cms) does not cause any disconnection syndromes. Occasionally a frontal cortical vein may have to be sacrificed to get adequate retraction of the frontal lobe and this may lead to convulsions or to venous infarction of the frontal lobe.  The size of the ventricle is inconsequential in this approach which can be used after the patient has been shunted.

The steps of the operation described is the way the author does and is  most comfortable with. 

The patient is positioned supine with the head elevated 20 degrees.  Three pin fixation is not used, only a head ring is used. It should be made sure that the head is not  tilted to the left  or right to a  great extent.  A question mark skin flap is turned with the medial limb on the midline.  Two thirds of the medial limb is anterior to the coronal suture and one third is behind it.  The posterior end is curved downwards towards the zygoma for 7 to 8 centimetres. The other skin flaps that may be used are bicoronal and horseshoe.

A free bone flap is turned with the medial end on the midline to expose only the lateral edge of the superior sagittal sinus (SSS). If there is a small ridge of bone lateral to the sinus, this must be ronguered or a Kerrison punch can be used to remove the inner table.  This is necessary to avoid excessive retraction of the frontal lobe and the need to work under a ledge of bone.  Some surgeons recommend going across the midline and others do not. The advantage  of going across the midline is that a little more space may be available  The author feels that the space available is determined by the SSS rather than by the extent of bone removal.  The disadvantages are the potential injury to the SSS when the whole of it is exposed and the chance of pressure on the sinus during retraction. This pressure can lead to venous stasis and raised intracranial pressure during surgery or venous infarction of either frontal lobe.  The ledge of bone left over the SSS prevents this  pressure.

After applying the hitch stitches on the dura, the dura is opened as  a flap hinged towards the sinus.  The lateral extent of the dural opening should be only about 2 cms from the midline so that the retracted frontal lobe stays under the dura and does not get hitched against the cut dural margin.  In some, due to cortical veins entering the SSS, the entire anteroposterior extent of the dura cannot be opened.  The restriction is usually posterior and the anterior two thirds of the dura can be comfortably opened and this exposure is adequate. 3 to 4 cms of longitudinal exposure of the frontal lobe is all that is necessary for retraction.

In many patients one or more cortical veins will be seen in the area exposed and they will be entering the SSS.  As far as possible, the retraction should be done between these veins.  An extra 3 to 5 mms of the vein can be mobilized by dissecting the arachnoid around the vein in the cortex.  If absolutely essential the smallest vein may be sacrificed.  The vein should be anterior to the coronal suture.  In the authorís experience sacrifice of a vein has been seldom necessary.

On gentle retraction of the frontal lobe, the falx will be seen.  There may be adhesions between arachnoid granulations and the falx or sinus and these have to be released.   In patients, in whom the brain is tight and the ventricles are enlarged the right frontal horn can be tapped and CSF let out.  When the ventricles are normal in size 20% mannitol can be given in a dose of 5 ml per kg body weight.  It is advisable not to give mannitol when the ventricles are enlarged as at the end of surgery there will be excessive shrinkage of the brain  due to a large quantity of CSF being let out during surgery.

The frontal lobe is retracted initially using a one cm retractor.  The angle of the microscope has to be changed to 60 degrees to the right during this step till the falx is seen well.  The microscope angle is then changed to have direct vision of the interhemispheric fissure.  The arachnoid in the fissure is dissected and CSF is let out.  Further retraction is carried out only after letting out the CSF which leads to further relaxation of the brain.  A 2 cm retractor is now used.  In most cases the two frontal lobes are easily separable but in some sharp dissection may be required.  Sharp dissection lessens the chances of injury to the cortex.  A branch of the pericallosal artery may be seen coursing in the cortex on one or both sides.  The cingulate gyri may be densely adherent to each other and in some the fissure may be angled to the left or right.  It is extremely important to identify the cingulate gyri and not mistake them for the corpus callosum.

The two pericallosal arteries will now be seen.  The arachnoid between and around the arteries are dissected and the corpus callosum will come into view.  The pericallosal arteries can be displaced to one side or the dissection  can be carried  out between the two arteries.  This will depend on the anatomy seen in each individual patient and there is no hard and fast rule.  Rarely a single pericallosal artery may be seen.  Occasionally a small cortical branch may have to be sacrificed in order to mobilise the pericallosal arteries and get enough space to expose the corpus callosum.

The corpus callosum will be white in color with a few small arteries and veins coursing over it. The cingulate gyrus should not be mistaken for the corpus callosum as then the entry into the ventricle becomes difficult and confusing.  An incision is made in the corpus callosum for a distance of 1.5 to 2.5 cms, depending on the type and size of the pathology in the ventricle.  The callosum is relatively avascular and the incision is deepened the ependyma will come into view.  A few small veins may be seen coursing over  the ependyma and these can be coagulated.  Initially a small opening is made in the ependyma in order to let out the CSF slowly.   The CSF should not be rapidly let out especially in patients with dilated ventricles as this will collapse the brain and may lead to the formation of a subdural haematoma. The ependyma is then opened to the extent necessary.

The next step is to determine whether the left or the right lateral ventricle has been opened into.  This is determined by locating the choroid plexus and following it forwards. The choroid plexus as it is traced forwards curves medially and enters the foramen of Munro. The thalamostriate and septal veins will also be visible.  In third ventricular tumors it really does not matter which ventricle has been entered into.

The approach to the third ventricle depends on the size and location of the lesion and whether the lesion has enlarged the foramen of  Munroe.  When the foramen of Monroe is enlarged which is so in the majority of the cases that the author operated upon, the lesion can be removed by the transforaminal approach.  This is the ideal approach as there is no destruction of neural tissue or sacrifice of veins.  If in case, there is necessity to enlarge the foramen of Monroe and this is not a common occurrence, one anterior column of the fomix can be sacrificed and the foramen enlarged anteriorly.  This should be done preferable on the non dominant side.  The other way is to enlarge the foramen posteriorly by sacrificing the thalamostriate  vein.  In many instances, it is possible to enlarge the foramen posteriorly without sacrificing the vein.  The foramen of Monroe may appear narrow on first appearance and there may be a bulge posteriorly.  In these it is best to make a small incision in the area of bulge, decompress the lesion and then the foramen of Monroe will open out and the further dissection and removal of the tumor can be carried out through the foramen of Munro.  

corpus callosum and pericallosal art exposed


colloid cyst exposed


pre op CT

post op CT

The subchoroidal approach can be used in mid third ventricular tumors where the foramen of Monroe is not enlarge and there  is no obvious bulge.  In this approach, the choroid plexus is mobilized from the choroidal fissure.  This will require cauterization of the choroid plexus and mobilizing the branches of the medial posterior choroidal arteries.  A microdissector passed under the tela chloridea will expose the plane of  cleavage between the medial wall of the thalamus and the roof of the third ventricle. The internal cerebral vein in  continuity with the septal vein will dissect away from the ipsilateral dorsolateral thalamus and will remain suspended in strands of arachnoid of the velum interposium.  The thalamostriate vein should be sacrificed only when absolutely essential.

The interfomiceal approach is the other option available and should be used only rarely and in specific instances. The approach should be strictly in the midline with a midline callosotomy and dissecting between the two fornices.  The fornices are very delicate structures and can be damaged easily  leading to post operative problems.  The ideal lesion in which this can be used is where there is a direct upward extension of the tumor and the fornices are spread apart by the tumor.  The maximum neural complications occur in the interforniceal approach and most neurosurgeons do not use this approach except in rare instances.  The complications that can occur are memory disturbances and a state of mutism.

Complications: The complications that may  specifically occur with the transcallosal approach are (1) immediate post operative convulsions  especially if a cortical vein has been sacrificed.  This is not often seen.  (2) Disconnection syndromes are extremely rare in anterior callosal sections limited to 2.5 cms. (3) Transient lower limb weakness may occur if there has been some pressure on the pericallosal artery during retraction.  The other complications like acute hydrocephalus, transient mutism, memory disturbances and hypothalamic disturbances are related to the surgical procedure in the lateral or third ventricle and are not related to the transcallosal approach.

Stereotactic aspiration: 

This new technique was originally claimed to have lower morbidity than excision, but severe complications have been reported.  Initial success is achieved in approximately 50% of cysts. A 100% recurrence rate has, however, been reported following aspiration procedures.  This was not surpirising since vital epithelium capable of producing mucoid was left intact. 

Ventriculoscopic surgery: 

      before aspiration        after aspiration

Endoscopy is a minimally invasive means of operating in the ventricular system.  Improved instrumentation has allowed the use of flexible and non-flexible endoscopes for fenestration and aspiration of colloid cysts, and lately, excision. 

Endoscopy for aspiration remains a treatment with the drawbacks of simple aspiration.  Aspiration with a generous fenestration of the cyst was described recently. A fenestration should allow continuous emptying of the colloid produced and thereby avoiding recurrences. 

These methods are recent, and have not yet been available for long-term follow-up. 


Prognosis following successful microsurgical removal is excellent.  The risks of microsurgery depend on the skills of the surgeon.  Different results have been reported.

Patients treated with shunting carry a risk of deterioration when shunts malfunction and are at risk of shunt infection. 

Aspiration has an unacceptable recurrence rate, and patients treated with aspiration procedures need to be followed and re-operated when a recurrence appears. Significant morbidity from recurrence following aspiration has been reported.  

The natural history of colloid cysts is not well known.  An increasing number of cysts are incidental findings, and will be followed without surgical intervention.  The safety of conservative treatment and risks of deterioration remain to be established.   

Simple aspiration has been challenged lately while endoscopy is becoming more of a routine tool in many departments.  Its use for colloid cyst surgery appears to become established, but long term follow-up is necessary to evaluate its safety in fenestration procedures.





















































































from Peer Reviewed Resources only