- Email: [email protected]
Outcome of epilepsy surgery in patients investigated with subdural electrodes
Epilepsy Research (2009) 85, 235—242
journal homepage: www.elsevier.com/locate/epilepsyres
Outcome of epilepsy surgery in patients investigated with subdural electrodes Keith W. MacDougall, Jorge G. Burneo, Richard S. McLachlan, David A. Steven ∗ Epilepsy Programme, Department of Clinical Neurological Sciences, Schulich School of Medicine and Dentistry, University of Western Ontario, University Hospital, London Health Sciences Centre, London, Ontario, Canada Received 5 January 2009; received in revised form 4 March 2009; accepted 15 March 2009 Available online 19 April 2009
KEYWORDS Epilepsy surgery; Intracranial electrodes; Subdural electrodes; Age
Summary Invasive intracranial electrodes (IE) are an important part of the work-up in many patients being considered for epilepsy surgery. Because IE are usually reserved for cases where seizure localization is ambiguous, one might expect that the eventual outcome of epilepsy surgery in these patients would be worse than in patients who did not require IE as part of their work-up. The purpose of this study was to specifically examine those patients who underwent insertion of subdural electrodes, to determine how many of these patients eventually underwent resective surgery of any type and to assess the eventual outcome. All cases admitted for subdural electrodes between January 2000 and June 2005 were reviewed. Surgical outcomes were reported using the Engel classification and a multivariate analysis was used to determine which factors were associated with successful surgery. 177 IE implantations were performed in 172 patients. Of these, 130 patients went on to have surgery. In the 113 of the 130 surgical patients in whom 1-year follow-up was available, 47% were seizure free at 1 year. Age was a major predictor of outcome with only 21% of patients over age 40 becoming seizure free with surgery compared to 58% in patients aged under 40 years (p = 0.0004). Other predictors of an Engel I outcome included having a temporal lobectomy or supplementary motor area resection. Good results from eventual resective surgery can be achieved in patients needing invasive recordings. Younger patients with temporal lobe epilepsy seem to have the highest likelihood of seizure freedom. © 2009 Elsevier B.V. All rights reserved.
Introduction ∗
Corresponding author at: Division of Neurosurgery, University Hospital, London Health Sciences Centre, 339 Windermere Road, Room A10-323, London, Ontario, Canada N6A 5A5. Tel.: +1 519 663 3297; fax: +1 519 663 3296. E-mail address: [email protected] (D.A. Steven).
The successful performance of resective epilepsy surgery depends on the pre-operative identification of the epileptogenic zone. In many patients, this can be achieved with a standard non-invasive pre-surgical work-up that includes a detailed history and physical examination, long-term scalp
0920-1211/$ — see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.eplepsyres.2009.03.014
236 video electroencephalography (EEG), magnetic resonance imaging (MRI), functional neuroimaging, and a neuropsychological evaluation (Sperling and Shewmon, 1997). However, even with this extensive work-up, it is not always possible to identify the epileptogenic zone, particularly in those patients with ambiguous or discordant scalp EEG data. In these patients, implanted intracranial electrodes (IE) can often provide further information and facilitate effective epilepsy surgery. This form of monitoring is generally carried out using subdural electrodes (Adelson et al., 1995; Cohen-Gadol and Spencer, 2003; Gonzalez-Feria and GarciaMarin, 1993; Lüders et al., 1989; Nair et al., 2008; Steven et al., 2007; Wyler et al., 1984), depth electrodes (Bancaud et al., 1969, 1970; Olivier et al., 1983; Pillay et al., 1992; Spencer, 1981) or a combination of both (Behrens et al., 1994; Spencer et al., 1990; van Veelen et al., 1990). At our center, most of the invasive monitoring is carried out using subdural strip electrodes. Although the outcomes for most types of resective epilepsy surgery are well known (Engel et al., 1993), less is known specifically about the outcome in those patients who require intracranial electrodes as part of their evaluation. Because IE are often utilized when the data from the non-invasive work-up is either ambiguous or discordant one could hypothesize that patients who require IE might be less likely to have a successful outcome following resective surgery. The purpose of this study was to specifically assess patients who underwent placement of subdural electrodes and to determine how many eventually underwent resective surgery of any type. In those who had surgery, we wished to examine the eventual outcome.
Methods All cases who underwent insertion of subdural electrodes between January 2000 and June 2005 were reviewed. Patients from this cohort who eventually had a resective surgical procedure and in whom at least 1 year of follow-up was available were included in the study. Relevant data was extracted from the patients’ charts. The indications for intracranial monitoring at our center include: bilateral, multifocal, normal or ambiguous EEG data in suspected cases of focal epilepsy; discordance between EEG and neuroimaging data; for mapping purposes when the seizure onset zone is suspected to be within or near an area of eloquent cortex; and in cases of suspected extra-temporal lobe onset with normal MRI findings. We do not routinely use IE in cases of temporal lobe epilepsy when the onset is clear on scalp EEG, even if the MRI is normal. The majority of our patients are investigated with subdural strip electrodes inserted through burr holes. We have found that this technique allows for excellent coverage of multiple locations without the need for craniotomy (Burneo et al., 2006b; Steven et al., 2007). This is particularly useful in cases where multilobar or bilateral coverage is desired. We have not routinely used subdural grids unless a detailed focal cortical localization is needed to plan surgery, particularly in peri-rolandic or tumor related cases. Our technique for inserting temporal subdural strip electrodes has been described elsewhere (Steven et al., 2007). Three lines of eight contacts each are inserted via a posterior temporal burr hole to cover the mesial, inferior and lateral surfaces of the temporal lobe. For frontal coverage, multiple subdural strips are inserted through a parasagittal precoronal burr hole. Typically, four 8—12 contact strips are placed over the lateral frontal convexity with the distal contacts of the more anterior electrodes placed such that they are beneath the orbitofrontal sur-
K.W. MacDougall et al. face. If midline coverage is desired, multiple bilaterally recording strips are placed along the falx. Electrodes covering the mesial or lateral parietal or occipital surfaces are similarly inserted via a posterior parasagittal or posterior temporal burr hole. In all cases, the placement of the strips is individualized to suit the hypothesis being tested. For example, when the seizure onset is suspected to be from the supplementary motor area (SMA), multiple strips are inserted over this area in addition to general frontal and parietal coverage. When used, subdural grids are inserted by craniotomy and depth electrodes are inserted with a Leksell stereotactic frame (Elekta AB, Stockholm, Sweden). Prior to August 2003, all subdural and depth electrodes were manufactured ‘‘in house.’’ Since January 2004, we have been using Ad-Tech subdural strip electrodes and grids (Ad-Tech Medical, Racine, WI). Following electrode implantation, patients were monitored with 24-h video EEG for a variable length of time that was predicated solely on the completeness of the invasive EEG data. The electrodes were removed after sufficient information had been gathered to make a decision regarding a resective operation. Resective surgery was usually offered if the non-invasive and invasive investigations revealed a surgically amenable epileptic focus. For the purposes of determining which variables might predict seizure freedom post-operatively, charts for all patients were reviewed to ascertain the location, size, type, and duration of electrode coverage. Information was also collected regarding the type of surgery performed, the resulting surgical pathology, and MRI findings. The seizure outcome from the resective procedure was classified using the Engel classification (Engel et al., 1993). A univariate analysis was carried out for each variable to test its association with an Engel I outcome. The unpaired Student’s t-test was used for the continuous variables and a Pearson 2 statistic was used for the categorical variables. For the categorical variables, if at least one cell on a 2 × 2 table had an expected value of less than 5, Fisher’s exact test was used to obtain p-values. The unadjusted odds ratio and 95% confidence intervals were obtained for each of the categorical variables. A significance level of ˛ = 0.05 was chosen for all statistical tests and all p-values in this report are two-tailed. Multivariate analysis using logistic regression was then used to obtain odds ratios for factors that predicted an Engel I outcome. For this analysis, standard demographic data (age and sex) as well as any risk factor identified as significant in the univariate analysis was included in the regression model. Adjusted odds ratios with 95% confidence intervals were used to summarize the results. All statistical analysis was completed with SAS® 8.02 (SAS Institute, Cary, NC).
Results One hundred and seventy-two patients underwent insertion of intracranial electrodes between January 2000 and June 2005. Eighty-four (49%) of the patients were females and the mean age at the time of implantation was 32.3 years (range 9—65). Five patients had electrodes inserted on two separate occasions in this period for a total of 177 cases. Multicontact subdural strip electrodes were placed in all 177 cases. The average number of strip electrodes was 8.2 (range 1—19) per case. In addition to strip electrodes, subdural grids were placed in 12 cases and depth electrodes in 5 cases. All combined, these implantations resulted in an average of 65.4 contacts per patient (range 16—174). Coverage was unilateral in 46 cases and bilateral in 131 cases. Temporal lobe electrodes were placed in 143 cases, frontal electrodes in 113 cases, parietal electrodes in 48 cases and occipital electrodes in 35 cases. The average duration of electrode implantation was 13.0 days (range 3—35). Com-
‘Outcome of epilepsy surgery’
237
plications related to the insertion of intracranial electrodes occurred in nine cases (5.1%). There were three (1.7%) infectious complications. Two patients had brain abscesses that required surgical drainage and intravenous antibiotics and one patient had subdural empyema that resulted in a vasculitis and middle cerebral artery (MCA) infarct. The latter complication resulted in a permanent expressive aphasia. Clinically significant hemorrhage occurred in two patients (1.1%). One patient had an epidural hematoma after insertion of a subdural grid and a second had an intracerebral hemorrhage after insertion of a subdural strip electrode over a previously resected brain tumor. Neither of these hemorrhages resulted in any permanent neurological sequelae. One patient suffered a cortical injury from a left temporal strip electrode that resulted in persistent mild word-finding difficulty. A transient third nerve palsy occurred in a single patient. In an additional two patients, a craniotomy wound infection occurred following a resective procedure performed with the subdural electrodes in situ. It was not possible to determine whether these infections were related to the electrode insertion or the subsequent craniotomy. There was no operative mortality. In the 167 patients who underwent a single implantation, resective surgery was performed in 123 patients (73.6%). In the five patients who underwent two implantations, three patients had surgery after both implantations, one had surgery after one of the implantations, and one did not have resective surgery after either implantation. Therefore a total of 172 patients underwent 177 implantations and of these 177 cases, 130 eventually underwent resective surgery and 47 did not. During the same time period 207 patients underwent resective epilepsy surgery with or without previous investigations with subdural electrodes. The reasons for not undergoing surgery are listed in Table 1, the most common reason being a bilateral or multifocal seizure onset.
Table 1 Reasons for not having resective epilepsy surgery in 47 cases investigated with intracranial electrodes. Reason
N
Multifocal/bilateral onset Patient decision Seizures not localizable Seizures arising from eloquent cortex Temporal origin with discordant amytal result Seizures stopped following implantation
27 (57%) 6 (13%) 6 (13%) 4 (8.5%) 3 (6.4%) 1 (2.1%)
The demographic data and the characteristics of the electrode implantation in surgical and non-surgical cases are presented in Table 2. The only statistically significant difference noted was that temporal electrodes were required more frequently in patients who did not have an eventual resective surgical procedure (94%) than those who did undergo surgery (76%). This may reflect a bilateral temporal lobe seizure origin in many of these patients. Table 3 lists the distribution of procedures in the 130 cases where surgery was performed. The most commonly performed surgical procedure was temporal lobectomy (47%) followed by focal cortical resection (40%) and SMA resection (10%). Note that in many cases, more than one type of surgery was performed during a single operation. For example, patients with a frontal corticectomy also occasionally had multiple subpial transections (MST) performed. Similarly, patients with posterior temporal neocortical seizures often had both an anterior temporal lobectomy and a posterior temporal corticectomy. In this situation, corticectomy was defined as the removal of cortex beyond the confines of a standard anterior temporal lobec-
Table 2 Demographics and nature of implantation in 177 cases investigated with intracranial electrodes; comparison between cases that subsequently had a resective surgical procedure and cases that did not. Non-surgical group
Surgical group
p-Value
N Age (years) Males MRI lesion present
47 34.4 25 (53%) 26 (55%)
130 31.5 65 (50%) 84 (65%)
0.18 0.7 0.26
Electrode type (# of patients) Strips Grid Depth
47 (100%) 2 (4.2%) 3 (6.4%)
130 (100%) 10 (7.7%) 2 (1.5%)
N/A 0.52 0.12
Number of Strips Number of electrode contacts
8.9 70.8
7.9 63.5
0.06 0.08
Coverage Bilateral Frontal Temporal Parietal Occipital
38 (81%) 30 (64%) 44 (94%) 12 (26%) 14 (30%)
93 (72%) 83 (64%) 99 (76%) 36 (28%) 27 (21%)
0.21 0.99 0.009 0.78 0.21
Duration of implantation (days)
13.8
12.7
0.28
238
K.W. MacDougall et al.
Table 3 Distribution of surgical procedures performed in 130 cases investigated with intracranial electrodes. MST: multiple subpial transections. Type of surgery
N
Temporal lobectomy Frontal lobectomy Supplementary motor resection Orbitofrontal resection Focal cortical resection Amygdalohippocampectomy Callosotomy MST Total
%
61 8 13 9 52 3 2 9
47 6 10 7 40 2 2 7
130
Table 4 Pathology encountered in the surgical specimens obtained from 130 cases where intracranial electrodes had been previously inserted. Pathology
N (%)
Hippocampal sclerosis Gliosis
29 (22%) 22 (17%)
Oligodendroglioma Low grade astrocytoma High grade astrocytoma Cavernoma Meningioma DNET
6 (4.6%) 6 (4.6%) 1 (0.7%) 2 (1.5%) 1 (0.7%) 1 (0.7%)
Malformations of cortical development Hemorrhage/contusion Normal Atrophy
17 (13%) 2 (1.5%) 29 (22%) 1 (0.7%)
Unavailable, N/A
13 (10%)
Total
130
tomy (Girvin, 1991). In terms of the resective procedure, 64 cases (49%) were strictly temporal and 66 (51%) were primarily extra-temporal. Pathological specimens were obtained in 117 cases and their diagnoses are presented in Table 4. Complications related to the resective procedure occurred in three cases (2.3%). One patient had a thalamic hemorrhage following a temporal lobectomy that resulted in a perma-
Table 5
nent aphasia. One patient had a cerebrospinal fluid leak that was repaired with re-suturing the wound and another patient had aseptic meningitis that responded to analgesics and steroids. As mentioned above, in two patients a craniotomy wound infection occurred following a resective procedure performed with the subdural electrodes in situ. It was not possible to determine whether these infections were related to the electrode insertion or the subsequent craniotomy. In all there were 14 complications (4.6%) in the 307 surgical procedures performed in this series. Only three (0.98%) of these complications resulted in permanent neurological sequelae. The outcome with respect to seizures in cases where a resection was performed is presented in Table 5. One-year follow-up was available in 113 of the 130 surgical cases (87%). Of the 17 in whom follow-up was unavailable, 13 were residents of another province of Canada or country and could not be contacted. Only 4 (3.4%) of the 117 inprovince patients were lost to follow-up. After 1 year, 47% of patients had an Engel I outcome. This effect persisted until at least the third year of follow-up where 44% of patients remained seizure free. Temporal resections (53%) were more likely to be associated with seizure freedom than extratemporal resections (41%), although the difference was not statistically significant (p = 0.21). We performed a univariate analysis of patient demographics, electrode placement, the presence or absence of a radiological lesion, type of surgery, and pathology to see which factors were associated with an Engel I outcome (Table 6). Although gender was not predictive, seizure-free patients were significantly younger than those that had persistent seizures. This effect was even more pronounced when age was categorized by decade (Fig. 1). In particular, only 21% of the 21 patients who were over the age of 40 at the time of electrode insertion were seizure free, compared to 58% of the patients aged 40 or less (p = 0.0004). There was no difference between the two groups with respect to the number of line electrodes, the number of electrode contacts, or the duration of implantation. Similarly, the placement of bilateral electrodes or the particular lobes over which the electrodes were placed made no difference. Patients who were investigated with subdural grids were much more likely to have persistent seizures than those without (p = 0.02), although this relationship was not statistically significant in the multivariate analysis (Table 7). Interestingly, the presence of a pre-operative radiological lesion, including hippocampal sclerosis, was not predictive of an Engel I outcome. Although temporal resections
Results of surgery for patients previously investigated with intracranial electrodes.
Surgical outcome Engel classification I II III IV Total
6 months 57 18 12 27 114
% 50 15.8 10.5 23.7 100
1 year 53 18 14 28 113
% 46.9 15.9 12.4 24.8 100
2 years 33 15 13 19 80
% 41.3 18.8 16.3 23.8 100
3 years 18 4 7 12 41
% 43.9 9.8 17.1 29.3 100
‘Outcome of epilepsy surgery’
239
Table 6 Comparison between seizure-free patients and those with recurrent seizures 1 year after epilepsy surgery in patients investigated with intracranial electrodes. Engel I (N = 53)
Engel ≥ 2 (N = 60)
Odds ratio
p
Demographics Mean age (years) Female sex Lesion on MRI
27.7 22 (42%) 35 (66%)
36.7 32 (53%) 42 (70%)
— 0.62 0.83
0.0002 0.21 0.65
Implantation Subdural grid Bilateral implantation Mean number of strips Mean number of contacts Duration of implantation (days)
1 (1.9%) 38 (72%) 8.1 63 12.6
9 (15%) 41 (68%) 7.4 63 12.4
0.11 1.17 — — —
0.01 0.71 0.16 0.99 0.91
Type of surgery Temporal lobectomy Frontal lobectomy SMA resection Orbitofrontal resection Any temporal resection Any extra-temporal resection
32 (60%) 2 (3.8%) 8 (15%) 5 (9.4%) 31 (58%) 22 (42%)
24 (40%) 5 (8.3%) 2 (3.3%) 3 (5.0%) 28 (47%) 32 (53%)
2.29 0.43 4.78 0.83 1.61 0.62
0.03 0.44 0.04 0.47 0.21 0.21
Pathology Hippocampal sclerosis Tumor Dysplasia Non-specific/normal
14 (26%) 1 (1.9%) 8 (15%) 20 (38%)
13 (22%) 2 (3.3%) 4 (6.7%) 18 (30%)
1.3 0.56 2.09 1.41
0.55 0.99 0.15 0.39
(53%) were more likely to be associated with seizure freedom than extra-temporal resections (41%), the difference was not statistically significant (p = 0.21). However, patients who underwent temporal lobectomy or resection of the SMA were more likely to be seizure free than patients who underwent other types of procedures. In the multivariate analysis younger age, temporal lobectomy and SMA resection all remained strongly associated with seizure freedom (Table 7).
Discussion In this series, the proportion of implanted patients who eventually underwent a resection was lower (73%) than that reported elsewhere. In a large study by Lee et al. resective surgery was offered to 180 of 183 (98%) patients after intracranial monitoring (Lee et al., 2004). Other reported series have a similarly high proportion of patients going on to surgery (84—100%) (Berg et al., 2003; Cukiert et al., 2001; Jennum et al., 1993; Lee et al., 2000; Murphy et al., 2002). Our lower resection rate may reflect the high proportion of patients with non-lesional epilepsy in the current series (38%). In addition, we have found that the complication rate of subdural strips inserted through burr holes to be very low (Burneo et al., 2006b) and, as a result, we may have a lower threshold for investigating potential surgical cases with IE. Finally, it is important to note that while a ‘‘negative’’ inves-
Table 7 Multivariate analysis using demographic data (age and sex) as well as risk factors identified as significant in the univariate analysis.
Figure 1 Percentage of patients seizure free at 1 year categorized by age at surgery. Error bars represent 95% confidence interval for the mean.
Variable Age (years) Male sex Temporal lobectomy SMA resection Subdural grid
Odds ratio (95% C.I.)
p
0.95 1.75 2.57 7.46 0.12
0.003 0.21 0.04 0.03 0.08
(0.92, (0.74, (1.03, (1.24, (0.01,
0.98) 4.13) 6.40) 44.7) 1.34)
240 tigation can be disappointing for the patient it nonetheless provides important information regarding the nature of their epilepsy and can serve to reassure them that they have been fully investigated and that resective surgery will not be an option. Intracranial electrodes are often reserved for cases where the non-invasive investigations, particularly the EEG, are discordant or ambiguous (Wyler et al., 1993). It is therefore not surprising that some investigators have found that the need for intracranial monitoring is associated with a lower likelihood of an Engel I outcome after epilepsy surgery (Tonini et al., 2004; Wyllie et al., 1998). On the other hand, in spite of the presumed complexity of cases requiring IE, we found that almost 50% of patients who underwent any resection and 60% of patients who had a temporal lobectomy were seizure free after surgery. Although these rates are somewhat lower than the results of epilepsy surgery in general (Engel et al., 1993; Téllez-Zenteno et al., 2005), they are comparable and a substantial proportion of these patients do become seizure free. Very few publications have focused exclusively on the seizure outcome in this group of patients. Jennum et al. found that 55% of extra-temporal and 70% of temporal cases were seizure free when guided by subdural electrodes, however it appears that the authors used subdural electrodes in all of their cases (Jennum et al., 1993). A smaller study by Cukiert et al. reported a 75% rate of seizure freedom in 16 adolescent and young adult patients with extra-temporal non-lesional epilepsy investigated with subdural grids (Cukiert et al., 2001). Because of the variability in the technique and indications for IE implantation between centers, a direct comparison of results may be misleading. Not surprisingly, we found that patients with temporal lobe epilepsy had the highest likelihood of seizure freedom. The strong association between SMA resection and a good outcome was somewhat surprising. Eight of our 10 patients with SMA resections in whom 1-year follow-up was available were seizure free. Although some authors have reported seizure freedom in up to 75% of patients (Jobst et al., 2000), more commonly lower success rates are reported (Aghakhani et al., 2004; Baumgartner et al., 1996; Olivier, 1996). One possible reason is that the resections in all 10 of our patients were restricted to the SMA, implying a very focal onset. Olivier reported an Engel I outcome in 22% of 28 patients who underwent resection of the SMA (Olivier, 1996). Although it is not clear in that study whether or not additional resection was carried out beyond the SMA, a subsequent publication from the same group reported that only one of six patients had their resection restricted to the SMA (Aghakhani et al., 2004). It is therefore possible that the higher success rate noted in the current series was related to selection of less complex cases. However, due to the small numbers of patients with SMA resections in this series, no firm conclusions should be drawn from this. The association between older age at surgery and poor outcome has been noted by some authors (Blume et al., 1994; Ficker et al., 1999; Jeong et al., 2005; Sirven et al., 2000), however others have not noted this relationship (Armon et al., 1996; Burneo et al., 2006a; Grivas et al., 2006; McIntosh et al., 2004; Salanova et al., 2002). While we noted an association when age was analyzed as a continuous variable, the nature of this association was
K.W. MacDougall et al. most evident when age was categorized by decade (Fig. 1). In particular, we found that patients older than 40 years had a dramatically lower likelihood of seizure freedom than patients aged 40 or less. This finding is very similar to results obtained by Cohen-Gadol et al. who noted that patients with cortical dysplasia aged under 18 and over 40 had a worse prognosis than patients with ages between these two parameters (Cohen-Gadol et al., 2004). They identified a quadratic relationship between the likelihood of seizure freedom and age, with the risk of surgical failure rising sharply at the extremes. When they categorized age as a continuous variable, no such association was noted. Although it is generally held that analyzing age as a continuous variable is preferable to dichotomization (Altman and Royston, 2006; Austin and Brunner, 2004; Royston et al., 2006), it is possible that a steep increase in risk over a certain age might be best appreciated if such a cut-off can be identified (Mazumdar and Glassman, 2000). In fact, a number of investigators have found that patients over certain age threshold have either a trend towards or statistically significant decreased likelihood of seizure freedom following epilepsy surgery (Boling et al., 2001; Cohen-Gadol et al., 2004; McLachlan et al., 1992). However, even in those studies, 30—75% of patients in the older age groups were seizure free and 75—83% had a greater than 90% reduction in seizure frequency at 1 year. In contrast, we noted that in this group that required IE only 21% of patients aged greater than 40 were seizure free and only 42% had a worthwhile improvement (Engel Grades I—III) 1 year after surgery. While our patient population as a whole had a slightly lower likelihood of seizure freedom than epilepsy surgery patients in general, the outcome in the older patients was disproportionally poorer than previously reported. While the reason for this is unclear, it is possible that older patients with complex epilepsies, by virtue of a longer duration of epilepsy, might me more prone to develop multifocality and therefore not respond to focal resective surgery. On the other hand, because both age and complexity may affect outcome independently, their combined risk may be additive and therefore our observation may simply represent the combined risk of these two variables. In conclusion, intracranial electrodes are an essential component of the pre-surgical investigation of many patients undergoing resective epilepsy surgery. Although the results of eventual resective epilepsy surgery are somewhat poorer than patients who do not require IE, good results can be achieved in almost 50% of patients in general and 60% of patients with temporal lobe epilepsy. Patients over age 40 seem to be an exception, as it appears that they have a significantly lower likelihood of surgical success. Although the latter finding requires further investigation and confirmation, it is another line of evidence that supports the earlier surgical management of focal epilepsies.
References Adelson, P.D., Black, P.M., Madsen, J.R., Kramer, U., Rockoff, M.A., Riviello, J.J., Helmers, S.L., Mikati, M., Holmes, G.L., 1995. Use of subdural grids and strip electrodes to identify a seizure focus in children. Pediatr. Neurosurg. 22, 174—180. Aghakhani, Y., Rosati, A., Olivier, A., Gotman, J., Andermann, F., Dubeau, F., 2004. The predictive localizing value of tonic limb
‘Outcome of epilepsy surgery’ posturing in supplementary sensorimotor seizures. Neurology 62, 2256—2261. Altman, D.G., Royston, P., 2006. The cost of dichotomising continuous variables. BMJ 332, 1080. Armon, C., Radtke, R.A., Friedman, A.H., Dawson, D.V., 1996. Predictors of outcome of epilepsy surgery: multivariate analysis with validation. Epilepsia 37, 814—821. Austin, P.C., Brunner, L.J., 2004. Inflation of the type I error rate when a continuous confounding variable is categorized in logistic regression analyses. Stat. Med. 23, 1159—1178. Bancaud, J., Angelergues, R., Bernouilli, C., Bonis, A., BordasFerrer, M., Bresson, M., Buser, P., Covello, L., Morel, P., Szikla, G., Takeda, A., Talairach, J., 1969. Functional stereotaxic exploration (stereo-electroencephalography) in epilepsies. Rev. Neurol. (Paris) 120, 448. Bancaud, J., Angelergues, R., Bernouilli, C., Bonis, A., BordasFerrer, M., Bresson, M., Buser, P., Covello, L., Morel, P., Szikla, G., Takeda, A., Talairach, J., 1970. Functional stereotaxic exploration (SEEG) of epilepsy. Electroencephalogr. Clin. Neurophysiol. 28, 85—86. Baumgartner, C., Flint, R., Tuxhorn, I., Van Ness, P.C., Kosalko, J., Olbrich, A., Almer, G., Novak, K., Lüders, H.O., 1996. Supplementary motor area seizures: propagation pathways as studied with invasive recordings. Neurology 46, 508—514. Behrens, E., Zentner, J., van Roost, D., Hufnagel, A., Elger, C.E., Schramm, J., 1994. Subdural and depth electrodes in the presurgical evaluation of epilepsy. Acta Neurochir. (Wien) 128, 84—87. Berg, A.T., Vickrey, B.G., Langfitt, J.T., Sperling, M.R., Walczak, T.S., Shinnar, S., Bazil, C.W., Pacia, S.V., Spencer, S.S., 2003. The multicenter study of epilepsy surgery: recruitment and selection for surgery. Epilepsia 44, 1425—1433. Blume, W.T., Desai, H.B., Girvin, J.P., McLachlan, R.S., Lemieux, J.F., 1994. Effectiveness of temporal lobectomy measured by yearly follow-up and multivariate analysis. J. Epilepsy 7, 203—214. Boling, W., Andermann, F., Reutens, D., Dubeau, F., Caporicci, L., Olivier, A., 2001. Surgery for temporal lobe epilepsy in older patients. J. Neurosurg. 95, 242—248. Burneo, J.G., Black, L., Martin, R., Devinsky, O., Pacia, S., Faught, E., Vasquez, B., Knowlton, R.C., Luciano, D., Doyle, W., Najjar, S., Kuzniecky, R.I., 2006a. Race/ethnicity, sex, and socioeconomic status as predictors of outcome after surgery for temporal lobe epilepsy. Arch. Neurol. 63, 1106—1110. Burneo, J.G., Steven, D.A., McLachlan, R.S., Parrent, A.G., 2006b. Morbidity associated with the use of intracranial electrodes for epilepsy surgery. Can. J. Neurol. Sci. 33, 223—227. Cohen-Gadol, A.A., Ozduman, K., Bronen, R.A., Kim, J.H., Spencer, D.D., 2004. Long-term outcome after epilepsy surgery for focal cortical dysplasia. J. Neurosurg. 101, 55—65. Cohen-Gadol, A.A., Spencer, D.D., 2003. Use of an anteromedial subdural strip electrode in the evaluation of medial temporal lobe epilepsy. Technical note. J. Neurosurg. 99, 921—923. Cukiert, A., Buratini, J.A., Machado, E., Sousa, A., Vieira, J.O., Argentoni, M., Forster, C., Baldauf, C., 2001. Results of surgery in patients with refractory extratemporal epilepsy with normal or nonlocalizing magnetic resonance findings investigated with subdural grids. Epilepsia 42, 889—894. Engel Jr., J., Van Ness, P.C., Rasmussen, T.B., Ojemann, L.M., 1993. Outcome with respect to epileptic seizures. In: Engel Jr., J. (Ed.), Surgical Treatment of the Epilepsies. Raven Press, New York, pp. 609—621. Ficker, D.M., So, E.L., Mosewich, R.K., Radhakrishnan, K., Cascino, G.D., Sharbrough, F.W., 1999. Improvement and deterioration of seizure control during the postsurgical course of epilepsy surgery patients. Epilepsia 40, 62—67. Girvin, J.P., 1991. Temporal lobectomy. In: Apuzzo, M.L.J. (Ed.), Neurosurgical Aspects of Epilepsy. American Association of Neurological Surgeons, Park Ridge, pp. 157—170.
241 Gonzalez-Feria, L., Garcia-Marin, V., 1993. Sub-occipital approach for implantation of recording multi-electrodes over the medial surface of the temporal lobe. Acta Neurochir. Suppl. (Wien) 58, 190—192. Grivas, A., Schramm, J., Kral, T., von Lehe, M., Helmstaedter, C., Elger, C.E., Clusmann, H., 2006. Surgical treatment for refractory temporal lobe epilepsy in the elderly: seizure outcome and neuropsychological sequels compared with a younger cohort. Epilepsia 47, 1364—1372. Jennum, P., Dhuna, A., Davies, K., Fiol, M., Maxwell, R., 1993. Outcome of resective surgery for intractable partial epilepsy guided by subdural electrode arrays. Acta Neurol. Scand. 87, 434— 437. Jeong, S.W., Lee, S.K., Hong, K.S., Kim, K.K., Chung, C.K., Kim, H., 2005. Prognostic factors for the surgery for mesial temporal lobe epilepsy: longitudinal analysis. Epilepsia 46, 1273—1279. Jobst, B.C., Siegel, A.M., Thadani, V.M., Roberts, D.W., Rhodes, H.C., Williamson, P.D., 2000. Intractable seizures of frontal lobe origin: clinical characteristics, localizing signs, and results of surgery. Epilepsia 41, 1139—1152. Lee, S.K., Kim, K.K., Nam, H., Oh, J.B., Yun, C.H., Chung, C.K., 2004. Adding or repositioning intracranial electrodes during presurgical assessment of neocortical epilepsy: electrographic seizure pattern and surgical outcome. J. Neurosurg. 100, 463—471. Lee, W.S., Lee, J.K., Lee, S.A., Kang, J.K., Ko, T.S., 2000. Complications and results of subdural grid electrode implantation in epilepsy surgery. Surg. Neurol. 54, 346—351. Lüders, H., Hahn, J., Lesser, R.P., Dinner, D.S., Morris III, H.H., Wyllie, E., Friedman, L., Friedman, D., Skipper, G., 1989. Basal temporal subdural electrodes in the evaluation of patients with intractable epilepsy. Epilepsia 30, 131—142. Mazumdar, M., Glassman, J.R., 2000. Categorizing a prognostic variable: review of methods, code for easy implementation and applications to decision-making about cancer treatments. Stat. Med. 19, 113—132. McIntosh, A.M., Kalnins, R.M., Mitchell, L.A., Fabinyi, G.C., Briellmann, R.S., Berkovic, S.F., 2004. Temporal lobectomy: long-term seizure outcome, late recurrence and risks for seizure recurrence. Brain 127, 2018—2030. McLachlan, R.S., Chovaz, C.J., Blume, W.T., Girvin, J.P., 1992. Temporal lobectomy for intractable epilepsy in patients over age 45 years. Neurology 42, 662—665. Murphy, M.A., O’Brien, T.J., Cook, M.J., 2002. Insertion of depth electrodes with or without subdural grids using frameless stereotactic guidance systems—–technique and outcome. Br. J. Neurosurg. 16, 119—125. Nair, D.R., Burgess, R., McIntyre, C.C., Lüders, H., 2008. Chronic subdural electrodes in the management of epilepsy. Clin. Neurophysiol. 119, 11—28. Olivier, A., 1996. Surgical strategies for patients with supplementary sensorimotor area epilepsy. The Montreal experience. Adv. Neurol. 70, 429—443. Olivier, A., Gloor, P., Quesney, L.F., Andermann, F., 1983. The indications for and the role of depth electrode recording in epilepsy. Appl Neurophysiol 46, 33—36. Pillay, P.K., Barnett, G., Awad, I., 1992. MRI-guided stereotactic placement of depth electrodes in temporal lobe epilepsy. Br. J. Neurosurg. 6, 47—53. Royston, P., Altman, D.G., Sauerbrei, W., 2006. Dichotomizing continuous predictors in multiple regression: a bad idea. Stat. Med. 25, 127—141. Salanova, V., Markand, O., Worth, R., 2002. Temporal lobe epilepsy surgery: outcome, complications, and late mortality rate in 215 patients. Epilepsia 43, 170—174. Sirven, J.I., Malamut, B.L., O’Connor, M.J., Sperling, M.R., 2000. Temporal lobectomy outcome in older versus younger adults. Neurology 54, 2166—2170.
242 Spencer, S.S., 1981. Depth electroencephalography in selection of refractory epilepsy for surgery. Ann. Neurol. 9, 207—214. Spencer, S.S., Spencer, D.D., Williamson, P.D., Mattson, R., 1990. Combined depth and subdural electrode investigation in uncontrolled epilepsy. Neurology 40, 74—79. Sperling, M.R., Shewmon, D.A., 1997. General principles for presurgical evaluation. In: Engel, J.J., Pedley, T.A. (Eds.), Epilepsy: A Comprehensive Textbook. Lippincott-Raven, Philadelphia, pp. 1697—1705. Steven, D.A., Andrade-Souza, Y.M., Burneo, J.G., McLachlan, R.S., Parrent, A.G., 2007. Insertion of subdural strip electrodes for the investigation of temporal lobe epilepsy. Technical note. J. Neurosurg. 106, 1102—1106. Téllez-Zenteno, J.F., Dhar, R., Wiebe, S., 2005. Long-term seizure outcomes following epilepsy surgery: a systematic review and meta-analysis. Brain 128, 1188—1198. Tonini, C., Beghi, E., Berg, A.T., Bogliun, G., Giordano, L., Newton, R.W., Tetto, A., Vitelli, E., Vitezic, D., Wiebe, S., 2004. Predic-
K.W. MacDougall et al. tors of epilepsy surgery outcome: a meta-analysis. Epilepsy Res. 62, 75—87. van Veelen, C.W., Debets, R.M., van Huffelen, A.C., van Emde Boas, W., Binnie, C.D., Storm van Leeuwen, W., Velis, D.N., van Dieren, A., 1990. Combined use of subdural and intracerebral electrodes in preoperative evaluation of epilepsy. Neurosurgery 26, 93—101. Wyler, A.R., Ojemann, G.A., Lettich, E., Ward Jr., A.A., 1984. Subdural strip electrodes for localizing epileptogenic foci. J. Neurosurg. 60, 1195—1200. Wyler, A.R., Wilkus, R.J., Blume, W.T., 1993. Strip Electrodes. In: Engel, J. (Ed.), Surgical Treatment of the Epilepsies. Raven Press, New York, pp. 387—397. Wyllie, E., Comair, Y.G., Kotagal, P., Bulacio, J., Bingaman, W., Ruggieri, P., 1998. Seizure outcome after epilepsy surgery in children and adolescents. Ann. Neurol. 44, 740—748.
journal homepage: www.elsevier.com/locate/epilepsyres
Outcome of epilepsy surgery in patients investigated with subdural electrodes Keith W. MacDougall, Jorge G. Burneo, Richard S. McLachlan, David A. Steven ∗ Epilepsy Programme, Department of Clinical Neurological Sciences, Schulich School of Medicine and Dentistry, University of Western Ontario, University Hospital, London Health Sciences Centre, London, Ontario, Canada Received 5 January 2009; received in revised form 4 March 2009; accepted 15 March 2009 Available online 19 April 2009
KEYWORDS Epilepsy surgery; Intracranial electrodes; Subdural electrodes; Age
Summary Invasive intracranial electrodes (IE) are an important part of the work-up in many patients being considered for epilepsy surgery. Because IE are usually reserved for cases where seizure localization is ambiguous, one might expect that the eventual outcome of epilepsy surgery in these patients would be worse than in patients who did not require IE as part of their work-up. The purpose of this study was to specifically examine those patients who underwent insertion of subdural electrodes, to determine how many of these patients eventually underwent resective surgery of any type and to assess the eventual outcome. All cases admitted for subdural electrodes between January 2000 and June 2005 were reviewed. Surgical outcomes were reported using the Engel classification and a multivariate analysis was used to determine which factors were associated with successful surgery. 177 IE implantations were performed in 172 patients. Of these, 130 patients went on to have surgery. In the 113 of the 130 surgical patients in whom 1-year follow-up was available, 47% were seizure free at 1 year. Age was a major predictor of outcome with only 21% of patients over age 40 becoming seizure free with surgery compared to 58% in patients aged under 40 years (p = 0.0004). Other predictors of an Engel I outcome included having a temporal lobectomy or supplementary motor area resection. Good results from eventual resective surgery can be achieved in patients needing invasive recordings. Younger patients with temporal lobe epilepsy seem to have the highest likelihood of seizure freedom. © 2009 Elsevier B.V. All rights reserved.
Introduction ∗
Corresponding author at: Division of Neurosurgery, University Hospital, London Health Sciences Centre, 339 Windermere Road, Room A10-323, London, Ontario, Canada N6A 5A5. Tel.: +1 519 663 3297; fax: +1 519 663 3296. E-mail address: [email protected] (D.A. Steven).
The successful performance of resective epilepsy surgery depends on the pre-operative identification of the epileptogenic zone. In many patients, this can be achieved with a standard non-invasive pre-surgical work-up that includes a detailed history and physical examination, long-term scalp
0920-1211/$ — see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.eplepsyres.2009.03.014
236 video electroencephalography (EEG), magnetic resonance imaging (MRI), functional neuroimaging, and a neuropsychological evaluation (Sperling and Shewmon, 1997). However, even with this extensive work-up, it is not always possible to identify the epileptogenic zone, particularly in those patients with ambiguous or discordant scalp EEG data. In these patients, implanted intracranial electrodes (IE) can often provide further information and facilitate effective epilepsy surgery. This form of monitoring is generally carried out using subdural electrodes (Adelson et al., 1995; Cohen-Gadol and Spencer, 2003; Gonzalez-Feria and GarciaMarin, 1993; Lüders et al., 1989; Nair et al., 2008; Steven et al., 2007; Wyler et al., 1984), depth electrodes (Bancaud et al., 1969, 1970; Olivier et al., 1983; Pillay et al., 1992; Spencer, 1981) or a combination of both (Behrens et al., 1994; Spencer et al., 1990; van Veelen et al., 1990). At our center, most of the invasive monitoring is carried out using subdural strip electrodes. Although the outcomes for most types of resective epilepsy surgery are well known (Engel et al., 1993), less is known specifically about the outcome in those patients who require intracranial electrodes as part of their evaluation. Because IE are often utilized when the data from the non-invasive work-up is either ambiguous or discordant one could hypothesize that patients who require IE might be less likely to have a successful outcome following resective surgery. The purpose of this study was to specifically assess patients who underwent placement of subdural electrodes and to determine how many eventually underwent resective surgery of any type. In those who had surgery, we wished to examine the eventual outcome.
Methods All cases who underwent insertion of subdural electrodes between January 2000 and June 2005 were reviewed. Patients from this cohort who eventually had a resective surgical procedure and in whom at least 1 year of follow-up was available were included in the study. Relevant data was extracted from the patients’ charts. The indications for intracranial monitoring at our center include: bilateral, multifocal, normal or ambiguous EEG data in suspected cases of focal epilepsy; discordance between EEG and neuroimaging data; for mapping purposes when the seizure onset zone is suspected to be within or near an area of eloquent cortex; and in cases of suspected extra-temporal lobe onset with normal MRI findings. We do not routinely use IE in cases of temporal lobe epilepsy when the onset is clear on scalp EEG, even if the MRI is normal. The majority of our patients are investigated with subdural strip electrodes inserted through burr holes. We have found that this technique allows for excellent coverage of multiple locations without the need for craniotomy (Burneo et al., 2006b; Steven et al., 2007). This is particularly useful in cases where multilobar or bilateral coverage is desired. We have not routinely used subdural grids unless a detailed focal cortical localization is needed to plan surgery, particularly in peri-rolandic or tumor related cases. Our technique for inserting temporal subdural strip electrodes has been described elsewhere (Steven et al., 2007). Three lines of eight contacts each are inserted via a posterior temporal burr hole to cover the mesial, inferior and lateral surfaces of the temporal lobe. For frontal coverage, multiple subdural strips are inserted through a parasagittal precoronal burr hole. Typically, four 8—12 contact strips are placed over the lateral frontal convexity with the distal contacts of the more anterior electrodes placed such that they are beneath the orbitofrontal sur-
K.W. MacDougall et al. face. If midline coverage is desired, multiple bilaterally recording strips are placed along the falx. Electrodes covering the mesial or lateral parietal or occipital surfaces are similarly inserted via a posterior parasagittal or posterior temporal burr hole. In all cases, the placement of the strips is individualized to suit the hypothesis being tested. For example, when the seizure onset is suspected to be from the supplementary motor area (SMA), multiple strips are inserted over this area in addition to general frontal and parietal coverage. When used, subdural grids are inserted by craniotomy and depth electrodes are inserted with a Leksell stereotactic frame (Elekta AB, Stockholm, Sweden). Prior to August 2003, all subdural and depth electrodes were manufactured ‘‘in house.’’ Since January 2004, we have been using Ad-Tech subdural strip electrodes and grids (Ad-Tech Medical, Racine, WI). Following electrode implantation, patients were monitored with 24-h video EEG for a variable length of time that was predicated solely on the completeness of the invasive EEG data. The electrodes were removed after sufficient information had been gathered to make a decision regarding a resective operation. Resective surgery was usually offered if the non-invasive and invasive investigations revealed a surgically amenable epileptic focus. For the purposes of determining which variables might predict seizure freedom post-operatively, charts for all patients were reviewed to ascertain the location, size, type, and duration of electrode coverage. Information was also collected regarding the type of surgery performed, the resulting surgical pathology, and MRI findings. The seizure outcome from the resective procedure was classified using the Engel classification (Engel et al., 1993). A univariate analysis was carried out for each variable to test its association with an Engel I outcome. The unpaired Student’s t-test was used for the continuous variables and a Pearson 2 statistic was used for the categorical variables. For the categorical variables, if at least one cell on a 2 × 2 table had an expected value of less than 5, Fisher’s exact test was used to obtain p-values. The unadjusted odds ratio and 95% confidence intervals were obtained for each of the categorical variables. A significance level of ˛ = 0.05 was chosen for all statistical tests and all p-values in this report are two-tailed. Multivariate analysis using logistic regression was then used to obtain odds ratios for factors that predicted an Engel I outcome. For this analysis, standard demographic data (age and sex) as well as any risk factor identified as significant in the univariate analysis was included in the regression model. Adjusted odds ratios with 95% confidence intervals were used to summarize the results. All statistical analysis was completed with SAS® 8.02 (SAS Institute, Cary, NC).
Results One hundred and seventy-two patients underwent insertion of intracranial electrodes between January 2000 and June 2005. Eighty-four (49%) of the patients were females and the mean age at the time of implantation was 32.3 years (range 9—65). Five patients had electrodes inserted on two separate occasions in this period for a total of 177 cases. Multicontact subdural strip electrodes were placed in all 177 cases. The average number of strip electrodes was 8.2 (range 1—19) per case. In addition to strip electrodes, subdural grids were placed in 12 cases and depth electrodes in 5 cases. All combined, these implantations resulted in an average of 65.4 contacts per patient (range 16—174). Coverage was unilateral in 46 cases and bilateral in 131 cases. Temporal lobe electrodes were placed in 143 cases, frontal electrodes in 113 cases, parietal electrodes in 48 cases and occipital electrodes in 35 cases. The average duration of electrode implantation was 13.0 days (range 3—35). Com-
‘Outcome of epilepsy surgery’
237
plications related to the insertion of intracranial electrodes occurred in nine cases (5.1%). There were three (1.7%) infectious complications. Two patients had brain abscesses that required surgical drainage and intravenous antibiotics and one patient had subdural empyema that resulted in a vasculitis and middle cerebral artery (MCA) infarct. The latter complication resulted in a permanent expressive aphasia. Clinically significant hemorrhage occurred in two patients (1.1%). One patient had an epidural hematoma after insertion of a subdural grid and a second had an intracerebral hemorrhage after insertion of a subdural strip electrode over a previously resected brain tumor. Neither of these hemorrhages resulted in any permanent neurological sequelae. One patient suffered a cortical injury from a left temporal strip electrode that resulted in persistent mild word-finding difficulty. A transient third nerve palsy occurred in a single patient. In an additional two patients, a craniotomy wound infection occurred following a resective procedure performed with the subdural electrodes in situ. It was not possible to determine whether these infections were related to the electrode insertion or the subsequent craniotomy. There was no operative mortality. In the 167 patients who underwent a single implantation, resective surgery was performed in 123 patients (73.6%). In the five patients who underwent two implantations, three patients had surgery after both implantations, one had surgery after one of the implantations, and one did not have resective surgery after either implantation. Therefore a total of 172 patients underwent 177 implantations and of these 177 cases, 130 eventually underwent resective surgery and 47 did not. During the same time period 207 patients underwent resective epilepsy surgery with or without previous investigations with subdural electrodes. The reasons for not undergoing surgery are listed in Table 1, the most common reason being a bilateral or multifocal seizure onset.
Table 1 Reasons for not having resective epilepsy surgery in 47 cases investigated with intracranial electrodes. Reason
N
Multifocal/bilateral onset Patient decision Seizures not localizable Seizures arising from eloquent cortex Temporal origin with discordant amytal result Seizures stopped following implantation
27 (57%) 6 (13%) 6 (13%) 4 (8.5%) 3 (6.4%) 1 (2.1%)
The demographic data and the characteristics of the electrode implantation in surgical and non-surgical cases are presented in Table 2. The only statistically significant difference noted was that temporal electrodes were required more frequently in patients who did not have an eventual resective surgical procedure (94%) than those who did undergo surgery (76%). This may reflect a bilateral temporal lobe seizure origin in many of these patients. Table 3 lists the distribution of procedures in the 130 cases where surgery was performed. The most commonly performed surgical procedure was temporal lobectomy (47%) followed by focal cortical resection (40%) and SMA resection (10%). Note that in many cases, more than one type of surgery was performed during a single operation. For example, patients with a frontal corticectomy also occasionally had multiple subpial transections (MST) performed. Similarly, patients with posterior temporal neocortical seizures often had both an anterior temporal lobectomy and a posterior temporal corticectomy. In this situation, corticectomy was defined as the removal of cortex beyond the confines of a standard anterior temporal lobec-
Table 2 Demographics and nature of implantation in 177 cases investigated with intracranial electrodes; comparison between cases that subsequently had a resective surgical procedure and cases that did not. Non-surgical group
Surgical group
p-Value
N Age (years) Males MRI lesion present
47 34.4 25 (53%) 26 (55%)
130 31.5 65 (50%) 84 (65%)
0.18 0.7 0.26
Electrode type (# of patients) Strips Grid Depth
47 (100%) 2 (4.2%) 3 (6.4%)
130 (100%) 10 (7.7%) 2 (1.5%)
N/A 0.52 0.12
Number of Strips Number of electrode contacts
8.9 70.8
7.9 63.5
0.06 0.08
Coverage Bilateral Frontal Temporal Parietal Occipital
38 (81%) 30 (64%) 44 (94%) 12 (26%) 14 (30%)
93 (72%) 83 (64%) 99 (76%) 36 (28%) 27 (21%)
0.21 0.99 0.009 0.78 0.21
Duration of implantation (days)
13.8
12.7
0.28
238
K.W. MacDougall et al.
Table 3 Distribution of surgical procedures performed in 130 cases investigated with intracranial electrodes. MST: multiple subpial transections. Type of surgery
N
Temporal lobectomy Frontal lobectomy Supplementary motor resection Orbitofrontal resection Focal cortical resection Amygdalohippocampectomy Callosotomy MST Total
%
61 8 13 9 52 3 2 9
47 6 10 7 40 2 2 7
130
Table 4 Pathology encountered in the surgical specimens obtained from 130 cases where intracranial electrodes had been previously inserted. Pathology
N (%)
Hippocampal sclerosis Gliosis
29 (22%) 22 (17%)
Oligodendroglioma Low grade astrocytoma High grade astrocytoma Cavernoma Meningioma DNET
6 (4.6%) 6 (4.6%) 1 (0.7%) 2 (1.5%) 1 (0.7%) 1 (0.7%)
Malformations of cortical development Hemorrhage/contusion Normal Atrophy
17 (13%) 2 (1.5%) 29 (22%) 1 (0.7%)
Unavailable, N/A
13 (10%)
Total
130
tomy (Girvin, 1991). In terms of the resective procedure, 64 cases (49%) were strictly temporal and 66 (51%) were primarily extra-temporal. Pathological specimens were obtained in 117 cases and their diagnoses are presented in Table 4. Complications related to the resective procedure occurred in three cases (2.3%). One patient had a thalamic hemorrhage following a temporal lobectomy that resulted in a perma-
Table 5
nent aphasia. One patient had a cerebrospinal fluid leak that was repaired with re-suturing the wound and another patient had aseptic meningitis that responded to analgesics and steroids. As mentioned above, in two patients a craniotomy wound infection occurred following a resective procedure performed with the subdural electrodes in situ. It was not possible to determine whether these infections were related to the electrode insertion or the subsequent craniotomy. In all there were 14 complications (4.6%) in the 307 surgical procedures performed in this series. Only three (0.98%) of these complications resulted in permanent neurological sequelae. The outcome with respect to seizures in cases where a resection was performed is presented in Table 5. One-year follow-up was available in 113 of the 130 surgical cases (87%). Of the 17 in whom follow-up was unavailable, 13 were residents of another province of Canada or country and could not be contacted. Only 4 (3.4%) of the 117 inprovince patients were lost to follow-up. After 1 year, 47% of patients had an Engel I outcome. This effect persisted until at least the third year of follow-up where 44% of patients remained seizure free. Temporal resections (53%) were more likely to be associated with seizure freedom than extratemporal resections (41%), although the difference was not statistically significant (p = 0.21). We performed a univariate analysis of patient demographics, electrode placement, the presence or absence of a radiological lesion, type of surgery, and pathology to see which factors were associated with an Engel I outcome (Table 6). Although gender was not predictive, seizure-free patients were significantly younger than those that had persistent seizures. This effect was even more pronounced when age was categorized by decade (Fig. 1). In particular, only 21% of the 21 patients who were over the age of 40 at the time of electrode insertion were seizure free, compared to 58% of the patients aged 40 or less (p = 0.0004). There was no difference between the two groups with respect to the number of line electrodes, the number of electrode contacts, or the duration of implantation. Similarly, the placement of bilateral electrodes or the particular lobes over which the electrodes were placed made no difference. Patients who were investigated with subdural grids were much more likely to have persistent seizures than those without (p = 0.02), although this relationship was not statistically significant in the multivariate analysis (Table 7). Interestingly, the presence of a pre-operative radiological lesion, including hippocampal sclerosis, was not predictive of an Engel I outcome. Although temporal resections
Results of surgery for patients previously investigated with intracranial electrodes.
Surgical outcome Engel classification I II III IV Total
6 months 57 18 12 27 114
% 50 15.8 10.5 23.7 100
1 year 53 18 14 28 113
% 46.9 15.9 12.4 24.8 100
2 years 33 15 13 19 80
% 41.3 18.8 16.3 23.8 100
3 years 18 4 7 12 41
% 43.9 9.8 17.1 29.3 100
‘Outcome of epilepsy surgery’
239
Table 6 Comparison between seizure-free patients and those with recurrent seizures 1 year after epilepsy surgery in patients investigated with intracranial electrodes. Engel I (N = 53)
Engel ≥ 2 (N = 60)
Odds ratio
p
Demographics Mean age (years) Female sex Lesion on MRI
27.7 22 (42%) 35 (66%)
36.7 32 (53%) 42 (70%)
— 0.62 0.83
0.0002 0.21 0.65
Implantation Subdural grid Bilateral implantation Mean number of strips Mean number of contacts Duration of implantation (days)
1 (1.9%) 38 (72%) 8.1 63 12.6
9 (15%) 41 (68%) 7.4 63 12.4
0.11 1.17 — — —
0.01 0.71 0.16 0.99 0.91
Type of surgery Temporal lobectomy Frontal lobectomy SMA resection Orbitofrontal resection Any temporal resection Any extra-temporal resection
32 (60%) 2 (3.8%) 8 (15%) 5 (9.4%) 31 (58%) 22 (42%)
24 (40%) 5 (8.3%) 2 (3.3%) 3 (5.0%) 28 (47%) 32 (53%)
2.29 0.43 4.78 0.83 1.61 0.62
0.03 0.44 0.04 0.47 0.21 0.21
Pathology Hippocampal sclerosis Tumor Dysplasia Non-specific/normal
14 (26%) 1 (1.9%) 8 (15%) 20 (38%)
13 (22%) 2 (3.3%) 4 (6.7%) 18 (30%)
1.3 0.56 2.09 1.41
0.55 0.99 0.15 0.39
(53%) were more likely to be associated with seizure freedom than extra-temporal resections (41%), the difference was not statistically significant (p = 0.21). However, patients who underwent temporal lobectomy or resection of the SMA were more likely to be seizure free than patients who underwent other types of procedures. In the multivariate analysis younger age, temporal lobectomy and SMA resection all remained strongly associated with seizure freedom (Table 7).
Discussion In this series, the proportion of implanted patients who eventually underwent a resection was lower (73%) than that reported elsewhere. In a large study by Lee et al. resective surgery was offered to 180 of 183 (98%) patients after intracranial monitoring (Lee et al., 2004). Other reported series have a similarly high proportion of patients going on to surgery (84—100%) (Berg et al., 2003; Cukiert et al., 2001; Jennum et al., 1993; Lee et al., 2000; Murphy et al., 2002). Our lower resection rate may reflect the high proportion of patients with non-lesional epilepsy in the current series (38%). In addition, we have found that the complication rate of subdural strips inserted through burr holes to be very low (Burneo et al., 2006b) and, as a result, we may have a lower threshold for investigating potential surgical cases with IE. Finally, it is important to note that while a ‘‘negative’’ inves-
Table 7 Multivariate analysis using demographic data (age and sex) as well as risk factors identified as significant in the univariate analysis.
Figure 1 Percentage of patients seizure free at 1 year categorized by age at surgery. Error bars represent 95% confidence interval for the mean.
Variable Age (years) Male sex Temporal lobectomy SMA resection Subdural grid
Odds ratio (95% C.I.)
p
0.95 1.75 2.57 7.46 0.12
0.003 0.21 0.04 0.03 0.08
(0.92, (0.74, (1.03, (1.24, (0.01,
0.98) 4.13) 6.40) 44.7) 1.34)
240 tigation can be disappointing for the patient it nonetheless provides important information regarding the nature of their epilepsy and can serve to reassure them that they have been fully investigated and that resective surgery will not be an option. Intracranial electrodes are often reserved for cases where the non-invasive investigations, particularly the EEG, are discordant or ambiguous (Wyler et al., 1993). It is therefore not surprising that some investigators have found that the need for intracranial monitoring is associated with a lower likelihood of an Engel I outcome after epilepsy surgery (Tonini et al., 2004; Wyllie et al., 1998). On the other hand, in spite of the presumed complexity of cases requiring IE, we found that almost 50% of patients who underwent any resection and 60% of patients who had a temporal lobectomy were seizure free after surgery. Although these rates are somewhat lower than the results of epilepsy surgery in general (Engel et al., 1993; Téllez-Zenteno et al., 2005), they are comparable and a substantial proportion of these patients do become seizure free. Very few publications have focused exclusively on the seizure outcome in this group of patients. Jennum et al. found that 55% of extra-temporal and 70% of temporal cases were seizure free when guided by subdural electrodes, however it appears that the authors used subdural electrodes in all of their cases (Jennum et al., 1993). A smaller study by Cukiert et al. reported a 75% rate of seizure freedom in 16 adolescent and young adult patients with extra-temporal non-lesional epilepsy investigated with subdural grids (Cukiert et al., 2001). Because of the variability in the technique and indications for IE implantation between centers, a direct comparison of results may be misleading. Not surprisingly, we found that patients with temporal lobe epilepsy had the highest likelihood of seizure freedom. The strong association between SMA resection and a good outcome was somewhat surprising. Eight of our 10 patients with SMA resections in whom 1-year follow-up was available were seizure free. Although some authors have reported seizure freedom in up to 75% of patients (Jobst et al., 2000), more commonly lower success rates are reported (Aghakhani et al., 2004; Baumgartner et al., 1996; Olivier, 1996). One possible reason is that the resections in all 10 of our patients were restricted to the SMA, implying a very focal onset. Olivier reported an Engel I outcome in 22% of 28 patients who underwent resection of the SMA (Olivier, 1996). Although it is not clear in that study whether or not additional resection was carried out beyond the SMA, a subsequent publication from the same group reported that only one of six patients had their resection restricted to the SMA (Aghakhani et al., 2004). It is therefore possible that the higher success rate noted in the current series was related to selection of less complex cases. However, due to the small numbers of patients with SMA resections in this series, no firm conclusions should be drawn from this. The association between older age at surgery and poor outcome has been noted by some authors (Blume et al., 1994; Ficker et al., 1999; Jeong et al., 2005; Sirven et al., 2000), however others have not noted this relationship (Armon et al., 1996; Burneo et al., 2006a; Grivas et al., 2006; McIntosh et al., 2004; Salanova et al., 2002). While we noted an association when age was analyzed as a continuous variable, the nature of this association was
K.W. MacDougall et al. most evident when age was categorized by decade (Fig. 1). In particular, we found that patients older than 40 years had a dramatically lower likelihood of seizure freedom than patients aged 40 or less. This finding is very similar to results obtained by Cohen-Gadol et al. who noted that patients with cortical dysplasia aged under 18 and over 40 had a worse prognosis than patients with ages between these two parameters (Cohen-Gadol et al., 2004). They identified a quadratic relationship between the likelihood of seizure freedom and age, with the risk of surgical failure rising sharply at the extremes. When they categorized age as a continuous variable, no such association was noted. Although it is generally held that analyzing age as a continuous variable is preferable to dichotomization (Altman and Royston, 2006; Austin and Brunner, 2004; Royston et al., 2006), it is possible that a steep increase in risk over a certain age might be best appreciated if such a cut-off can be identified (Mazumdar and Glassman, 2000). In fact, a number of investigators have found that patients over certain age threshold have either a trend towards or statistically significant decreased likelihood of seizure freedom following epilepsy surgery (Boling et al., 2001; Cohen-Gadol et al., 2004; McLachlan et al., 1992). However, even in those studies, 30—75% of patients in the older age groups were seizure free and 75—83% had a greater than 90% reduction in seizure frequency at 1 year. In contrast, we noted that in this group that required IE only 21% of patients aged greater than 40 were seizure free and only 42% had a worthwhile improvement (Engel Grades I—III) 1 year after surgery. While our patient population as a whole had a slightly lower likelihood of seizure freedom than epilepsy surgery patients in general, the outcome in the older patients was disproportionally poorer than previously reported. While the reason for this is unclear, it is possible that older patients with complex epilepsies, by virtue of a longer duration of epilepsy, might me more prone to develop multifocality and therefore not respond to focal resective surgery. On the other hand, because both age and complexity may affect outcome independently, their combined risk may be additive and therefore our observation may simply represent the combined risk of these two variables. In conclusion, intracranial electrodes are an essential component of the pre-surgical investigation of many patients undergoing resective epilepsy surgery. Although the results of eventual resective epilepsy surgery are somewhat poorer than patients who do not require IE, good results can be achieved in almost 50% of patients in general and 60% of patients with temporal lobe epilepsy. Patients over age 40 seem to be an exception, as it appears that they have a significantly lower likelihood of surgical success. Although the latter finding requires further investigation and confirmation, it is another line of evidence that supports the earlier surgical management of focal epilepsies.
References Adelson, P.D., Black, P.M., Madsen, J.R., Kramer, U., Rockoff, M.A., Riviello, J.J., Helmers, S.L., Mikati, M., Holmes, G.L., 1995. Use of subdural grids and strip electrodes to identify a seizure focus in children. Pediatr. Neurosurg. 22, 174—180. Aghakhani, Y., Rosati, A., Olivier, A., Gotman, J., Andermann, F., Dubeau, F., 2004. The predictive localizing value of tonic limb
‘Outcome of epilepsy surgery’ posturing in supplementary sensorimotor seizures. Neurology 62, 2256—2261. Altman, D.G., Royston, P., 2006. The cost of dichotomising continuous variables. BMJ 332, 1080. Armon, C., Radtke, R.A., Friedman, A.H., Dawson, D.V., 1996. Predictors of outcome of epilepsy surgery: multivariate analysis with validation. Epilepsia 37, 814—821. Austin, P.C., Brunner, L.J., 2004. Inflation of the type I error rate when a continuous confounding variable is categorized in logistic regression analyses. Stat. Med. 23, 1159—1178. Bancaud, J., Angelergues, R., Bernouilli, C., Bonis, A., BordasFerrer, M., Bresson, M., Buser, P., Covello, L., Morel, P., Szikla, G., Takeda, A., Talairach, J., 1969. Functional stereotaxic exploration (stereo-electroencephalography) in epilepsies. Rev. Neurol. (Paris) 120, 448. Bancaud, J., Angelergues, R., Bernouilli, C., Bonis, A., BordasFerrer, M., Bresson, M., Buser, P., Covello, L., Morel, P., Szikla, G., Takeda, A., Talairach, J., 1970. Functional stereotaxic exploration (SEEG) of epilepsy. Electroencephalogr. Clin. Neurophysiol. 28, 85—86. Baumgartner, C., Flint, R., Tuxhorn, I., Van Ness, P.C., Kosalko, J., Olbrich, A., Almer, G., Novak, K., Lüders, H.O., 1996. Supplementary motor area seizures: propagation pathways as studied with invasive recordings. Neurology 46, 508—514. Behrens, E., Zentner, J., van Roost, D., Hufnagel, A., Elger, C.E., Schramm, J., 1994. Subdural and depth electrodes in the presurgical evaluation of epilepsy. Acta Neurochir. (Wien) 128, 84—87. Berg, A.T., Vickrey, B.G., Langfitt, J.T., Sperling, M.R., Walczak, T.S., Shinnar, S., Bazil, C.W., Pacia, S.V., Spencer, S.S., 2003. The multicenter study of epilepsy surgery: recruitment and selection for surgery. Epilepsia 44, 1425—1433. Blume, W.T., Desai, H.B., Girvin, J.P., McLachlan, R.S., Lemieux, J.F., 1994. Effectiveness of temporal lobectomy measured by yearly follow-up and multivariate analysis. J. Epilepsy 7, 203—214. Boling, W., Andermann, F., Reutens, D., Dubeau, F., Caporicci, L., Olivier, A., 2001. Surgery for temporal lobe epilepsy in older patients. J. Neurosurg. 95, 242—248. Burneo, J.G., Black, L., Martin, R., Devinsky, O., Pacia, S., Faught, E., Vasquez, B., Knowlton, R.C., Luciano, D., Doyle, W., Najjar, S., Kuzniecky, R.I., 2006a. Race/ethnicity, sex, and socioeconomic status as predictors of outcome after surgery for temporal lobe epilepsy. Arch. Neurol. 63, 1106—1110. Burneo, J.G., Steven, D.A., McLachlan, R.S., Parrent, A.G., 2006b. Morbidity associated with the use of intracranial electrodes for epilepsy surgery. Can. J. Neurol. Sci. 33, 223—227. Cohen-Gadol, A.A., Ozduman, K., Bronen, R.A., Kim, J.H., Spencer, D.D., 2004. Long-term outcome after epilepsy surgery for focal cortical dysplasia. J. Neurosurg. 101, 55—65. Cohen-Gadol, A.A., Spencer, D.D., 2003. Use of an anteromedial subdural strip electrode in the evaluation of medial temporal lobe epilepsy. Technical note. J. Neurosurg. 99, 921—923. Cukiert, A., Buratini, J.A., Machado, E., Sousa, A., Vieira, J.O., Argentoni, M., Forster, C., Baldauf, C., 2001. Results of surgery in patients with refractory extratemporal epilepsy with normal or nonlocalizing magnetic resonance findings investigated with subdural grids. Epilepsia 42, 889—894. Engel Jr., J., Van Ness, P.C., Rasmussen, T.B., Ojemann, L.M., 1993. Outcome with respect to epileptic seizures. In: Engel Jr., J. (Ed.), Surgical Treatment of the Epilepsies. Raven Press, New York, pp. 609—621. Ficker, D.M., So, E.L., Mosewich, R.K., Radhakrishnan, K., Cascino, G.D., Sharbrough, F.W., 1999. Improvement and deterioration of seizure control during the postsurgical course of epilepsy surgery patients. Epilepsia 40, 62—67. Girvin, J.P., 1991. Temporal lobectomy. In: Apuzzo, M.L.J. (Ed.), Neurosurgical Aspects of Epilepsy. American Association of Neurological Surgeons, Park Ridge, pp. 157—170.
241 Gonzalez-Feria, L., Garcia-Marin, V., 1993. Sub-occipital approach for implantation of recording multi-electrodes over the medial surface of the temporal lobe. Acta Neurochir. Suppl. (Wien) 58, 190—192. Grivas, A., Schramm, J., Kral, T., von Lehe, M., Helmstaedter, C., Elger, C.E., Clusmann, H., 2006. Surgical treatment for refractory temporal lobe epilepsy in the elderly: seizure outcome and neuropsychological sequels compared with a younger cohort. Epilepsia 47, 1364—1372. Jennum, P., Dhuna, A., Davies, K., Fiol, M., Maxwell, R., 1993. Outcome of resective surgery for intractable partial epilepsy guided by subdural electrode arrays. Acta Neurol. Scand. 87, 434— 437. Jeong, S.W., Lee, S.K., Hong, K.S., Kim, K.K., Chung, C.K., Kim, H., 2005. Prognostic factors for the surgery for mesial temporal lobe epilepsy: longitudinal analysis. Epilepsia 46, 1273—1279. Jobst, B.C., Siegel, A.M., Thadani, V.M., Roberts, D.W., Rhodes, H.C., Williamson, P.D., 2000. Intractable seizures of frontal lobe origin: clinical characteristics, localizing signs, and results of surgery. Epilepsia 41, 1139—1152. Lee, S.K., Kim, K.K., Nam, H., Oh, J.B., Yun, C.H., Chung, C.K., 2004. Adding or repositioning intracranial electrodes during presurgical assessment of neocortical epilepsy: electrographic seizure pattern and surgical outcome. J. Neurosurg. 100, 463—471. Lee, W.S., Lee, J.K., Lee, S.A., Kang, J.K., Ko, T.S., 2000. Complications and results of subdural grid electrode implantation in epilepsy surgery. Surg. Neurol. 54, 346—351. Lüders, H., Hahn, J., Lesser, R.P., Dinner, D.S., Morris III, H.H., Wyllie, E., Friedman, L., Friedman, D., Skipper, G., 1989. Basal temporal subdural electrodes in the evaluation of patients with intractable epilepsy. Epilepsia 30, 131—142. Mazumdar, M., Glassman, J.R., 2000. Categorizing a prognostic variable: review of methods, code for easy implementation and applications to decision-making about cancer treatments. Stat. Med. 19, 113—132. McIntosh, A.M., Kalnins, R.M., Mitchell, L.A., Fabinyi, G.C., Briellmann, R.S., Berkovic, S.F., 2004. Temporal lobectomy: long-term seizure outcome, late recurrence and risks for seizure recurrence. Brain 127, 2018—2030. McLachlan, R.S., Chovaz, C.J., Blume, W.T., Girvin, J.P., 1992. Temporal lobectomy for intractable epilepsy in patients over age 45 years. Neurology 42, 662—665. Murphy, M.A., O’Brien, T.J., Cook, M.J., 2002. Insertion of depth electrodes with or without subdural grids using frameless stereotactic guidance systems—–technique and outcome. Br. J. Neurosurg. 16, 119—125. Nair, D.R., Burgess, R., McIntyre, C.C., Lüders, H., 2008. Chronic subdural electrodes in the management of epilepsy. Clin. Neurophysiol. 119, 11—28. Olivier, A., 1996. Surgical strategies for patients with supplementary sensorimotor area epilepsy. The Montreal experience. Adv. Neurol. 70, 429—443. Olivier, A., Gloor, P., Quesney, L.F., Andermann, F., 1983. The indications for and the role of depth electrode recording in epilepsy. Appl Neurophysiol 46, 33—36. Pillay, P.K., Barnett, G., Awad, I., 1992. MRI-guided stereotactic placement of depth electrodes in temporal lobe epilepsy. Br. J. Neurosurg. 6, 47—53. Royston, P., Altman, D.G., Sauerbrei, W., 2006. Dichotomizing continuous predictors in multiple regression: a bad idea. Stat. Med. 25, 127—141. Salanova, V., Markand, O., Worth, R., 2002. Temporal lobe epilepsy surgery: outcome, complications, and late mortality rate in 215 patients. Epilepsia 43, 170—174. Sirven, J.I., Malamut, B.L., O’Connor, M.J., Sperling, M.R., 2000. Temporal lobectomy outcome in older versus younger adults. Neurology 54, 2166—2170.
242 Spencer, S.S., 1981. Depth electroencephalography in selection of refractory epilepsy for surgery. Ann. Neurol. 9, 207—214. Spencer, S.S., Spencer, D.D., Williamson, P.D., Mattson, R., 1990. Combined depth and subdural electrode investigation in uncontrolled epilepsy. Neurology 40, 74—79. Sperling, M.R., Shewmon, D.A., 1997. General principles for presurgical evaluation. In: Engel, J.J., Pedley, T.A. (Eds.), Epilepsy: A Comprehensive Textbook. Lippincott-Raven, Philadelphia, pp. 1697—1705. Steven, D.A., Andrade-Souza, Y.M., Burneo, J.G., McLachlan, R.S., Parrent, A.G., 2007. Insertion of subdural strip electrodes for the investigation of temporal lobe epilepsy. Technical note. J. Neurosurg. 106, 1102—1106. Téllez-Zenteno, J.F., Dhar, R., Wiebe, S., 2005. Long-term seizure outcomes following epilepsy surgery: a systematic review and meta-analysis. Brain 128, 1188—1198. Tonini, C., Beghi, E., Berg, A.T., Bogliun, G., Giordano, L., Newton, R.W., Tetto, A., Vitelli, E., Vitezic, D., Wiebe, S., 2004. Predic-
K.W. MacDougall et al. tors of epilepsy surgery outcome: a meta-analysis. Epilepsy Res. 62, 75—87. van Veelen, C.W., Debets, R.M., van Huffelen, A.C., van Emde Boas, W., Binnie, C.D., Storm van Leeuwen, W., Velis, D.N., van Dieren, A., 1990. Combined use of subdural and intracerebral electrodes in preoperative evaluation of epilepsy. Neurosurgery 26, 93—101. Wyler, A.R., Ojemann, G.A., Lettich, E., Ward Jr., A.A., 1984. Subdural strip electrodes for localizing epileptogenic foci. J. Neurosurg. 60, 1195—1200. Wyler, A.R., Wilkus, R.J., Blume, W.T., 1993. Strip Electrodes. In: Engel, J. (Ed.), Surgical Treatment of the Epilepsies. Raven Press, New York, pp. 387—397. Wyllie, E., Comair, Y.G., Kotagal, P., Bulacio, J., Bingaman, W., Ruggieri, P., 1998. Seizure outcome after epilepsy surgery in children and adolescents. Ann. Neurol. 44, 740—748.