OBJECTIVE:

We aimed to evaluate the long-term outcome of resective epilepsy surgery in patients with Lennox-Gastaut syndrome (LGS).

METHODS:

We reviewed the case reports of 90 patients with LGS who had undergone resective surgery between 2003 and 2014 at the Severance Children’s Hospital and managed them for a minimum period of 2 years.

RESULTS:

At the time of surgery, the patients were between 3.0 and 23.5 years old (mean ± SD: 9.3 ± 4.4). The time from seizure onset to surgery ranged from 0.7 to 20.1 years (7.2 ± 4.3). On postoperative follow-up for an average period of 6.1 ± 2.2 years (range: 2.1–11.4 years), 45 patients (50.0%) had no seizures, and 15 (16.7%) reported infrequent seizures. Seizure-free outcomes were achieved in 15 of the 21 (71.4%) hemispherectomies, 23 of the 51 (45.1%) multilobar resections, and 7 of the 18 (38.9%) single lobar resections. On high-resolution MRIs, 20 patients (22.2%) had negative findings, 8 of whom (40.0%) became seizure-free after resective surgery. Malformation of cortical development was the most common pathologic finding and was noted in 57 patients (63.3%). Seizure-free patients achieved better adaptive behavior and social competence than did patients with persistent seizures at the second (2–3 years after surgery) and third (4–6 years after surgery) follow-ups, as indicated by social quotients (P < .05).

CONCLUSIONS:

Resective surgery is a viable option in some patients to treat seizures that are associated with LGS, with a high probability of seizure control and better adaptive function.

What’s Known on This Subject:

Approximately one-half of the patients with Lennox-Gastaut syndrome are rendered seizure free with resective epilepsy surgery, and they show better adaptive function in serial follow-ups.

What This Study Adds:

In patients with seizure-free status, an improvement in electroencephalography profiles and a decrease in the number of administered anticonvulsants were observed. Resective surgery can be a viable option to treat patients with Lennox-Gastaut syndrome.

Lennox-Gastaut syndrome (LGS), a severe childhood-onset epileptic encephalopathy, can cause progressive cognitive decline and behavioral problems.1,3 LGS has been regarded as refractory to medical treatment, including treatment with various new antiepileptic drugs (AEDs), altered diet, vagal nerve stimulation, palliative surgery (such as corpus callosotomy), and resective surgery2,4,10; experience with resective surgery is limited because of the small number of patients and short follow-up periods.11,13 LGS exhibits diverse etiologies, from genetic and/or metabolic to structural causes with destructive lesions or cortical malformations. Predominantly, LGS exhibits bilateral and diffuse epileptic foci; however, unilateral diffuse and localized epileptic pathology are also observed. Patients with focal epileptogenic regions have been subjected to resective surgery, which seems promising in terms of seizure and developmental outcomes.11,13 

We examined the case patients with LGS who had resective surgery to evaluate seizure and developmental outcomes in a more extended follow-up period than in our previous study.13 

We retrospectively reviewed the medical records of 90 patients with LGS who underwent resective surgery from 2003 to 2014 at Severance Children’s Hospital. LGS is characterized by the clinical triad of the following: (1) multiple types of generalized seizures, including generalized tonic, atonic, and myoclonic seizures as well as atypical absences and spasms; (2) generalized slow spike-and-wave (GSSW) discharges and/or generalized paroxysmal fast activities (GPFAs) on EEG; and (3) progressive developmental regression after the onset of a seizure. Inclusion criteria for patients included the following: (1) resective surgery because of refractoriness to ≥3 medical management strategies, including AEDs and a ketogenic diet; (2) daily disabling seizures (generalized tonic, tonic-clonic, or clonic), with generalized convulsions and/or myoclonic or atonic seizures mixed with atypical absences or epileptic spasms; (3) typical EEG features with GSSW and/or GPFA; and (4) cognitive decline after seizure onset. We presumed focal abnormalities to be present in patients who had the following: (1) persistent focal polymorphic slow waves, (2) frequent focal rhythmic slow sharp and waves, and/or (3) frequent focal interictal epileptiform discharges >3 times those from the other side and/or areas of the brain in long-term EEG recordings. We conducted visual analyses of the long-term EEGs of patients and focused on interictal focal abnormalities other than generalized epileptiform discharges. With the exception of generalized epileptiform discharge, asymmetries that were quantitatively assessed >3 times in frequency were deemed to exhibit asymmetries when these focal abnormalities were localized in a certain area, and we decided to call these features “focal abnormalities.” We included patients of older age with LGS who showed GSSW and/or GPFA in the course of disease progression in serial EEGs, although well-formed GSSW and/or GPFA was absent at the time of surgery. The exclusion criteria were as follows: (1) progressive, degenerative neurologic disorders other than epileptic encephalopathy; (2) proven metabolic disorders, including mitochondrial cytopathy; (3) only nonmotor seizures or epileptic spasms; and (4) electroclinical syndromes other than LGS. The clinical profiles of the patients are shown in Table 1.

TABLE 1

Patient Profiles

Result
Sex, n (%)  
 Male 61 (67.8) 
 Female 29 (32.2) 
Onset of seizure, y, median (IQR) 1.0 (0.3–3.0) 
Age at operation, y, median (IQR) 8.7 (6.0–11.8) 
Latent period, y, median (IQR) 6.2 (4.4–9.5) 
Follow-up duration, y, median (IQR) 7.6 (5.7–10.9) 
History of ISs, n (%) 26 (28.9) 
Seizure type, n (%)  
 Generalized tonic 90 (100.0) 
 Other convulsive 69 (76.7) 
 Atonic 34 (37.8) 
 Myoclonic 28 (31.1) 
 Atypical absence 20 (22.2) 
 Focal 16 (17.8) 
 Epileptic spasm 14 (15.6) 
EEG feature, n (%)  
 Abnormal background 90 (100.0) 
 Persistent focal slowing 64 (63.1) 
 Epileptiform discharges  
  Generalized epileptiform dischargesa 90 (100.0) 
  Focal epileptiform dischargesb 70 (77.8) 
 Electrographic seizure  
  Generalized seizure 90 (100.0) 
  Focal seizure 22 (24.4) 
MRI finding, n (%)  
 Abnormal focal 70 (77.8) 
  MCD 36 (40.0) 
  Destructive 27 (30.0) 
  Tuberous sclerosis 3 (3.3) 
  Otherc 4 (4.5) 
 Abnormal nonfocal  
  Cortical atrophy 3 (3.3) 
 Normal 17 (18.9) 
Result
Sex, n (%)  
 Male 61 (67.8) 
 Female 29 (32.2) 
Onset of seizure, y, median (IQR) 1.0 (0.3–3.0) 
Age at operation, y, median (IQR) 8.7 (6.0–11.8) 
Latent period, y, median (IQR) 6.2 (4.4–9.5) 
Follow-up duration, y, median (IQR) 7.6 (5.7–10.9) 
History of ISs, n (%) 26 (28.9) 
Seizure type, n (%)  
 Generalized tonic 90 (100.0) 
 Other convulsive 69 (76.7) 
 Atonic 34 (37.8) 
 Myoclonic 28 (31.1) 
 Atypical absence 20 (22.2) 
 Focal 16 (17.8) 
 Epileptic spasm 14 (15.6) 
EEG feature, n (%)  
 Abnormal background 90 (100.0) 
 Persistent focal slowing 64 (63.1) 
 Epileptiform discharges  
  Generalized epileptiform dischargesa 90 (100.0) 
  Focal epileptiform dischargesb 70 (77.8) 
 Electrographic seizure  
  Generalized seizure 90 (100.0) 
  Focal seizure 22 (24.4) 
MRI finding, n (%)  
 Abnormal focal 70 (77.8) 
  MCD 36 (40.0) 
  Destructive 27 (30.0) 
  Tuberous sclerosis 3 (3.3) 
  Otherc 4 (4.5) 
 Abnormal nonfocal  
  Cortical atrophy 3 (3.3) 
 Normal 17 (18.9) 
a

GSSW discharges and GPFAs.

b

Localized paroxysmal fast activities, persistent focal sharp and spike waves, and focal rhythmic slow sharp and wave discharges.

c

Three dysembryoplastic neuroepithelial tumors and 1 arachnoid cyst in the right cerebellopontine angle.

Data on individual patient history, clinical findings, findings in imaging studies, seizure outcomes of surgery, and developmental outcomes were collected. Seizure outcomes were assessed on the basis of caregivers’ reports and seizure diaries at each scheduled visit according to the Engel Epilepsy Surgery Outcome Scale.14 

Presurgical evaluation included long-term video EEG monitoring, high-resolution MRI with a specific epilepsy protocol, 18-fluorodeoxyglucose positron emission tomography (18-FDGPET) scans, and interictal single-photon emission computed tomography (SPECT) according to feasibility. The decision-making process for performing the surgery is shown in a flow diagram (Fig 1). There were 27 patients with previous corpus callosotomy, 3 with corpus callosotomy after resective surgery, and 9 with vagus nerve stimulation (4 before and 5 after). With the exception of those who underwent corpus callosotomies and vagus nerve stimulation, 10 underwent repeated resective surgeries. Histopathologic examinations were completed according to the International League Against Epilepsy and Palmini classifications.15,16 

FIGURE 1

Flow diagram of approaching patients with LGS for resective surgery. PET, positron emission tomography.

FIGURE 1

Flow diagram of approaching patients with LGS for resective surgery. PET, positron emission tomography.

Seizure outcome was assessed by using Engel classification at the last follow-up visit after a minimum of 3 years, with a median duration of 7.6 years (interquartile range [IQR]; 5.7–10.9 years). In repeated surgery cases, seizure outcome was assessed as the outcome of the final seizure after the last resective surgery. Global adaptive behavior functioning and social competence skills, including communication, general self-help ability, locomotion, occupation, self-direction, self-help eating and/or dressing, and socialization, were evaluated at preoperation and 3 times postoperation (<1 year, 2–3 years, and 4–6 years after surgery) by using the Social Maturity Scale17 in patients who had successful follow-up evaluations.

We also analyzed the relationship between postoperative seizure outcomes and preoperative clinical data, including age at seizure onset, age at the time of surgery, duration of epilepsy before surgery, seizure type, presurgical MRI findings, and type of surgery.

Statistical analysis was conducted by using SPSS version 20.0 (IBM SPSS Statistics, IBM Corporation). Mann–Whitney tests, Pearson χ2 tests, Fisher’s exact tests, and t tests were applied to the data to compare variables; P < .05 was regarded as statistically significant.

The study was approved by the institutional review boards at Severance Children’s Hospital.

The EEGs of 90 patients revealed slow and disorganized backgrounds and persistent focal slowing in 64 patients, which was concordant with the resective areas (Table 1). Generalized epileptiform discharges, such as GSSWs and GPFAs, were observed in all patients. Focal epileptiform discharges, such as localized paroxysmal fast activities, persistent focal sharp waves or spikes, and focal rhythmic slow and sharp wave discharges, were observed in 77 patients (77.8%). The EEGs of all patients revealed generalized electrographic seizures, and those of 22 patients (24.4%) revealed focal electrographic seizures.

Malformations of cortical development (MCDs) (36 patients; 40.0%), including focal cortical dysplasia (FCD), polymicrogyria, pachygyria, schizencephaly combined with heterotopia, and destructive lesions (27 patients; 30.0%), were commonly seen on MRIs. Of the 70 cases with abnormal MRI findings, 60 (85.7%; 66.6% of the 90 total cases) had unilateral focal abnormalities. 18-FDGPET scans were performed in 81 patients; 56 patients (69.1%) had decreased 18-fluorodeoxyglucose uptake in the concordant hemisphere of the lesion. Interictal SPECT scanning was performed in 77 patients; 52 patients (67.5%) had decreased blood perfusion in the concordant hemisphere of the lesion. We used the term “concordant” for the cases that had results that were similar to those in the presurgical studies (EEG, MRI, positron emission tomography, and SPECT). Twenty-eight cases (31.1%) had concordant results in all of the investigations. Results that were concordant with 3 investigations were shown in 31 cases (34.4%), and results that were concordant with 2 investigations were shown in 22 cases (24.4%). In more than one-half of the cases, at least 3 test results were concordant.

Overall, 45 patients (50.0%) achieved seizure-free status after resective surgery in the mean follow-up period of 6.1 ± 2.2 years (Table 2). The MRIs of 20 patients revealed no focal lesions, and of these, 8 (40.0%) were seizure free after surgery. Of the different types of surgeries, hemispherotomy resulted in the most favorable outcome in seizure control; 15 of the 21 patients (71.4%) in the hemispherotomy group and 30 of the 69 patients (43.5%) in the single- and multilobar resection groups (P = .022) became seizure free (Engel class I). In the multilobar resection group, 9 case patients underwent subtotal hemispherotomy (3 seizure free; 33.3%), 26 underwent frontotemporal lobectomy (10 seizure free; 38.5%), 6 underwent posterior quadrantectomy (4 seizure free; 66.7%), 4 underwent temporo-occipital lobectomy (4 seizure free; 100.0%), 3 underwent parieto-occipital lobectomy (2 seizure free; 66.6%), 2 underwent frontoparietal corticectomy (none were seizure free), and 1 underwent multiple tubectomy (none were seizure free).

TABLE 2

Outcome of Surgery by Engel Classification According to Surgery Type, Concordance, and Latent Period

Class I, n (%)Class II, n (%)Class III, n (%)Class IV, n (%)Total, n (%)
Surgery type (P < .05)      
 Hemispherotomy 15 (71.4) 1 (4.8) 2 (9.5) 3 (14.3) 21 (100.0) 
 Lobar resection 30 (43.5) 7 (10.1) 5 (7.2) 27 (39.1) 69 (100.0) 
  Single lobar 7 (38.9) 2 (11.1) 1 (5.6) 8 (44.4) 18 (100.0) 
  Multilobara 23 (45.1) 5 (9.8) 4 (7.8) 19 (37.3) 51 (100.0) 
Concordance (P < .01)      
 >2 45 (55.6) 8 (9.9) 5 (6.2) 23 (28.3) 81 (100.0) 
  All 16 (57.1) 5 (17.9) 0 (0.0) 7 (25.0) 28 (100.0) 
  3 17 (54.8) 3 (9.7) 1 (3.2) 10 (32.3) 31 (100.0) 
  2 12 (54.5) 0 (0.0) 4 (18.2) 6 (27.3) 22 (100.0) 
  1 0 (0.0) 0 (0.0) 2 (22.2) 7 (77.8) 9 (100.0) 
Latent period (P > .05)      
 Within 5 y 20 (62.5) 3 (9.4) 1 (3.1) 8 (25.0) 32 (100.0) 
 5–10 y 16 (47.1) 3 (8.8) 1 (2.9) 14 (41.2) 34 (100.0) 
 >10 y 9 (37.5) 2 (8.3) 5 (20.8) 8 (33.3) 24 (100.0) 
Total 45 (50.0) 8 (8.9) 7 (7.8) 30 (33.3) 90 (100.0) 
Class I, n (%)Class II, n (%)Class III, n (%)Class IV, n (%)Total, n (%)
Surgery type (P < .05)      
 Hemispherotomy 15 (71.4) 1 (4.8) 2 (9.5) 3 (14.3) 21 (100.0) 
 Lobar resection 30 (43.5) 7 (10.1) 5 (7.2) 27 (39.1) 69 (100.0) 
  Single lobar 7 (38.9) 2 (11.1) 1 (5.6) 8 (44.4) 18 (100.0) 
  Multilobara 23 (45.1) 5 (9.8) 4 (7.8) 19 (37.3) 51 (100.0) 
Concordance (P < .01)      
 >2 45 (55.6) 8 (9.9) 5 (6.2) 23 (28.3) 81 (100.0) 
  All 16 (57.1) 5 (17.9) 0 (0.0) 7 (25.0) 28 (100.0) 
  3 17 (54.8) 3 (9.7) 1 (3.2) 10 (32.3) 31 (100.0) 
  2 12 (54.5) 0 (0.0) 4 (18.2) 6 (27.3) 22 (100.0) 
  1 0 (0.0) 0 (0.0) 2 (22.2) 7 (77.8) 9 (100.0) 
Latent period (P > .05)      
 Within 5 y 20 (62.5) 3 (9.4) 1 (3.1) 8 (25.0) 32 (100.0) 
 5–10 y 16 (47.1) 3 (8.8) 1 (2.9) 14 (41.2) 34 (100.0) 
 >10 y 9 (37.5) 2 (8.3) 5 (20.8) 8 (33.3) 24 (100.0) 
Total 45 (50.0) 8 (8.9) 7 (7.8) 30 (33.3) 90 (100.0) 

Pearson χ2 and Fisher’s exact tests were used as appropriate.

a

Nine subtotal hemispherectomies, 26 frontotemporal lobectomies, 6 posterior quadrantectomies, 4 temporo-occipital lobectomies, 3 parieto-occipital lobectomies, 2 frontoparietal cortisectomies, and 1 multiple tubectomy.

Of the 28 patients with localized and concordant findings on EEG, MRI, positron emission tomography, and SPECT, 16 (57.1%) achieved seizure-free status. Of the 9 patients with discordant findings, none became seizure free. The patients with >2 concordant findings in the presurgical studies were more likely to achieve seizure-free status after surgery (P = .003). Patients with focal MRI findings had a higher seizure-free rate (52.9%) than those with nonfocal findings (40.0%; P = .016).

An early latent period from seizure onset to surgery revealed a tendency of favorable surgery outcome but was not statistically significant (P = .113). Twenty of 32 patients (62.5%) who underwent surgery within 5 years of seizure onset achieved seizure-free status. Of 24 patients who had surgery after 10 years from seizure onset, only 9 (37.5%) were seizure free after surgery (P = .056). Only minor bleeding was observed in some cases during surgery without fatal complications.

There were 26 patients with a history of infantile spasms (ISs), and 15 of these became seizure free (57.7%). Of the 64 patients with no history of ISs, 30 achieved seizure-free status (46.9%). Patients with a history of ISs had a higher seizure-free rate, although this was not statistically significant.

By 2009, 44 case patients underwent surgery, and 46 underwent surgery after 2010. In both periods, seizure-free status was observed in 50% of the cases.

The pathology of the suspected lesions was examined in 80 patients who underwent surgery; 10 patients were not examined because of hemispherotomy. MCD was the most common finding and was found in 57 patients (63.3%). FCD types I, II, and III (according to the International League Against Epilepsy classification15) were noted in 6 (6.7%), 20 (22.2%), and 4 (4.4%) patients, respectively; mild MCD according to the Palmini classification16 was noted in 27 patients (30.0%). Other findings were gliosis (12 cases), tuberous sclerosis (3 cases), leukomalacia (3 cases), dysembryoplastic neuroepithelial tumor (1 case), and nonspecific findings (4 cases). Of the 20 cases with MRI-negative results, FCD in 8 cases, MCD in 5 cases, gliosis in 4 cases, and nonspecific findings in 3 cases were noted. There was no statistical difference between the pathologic findings and surgery outcomes.

The lead time from seizure onset to surgery was inversely correlated with the presurgical adaptive function as a social quotient (SQ) when using the Social Maturity Scale (P = .037; Fig 2). The SQ level in the seizure-free group was significantly higher than in the seizure-persistent group at the second (short-term follow-up; 2–3 years after surgery; P = .022) and third (long-term follow-up; 4–6 years after surgery; P = .001) follow-up evaluations after surgery (Fig 3). No significant differences were found at the baseline presurgical evaluation and first follow-up evaluation after surgery (within 1 year after surgery). Presurgical and postsurgical evaluation results for the SQ level (mean ± SD) in the seizure-free versus seizure-persistent groups are as follows: 42.0 ± 20.5 vs 36.3 ± 22.2 in the presurgical state, 44.0 ± 22.9 vs 38.2 ± 17.7 at the first follow-up, 46.3 ± 22.4 vs 31.6 ± 19.3 at the second follow-up, and 45.3 ± 28.0 vs 22.3 ± 16.0 at the third follow-up. In patients with persistent seizures, the SQ was significantly reduced at the third follow-up compared with the presurgical state (mean ± SD: 36.3 ± 22.2 vs 22.3 ± 16.0; P = .002), whereas the SQ scores were maintained in seizure-free cases.

FIGURE 2

Correlation of the SQ and lead time from seizure onset to surgery (latent). The lead time from seizure onset to surgery was inversely correlated with the presurgical adaptive function as an SQ via the Vineland Social Maturity Scale (n = 83; P = .037).

FIGURE 2

Correlation of the SQ and lead time from seizure onset to surgery (latent). The lead time from seizure onset to surgery was inversely correlated with the presurgical adaptive function as an SQ via the Vineland Social Maturity Scale (n = 83; P = .037).

FIGURE 3

Comparison of the serial SQ after resective surgery according to the surgery outcome. a The level of SQ in the seizure-free group was significantly higher than in the seizure-persistent group at the follow-up evaluations between 2 and 3 years after surgery (P = .022). b The level of SQ in the seizure-free group was significantly higher than in the seizure-persistent group at the follow-up evaluations between 4 and 6 years after surgery (P = .001). Adaptive function as an SQ was evaluated by using the Vineland Social Maturity Scale.

FIGURE 3

Comparison of the serial SQ after resective surgery according to the surgery outcome. a The level of SQ in the seizure-free group was significantly higher than in the seizure-persistent group at the follow-up evaluations between 2 and 3 years after surgery (P = .022). b The level of SQ in the seizure-free group was significantly higher than in the seizure-persistent group at the follow-up evaluations between 4 and 6 years after surgery (P = .001). Adaptive function as an SQ was evaluated by using the Vineland Social Maturity Scale.

The number of AEDs was reduced from 5.1 (range: 1–11) before surgery to 2.9 (range: 0–7) after surgery (P < .05). In patients with seizure-free status, 4.2 anticonvulsants were used before surgery and decreased to 1.6 after surgery. In patients with persistent seizures, 5.9 AEDs were used before surgery and decreased to 4.2 after surgery. In patients with Engel class IV, 6.4 AEDs were used before surgery but decreased to 4.4 after surgery. In 45 patients with seizure-free status, 14 discontinued using AEDs.

There was a difference between seizure-free patients and patients with persistent seizures with respect to the last postoperative EEG (Table 3). Focal slowing and GPFA were more common in seizure-free patients than in patients with persistent seizures, but there were no statistical differences. GSSW remained more common in patients with persistent seizures (80%) and had statistical differences (P < .05). Focal epileptiform discharges were more frequently absent in seizure-free patients (70%), and this difference was statistically significant (P < .05).

TABLE 3

Last EEG Findings After Surgery

Seizure FreeSeizures PersistTotalP
Focal slowing    .274 
 None 19 (57.6) 14 (42.4) 33 (100.0)  
 Persist 26 (45.6) 31 (54.4) 57 (100.0)  
GSSWs    .011* 
 None 42 (56.0) 33 (44.0) 75 (100.0)  
 Persist 3 (20.0) 12 (80.0) 15 (100.0)  
GPFAs    .134 
 None 41 (53.2) 36 (46.8) 77 (100.0)  
 Persist 4 (30.8) 9 (69.2) 13 (100.0)  
Focal epileptiform discharges    .043* 
 None 14 (70.0) 6 (30.0) 20 (100.0)  
 Persist 31 (44.3) 39 (55.7) 70 (100.0)  
Seizure FreeSeizures PersistTotalP
Focal slowing    .274 
 None 19 (57.6) 14 (42.4) 33 (100.0)  
 Persist 26 (45.6) 31 (54.4) 57 (100.0)  
GSSWs    .011* 
 None 42 (56.0) 33 (44.0) 75 (100.0)  
 Persist 3 (20.0) 12 (80.0) 15 (100.0)  
GPFAs    .134 
 None 41 (53.2) 36 (46.8) 77 (100.0)  
 Persist 4 (30.8) 9 (69.2) 13 (100.0)  
Focal epileptiform discharges    .043* 
 None 14 (70.0) 6 (30.0) 20 (100.0)  
 Persist 31 (44.3) 39 (55.7) 70 (100.0)  
*

P < .05.

LGS is 1 of the most severe forms of intractable childhood epilepsies.1,2 In general, the remission of seizures with anticonvulsants in the natural course of LGS is known to be <10%.2,9,18 For patients with LGS with focal epileptic pathology, resective surgery has been regarded as an important management strategy for freedom from seizures.11,13 

Our previous study of 27 cases revealed 59.3% of seizure-free outcomes with a mean duration of follow-up of 33.1 months (SD: ± 20.3 months). In this study, we observed long-term, seizure-free outcomes after resective surgery in patients with LGS at a 50.0% rate in 90 patients with a follow-up duration of 73.2 months (SD: ± 26.4 months). This long-term outcome with an extended follow-up duration in a large number of patients was comparable to the findings in our previous study, which revealed that a seizure-free outcome is expected to be maintained for a prolonged period of time. In the study of Liu et al,12 7 of 18 patients with LGS (38.9%) obtained seizure-free status after resective surgery. This study involved observations for at least 1 year and up to 9 years. In a study by Kang et al,19 73.3% (22 of 30) of the patients with IS and FCD achieved seizure-free status after resective surgery. IS and LGS are common encephalopathies; however, higher seizure-free rates were observed in IS, probably because the study was restricted to IS patients with FCD.

Patients with single- and multilobar resection achieved significantly lower seizure-free rates than those with hemispherectomies. The seizure-free rate in those with hemispherotomy in LGS (71.4%) is higher than with other types of surgery.20 Our cases also had better surgical outcomes with wider resection. Wider resection in epilepsy surgery is important for success of the operation, which implicates that the epileptogenic cortex is wider in distribution than what is estimated in multimodal diagnostic studies. The seizure-free rate in multilobar resection was higher than in single-lobar resection, although there was no statistical difference. An undefined epileptogenic zone should be a concern for the surgery.

Another prognostic factor that influences surgical outcome is the degree of concordance in the preoperative investigations, which include long-term video EEG monitoring, MRI, 18-FDGPET, and SPECT. MRI scanning was performed in all patients, and 18-FDGPET and SPECT were performed in 81 and 77 patients, respectively. A higher degree of anatomic matching in the findings in the investigations was related to better surgical prognosis (P = .003). When all presurgical investigations had the same localization, 57.1% of the patients became seizure free. In contrast, in case of discordant results, none of the 9 patients became seizure free. Similarly, authors of other studies have also reported that higher concordance rates correspond to better surgical outcome.21,22 Patients with focal MRI findings had a significantly higher seizure-free rate (52.9%) than those with nonfocal findings (40.0%; P = .016). The positivity of MRI lesion result is 1 of the important factors for a successful surgery.23 Along with the MRI results, the localization of epileptic focus can be achieved with an 18-FDGPET scan or SPECT to increase the postsurgical seizure-free rate. Resective surgery should be carefully considered in cases that are discordant from multimodal diagnostic findings.22,23 

Changes in EEG results after surgery are also notable. All the patients showed GSSW and/or GPFA in the EEGs that were performed before surgery. In the last EEG performed after surgery, GSSW disappeared in 75 cases, and GPFA disappeared in 77 cases. It can be expected that abnormal epileptiform discharges will be resolved because of the resection of the primary foci. This means that physicians should consider the surgical treatment option despite generalized EEG patterns.24 GSSW disappeared more frequently in seizure-free patients, which is related to the prognosis of seizures. This may render superior cognitive abilities in seizure-free patients rather than in patients with persistent seizures. Focal epileptiform discharge still persisted in 70 patients but was more often absent in seizure-free patients. This can also affect cognition. In addition, the number of AEDs that were taken after surgery was reduced, which can be thought of as a result of the improvement of the EEG.

LGS is accompanied by a characteristic developmental delay.2 We found that a shorter latent period from seizure onset to surgery led to a better level of adaptive behavioral functioning similar to that found in patients with ISs.19 Thus, it could also be associated with better seizure outcomes; early surgery after seizure onset was associated with a favorable outcome of seizure control in our study, although this was not statistically significant. It would be useful to study the effect of surgery on the neurocognitive and motor outcomes of patients. Unfortunately, it was difficult to determine the IQ for most patients with mental retardation and LGS. Thus, adaptive function was used as the outcome measure in this study. Despite the small number of cases, there was a statistically significant difference in the level of adaptive behavioral functioning between the seizure-free and seizure-persistent group at the within-2-to-3– and 4-to-6–year follow-up. There was, however, no significant improvement in adaptive functioning between the presurgical evaluation and the first year of follow-up in both the seizure-free and seizure-persistent groups. Follow-up for a longer period, and in a higher number of cases, is required to determine long-term improvements in adaptive functioning. Although it is difficult to reach an average range of adaptive functioning after surgery, effective surgery can preserve the baseline level in patients with LGS (ie, without further deterioration), whereas a gradual improvement can be expected in the seizure-free patients. Thus, early surgery can lead to a better prognosis in adaptive functioning and help improve the quality of life of patients and their families.25 

There are some limitations of our study. There is no control group without surgery with a 2-year follow-up; although a comparison with this group is preferable, the caregivers, who did not undergo surgery, did not agree to undergo neurocognitive tests. The patient cohort is also heterogeneous; in fact, the etiologies and age distribution for patients with LGS is extremely varied. The patient group would be extremely small if we had divided patients according to etiology or age.

Early and intensive investigations to determine underlying focal etiology are important for deciding surgical candidates, and careful preoperative investigations can be used to steer the course of LGS in patients toward seizure-free status and better outcomes in adaptive functioning and social competence. Resective surgery should be actively used to treat patients with LGS with suspected focal epileptic pathology.

     
  • 18-FDGPET

    18-fluorodeoxyglucose positron emission tomography

  •  
  • AED

    antiepileptic drug

  •  
  • FCD

    focal cortical dysplasia

  •  
  • GPFA

    generalized paroxysmal fast activity

  •  
  • GSSW

    generalized slow spike-and-wave

  •  
  • IQR

    interquartile range

  •  
  • IS

    infantile spasm

  •  
  • LGS

    Lennox-Gastaut syndrome

  •  
  • MCD

    malformation of cortical development

  •  
  • SPECT

    single-photon emission computed tomography

  •  
  • SQ

    social quotient

Dr J. Kang contributed to the analysis and interpretation of data, drafted the initial manuscript, and revised the manuscript; Dr Eom contributed to the analysis and interpretation of psychological data and drafted the initial manuscript; Drs Hong, Kwon, and H. Kang contributed to the acquisition and analysis of data; Drs Park and J. Lee contributed to the analysis of data; Drs Ko, Y. Lee, and Kim contributed to the acquisition of data; Dr Kim contributed to the study concept and design and the critical revision of the manuscript; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

FUNDING: Supported by the National Research Foundation of Korea (grant NRF-2012R1A2A2A01012608) funded by the Government of South Korea.

We thank Dr In Sun Kwon for the statistical analysis of the data in the article.

1
Arzimanoglou
A
,
French
J
,
Blume
WT
, et al
.
Lennox-Gastaut syndrome: a consensus approach on diagnosis, assessment, management, and trial methodology.
Lancet Neurol
.
2009
;
8
(
1
):
82
93
[PubMed]
2
Camfield
PR
.
Definition and natural history of Lennox-Gastaut syndrome.
Epilepsia
.
2011
;
52
(
suppl 5
):
3
9
[PubMed]
3
Archer
JS
,
Warren
AE
,
Jackson
GD
,
Abbott
DF
.
Conceptualizing Lennox-Gastaut syndrome as a secondary network epilepsy.
Front Neurol
.
2014
;
5
:
225
[PubMed]
4
Kwan
P
,
Arzimanoglou
A
,
Berg
AT
, et al
.
Definition of drug resistant epilepsy: consensus proposal by the ad hoc Task Force of the ILAE Commission on Therapeutic Strategies [published correction appears in Epilepsia. 2010;51(9):1922].
Epilepsia
.
2010
;
51
(
6
):
1069
1077
[PubMed]
5
Carmant
L
,
Whiting
S
.
Lennox-Gastaut syndrome: an update on treatment.
Can J Neurol Sci
.
2012
;
39
(
6
):
702
711
[PubMed]
6
Hancock
EC
,
Cross
HH
.
Treatment of Lennox-Gastaut syndrome.
Cochrane Database Syst Rev
.
2009
;(
3
):
CD003277
[PubMed]
7
Michoulas
A
,
Farrell
K
.
Medical management of Lennox-Gastaut syndrome.
CNS Drugs
.
2010
;
24
(
5
):
363
374
[PubMed]
8
Cross
JH
.
The ketogenic diet in the treatment of Lennox-Gastaut syndrome.
Dev Med Child Neurol
.
2012
;
54
(
5
):
394
395
[PubMed]
9
Montouris
GD
.
Rational approach to treatment options for Lennox-Gastaut syndrome.
Epilepsia
.
2011
;
52
(
suppl 5
):
10
20
[PubMed]
10
Kossoff
EH
,
Shields
WD
.
Nonpharmacologic care for patients with Lennox-Gastaut syndrome: ketogenic diets and vagus nerve stimulation.
Epilepsia
.
2014
;
55
(
suppl 4
):
29
33
[PubMed]
11
Douglass
LM
,
Salpekar
J
.
Surgical options for patients with Lennox-Gastaut syndrome.
Epilepsia
.
2014
;
55
(
suppl 4
):
21
28
[PubMed]
12
Liu
SY
,
An
N
,
Fang
X
, et al
.
Surgical treatment of patients with Lennox-Gastaut syndrome phenotype.
Scientific World Journal
.
2012
;
2012
:
614263
13
Lee
YJ
,
Kang
HC
,
Lee
JS
, et al
.
Resective pediatric epilepsy surgery in Lennox-Gastaut syndrome.
Pediatrics
.
2010
;
125
(
1
). Available at: www.pediatrics.org/cgi/content/full/125/1/e58
[PubMed]
14
Engel
J
 Jr
.
Epilepsy surgery.
Curr Opin Neurol
.
1994
;
7
(
2
):
140
147
[PubMed]
15
Blümcke
I
,
Thom
M
,
Aronica
E
, et al
.
The clinicopathologic spectrum of focal cortical dysplasias: a consensus classification proposed by an ad hoc Task Force of the ILAE Diagnostic Methods Commission.
Epilepsia
.
2011
;
52
(
1
):
158
174
[PubMed]
16
Palmini
A
,
Najm
I
,
Avanzini
G
, et al
.
Terminology and classification of the cortical dysplasias.
Neurology
.
2004
;
62
(
6, suppl 3
):
S2
S8
[PubMed]
17
Gould
J
.
The use of the Vineland Social Maturity Scale, the Merrill-Palmer Scale of mental tests (non-verbal items) and the Reynell Developmental Language Scales with children in contact with the services for severe mental retardation.
J Ment Defic Res
.
1977
;
21
(
3
):
213
226
[PubMed]
18
Montouris
GD
,
Wheless
JW
,
Glauser
TA
.
The efficacy and tolerability of pharmacologic treatment options for Lennox-Gastaut syndrome [published correction appears in Epilepsia. 2015;56(6):984].
Epilepsia
.
2014
;
55
(
suppl 4
):
10
20
[PubMed]
19
Kang
JW
,
Rhie
SK
,
Yu
R
, et al
.
Seizure outcome of infantile spasms with focal cortical dysplasia.
Brain Dev
.
2013
;
35
(
8
):
816
820
[PubMed]
20
Griessenauer
CJ
,
Salam
S
,
Hendrix
P
, et al
.
Hemispherectomy for treatment of refractory epilepsy in the pediatric age group: a systematic review.
J Neurosurg Pediatr
.
2015
;
15
(
1
):
34
44
[PubMed]
21
Pittau
F
,
Grouiller
F
,
Spinelli
L
,
Seeck
M
,
Michel
CM
,
Vulliemoz
S
.
The role of functional neuroimaging in pre-surgical epilepsy evaluation.
Front Neurol
.
2014
;
5
:
31
[PubMed]
22
Seo
JH
,
Holland
K
,
Rose
D
, et al
.
Multimodality imaging in the surgical treatment of children with nonlesional epilepsy.
Neurology
.
2011
;
76
(
1
):
41
48
[PubMed]
23
Arya
R
,
Leach
JL
,
Horn
PS
, et al
.
Clinical factors predict surgical outcomes in pediatric MRI-negative drug-resistant epilepsy.
Seizure
.
2016
;
41
:
56
61
[PubMed]
24
Wyllie
E
,
Lachhwani
DK
,
Gupta
A
, et al
.
Successful surgery for epilepsy due to early brain lesions despite generalized EEG findings.
Neurology
.
2007
;
69
(
4
):
389
397
[PubMed]
25
Perry
MS
,
Duchowny
M
.
Surgical versus medical treatment for refractory epilepsy: outcomes beyond seizure control.
Epilepsia
.
2013
;
54
(
12
):
2060
2070
[PubMed]

Competing Interests

POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.