SEARCH

SEARCH BY CITATION

Keywords:

  • Refractory epilepsy;
  • Invasive monitoring;
  • Epilepsy surgery

Summary

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Summary
  8. Disclosure
  9. References

Purpose:  Intracranial monitoring (IM) is a key diagnostic procedure for select patients with treatment-resistant epilepsy (TRE). Seizure focus resection may improve seizure control in both lesional and nonlesional TRE. IM itself is not considered to have therapeutic potential. We describe a cohort of patients with improved seizure control following IM without resective surgery.

Methods:  Over 12.5 years, 161 children underwent 496 surgeries including intracranial monitoring. We retrospectively reviewed the patients’ charts, operative reports, and radiologic scans, under an institutional review board–approved protocol.

Key Findings:  Seventeen patients underwent only IM, without additional resective surgery, and seven had a dramatic improvement in their epilepsy; six of the seven patients are seizure-free (Engel class I), and one rarely has seizures (Engel class II). All seven patients had frequent seizures that led to IM: either daily (five patients) or 1–2 per week (two patients). The mean age (± standard deviation, SD) at seizure onset was 1.6 ± 1.3 years (range 0.5–4 years). Etiologies were tuberous sclerosis (3 patients), trauma (1 patient), and unknown (3 patients). Mean age at surgery (± SD) was 4.1 ± 2 years (range 1–7 years), and duration of epilepsy 2.5 ± 1.1 years (range 0.5–4 years). Duration of IM was 11.7 ± 5.6 days (5–19 days). Six patients had bilateral and one unilateral invasive electrodes. At last follow-up, four patients required fewer antiepileptic drugs (AEDs), one had the same medication but a higher dose, and two patients were taking additional AEDs. Follow-up was 30.6 ± 9.5 months (range 19–41 months).

Significance:  Although uncommon, patients with TRE may improve after IM alone. The explanation for this observation remains unclear; however, perioperative medications including steroids, direct cortical manipulation, or other factors may influence the epileptogenic network.

Approximately 30% of children who have epilepsy are resistant to medical treatment (Beleza, 2009). Epilepsy surgery, including resection of the epileptogenic focus, a disconnection procedure (corpus callosotomy or hemispherectomy), or a palliative procedure such as a vagus nerve stimulator (VNS), is now widely accepted as an effective therapeutic option in a select subset of these patients (Privitera et al., 2002; Centeno et al., 2006).

The preoperative evaluation of patients with treatment resistant epilepsy (TRE) has two primary goals: localizing the epileptogenic focus and mapping vital cortical areas. Various noninvasive methods are used for this purpose, including scalp video–electroencephalography (EEG), positron-emission tomography (PET), single-photon emission computed tomography (SPECT), magnetoencephalography (MEG), Wada test (intracarotid sodium amobarbital procedure, ISAP), and functional magnetic resonance imaging (fMRI) (Duchowny et al., 2000; Cross, 2002; Donaire et al., 2009). However, despite advancements in noninvasive techniques, cortical recording and stimulation provide critical information with low morbidity when the noninvasive evaluation cannot fully achieve the primary goals (Weiner et al., 2006).

Invasive monitoring (IM) includes placement of subdural strips, grids, and, in certain cases, depth electrodes. IM directly records EEG activity from suspected brain areas, enabling direct mapping of epileptic networks, as well as functional mapping (Bollo et al., 2009). IM is critical when evaluating patients with multifocal epilepsy (Bauman et al., 2008, Carlson et al., 2011. IM is considered a diagnostic, not a therapeutic, procedure. The low morbidity of IM and high efficacy of focal resections support the use of IM as a part of the surgical approach.

We describe a small series of children with TRE who underwent IM as part of a surgical evaluation. All patients had frequent seizures, and, for various reasons, had IM without subsequent resective surgery. All patients had a dramatic improvement in their seizure control following IM alone. The clinical features and possible etiologies for this phenomenon are presented.

Methods

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Summary
  8. Disclosure
  9. References

Between September 1997 and March 2009, 161 children underwent various IM procedures at the NYU Langone Medical Center (NYULMC). Many of the patients had a staged procedure, consisting of a primary invasive monitoring period, followed by a second surgical step that included resection of an epileptic focus (Bauman et al., 2008). In some patients, during the second stage, additional electrodes were reimplanted, and this was followed by a third stage of resection and removal of electrodes (one patient underwent a fourth stage). In total, 161 patients underwent 200 multistage procedures and 496 surgeries involving intracranial monitoring. The surgical technique, anesthetic protocol, and postoperative routines were similar in all patients and procedures. As a routine, every patient had a postoperative MRI (including axial T2, fluid-attenuated inversion recovery (FLAIR), and noncontrast T1-weighted sequences) within 24 h following surgery. Invasive monitoring included subdural grids and strips, and depth electrodes with intercontact distances of 5–10 mm (Ad-Tech Medical Instrument Corporation, Racine, WI, U.S.A.). General anesthesia was induced with propofol, fentanyl, and a muscle relaxant (vecuronium and/or rocuronium), and maintained with a continuous drip of propofol and remifentanil. Often, patients receive inhalational anesthetics (i.e., sevoflurane and/or isoflurane). During surgery, all patients received 6 mg of dexamethasone, and a dose of prophylactic antibiotics (typically cefazolin or clindamycin). All patients had a subgaleal 7 mm Jackson-Pratt (JP) drain left at the surgical site (as per our routine for all craniotomies involving IM). Following surgery, patients were started on 16 mg/day dexamethasone, tapered over about 1 week.

Of the 161 patients described above, 17 (10.5%) underwent IM, without subsequent resection or disconnection. Of these 17 patients, 5 had poorly localized TRE, and were subsequently treated with implantation of a VNS. Four other patients with poorly localized epilepsy are currently being treated medically with no change in their epilepsy. One patient was lost to follow-up. The remaining seven patients are the focus of this study.

Data on these seven patients was retrospectively reviewed, and includes patients’ files from the Comprehensive Epilepsy Center (CEC) at NYU Langone Medical Center, operative notes and charts from the Division of Pediatric Neurosurgery at the NYULMC, and preoperative and postoperative MRI scans performed at the NYULMC. The epilepsy status was evaluated by an epileptologist during routine follow-up visits, and was classified according to the Engel outcome scale (Engel, 1987). Surgical outcome, including complications, was based on the follow-up notes from the surgical chart. Radiologic evaluation was based on a retrospective review of the patients’ MRI scans. Institutional review board approval was given prior to performance of this study.

Results

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Summary
  8. Disclosure
  9. References

Basic patient data

The series consists of seven pediatric patients (six male, and one female) (Table 1). None of the patients had any prior epilepsy surgery. The etiology of epilepsy was tuberous sclerosis complex (TSC) (three patients), posttraumatic (one patient), and unknown (three patients). All patients had frequent seizures: Five had daily seizures (usually multiple times a day) and two had seizures once to twice a week. Two patients had daily tonic–clonic seizures and five children had complex partial seizures (CPS), three of whom had secondary generalization. The age at epilepsy onset was 6 months to 4 years of age (1.6 ± 1.3 years), and the duration of epilepsy until the IM procedure varied between 6 months and 4 years (2.5 ± 1.1 years). Patients were taking two to three AEDs during the period prior to the IM procedure (mean 2.4 drugs). Patients failed 3–7 (5.2 ± 1.7) AEDs in various combinations prior to surgery. Preoperative noninvasive EEG findings are summarized in Table 1. Children were between 1 and 7 years of age at the time of IM (4.1 ± 1.9 years of age). Regarding the ten children who underwent IM only but were not seizure-free, their age during IM was 2 to 10.5 years old (6 ± 3), and was not significantly different from the six seizure-free children.

Table 1.   Patient demographics
 GenderEpilepsy etiologyAge at SZ onset (years)SZ typeSZ frequencyAED (preoperative)Age at surgery (years)Preoperative noninvasive VEEGDuration of epilepsy before surgery (years)
  1. Dev, developmental; SZ, seizure; CPS, complex partial seizures; G, generalization; AED, antiepileptic drugs; VPA, valproic acid; LTG; lamotrigine; LVT, levetiracetam; L, left; R, right.

  2. AED dosages are total per day.

1MUnknown2.5CPSDailyVPA 750 mg LVT 1,500 mg LTG 50 mg5.5Interictal: bilateral synchronous and multifocal spike waves in the frontotemporal regions, L > R3
2MTSC0.5CPSDailyVigabatrin 500 mg Zonisamide 150 mg1Interictal: bilateral focal slowing over the temporal regions, multifocal independent spikes and asymmetric sleep spindles0.5
3FDev delay1.2CPS + 2ry G1–2/weekLTG 200 mg Zonisamide 125 mg4Generalized spikes with possible L frontal onset2.8
4MTSC0.5TonicDailyTopiramate 175 mg Vigabatrin 1,000 mg4.5Abnormal disorganized and generalized slowing Generalized polyspikes Bilateral frontopolar, R anterior temporal, R parietal and L temporal spikes and waves4
5MUnknown2CPS + 2ry G1–2/weekLTG 50 mg VPA 250 mg Clonazepam 0.25 mg4.5Multiple seizures with central vertex onset, sometimes lateralizing to the frontal region2.5
6MPosttrauma4CPS + 2ry GDailyPhenobarbital 15 ml Phenytoin sodium 6cc VPA 1,125 mg7L posterior temporal and frontotemporal spikes3
7MTS0.5Atonic + GTSDailyVigabatrin 2,000 mg Topiramate 325 mg2.5Multifocal interictal epileptic activity2

Results of preoperative MRI scans were normal in two patients, showed multiple bilateral tubers (in the three TSC patients), and showed nonspecific white matter signal changes in two patients.

Operative and perioperative course

Six patients had bilateral craniotomies with bilateral subdural strip placement; one had a unilateral craniotomy for electrode implantation. Three patients had depth electrodes in addition to subdural strips/grids. Typically, strips were placed over five regions including the lateral surface of the brain (specific locations stated in Table 2). The IM duration was 5–19 days (11.7 ± 5.6 days). One intraoperative acute subdural hematoma was surgically controlled, but led to postponing of placement of the invasive monitoring to the next day. No other perioperative or late complications occurred, and postoperative MRI scans (done on the first postoperative day) showed no surgical complications. Typically, MRI scans showed a small amount of subdural air, and in some cases, a minimal amount of subdural blood. All subsequent MRI scans showed complete resolution of these findings. There were no findings suggestive of brain swelling.

Table 2.   Surgical details and outcome
 Sides monitoredElectrode locationDepth electrodesDuration of IM (day)Invasive EEGSZ during IMFollow-up duration (month)Engel outcome classAED at last f/uEEG at f/u
  1. PLE, poorly localized epileptic activity; RDE, rare diffuse epileptic activity; f/u, follow-up; NVEEG, normal VEEG; NEEG, normal EEG; T, temporal; F, frontal; P, parietal; PO, parietooccipital; TO, temporooccipital; O, occipital; d, depth; g, grid; VPA, valproic acid; LTG; lamotrigine; LVT, levetiracetam.

12LT: 2T, 1PO, 2F, 1C RT: 1PO, 1T, 1C, 2FNo6PLEMultiple25I0Interictal: left frontal spikes and sharp waves
221F, 1C, 1PO, 2T (both sides)No5PLEMultiple41IVigabatrin 750 mg Zonisamide 150 mgNEEG
32LT: 3F, 1PO, 1C, 2T, 2Fd RT: 2F, 1PO, 1C, 2T, 1FdYes (3)16No epileptic activityNo19II0NEEG
42LT: 1F, 1C, 1P, 1O, 2T RT: 1F, 1C, 1P, 2O, 2TNo7PLEMultiple35IVPA 1,750 mg Topiramate 250 mg Vigabatrin 2,500 mgNVEEG
52LT:2F, 2Fd, 1FT, 1C, 2T, 1PO RT: 2F, 2Fd, 1C, 1PO, 2TYes (4)15RDE1 non typical SZ34ILVT 250 mg LTG 170 mgNVEEG
61 (left)3T, 3F, 1PO, 2Fd, 1FPg, 1FTgYes (2)19Mild focal activityNo41ILTG 200 mgNVEEG
72LT: 1Fg, 1Pg, 1Tg, 1FP, 1TO RT: 1F, 1C, 1P, 2TNo14PLEMultiple19IVigabatrin 2,000 mg Topiramate 100 mg LVT 800 mgFrequent multifocal epileptic activity

Invasive monitoring findings

During the monitoring period, one patient had no epileptiform activity on EEG and no clinical seizures (over 16 days), one patient had rare interictal epileptiform activity with no clinical seizures (19 days of monitoring), one patient had rare interictal epileptiform activity and seizures that were not the patient’s typical clinical seizure (15 days of monitoring), and four patients had diffuse and poorly localized epileptic activity (over 5, 6, 7, and 14 days).

Follow-up

Patients were followed for 19–41 months after the IM procedure (30.6 ± 9.5 months). None of the patients underwent any additional surgery.

Epilepsy outcomes were: Engel class I (seizure-free) in six patients (at 15–36 months), and Engel class II (>90% reduction in seizure frequency) in one patient (followed for 12 months). Two patients have discontinued AEDs, two are on fewer drugs, one is on the same AEDs but higher doses, and two patients are on additional AEDs as compared to preoperatively.

Monitoring duration was 6–19 days (14 ± 5.6 days) in the four patients who eventually took fewer AEDs, as opposed to 5–14 days (8.6 ± 4.7 days) in the three patients who were receiving additional AEDs or higher doses of the AEDs they were already taking. In addition, three of the patients who were taking fewer AEDs had depth electrodes. None of the other three children who were on more AEDs or on a higher dose had depth electrodes.

Six patients had EEG documented during the follow-up period (Table 2). Three patients had normal video-EEG (VEEG), and two had normal EEG recording. One patient had interictal focal discharges.

Discussion

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Summary
  8. Disclosure
  9. References

This study suggests that the process of invasive monitoring can have a therapeutic antiseizure effect. In a series of 161 patients, 17 underwent IM without resection and 11 had subsequent follow-up and did not undergo implantation of a VNS. Six of the eleven patients remain seizure free and one patient has had a >90% reduction in seizure frequency following IM; the remaining four children had no significant change in seizure frequency following IM. For these seven children, different etiologies were seen, including three patients with multifocal tubers secondary to TSC. Prior to IM, patients had seizures ranging from once per week to multiple per day and had failed as many as seven AEDs. The temporal relationship of seizures in four patients with fewer AEDs strongly supports that IM and/or the associated perioperative procedures led to the dramatic reduction in seizures.

Improvement following IM with subdural electrodes has been reported previously in a series of six patients (two children) (Katariwala et al., 2001). Three patients had a relapse of rare seizures (Engel class II), and three were in complete remission (Engel class I, for 11 months to 15 years). In three patients, subtle radiologic findings suggesting brain injury following IM were credited with improving seizures; however, the etiology remains unknown in the other three patients. A recent case report described a 37-year-old patient who had multiple focal seizures for >10 years. She underwent stereotactic placement of a hippocampal depth electrode, as well as a left temporal subdural grid and subtemporal subdural strips that were removed several days later. The patient has remained seizure free ever since, with no change in her AED regimen (Schulze-Bonhage et al., 2010).

A definitive mechanism resulting in improvement in seizure frequency cannot be identified for the seven patients reported in our series. Neither could we find a good explanation for the association between longer monitoring periods and the final lowering of AEDs, or the fact that all three patients with depth electrodes eventually needed to take fewer AEDs. The small number of patients in this cohort adds to the uncertainty of these observations. However, potential mechanisms for the good seizure outcomes would include intrinsic properties of the patient’s epilepsy (i.e., normal variations in seizure frequency/spontaneous remission), responses to different AEDs or doses (potentially explains three of the seven presented cases), response to perioperative medications (e.g., corticosteroids, opioids, anesthetics), local responses to intracranial electrodes (e.g., cerebral contusion, local inflammation), responses to the neurosurgical procedure (e.g., hypothermia during surgery, brain irrigation with cold fluids, changes in intracranial pressure secondary to surgery, postoperative pneumocephalus), or systemic responses to surgery (e.g., postoperative fever).

Natural fluctuation in the disease nature

Epilepsy, by its nature, is paroxysmal and, as such, seizure frequency may fluctuate over time. A well-controlled disease may become resistant, and TRE may go into remission with prolonged periods of seizure freedom (Berg et al., 2006; Sillanpaa & Schmidt, 2006; Camfield & Camfield, 2007; Choi et al., 2008). Therefore, spontaneous improvement may occur among both children and adults with TRE. However, the abrupt cessation of seizures in our patients following IM with remissions of 1 year or longer raise the possibility that this improvement may be secondary to more than just natural variation in seizure frequency.

Anesthetic drugs and opioids

Anesthetic drugs have an antiepileptic effect. Propofol has antiepileptic properties related to γ-aminobutyric acid (GABA)–mediated presynaptic and postsynaptic inhibition, and thus decreases the release of glutamate and aspartate (Meyer et al., 2009, 2006). Propofol also blocks sodium and calcium current channels (Voss et al., 2008). The effect of propofol is dose dependent, and in low doses it may actually induce seizures (Wang et al., 1997). Propofol is a short-acting drug used to treat status epilepticus (Soriano et al., 2000).

Benzodiazepine drugs act as potent anticonvulsant drugs via activation of the GABAA receptor, inhibiting glutamate secretion (Voss et al., 2008). Midazolam is often used for sedation, and is a potent anticonvulsant used for treating status epilepticus.

However, the half-life of these drugs is short (minutes to hours), and we did not find any reports regarding a long-term effect of anesthetic drugs on epilepsy.

Volatile anesthetics shift the EEG pattern to burst suppression (Iijima et al., 2000). However, sevoflurane may induce a marked increase in electrocorticography (ECoG) spike activity in patients with temporal lobe epilepsy (Watts et al., 1999; Hisada et al., 2001). Therefore, despite the anticonvulsant effect of volatile anesthetics, some drugs may provoke epileptiform activity, and certain patients may experience seizures secondary to these drugs. Because the half-life of these drugs is short, most clinical convulsions occur at the beginning or end of the anesthesia. Rarely do seizures occur late in the recovery period (weeks after), and long-term effects of volatile anesthetics on epilepsy have not been reported.

Opioids have both proconvulsant and anticonvulsant properties (Voss et al., 2008). Opioid-induced epileptiform activity is mediated through μ receptors. Postulated mechanisms include disinhibition of GABAergic interneurons and inhibition of hyperpolarization-activated potassium currents (Voss et al., 2008). Most experimental work has emphasized that opioid-induced seizures originate in deep structures (limbic and mesencephalic areas, or medial temporal lobes) (Voss et al., 2008). Remifentanil has been shown to significantly increase the frequency of spikes in patients undergoing ECoG during anterior temporal lobectomies for refractory epilepsy (Wass et al., 2001). Conversely, the kappa ligand dynorphin is thought to be the principal anticonvulsant endogenous opioid (Voss et al., 2008).

Therefore, despite various anticonvulsive and proconvulsive properties of anesthetic drugs and opioids, we did not find any literature on the long-term effects of these drugs on epilepsy patients. The half-life of these drugs is short, and they were widely used among our large group of epilepsy patients who were undergoing surgery. As noted, some of the cases discussed here continued to have seizures while awaiting electrode removal, and thus we do not think that these drugs had an influence on the long-term outcome.

Steroids

Steroids have a therapeutic effect on seizures and are used for infantile spasms, as well as for other epilepsy syndromes (Lerman et al., 1991; Hart et al., 1994; Go, 1999; Tsuru et al., 2000; Oguni et al., 2002; Buzatu et al., 2009). The mechanisms underlying the antiepileptic action of steroids is unknown. Suggestions include alterations in neurochemical transmission as a result of alterations in serotonin turnover or GABA uptake. These effects may be mediated through the glucocorticoid and/or mineralocorticoid receptors. Steroid use during IM can decrease the frequency of seizures; however, there may be greater mass effect among patients who did not receive steroids during the IM period (Araki et al., 2006). Despite the notion of a steroid effect in epilepsy, a recent Cochrane study found no evidence for the efficacy or safety of corticosteroids in treating childhood epilepsies; therefore, a routine trial of steroids is not justified prior to surgery (Gayatri et al., 2007).

Disruption of the epileptogenic network

The epileptic network is a complex anatomic network, consisting of reciprocal interactions between a primary focus and many other cortical areas (Madhavan et al., 2007). The network may be affected by removing the epileptogenic focus (the essential goal of surgery); however, even disruption of another component in the network may theoretically alter the pattern of the disease (Kahane & Depaulis, 2010; Schulze-Bonhage et al., 2010).

During IM, minimal tissue injury may occur secondary to small contusions of the brain adjacent to the electrodes, or as a result of a local inflammation secondary to tissue reaction to the electrodes (Al-Otaibi et al., 2009). Other possible factors that may interact with the epileptogenic network are subdural air or blood. Both pneumocephalus and subdural hematomas (acute or chronic) may cause seizures (Rohde et al., 2002; Suri et al., 1998). Although uncommon, this suggests a local cortical irritation in response to air or blood products (such as hemosiderin).

Depth electrodes may directly injure the epileptogenic focus or other areas in the epileptic network, in a fashion similar to the direct traumatic effect following placement of a deep brain stimulator used for treating various diseases (Cersosimo et al., 2009). In the context of this series, we postulate that a similar irritation or direct injury may affect the epileptic network in some way, and thus alter the epileptic disease.

Hypothermia and cold brain irrigation

Moderate systemic hypothermia (to 31–35°C) has been shown to abort refractory status epilepticus (Corry et al., 2008). Similarly, local brain irrigation with cold fluids (4°C) is commonly used to abort intraoperative epileptic discharges (Ablah et al., 2009). However, the long-term effects of these factors are unknown. Despite the wide use of mildly cooled fluid irrigations during surgery, and mild ambient air-cooling in the operating room, the impact of these factors on the epilepsy network is unknown.

Changes in intracranial pressure

Following surgery, a subgaleal JP drain may induce a negative intracranial pressure. Intracranial hypotension may increase regional cerebral blood flow (CBF) and cause relative hyperemia (Chen et al., 1999). The clinical impact of this effect is unknown. Rarely, and for reasons that are unclear, intracranial hypotension may induce seizures (Agrawal & Durity, 2006). Van Roost et al. (2003) presented a series of “pseudohypoxic” changes in the basal ganglia and thalami that were related to negative ICP induced by subgaleal drains. The actual cause of this phenomenon is unclear; however, it suggests that negative pressure induced by subgaleal drains, may irreversibly affect intracranial neural function. The maximal negative pressure a 100 cc vacuum bulb can induce is −100 cm H2O. In the presented series, six patients had bilateral drains; however, it is impossible to evaluate the actual effect on the intracranial pressure.

Fever and infection

High fever may alter the course of epilepsy. Previous reports have demonstrated the effect of various viral illnesses causing high-grade fever among children with infantile spasms (Pintaudi et al., 2007; Sugiura et al., 2007; Yamamoto et al., 2007). High-grade fevers may induce elevated levels of corticotropin and corticosteroids, secondarily affecting the epileptic disease. Among the current series, no patient had high-grade fever following surgery. Even in the absence of fever, a mild infection introduced during surgery but treated with prophylactic perioperative antibiotics, may have caused a disruption in the epileptogenic network.

Limitations

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Summary
  8. Disclosure
  9. References

One major limitation of this report is its reliance on accurate parental awareness of seizures. Despite the refractory nature of the disease prior to surgery, and the familiarity of the parents with seizures, there may have been unrecognized seizures. In a study of 78 children with epilepsy with 1,244 EEG documented seizures, only 472 (38%) were correctly reported by parents, and only 60% of CPS were identified (Akman et al., 2009). Among the seven patients in the current series, four had follow-up scalp EEG recordings, two of which documented interictal epileptiform activity. It is possible that despite an improvement in the frequency and intensity of clinical seizures, some seizures may have been overlooked by the parents. The clinical implication of this is important. If seizures are underdiagnosed but the EEG remains pathologic, the question remains as to the importance of eradicating these undetected seizures by continuing aggressive medical treatment. Currently, we are closely following these patients.

Another limitation is an add-on treatment with a new AED. Four of seven patients had decreased the number of AEDs (two are off all AEDs), and three increased the dose or number of AEDs. Therefore, the seizure control may be attributed to change in AEDs or their dose. However, this does not explain the seizure control in the other four patients.

Summary

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Summary
  8. Disclosure
  9. References

Despite the lack of a pathophysiologic mechanism, the dramatic improvement in the patients’ epilepsy was temporally related to the diagnostic surgical procedure. Although not seen in all patients who were undergoing IM without resection, this retrospective study suggests that, in some patients, the process of IM improves seizure control. At present, one can only speculate as to the pathophysiologic mechanism, and if it is truly related to the surgical procedure or to other medical factors.

Disclosure

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Summary
  8. Disclosure
  9. References

None of the authors has any conflict of interest to disclose. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

References

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Summary
  8. Disclosure
  9. References