Abstract
Intracranial EEG in Predicting Surgical Outcome in Frontal Lobe Epilepsy
Holtkamp M, Sharan A, Sperling MR. Epilepsia 2012;53:1739–1745.
PURPOSE: Surgery in frontal lobe epilepsy (FLE) has a worse prognosis regarding seizure freedom than anterior lobectomy in temporal lobe epilepsy. The current study aimed to assess whether intracranial interictal and ictal EEG findings in addition to clinical and scalp EEG data help to predict outcome in a series of patients who needed invasive recording for FLE surgery. METHODS: Patients with FLE who had resective surgery after chronic intracranial EEG recording were included. Outcome predictors were compared in patients with seizure freedom (group 1) and those with recurrent seizures (group 2) at 19–24 months after surgery. KEY FINDINGS: Twenty-five patients (16 female) were included in this study. Mean age of patients at epilepsy surgery was 32.3 ± 15.6 years (range 12–70); mean duration of epilepsy was 16.9 ± 13.4 years (range 1–48). In each outcome group, magnetic resonance imaging revealed frontal lobe lesions in three patients. Fifteen patients (60%) were seizure-free (Engel class 1), 10 patients (40%) continued to have seizures (two were class II, three were class III, and five were class IV). Lack of seizure freedom was seen more often in patients with epilepsy surgery on the left frontal lobe (group 1, 13%; group 2, 70%; p = 0.009) and on the dominant (27%; 70%; p = 0.049) hemisphere as well as in patients without aura (29%; 80%; p = 0.036), whereas sex, age at surgery, duration of epilepsy, and presence of an MRI lesion in the frontal lobe or extrafrontal structures were not different between groups. Electroencephalographic characteristics associated with lack of seizure freedom included presence of interictal epileptiform discharges in scalp recordings (31%; 90%; p = 0.01). Detailed analysis of intracranial EEG revealed widespread (>2 cm) (13%; 70%; p = 0.01) in contrast to focal seizure onset as well as shorter latency to onset of seizure spread (5.8 ± 6.1 s; 1.5 ± 2.3 s; p = 0.016) and to ictal involvement of brain structures beyond the frontal lobe (23.5 ± 22.4 s; 5.8 ± 5.4 s; p = 0.025) in patients without seizure freedom. The distribution of ictal onset patterns was similar in both groups, and fast rhythmic activity in the beta to gamma range was found in 57% of seizure-free patients compared to 70% of patients with recurrent seizures. Analysis of the temporal relation between first clinical alterations and EEG seizure onset did not reveal significant differences between both groups of patients. In multivariate analysis, resection in the left hemisphere (odds ratio [OR] 12.197 95% confidence interval [95% CI] 1.33–111.832; p = 0.027) and onset of seizure spread (odds ratio [OR] 0.733, 95% CI 0.549–0.978, p = 0.035) were independent predictors of ongoing seizures. SIGNIFICANCE: Widespread epileptogenicity as indicated by rapid onset of spread of ictal activity likely explains lack of seizure freedom following frontal resective surgery. The negative prognostic effect of surgery on the left hemisphere is less clear. Future study is needed to determine if neuronal network properties in this hemisphere point to intrinsic interhemispheric differences or if neurosurgeons are restrained by proximity to eloquent cortex.
Outcome of Frontal Lobe Epilepsy Surgery
Lazow SP, Thadani VM, Gilbert KL, Morse RP, Bujarski KA, Kulandaivel K, Roth RM, Scott RC, Roberts DW, Jobst BC. Epilepsia 2012;53:1746–1755.
PURPOSE: There is still controversy in deciding which patients with frontal lobe epilepsy (FLE) should undergo resective surgery, even though it is a well-established therapy. The aim of this study is to define multiple outcome measures and determine whether there are certain subpopulations of preferred surgical candidates that have a more favorable seizure prognosis. METHODS: Fifty-eight patients underwent resective FLE surgery with a mean follow-up period of 79.3 months (range 12–208 months). Patient demographics, clinical seizure characteristics, seizure-onset zone within the frontal lobes, and diagnostic tests were tabulated. Engel class, International League Against Epilepsy (ILAE) class, postoperative seizure patterns, time to first recurrent seizure, and seizures and employment during the last year of follow-up were used as outcome measures. Neuropsychological performance and Beck Depression Inventory (BDI) scores were used to define neuropsychological outcome and examined as predictors of seizure outcome. KEY FINDINGS: Thirty-three (57%) patients with resective surgery had an Engel class I outcome and 29 (50%) had an ILAE class I outcome. Mean time to first seizure after surgery was 33.3 months (range 0–208). Only 14 patients (24%) were completely seizure-free without auras (Engel IA) throughout the entire follow-up period. The most common pattern of seizure recurrence was mixed, with prolonged periods of seizure freedom intermixed with recurrences. In addition, 32% of patients made gains in employment and 52% were able to reduce use of antiepileptic drugs (AEDs), although only 9% discontinued AEDs. No significant association was found between class I or class IA outcome and the presence of a focal magnetic resonance imaging (MRI) abnormality, any specific localization of seizure focus within the frontal lobe, or neuropsychological change. SIGNIFICANCE: Findings indicate that that long-term outcome is generally favorable in FLE resective surgery, and support the need for considering multiple outcome measures to more fully characterize clinically relevant postsurgical changes. Outcome can be favorable even in MRI-negative patients.
Improved Outcomes with Earlier Surgery for Intractable Frontal Lobe Epilepsy
Simasathien T, Vadera S, Najm I, Gupta A, Bingaman W, Jehi L. Ann Neurol 2013;73:646–654.
OBJECTIVE: To explore the prognostic implications of epilepsy duration and age at surgery for seizure outcomes after frontal lobe epilepsy (FLE) surgery. METHODS: We reviewed 158 patients who underwent FLE surgery from 1995 to 2010. The primary outcome was seizure freedom at last follow-up (Engel class IA). Analyses employed Cox proportional and multiphase hazard modeling. RESULTS: The mean age at surgery was 20.4 years, and mean epilepsy duration was 12.0 years. The estimated chance of seizure freedom was 66% (95% confidence interval [CI]562–68) at 1 postoperative year, 52% (95% CI548–56) at 2 years, and 44% (95% CI539–49) at 5 years and beyond. Seventy-five percent of recurrences occurred within 6 postoperative months. Both younger age at surgery (<18 years) and shorter epilepsy duration (<5 years) correlated with better seizure outcomes on univariate analysis, but only epilepsy duration remained statistically significant after multivariate modeling. Independent poor prognostic indicators included left-sided resections and acute postoperative seizures (APOSs; whole model log-rank test p < 0.0001). APOSs were particularly predictive of early epilepsy recurrence, starting within 6 postoperative months (adjusted risk ratio [RR]54.42, p < 0.0001), whereas long epilepsy duration correlated with late recurrences (RR56.25, p < 0.0001). Worse outcomes were seen with longer epilepsy duration for duration cutoffs of 2, 5, and 10 years independently for adults and children, although statistical significance was only achieved in children (66% seizure free at 5 postoperative years if operated on within 5 years of epilepsy onset vs 31% if later; p50.01). INTERPRETATION: Early resection may improve seizure outcomes of FLE surgery, particularly in children.
Commentary
The frontal lobe is the largest lobe of the brain and, after the temporal lobe, the second most common lobe from which seizures arise. Although the temporal lobe has a nice, discrete, highly epileptogenic structure from which most seizures arise (the hippocampus), the frontal lobe is not so lucky. Primary motor and language function reside here as well. Thus, it is not surprising that surgery in the vast frontal lobe is not as successful as that in the temporal lobe. How often is it successful and in which patients? A recent systematic review of 21 studies and 1,199 patients undergoing resective surgery for frontal lobe epilepsy (FLE) reported Engel class I outcome (free of disabling seizures for at least 2 years) in about 45% of patients (1). The most consistent predictor of better response was presence of a lesion, and its complete resection. Of those with abnormal MRI, 61% were seizure free, compared to only 39% of those with normal MRI (p < 0.001). Of note, contrary to the studies discussed here, duration of epilepsy and side of surgery were not predictors in this very large analysis. The three single-center studies reviewed here differ a bit from that review, and all report better outcomes, especially for the nonlesional cases.
Study 1: Holtkamp et al
Main Findings and Significance
The authors report on surgical outcomes 19 to 24 months after resection in 25 patients with FLE who had resection after intracranial EEG recording at Thomas Jefferson University in Philadelphia. Only 6 patients had lesions on imaging; 3 became seizure free and 3 did not. Overall, 60% were seizure free, with or without auras. In multivariate analysis, lack of seizure freedom was predicted by only two factors: resection in the left hemisphere and rapid seizure spread. Only 2 of 9 patients with left frontal surgery were seizure free, versus 13 of 16 patients with right frontal surgery (p = 0.009). In the seizure-free group, time to electrographic spread to at least 2 cm outside the seizure onset zone averaged 5.8 s, versus only 1.5 s in the group with persistent seizures; this is similar to prior studies on intracranial EEG in neocortical epilepsy (2). Although a more focal onset (<2 cm) correlated with better outcome in univariate analysis as expected, ictal EEG onset frequency did not correlate with outcome, with low voltage beta to gamma range activity seen at onset in more than half the patients in each group, contrary to some prior studies, where faster activity correlated with better outcome (3, 4). At least some interictal discharges were found outside the seizure onset zone in about half of patients, with no correlation with outcome, even when bilateral; however, the majority of epileptiform activity arose from the seizure onset zone in almost all patients.
Limitations
The main limitations of this study are its retrospective nature and small sample size. One of the main findings is that surgery on the left frontal lobe has a worse prognosis than the right, yet there is no report of resection size or how often resection had to be limited due to eloquent cortex. Less aggressive resection is the most likely explanation for this finding. Hemispheric dominance was considered contrary to handedness in the manuscript, rather than determined by functional imaging, Wada testing or brain mapping. There was no discussion of the details of implantation, such as number of contacts or whether depth electrodes were utilized, though the two examples provided had only subdural strips and grids. Post-resection MRI was not utilized to see what was actually resected and how that related to the seizure onset zone.
Study 2: Lazow et al
Main Findings and Significance
This study from Dartmouth was larger (58 patients with FLE undergoing resection), had longer but variable follow-up averaging >6 years and included outcome related to employment, neuropsychological functioning, and mood. Just over half the patients had lesions on imaging. A surprisingly high number of patients (54/58) underwent intracranial EEG monitoring. Fifty-seven percent had an Engel class I outcome, and 73% were Engel I or II (seizure free or rare disabling seizures). Contrary to many prior studies, the presence of an MRI lesion was not predictive of better outcome; 58% of nonlesional patients were seizure free for at least 1 year at last follow-up. Perhaps advances in functional imaging, neurosurgical techniques, and intracranial EEG (methodology or interpretation) are improving results in MRI-negative patients over time. Patients with cortical dysplasia on pathology also did well in this study, with 65% becoming seizure free. Sublobar location of the seizure focus, seizure onset pattern, neuropsychological testing results, and depression—all of which have been predictive in at least some prior epilepsy surgery studies—were all without prognostic significance in this series. Surgery on the left was just as successful as the right (56% vs 58% seizure free). The long-term longitudinal outcomes showed that a significant portion of the patients (41%) switched from one outcome group to another (seizure free for a year, then relapsed, or vice versa), often multiple times in a “relapsing, remitting” fashion, as has been seen in other surgical series with long-term outcome (5). One-third of patients made gains in employment. Postoperative employment was largely related to seizure freedom. Almost half the patients were able to reduce their AEDs; only 9% were off AEDs completely, but there is no mention of how many attempted this. Based on prior literature, primarily on temporal lobe epilepsy, about two-thirds of those that attempt to stop AEDs after epilepsy surgery (and prolonged seizure freedom) are successful (6, 7). This successful AED withdrawal rate after successful surgery may be slightly lower in extratemporal epilepsy, but a recent long-term study found that half of patients who were seizure free at 2 years after surgery were able to discontinue and remain off AEDs at 10 years (8).
Limitations
As in the prior study, the retrospective nature and modest sample size are the main weaknesses.
Study 3: Simasathien et al
Main Findings and Significance
This study was even larger, with 158 patients having undergone frontal lobe epilepsy surgery at the Cleveland Clinic over a 15-year period; follow-up averaged 4.3 years but was highly variable. A strength of this study was the use of blinded review of co-registered postoperative scans to determine if there was complete resection, defined as removal of the seizure onset zone and the area with the most frequent interictal epileptiform discharges, although these specific results are not given (presumably, it was not an independent predictor of outcome in this study, which is surprising). Mean age at surgery was rather young at 20 years old. One quarter of the patients had normal MRI, and half the patients underwent subdural EEG recordings. Pathology showed cortical dysplasia in 59%. Sustained seizure freedom was achieved in 66% at 1 year, 52% at 2 years, and 44% at 5 years. At last follow-up, just over half of patients were seizure free for the prior year. Most recurrences were unprovoked (70%), with only 17% occurring at the time of AED withdrawal. There were three multivariate predictors of seizure recurrence: 1) acute postoperative seizures (APOS) in the first week (strong predictor of early recurrence; only 6/25 [24%] of those with APOS achieved seizure freedom vs 73/133 [55%] without APOS); 2) surgery on the left (predictive of late recurrence, >6 months; 28/73 [38%] seizure free vs 51/85 [60%] with surgery on the right), and 3) epilepsy duration >5 years (predictive of late recurrence; <1/3 of patients became seizure free if >5 years duration vs 2/3 if <5 years). The predictive value of epilepsy duration was true in both lesional and nonlesional cases. This was a robust finding in this series and was also found in another recent study of 70 patients undergoing frontal lobe resection in Brazil (9). These studies suggest that earlier surgical intervention is more likely to lead to long-term seizure freedom, as well as the greatest chance of employment. Simasathien et al. suggest (reasonably) that early recurrence is due to incomplete resection of the original focus, whereas late recurrence is due to more complex factors such as secondary epileptogenesis. Interruption of an epileptogenic network without removal of some its key nodes is another possible explanation for late recurrence. The negative prognostic value of early postoperative seizures has been seen in several other surgical series, including at least two large frontal lobe epilepsy studies (9, 10).
Limitations
The limitations are the retrospective nature, the moderate sample size (though very large for a single-center study), and the lack of data or discussion on completeness of resection and its importance or lack thereof.
Conclusions
All three studies are useful but retrospective and single-center, thus limited by modest sample sizes. None included magne-toencephalography, stereo-EEG, or analysis of high frequency oscillations, but future studies will. Surgical outcome is fairly good overall (with room for improvement) for these challenging patients with FLE in these expert tertiary centers, including in MRI-negative patients and those with cortical dysplasia, especially if complete resection is feasible. Shorter duration of epilepsy prior to surgery appears to correlate with a greater chance of seizure freedom, at least as best can be determined retrospectively and without randomization. Thus, as is true in temporal lobe epilepsy, if seizures are refractory in patients with frontal lobe epilepsy, surgery should be considered earlier rather than later, including in MRI-negative cases.