What Is Epilepsy Surgery? Types, Risks, and Recovery

Epilepsy surgery is any brain procedure aimed at stopping or reducing seizures that medications cannot control. It ranges from removing the small patch of brain tissue where seizures start to implanting a device that delivers electrical pulses to interrupt them. For the right candidates, surgery can be life-changing: roughly two-thirds of people who undergo the most common form of the operation become seizure-free within a year, and the benefits often extend to employment, mental health, and daily independence. Yet the path from diagnosis to the operating room is longer and more complex than most people expect, involving months of specialized testing, difficult personal decisions, and a recovery process that continues well beyond the surgical wound healing.

Who Qualifies for Epilepsy Surgery

The starting point is drug-resistant epilepsy, which the International League Against Epilepsy defines as the failure of two appropriately chosen and adequately tried antiseizure medications to achieve sustained seizure freedom.¹ Once that threshold is crossed, international expert consensus recommends that a surgical evaluation be offered regardless of seizure type, how long someone has had epilepsy, sex, or socioeconomic status, and the recommendation extends up to age 70.² Even people with significant psychiatric conditions, including psychogenic nonepileptic seizures alongside true epilepsy, should be referred. The only firm exclusion is active, uncooperative substance abuse that would make safe post-operative management impossible.²

Surgery may also be worth considering in people whose seizures are controlled on one or two medications if brain imaging shows a lesion sitting in a part of the brain that is not responsible for critical functions like speech or movement. In these cases, removing the lesion can sometimes eliminate the need for lifelong medication.

The Presurgical Workup

Before anyone gets near an operating room, an epilepsy team needs to answer two questions: exactly where do the seizures start, and can that area be treated without causing unacceptable harm? Answering those questions usually takes months and involves layers of testing that build on one another.

Standard brain MRI is the first imaging step, and finding a visible abnormality matters. A systematic review and meta-analysis found that people with a detectable lesion on MRI had a favorable surgical outcome about 75% of the time, compared with 58% for those whose MRI appeared normal.³ That gap has pushed researchers to develop advanced post-processing tools, including machine-learning algorithms that can detect subtle cortical abnormalities invisible to the human eye. These tools have increased the diagnostic yield by roughly 39% in cases initially read as MRI-negative.³ In children, automated surface-based analysis systems can identify features of focal cortical dysplasia even when conventional MRI looks normal.⁴

When non-invasive tests cannot pinpoint the seizure focus clearly enough, doctors may turn to invasive monitoring. Stereoelectroencephalography, or SEEG, involves threading thin electrodes through small holes in the skull directly into suspected brain regions. Electrical stimulation through these electrodes remains the gold standard for mapping where seizures begin and identifying nearby areas responsible for language, movement, or sensation.⁵ SEEG has largely replaced the older approach of placing electrode grids on the brain surface because it is better tolerated, allows more flexible electrode placement, and leads to more accurate identification of surgical candidates.⁶

For temporal lobe surgery, the team also needs to understand how each hemisphere handles memory and language. Functional MRI has increasingly taken over from the Wada test, an older procedure that involves injecting an anesthetic into one brain hemisphere at a time. Agreement between the two methods for memory lateralization is modest overall but improves substantially once cases with bilateral memory distribution are excluded.⁷ In practice, many centers now reserve the Wada test for cases where fMRI results are ambiguous or when bilateral memory makes it hard to predict which side is doing the heavy lifting.

Types of Epilepsy Surgery

The operations fall into a few broad categories, and the right choice depends on where seizures originate, whether a single focus can be identified, and what the person is willing to accept in terms of trade-offs.

Resective Surgery

This is the most established approach: the surgeon removes the brain tissue where seizures start. The most common version is anterior temporal lobectomy, which targets the front part of the temporal lobe, often including the hippocampus and amygdala. About 70% of patients become seizure-free within a year of this operation.⁸ Outcomes are somewhat better when a structural abnormality like mesial temporal sclerosis is present on pathology, with seizure freedom reaching about 74% in those cases compared with 60% when no clear pathology is found.⁸

Selective amygdalohippocampectomy is a more targeted version that removes only the inner structures of the temporal lobe, sparing the overlying cortex. A large meta-analysis found seizure-free rates of about 70% for this procedure, slightly higher than the 62% reported for standard anterior temporal lobectomy across the same pooled analysis.⁹ The difference likely reflects the fact that selective procedures tend to be offered when the seizure focus is well-localized to those inner structures.

Long-term durability does slip over the years. One study tracking outcomes for a decade estimated that the probability of seizure freedom was about 75% at two years, 67% at five years, and 51% at ten years.¹⁰ That decline does not mean the surgery “stops working” for most people; in many cases, breakthrough seizures are infrequent and far milder than what the person experienced before.

Disconnective Surgery

When seizures arise from a broad or diffuse area that cannot simply be cut out, disconnective procedures sever the pathways that allow seizures to spread. Corpus callosotomy, which divides the thick band of fibers connecting the two brain hemispheres, is the primary example. It is most effective against drop attacks, the sudden falls caused by atonic, tonic, or myoclonic seizures. In one series, about two-thirds of children had at least a 50% reduction in generalized seizures leading to falls, and roughly a third became seizure-free for those seizure types.¹¹ A more recent study confirmed that complete callosotomy eliminated drop attacks in about 71% of cases, though overall seizure freedom across all seizure types was harder to achieve and strongly correlated with younger age at surgery.¹²

Hemispherectomy, or its more modern variant hemispheric disconnection, is reserved for the most severe cases where one entire hemisphere is damaged and generating constant seizures. It sounds extreme, but in a Danish series, nearly 79% of hemispherectomy patients achieved seizure freedom, and more than half were eventually able to stop all antiseizure medications.¹³

Minimally Invasive and Radiosurgical Approaches

Laser interstitial thermal therapy, or LITT, uses a thin laser fiber guided by MRI to heat and destroy the seizure focus from inside the brain, without a traditional open craniotomy. It has gained popularity because the hospital stay is shorter and recovery is faster than with open resection. The trade-off is a lower seizure-free rate: about 58% across studies, compared with 62–70% for open resective approaches.⁹ Gamma knife radiosurgery (about 57% seizure-free) and radiofrequency thermocoagulation (about 47%) also fall below open surgery in seizure control but offer less invasive alternatives for people who are unwilling or unable to undergo craniotomy.⁹

Although open resection consistently achieves the highest long-term seizure freedom, it also carries greater risks for memory and language decline. LITT and neuromodulation have carved out a real niche for patients whose seizure focus sits near critical brain regions or who prioritize a faster physical recovery over the best possible seizure-free odds.¹⁴

Neurostimulation Devices

For people who are not candidates for removing or destroying brain tissue, implanted devices that deliver electrical stimulation can still reduce seizure frequency. Vagus nerve stimulation (VNS) involves a pacemaker-like device in the chest connected to the vagus nerve in the neck, reducing seizures in about half of patients.¹⁵ Responsive neurostimulation (RNS) takes a different approach: electrodes placed directly at the seizure focus continuously monitor brain activity and deliver a brief pulse of stimulation when they detect the start of a seizure. RNS is particularly suited for seizures that originate in brain areas too important to remove, like the speech or motor cortex. Studies comparing the two in temporal lobe epilepsy have found similar response rates, meaning VNS remains a reasonable option even when the seizure focus is well-localized.¹⁶

The Real Risks

Any brain surgery carries the standard surgical risks of infection, bleeding, and anesthesia complications. The risks specific to epilepsy surgery, however, are cognitive, and they depend heavily on which side and which structures are operated on.

Left temporal lobe resection poses the greatest threat to verbal memory. In one imaging study, 25 out of 29 patients who had left-sided surgery experienced some decline in verbal learning scores, and for seven of them, the decline was clinically significant.¹⁷ That means difficulty learning new word lists, remembering names, or recalling conversations. Right temporal surgery tends to affect visual or spatial memory instead, though these changes are usually less noticeable in daily life.

A broader meta-analysis found that verbal memory decline affected 17–37% of temporal lobe surgery patients, but among those who achieved seizure freedom, the memory decline did not independently reduce quality of life. About a fifth of seizure-free left temporal lobe surgery patients actually showed verbal memory improvement over time.¹⁸ The key insight is that uncontrolled seizures themselves damage memory over time, so for many people the surgical risk needs to be weighed against the ongoing cognitive toll of doing nothing.

Recovery and Life After Surgery

Physical recovery from open brain surgery typically takes a few weeks, with most people returning to light activities within a month and work within two to three months, depending on the job. But the deeper recovery, the cognitive, emotional, and social adjustment, unfolds over a much longer timeline.

Seizure medications are not stopped right away. A survey of U.S. epilepsy center neurologists found that most recommend patients remain seizure-free for at least two years before attempting to taper off medications.¹⁹ The decision to reduce or stop medication is highly individualized, depending on the type of surgery, pathology results, and post-operative brain monitoring.²⁰

Quality of life improves steadily in the years following successful surgery, and the improvement appears to continue for at least five years. Depression and anxiety scores tend to improve most sharply in the first post-operative year and then level off, while overall quality-of-life ratings keep climbing. Seizure freedom is the strongest predictor of improved quality of life.²¹ Employment tells a similar story: 80% of seizure-free patients in one study were in gainful employment after surgery, compared with about 28% of those still experiencing frequent seizures.²²

Cognitive Rehabilitation After Surgery

For people who do experience memory or attention problems after surgery, structured cognitive rehabilitation can make a real difference. Research on patients who underwent temporal lobe surgery found that those who received rehabilitation had about a quarter the risk of verbal memory loss compared with those who did not, with gains especially pronounced after right-sided operations.²³ Programs that target verbal learning, naming, and memory strategies have shown lasting improvements and even changes in brain activation patterns visible on follow-up functional imaging.²⁴ More recent work using computerized multi-domain rehabilitation platforms has confirmed benefits across memory, attention, and executive function, with left temporal lobectomy patients gaining the most in verbal memory specifically.²⁵

Despite these findings, cognitive rehabilitation is not universally offered to epilepsy surgery patients. If you or someone you know is facing temporal lobe surgery, it is worth asking the epilepsy team specifically about post-operative cognitive rehab, because many centers still treat it as optional rather than routine.

Epilepsy Surgery in Children

Children present a fundamentally different surgical picture than adults, largely because the developing brain can reorganize in ways that an adult brain cannot. This is most dramatically illustrated by hemispherectomy. Removing or disconnecting an entire hemisphere sounds catastrophic, yet children who undergo the procedure often recover motor function on the affected side through alternative neural pathways, and language development can proceed remarkably well when surgery happens early in life.²⁶ Visual field loss on the side opposite the removed hemisphere is the one deficit that tends to persist, though compensatory eye movements and scanning strategies can expand the functional field of view.²⁷

The brain’s capacity for this kind of rewiring fades with age. A study of complete corpus callosotomy found that seizure freedom was only achieved in children younger than seven, while older patients and adults could still benefit from significant seizure reduction but rarely became completely seizure-free for all seizure types.¹² This reinforces the case for early referral in pediatric epilepsy, though even in children the evaluation process can be slow. One study found median delays of about six months from seeing a neurologist to entering an epilepsy surgery program, and another six months to determine surgical candidacy.²⁸

The Delay Problem

One of the most frustrating aspects of epilepsy surgery is how long people wait to get it. A cohort study tracking adults found that the median interval from epilepsy diagnosis to surgery was nearly 17 years, with a median gap of about 10 years just from epilepsy diagnosis to being formally diagnosed with drug-resistant epilepsy.²⁹ Once drug resistance was recognized, patients with public health insurance waited nearly three times as long as those with private insurance to actually reach surgery (median 2.9 years versus 1.0 year).²⁹

These delays are not benign. Every year of uncontrolled seizures carries risks of injury, cognitive decline, social isolation, and lost employment. The expert consensus is clear that referral for a surgical evaluation should happen as soon as drug resistance is established, without waiting for additional medication failures.² If you have tried two medications without success, asking your neurologist about a referral to a comprehensive epilepsy center is a reasonable step.

Why People Decline Surgery

Even when surgery is recommended, many candidates choose not to go through with it. Research into patient decision-making reveals that the decision often comes down more to attitudes and emotions than to clinical factors. Patients who declined surgery tended to be less bothered by their epilepsy despite having seizures of comparable severity, more anxious about the procedure itself, and less likely to rely on their doctors’ recommendations. Those who chose surgery, by contrast, were more embarrassed by their seizures, more driven by the goal of being seizure-free, and less anxious about the specific risks of the operation.³⁰

A systematic review identified a triad of barriers: fear of surgery, ignorance of available treatment options, and tolerance of symptoms. Misperceptions were common among both patients and physicians, leading to what the authors described as undertreatment with serious health consequences.³¹ The takeaway is not that everyone should choose surgery, but that the decision should be based on accurate information about realistic outcomes rather than on vague fears or an underestimate of what uncontrolled epilepsy costs over a lifetime.

When Normal MRI Does Not Mean No Options

A common misconception is that epilepsy surgery requires a visible brain lesion on MRI. In reality, a substantial number of surgical candidates have normal-appearing scans. While outcomes are better when a lesion is found, more than half of MRI-negative patients still achieve a favorable surgical outcome.³ Advanced post-processing techniques that combine different MRI sequences can substantially increase the chance of finding a subtle abnormality. One study showed that combining standard imaging with FLAIR sequences raised the rate of identifying a concordant finding from about 31% to 46%, and newer machine-learning approaches push detection even further.³² Being told “your MRI is normal” should not be taken as the end of the conversation about surgery. If the clinical picture otherwise fits, a comprehensive epilepsy center with access to advanced imaging may see what a standard scan missed.

1
International Health Sciences Review. Drug-Resistant Epilepsy: Definitional criteria and indications for surgical treatment
2
PubMed Central. Timing of referral to evaluate for epilepsy surgery: Expert Consensus Recommendations from the Surgical Therapies Commission of the International League Against Epilepsy
3
PubMed Central. MRI-negative epilepsy: A systematic review and meta-analysis
4
PubMed. Advanced neuroimaging in pediatric epilepsy surgery: state of the art and future perspectives
5
PubMed Central. Stimulation Mapping Using Stereoelectroencephalography: Current and Future Directions
6
PubMed Central. Effect of stereo-EEG versus subdural EEG on functional and seizure outcome in pediatric and adult epilepsy surgery: A 21-year single-center experience
7
PubMed. Concordance between fMRI and Wada test for memory lateralization in temporal lobe epilepsy: A meta-analysis and systematic review
8
American Epilepsy Society. Surgery Outcome after Anterior Temporal Lobectomy for Intractable Epilepsy
9
PubMed. Comparative Efficacy of Surgical Strategies for Drug-Resistant Epilepsy: A Systematic Review and Meta-Analysis
10
PubMed Central. Long-term seizure outcome following resective surgery for epilepsy: to be or not to be completely cured?
11
PubMed Central. Corpus Callosotomy for Intractable Epilepsy Revisited: The Children’s Hospital of Michigan Series
12
PubMed Central. Complete Corpus Callosotomy Brings Worthwhile Seizure Reduction in Both Pediatric and Adult Patients
13
PubMed. Current state of hemispherectomy and callosotomy for pediatric refractory epilepsy in Denmark
14
PubMed Central. Comparative Review of Seizure and Cognitive Outcomes in Resective, Ablative, and Neuromodulatory Temporal Lobe Epilepsy Surgery
15
PubMed Central. Neurostimulation treatments for epilepsy: Deep brain stimulation, responsive neurostimulation and vagus nerve stimulation
16
PubMed. Vagus Nerve Stimulation versus Responsive Neurostimulator System in Patients with Temporal Lobe Epilepsy
17
Brain. Imaging memory in temporal lobe epilepsy: predicting the effects of temporal lobe resection
18
PubMed Central. Multidomain neuropsychiatric outcomes after epilepsy surgery: A systematic review and meta-analysis of depression, psychosis, cognition, and psychosocial adjustment
19
PubMed Central. Stopping antiepileptic drugs after epilepsy surgery: a survey of U.S. epilepsy center neurologists
20
PubMed. Management of antiepileptic drugs following epilepsy surgery: a meta-analysis
21
PubMed Central. Longitudinal trajectory of quality of life and psychological outcomes following epilepsy surgery
22
PubMed Central. Quality of life after epilepsy surgery
23
PubMed. The effects of cognitive rehabilitation on memory outcome after temporal lobe epilepsy surgery
24
Arquivos de Neuro-Psiquiatria. Potential role of a cognitive rehabilitation program following left temporal lobe epilepsy surgery
25
PubMed Central. Rehacom Cognitive Rehabilitation on the Neurocognitive Status of Patients with Temporal Lobe Epilepsy After Anterior Temporal Lobectomy
26
PubMed. Functional reorganization after hemispherectomy in humans and animal models: What can we learn about the brain’s resilience to extensive unilateral lesions?
27
PubMed. Functional recovery and neuroplasticity post-hemispherectomy in humans
28
PubMed. Potential delays in referral and assessment for epilepsy surgery in children with drug-resistant, early-onset epilepsy
29
PubMed Central. Delays in the Diagnosis and Surgical Treatment of Drug-resistant Epilepsy: A Cohort Study
30
PubMed Central. Epilepsy Surgery: Factors That Affect Patient Decision-Making in Choosing or Deferring a Procedure
31
PubMed. Perceptions of epilepsy surgery: a systematic review and an explanatory model of decision-making
32
American Journal of Neuroradiology. Systematic Assessment of Multispectral Voxel-Based Morphometry in Previously MRI-Negative Focal Epilepsy