An eloquent area ganglioglioma

October 2012

Presenting complaint

The patient is a 3 year boy who presented with a complex partial seizure.


The boy was born by normal delivery at term. Development has been normal and the parents have had no concerns thus far.

Current symptoms

Apart from this presenting seizure, the child is asymptomatic.

Clinical examination

General and neurological examination were normal.


The most relevant pictures are included below. In summary, there is a T1 hypointense lesion in the posterior-superior temporal lobe on the left that does not enhance with contrast. It is T2 hyperintense and vessels can be seen coursing through it.

Management to date:

The child was initially sent to a neurosurgeon elsewhere, who performed a biopsy of the lesion (operations notes not available). The histology of the biopsy sample is shown below:

Neoplasm with delicate vascularity.

Neuronal (arrows) and giant cells with no atypia or necrosis.

GFAP-staining glial component

Synaptophysin-staining neuronal component (arrows)

Neurofilament staining neuronal component (arrows) and abundant neurofilaments.

S100 was positive and the Ki67 showed a low proliferation index.

Histological diagnosis

I am indebted to our pathologist Dr Komala Pillay for the above pictures. Her diagnosis is that of a low grade glioneuronal tumour, favoring a ganglioglioma.


So, having established that this 3 year child with normal development and examination has a ganglioglioma in the posterior superior temporal lobe on the left, where to from here?

There are several concerns of course, including the location in the posterior superior left temporal lobe, inability to get a functional MRI to map speech in such a young child, and vessels coursing through the lesion (this is not a partial volume effect). What would be the acceptable morbidity of a surgical approach? Adjuvant therapy? No surgery at all, watch and wait? If so, what happens when it grows?

Let’s try to stimulate some discussion through this portal. If you have any comments (keep them brief if possible), please send them with your name to Tony Figaji who leads the ‘Case of the Month’ team. We will try to put up comments (with editing where needed) on this page.

Comments from some colleagues (edited):

Rick Abbott says: Very interesting case that definitely tests the surgeons confidence in neurodevelopment.  The histology of the tumor is very benign supporting its being present for an extended time.  Typically tumors of this type in the very young are relatively dominant which would support its having being present at nearly this size during the time of early speech development.  Both the presumed normal development of speech and the chronicity of the tumor would support there not being eloquent parenchyma being infiltrated by this tumor.  Finally, the child is at an age where the brain is very tolerant of aggressive surgeries around eloquent tissue and the tumor comes to the surface of a sulcus and the Rolandic fissure giving a good avenue into it.

Because of all these factors I would favor resection of the tumor.  I would not consider preoperative functional assessment of the area nor intraoperative speech mapping given the child’s age.  I would consider using both computer assisted intraoperative guidance and intraoperative imaging such as the ultrasound to minimize injury to normal brain parenchyma immediately adjacent to the tumor.

Shlomi Constantini says: It is indeed a challenging case. For this specific lesion, in such a location, I would personally wait for either intractability or an objective demonstration (MR) of tumor growth. For such a low grade pathology (GG), in this age, there is no concern about malignant transformation. Due to risk for speech and the diffuse nature of the lesion, morbidity can be significant. At an older age, FMR, tractography, and awake craniotomy technique may reduce this risk.

If intractability or tumor growth happen, both parents and surgeon may be more accepting potential morbidity. In such a case, I believe this lesion can be resected. I would personally like to use SonoWand (Mison) ultrasound guidance, or an intra-operative MR with In Operative ECORG.

John Kestle says: Interesting case. I think Id drag my feet on this one. The best treatment is resection but the location is a problem. I’d consider surgery if it grows, develops enhancement or if the child develops intractable epilepsy. Maybe that would take long enough that he’d be old enough to do some functional imaging or even an awake procedure.

Rick Boop says: This is a reference for a paper we published recently demonstrating the ability to map Wernicke’s area using MEG in sedated children as young as 18 months of age.  The case you present is a confluent tumor in an eloquent area but it is a confluent tumor which can be safely resected as long as the en passsage blood vessels are not violated.  Attached is a powerpoint of two older children with similar seizures and tumors in the same location.  The first is a 14 year old with a DNET we resected after Grid mapping co-registered to MEG.  The second is a 16 year old with a gangioglioma resected using fMRI guidance.  Both kids are seizure free with no language deficits.  In both we were able to spare visual tracts and avoid a field cut.

Passive language mapping with magnetoencephalography in pediatric epilepsy patients.  Clinical article. Van Poppel MD, Wheless JW, Clarke DF, McGregor A, McManis M, Perkins FF, Van Poppel K, Fulton, S,Boop FA. J Neurosurg Pediatrics, 2012 Aug;10(2):96-102.

William Harkness says: From an oncological point of view prognosis is uncertain but generally in a three year old with a lifetime ahead I think the treatment of choice would be excision. Clearly this would have to be discussed with the family and the option of a wait and see policy given. However, on balance I would be advising for early intervention.

From a seizure standpoint the child has had a single seizure and in the short term things may be controlled. However, in the presence of a structural lesion the chances of the child staying seizure free are probably around 15%. If the child had breakthrough seizures on 1st line meds then I would offer surgery as an option.

There is no hurry here and if the parents were ambivalent then I would suggest an interval scan in 6 months and see if szs remain controlled and if there is any change in the lesion. It may take the family a little time to become accustomed to the idea of surgical intervention and the dual diagnosis of tumour and epilepsy.

Of course the major consideration here is morbidity in particular speech. The child’s age is in its favour as firstly I believe speech will have moved and secondly even if speech is adjacent to the lesion, in my experience it won’t be in the lesion, then speech will relocate quite successfully. For this reason cerebral plasticity is on the child’s side but the family have to understand that from this aspect the earlier the surgical insult occurs the better….. In my view early surgery is the way forwards but after clear and adequate counselling and with the aim of total removal of the lesion and I would favour the use of intraoperative ultrasound.

Paul Steinbok says: It is probable that the tumour will grow at some time in the future and/or be associated with seizures that are difficult to control. The ideal treatment is resection of the tumour, which has a high chance of long term tumour free survival. The lesion is limited to the superior temporal gyrus and I am not convinced that important middle cerebral vessels run through it, so should be resectable. If one wished to be sure then a MRI angiogram could be done.

At this age we cannot consider an awake craniotomy for language testing or FMRI for language so it is difficult to quantify the risk of aphasia with resection.  Options are to observe, treat seizures and repeat scans periodically. Then indication for surgery would be growth of tumour and/or uncontrolled epilepsy. Pros of this option would be continued normal neurologic status until such time as surgery is required and maybe by that time one could assess the risk of aphasia better. Maybe instead of surgery at the time of progression the oncologists might consider chemotherapy  if the risk of aphasia is determined to be very high. Downside is that since the tumour is likely to grow, there is a high chance that surgery will be done at some time and the later the aphasia is produced the less recovery can be expected. The other option would be to resect the tumour now and accept risk of aphasia. If that occurs st this age one might expect fair recovery of language function due to plasticity of the brain.

I am leaning toward the surgical option, but neither option is clearly better and I would be guided by how the parents feel.

Taka Inagaki says: I had a similar case who came to us when the patient was 3-years old because of a seizure. The result of biopsy was benign astrocytoma according to our pathologist. But the MRI findings are very similar including the location. The seizure had been controlled well until she became 14-years old. MRI at that point showed enlarged mass with enhanced with Gd. I reoperated her and the tumor was thought to be Grade III astrocytoma at that time. She had received irradiation therapy after subtotal removal. She is now over 21-years old and doing fine. So I prefer to follow with medication for a while to see if the tumor starts to grow.

Wan Tew Seow says: I would attempt to try to remove this tumour completely – especially if I have access to an intraoperative MRI – this lesion is highly suitable to be done under iMRI as the lesion is seen well on the T2 images. More likely, I might end up doing a subtotal excision, looking at the large vessels in the lesion!

Thanks again to all our colleagues who put themselves on the line with their opinions. There is much that junior and senior colleagues can learn from their perspectives.

Tony Figaji