Resecting the epileptogenic tuber: What happens in the long term?


  • Asim Shahid

    Corresponding author
    1. Division of Pediatric Epilepsy, University Hospitals Case Medical Center, Rainbow Babies and Children's Hospital, Cleveland, Ohio, U.S.A
    • Address correspondence to Asim Shahid, Division of Pediatric Epilepsy, University Hospitals Case Medical Center, Rainbow Babies and Children's Hospital, 11100 Euclid Ave, Cleveland, OH 44106, U.S.A. E-mail:

    Search for more papers by this author


Tuberous sclerosis complex (TSC) is an autosomal dominant, neurocutaneous disorder causing benign noninvasive lesions in multiple organ systems. Cognitive delays and neuropsychiatric disorders such as autism are common. Epilepsy is very common and is associated with single or multiple epileptogenic cortical tubers. More than half of these individuals will become medically refractory over time. For these patients, epilepsy surgery with resection of the epileptogenic tuber or tubers not only can render a select patient population seizure-free, but also improves IQ and cognitive behaviors over time.

Tuberous sclerosis complex (TSC) is a genetic, neurocutaneous, multisystem disorder due to mutations in two different genes, TSC1 on chromosome 9 and TSC2 on chromosome 16 (Gomez et al., 1999). TSC has a reported frequency of one in 12,000–14,000 in children younger than 10 years of age (O'Callaghan et al., 1998). Cortical tubers, renal angiomyolipomas, and cardiac rhabdomyomas are associated with significant morbidity. About 90% of individuals with TSC will develop epilepsy (Gomez et al., 1999). Antiepileptic medications were typically the mainstay of epilepsy therapy until the 1950s and 1960s. However, since about 50–80% of the patients with epilepsy will eventually become medically refractory (Jansen et al., 2006), epilepsy surgery was seen as a potentially curative procedure with significant improvements in multiple domains.

The first case series of seven patients with TSC who underwent resective surgery for epilepsy was reported by Perot et al. (1966). Prior to that, most surgeries involving patients with TSC were to relieve hydrocephalus by the removal of giant cell astrocytomas.

Resective surgery to treat epilepsy in TSC is being increasingly used to not only improve seizure outcomes but also to improve cognitive outcome and more importantly to improve overall quality of life (Roth et al., 2011).

Multiple studies have been published in the last 15–20 years looking at ways of identifying epileptogenic tubers, the surgical processes involved in their resection, and long-term outcomes of patients who are undergoing tuberal resections. This article will review the presurgical and surgical strategies and also the outcomes after resection of the epileptogenic tuber.

Patient Selection and Presurgical Evaluation

It is important to mention the goals of the presurgical assessment, which should include: (1) Selection of the patients who may benefit from a resective surgery, followed by (2) identification of the epileptogenic tuber or tubers using a multimodality approach, and (3) baseline cognitive assessment.

The selection of patients involves history (age at seizure onset, seizure types, history of infantile spasms, and electrographic findings). Multifocal or generalized cases are usually poor surgical candidates who may not improve after surgery (Teutonico et al., 2008). Identification of the epileptogenic tuber or tubers is probably the most important step in this evaluation. An epileptogenic tuber may be defined as an area of the cortex that produces an ictal discharge or >50% of the interictal discharges localized to an area confirmed as tuber by an magnetic resonance imaging (MRI) (Koh et al., 2000; Roach & Sparagana, 2004; Jansen et al., 2005). The long-term postresection outcome correlates best with the amount of the epileptogenic zone that is removed. Noninvasive multimodality approach includes high-resolution electroencephalography (EEG) and MRI, fluorodeoxyglucose (FDG) or alpha-[11C]-methyl-l-tryptophan (AMT) positron emission tomography (PET) scan, magnetoencephalography (MEG), seizure semiology, and ictal single-photon emission computed tomography (SPECT) scans. PET scan plays an important role in localizing the epileptogenic tubers. Chugani et al. (1998) first reported the utility of the AMT PET scan in the serotonin-rich epileptogenic tubers in the interictal state. They were able to effectively differentiate the epileptogenic tubers from nonepileptogenic tubers using AMT PET as opposed to glucose metabolism PET scan, which shows cortical areas of hypometabolism corresponding to the location of all the tubers—epileptogenic and nonepileptogenic. AMT PET has been found to be extremely useful in patients with multifocal EEG abnormalities to effectively localize and resect the epileptogenic tuber for a more favorable seizure outcome. Seizure freedom rates may be much lower if tubers are nonlocalizable by AMT PET (Kagawa et al., 2005). Concordance of some or all of these studies is a good indicator of the identity and location of the epileptogenic tuber. Further confirmation of the epileptogenic tuber and extent of the resection may involve invasive evaluation with subdural grids, strips, depth electrodes, and electrocorticography (ECoG).

Characteristics of the Epileptogenic Tuber

A large tuber containing a nidus of calcification is most likely to be an epileptogenic tuber (Koh et al., 2000; Karenfort et al., 2002). Nonepileptogenic tubers may also be calcified, but a large tuber size distinguishes epileptogenic from nonepileptogenic tubers (Koh et al., 2000). Multiple studies have shown evidence of ictogenesis from within the tubers with the help of depth electrodes or subdural grids (Guerreiro et al., 1998; Koh et al., 2000; Mohamed et al., 2012). Resection of epileptogenic tubers alone leaving the peritubular cortex intact has been reported with good results (Mohamed et al., 2012). However, a resection of both the tuber and the perituberal cortex results in a more favorable long-term outcome. Factors such as single seizure type, single large tuber with concordant imaging, clinical and neurophysiologic data, and no mental retardation are indicators of a good outcome (Jarrar et al., 2004; Romanelli et al., 2004).

When recorded from within some tubers, the ictal EEG was lower voltage and faster frequency than the peritubular cortex, and the interictal discharges consisted of continuous periodic sharp waves on an attenuated background. The peritubular cortex had higher voltage, more polyspikes, and faster frequencies than tubular interictal discharges (Mohamed et al., 2012).

Surgical Approaches

Epileptogenic tubers are resected in a single stage surgery if the preoperative testing or intraoperative ECoG shows concordant findings identifying a distinct tuber, or a two-stage procedure requiring implantation of subdural electrodes to confirm the area of the epileptogenic zone followed by resection. In single-stage surgeries, ECoG may also be used to define resection margins. A multistage procedure involves resection of the epileptogenic focus, followed by further resection to extend the area of previous resection, or resection of a secondary area of ictogenesis (Weiner et al., 2006).

Multistage approach

A multistage approach should be considered in epilepsy surgery for TSC, since some nonepileptogenic tubers may become epileptogenic after the resection of the dominant tuber. The first stage of this procedure corresponds to the placement of grids or strips in different lobes in the same hemisphere or both hemispheres based on presurgical data. The second stage is the resection of the primary focus, and in the third and the last stage, resection of the residual tuber or the secondary focus is undertaken. This secondary focus may become apparent and more active only after the dominant focus is resected. Failure to recognize this activation possibly results in lower rates of seizure freedom (Weiner et al., 2006). This approach is especially useful in cases where bihemispheric or eloquent cortex involvement is suspected, which could result in these cases being rejected for possible surgery (Romanelli et al., 2002; Weiner et al., 2006). Romanelli et al. (2002) reported two cases where the presurgical evaluation failed to lateralize and localize the epileptogenic areas. Invasive monitoring either in different locations in the same hemisphere, or both hemispheres, was followed by a resection of the primary tuber. Further monitoring revealed activation of residual tubers resulting in the third stage of the surgery or resection of “new” epileptogenic areas. Both patients had an excellent long-term outcome.

Does the Tuber Location Influence the Resection?

Location of the tuber has no effect on the outcome (Madhavan et al., 2007). Epileptogenic tubers located in eloquent cortex can be safely removed without any long-term deficits, thus proving the nonfunctionality of these tubers (Moshel et al., 2010). Multifocal or generalized EEG findings is not a contraindication to surgery, as long as focal seizures and good EEG and imaging concordance is seen (Guerreiro et al., 1998). Extensive resection of the epileptogenic zone is a predictor of good outcome (Fallah et al., 2013).

A recent meta-analysis of 20 studies looking at outcome after epilepsy surgery in TSC reported absence of generalized seizure semiology, no or mild developmental delay, unifocal ictal surface EEG abnormalities, and EEG/MRI concordance as statistically significant predictors of good outcome (Fallah et al., 2013).

Neuropsychological Outcome

An improvement in performance intelligence quotient (IQ) is seen after resection of the epileptogenic tuber with or without corpus callosotomy at 2-year follow-up in a study by Liang et al., 2010. A comparison of patients treated surgically with the ones treated medically showed an overall improvement of IQ and quality of life at the end of a 2-year follow-up (Liang et al., 2010).

When neuropsychological evaluations are performed preoperatively and postoperatively, good outcomes are related to seizure cessation or if there is a decrease in seizure burden. Developmental progress was seen in a majority of seven patients who underwent epilepsy surgery (Zaroff et al., 2005). TSC patients with low tuber counts, lateralized and localized epileptiform activity, and less severe cognitive impairment support a more favorable postoperative outcome. Poor outcome may be associated with multiple seizure types, severe cognitive impairment, and bilateral or diffuse epileptogenicity (Teutonico et al., 2008). Another meta-analysis of 13 studies also found that patients with unilateral EEG abnormalities with seizure onset later than 1 year of age and lobectomy had a higher rate of seizure freedom (Zhang et al., 2013). This would also result in an improvement in cognitive and behavioral outcomes in the long term. Medically refractory West syndrome may not be a contraindication to surgery, even with multiple epileptogenic foci in one or both hemispheres. Outcome is even better when tuber resection is paired with corpus callosotomy in bilateral cases as shown by Liu et al., 2012. In their study, 17 patients with refractory West syndrome underwent resection of the epileptogenic tuber, with a subgroup requiring corpus callosotomy. Eleven patients became seizure-free, with a mean follow-up of 2.99 ± (standard deviation) 1.26 years. A significant improvement in not only IQ, but also in language and social domains were also reported by the families postoperatively (Liu et al., 2012).

Patients with moderate to severe cognitive disability had an excellent outcome at 1-year follow-up, but at 5-year follow-up they were classified as Engel class III/IV (Jarrar et al., 2004). This could possibly be due to the fact that a higher tuber burden is related to the severity of the cognitive impairment (Teutonico et al., 2008), with other tubers becoming epileptogenic over time after resection of the dominant tuber.


Long-term outcome after resection of the epileptogenic tuber depends on multiple factors that include presurgical seizure types, baseline cognitive level, and identification of the epileptogenic tuber or tubers. Predictors of good outcome include the following: Normal or mild cognitive impairment prior to surgery, focal electrographic ictal and interictal findings, concordance between EEG and MRI or additional imaging studies like PET scan, presence of a dominant tuber, and extensive resection of the epileptogenic zone. Improvement of seizures is one facet of the surgical outcome, the other being improvement in cognition and IQ. History of infantile spasms and multiple epileptogenic foci is not a contraindication for resective surgery.


The author has no conflicts of interest. The author confirms that he has read the Journal's position on issues involved in ethical publication and affirms that this report is consistent with those guidelines.