Factors limiting the application of electrical impedance tomography for identification of regional conductivity changes using scalp electrodes during epileptic seizures in humans.

Electrical impedance tomography (EIT) has the potential to produce images during epileptic seizures. This might improve the accuracy of the localization of epileptic foci in patients undergoing presurgical assessment for curative neurosurgery. It has already been shown that impedance increases by up to 22% during induced epileptic seizures in animal models, using cortical or implanted electrodes in controlled experiments. The purpose of this study was to determine if reproducible raw impedance changes and EIT images could be collected during epileptic seizures in patients who were undergoing observation with video-electroencephalography (EEG) telemetry as part of evaluation prior to neurosurgery to resect the region of brain causing the epilepsy. A secondary purpose was to develop an objective method for processing and evaluating data, as seizures arose at unpredictable times from a noisy baseline. Four-terminal impedance measurements from 258 combinations were collected continuously using 32 EEG scalp electrodes in 22 seizure episodes from 7 patients during their presurgical assessment together with the standard EEG recordings. A reliable method for defining the pre-seizure baseline and recording impedance data and EIT images was developed, in which EIT and EEG could be acquired simultaneously after filtering of EIT artefact from the EEG signal. Fluctuations of several per cent over minutes were observed in the baseline between seizures. During seizures, boundary voltage changes diverged with a standard deviation of 1-54% from the baseline. No reproducible changes with the expected time course of some tens of seconds and magnitude of about 0.1% could be reliably measured. This demonstrates that it is feasible to acquire EIT images in parallel with standard EEG during presurgical assessment but, unfortunately, expected EIT changes on the scalp of about 0.1% are swamped by much larger movement and systematic artefact. Nevertheless, EIT has the unique potential to provide invaluable neuroimaging data for this purpose and may still become possible with improvements in electrode design and instrumentation.

Electrocorticography and stimulation.

Although acute electrocorticography (ECoG) is routinely used during epilepsy surgery there is little agreement as to its value nor criteria for its interpretation. Specific issues are reviewed on the basis of the literature and personal studies: does failure to resect the entire irritative zone prejudice seizure control, and are residual discharges predictive of failure; does activation of the ECoG by intravenous barbiturates provide information of clinical value; does intraoperative electrical stimulation help to improve localisation or avoid postoperative deficits; is the ECoG of value for monitoring functional procedures; can the value of ECoG be increased by new interpretive approaches? It is suggested that resection of the entire area of interictal discharge is not essential for satisfactory surgical outcome, but a distinction may need to be made between those discharging regions that function as pacemakers and those in which ECoG spikes appear secondarily. There is also evidence that, apart from any consideration of determining the area resected, the topography of epileptiform discharge may be predictive of pathology and surgical outcome. It is concluded that more detailed topographic and quantitative analysis of the ECoG is required before its value in planning surgery can be determined or objective interpretive criteria established.

Sudden withdrawal of carbamazepine increases cardiac sympathetic activity in sleep.

OBJECTIVE: To evaluate the cardiac autonomic effects of abrupt withdrawal of carbamazepine (CBZ) during sleep in patients with epilepsy. BACKGROUND: The pathophysiology of sudden unexpected death in epilepsy (SUDEP) is uncertain, with ictal or peri-ictal cardiorespiratory compromise appearing probable. Risk factors for SUDEP include multiple antiepileptic drugs (AED), poor compliance, and abrupt AED withdrawal. The spectral analysis of the beat-to-beat heart rate variability (HRV) displays two main components: low frequency (LF), representing sympathetic and parasympathetic influence and high frequency (HF), representing parasympathetic influence. The LF/HF ratio is commonly regarded as an indicator of sympathovagal balance. METHOD: Twelve patients with medically intractable seizures underwent abrupt withdrawal of CBZ to facilitate seizure recording during controlled circuit TV-EEG monitoring. Continuous EKG recording was begun 24 hours before CBZ reduction. Spectral analysis of the HRV was performed during selected samples of non-REM sleep before and after CBZ reduction. Analyses were made at least 6 hours after from (complex) partial and 12 hours from generalized seizures. RESULTS: The mean LF/HF ratio before withdrawal of CBZ was 2.15 compared with a ratio of 2.65 on day 4 after withdrawal, an increase of 19% (geometric mean; 95% CI, 2% to 34%; Wilcoxon test, z = 2.36; p = 0.018). The ratio increased in 10 patients compared with a decrease in only one patient. CONCLUSION: Abrupt withdrawal of CBZ leads to enhanced sympathetic activity in sleep as evidenced by increased LF/HF ratios. Increased sympathetic activity in the setting of seizure-induced hypoxia could predispose to SUDEP.

Coexistence of temporal lobe and idiopathic generalized epilepsies.

OBJECTIVE: To assess the interrelation of idiopathic generalized epilepsy (IGE) and temporal lobe epilepsy (TLE) when they coexist in the same patient. METHODS: The authors reviewed the electroclinical features of 350 consecutive patients who had temporal resection between 1975 and 1997 at the Maudsley and King's College Hospitals, London. RESULTS: Two patients had the unusual combination of TLE and IGE (0.57%). In the first, the clinical onset of juvenile myoclonic epilepsy followed the surgical resolution of his partial seizures but had been heralded for at least 5 years by subclinical spontaneous and photically induced generalized spike-wave discharges. In the second, TLE and juvenile absence epilepsy had a long parallel course before surgery. After surgery he had no further partial seizures. CONCLUSION: These cases suggest that when an idiopathic absence or myoclonic syndrome manifests in a patient with symptomatic TLE, the phenotype may not be a merged syndrome. Rather, the two conditions can retain their inherent electroclinical profile, responsiveness to treatment, and prognosis.

A study of mortality after temporal lobe epilepsy surgery.

OBJECTIVE: To determine early and late mortality in a cohort of 305 consecutive patients who had temporal lobe epilepsy (TLE) surgery over a 20-year period. METHODS: Survival status, cause of death, and postoperative clinical details of those who died were ascertained in a cohort of 305 patients who had TLE surgery. Mortality was related to postoperative seizure status, operative pathology, and side of resection. RESULTS: The survival status of 299 patients was established. Twenty deaths occurred. Mortality was 1 per 136 person-years, with a standardized mortality ratio (SMR) of 4.5 (95% confidence interval [CI], 3.2 to 6.6). Six deaths were sudden and unexpected (SUDEP). The SUDEP rate was 1 per 455 person-years. The overall death and SUDEP rates were lower than those reported for similar patient populations with chronic epilepsy. Mortality in patients who had right-sided resections for mesial temporal sclerosis (MTS) remained considerably elevated with a mortality rate of 1 per 54 person-years, an SMR of 32.0 (95% CI, 24.7 to 40.5), and a SUDEP rate of 1 per 134 person-years. These patients had significantly lower seizure remission rates than left-sided patients, but the excess mortality was not simply explained by those patients whose partial seizures were uninfluenced by surgery. Patients who died had more severe or convulsive seizures despite an overall reduction in seizure frequency. CONCLUSIONS: The present findings confirm previous reports that TLE surgery lowers but does not normalize the overall mortality associated with chronic epilepsy. In patients with right-sided MTS, however, the postoperative mortality has remained similar to other groups with medically intractable seizures.

Telemetric EEG and video monitoring in epilepsy.

We report a clinical evaluation of EEG and video monitoring. In 181 consecutive records, the most common clinical indications were differentiation of epileptic and nonepileptic events (99), seizure recordings for locating a possible focus (23), seizure frequency determination (32), and investigation of possible trigger factors (19). Overall, useful information was obtained in 72% and the clinical question was answered in 67% of records. Prolonged EEG and video monitoring is therefore an important diagnostic aid in patients with proven or suspected epilepsy.

Chronic encephalitis and temporal lobe epilepsy: a variant of Rasmussen's syndrome?

The authors report two adult patients with chronic temporal lobe epilepsy and pathologic features consistent with Rasmussen's encephalitis. Although seizures persisted after temporal lobe surgery no progressive cognitive or neurologic deficit has emerged. Prominent auditory auras in each suggested a persisting epileptogenic focus in the superior temporal gyrus. The current findings expand the clinical spectrum of Rasmussen's encephalitis and suggest that chronic nonprogressive encephalitis may serve as the pathologic substrate of medically intractable temporal lobe epilepsy.

N-Acetylaspartate and creatine levels measured by (1)H MRS relate to recognition memory.

OBJECTIVES: To investigate the relationship between recognition memory and metabolite levels in medial structures of the temporal lobes in the living human brain. METHODS: Proton MRS ((1)H MRS) and the intracarotid amobarbital test were performed in 16 epileptic patients found suitable for temporal lobectomy. All patients had mesial temporal sclerosis. Metabolite ratios between N:-acetylaspartate (NAA), creatine and phosphocreatine (Cr + PCr), and choline-containing compounds (Cho) [NAA/(Cr + PCr), NAA/Cho, and NAA/(Cr + PCr + Cho)] were calculated for (1)H MRS voxels that included the amygdala, anterior half of the hippocampus, and underlying subiculum. Metabolite ratios were correlated with unilateral memory scores estimated by the intracarotid amobarbital test for words, objects, faces, and total score. RESULTS: The total memory score, memory for objects and faces, and NAA/(Cr + PCr) were significantly lower for the hemisphere ipsilateral to the resection. The asymmetry indexes for NAA/(Cr + PCr) correlated with asymmetry indexes for words (rho = 0.82, p = 0.0001) and total memory (rho = 0. 72, p = 0.002). Analysis of memory scores and metabolite ratios from all 32 hemispheres revealed a correlation between NAA/(Cr + PCr) and memory for words (rho = 0.45, p = 0.009). A correlation between memory for words and NAA/(Cr + PCr) existed in the contralateral (rho = 0.58, p = 0.019) and in the right (rho = 0.51, p = 0.045) hemispheres, and a trend was found in the left hemispheres (rho = 0. 48, p = 0.06). CONCLUSION: There is a correlation between memory for words and the NAA/(Cr + PCr) ratio from medial temporal structures in patients with mesial temporal sclerosis. The findings suggest that medial temporal structures and adjacent neocortex play a significant role in recognition memory in humans, particularly for words.

Sodium valproate: serial monitoring of EEG and serum levels.

The clinical and electroencephalographic (EEG) effects of sodium valproate were studied in four patients by means of serial 24-hour EEG recordings and simultaneous hourly determinations of serum drug concentrations. The patients all had frequent clinical seizures and generalized spike-wave discharges. Valproate appeared to reduce diurnal paroxysmal discharges (PD) and clinical seizures, but the effect on nocturnal PD was less marked. The extent and duration of the depression of PD and seizures varied. Altering the distribution of the total daily dose may change the pattern of clinical seizures and PD. Valproate concentrations fluctuated widely over 24 hours, and the significance of single estimations often cited in the literature appears dubious. Peak serum concentrations above 100 micrograms per milliliter may be necessary in some patients to achieve clinical and EEG improvement.

Significance of interictal bilateral temporal hypometabolism in temporal lobe epilepsy.

OBJECTIVE: To assess the clinical implications and the pathophysiologic determinants of interictal bitemporal hypometabolism (BTH) in temporal lobe epilepsy (TLE) not associated with bilateral MRI abnormalities or intracranial space-occupying lesions. METHODS: The authors compared the clinical, interictal, and ictal EEG, Wada test, and neuropsychology data of 15 patients with intractable complex partial seizures of temporal lobe origin and BTH with those of 13 consecutive patients with unilateral TLE associated with unilateral temporal hypometabolism (UTH) who remained seizure free for more than 3 years after anterior temporal lobectomy. 18F-fluorodeoxyglucose PET scans were analyzed visually and semiquantitatively, and ratios of counts in individual temporal areas to the rest of the cerebrum were compared with the corresponding values from 11 normal control subjects and with the nonepileptogenic hemisphere of the 13 patients with UTH. BTH was defined as more than 2.5 SDs below control values for two or more temporal areas on each side irrespective of any asymmetry. RESULTS: BTH reflected bilateral independent seizure onset in eight patients (53%). The topography of the metabolic depression was not a reliable predictor of epileptogenicity, but involvement of the inferior temporal gyrus was related specifically to ipsilateral seizure onset (70% sensitivity, 100% specificity). In patients with unilateral TLE, contralateral hypometabolism was associated with longer disease duration and worst memory performance during the Wada test, which amounted to global amnesia after ipsilateral injection in three patients, precluding surgical treatment. Contralateral seizure spread in the ictal EEG was significantly faster in patients with BTH. CONCLUSIONS: In TLE, symmetric or asymmetric BTH may signal bilateral independent seizure onset in approximately half the patients, especially when involving the inferior temporal gyrus. Alternatively, it may reflect an advanced stage of the disease process, characterized by a breakdown of the inhibitory mechanisms in the contralateral hemisphere, and secondary memory deficit associated with higher risk of postoperative memory decline. Patients with TLE and BTH but without bilateral MRI changes may still be operated on successfully, but surgical suitability should be proved by comprehensive intracranial EEG studies and Wada test.

Vagal nerve stimulation in epileptic encephalopathies.

OBJECTIVE: To study the effect of vagal nerve stimulation (VNS) in children with epileptic encephalopathies. METHODS AND MATERIALS: All children receiving VNS during a 2-year period at our center were studied prospectively for changes in seizure frequency, electroencephalogram (EEG), adaptive behavior, quality of life, and where appropriate, verbal/nonverbal performance. Assessments were made before and for at least 1 year after implant. RESULTS: Sixteen children were studied. One device was removed because of infection. Of the remaining 15 children, 4 had a >50% reduction and 2 had a >50% increase in seizure frequency at 1 year after implant. Median reduction in seizure frequency was 17%. There was no trend toward improvement of the EEG or adaptive behavior. Quality of life was unchanged in most areas, except in perceived treatment side effects and general behavior that were improved. In 6 children undergoing further assessment, there was a significant improvement in verbal performance; this did not correlate with reduction in seizure frequency. CONCLUSION: VNS did not significantly improve seizure frequency, severity, adaptive behavior, or the EEG during the first year of treatment for the group as a whole, although 4 children (27%) had a worthwhile reduction in seizure frequency. There were significant improvements in perceived treatment side effects and general behavior.

Clinical pharmacokinetics of newer antiepileptic drugs. Lamotrigine, vigabatrin, gabapentin and oxcarbazepine.

The clinical pharmacokinetics of the 4 antiepileptic drugs lamotrigine, vigabatrin, gabapentin and oxcarbazepine have been reviewed in this paper. All the drugs have linear kinetics and reliable absorption, although the saturation of transport across the gut may occur at high doses with gabapentin. All the drugs can be conveniently given as a twice daily dosage apart from gabapentin, which has a short half-life and a midday dose is needed. Unlike many of the older drugs, lamotrigine, vigabatrin and gabapentin have a predominantly renal excretion and are not metabolised through the cytochrome P450 system. They do not induce their own metabolism or that of other commonly used anticonvulsants. Similarly, clinically important interactions with other major classes of drugs metabolised this way, such as anticoagulants or steroid hormones, do not occur. Oxcarbazepine, however, can cause oral contraceptive pill failure. Oxcarbazepine is immediately metabolised to a hydroxy metabolite and could be considered a prodrug. It appears to have fewer pharmacokinetic interactions than carbamazepine. Valproic acid (sodium valproate) inhibits the glucuronidation of lamotrigine and increases its half-life; when used together, dosage modification of lamotrigine is needed to avoid toxicity.

A simple algorithm for a digital three-pole Butterworth filter of arbitrary cut-off frequency: application to digital electroencephalography.

Algorithms for low-pass and high-pass three-pole recursive Butterworth filters of a given cut-off frequency have been developed. A band-pass filter can be implemented by sequential application of algorithms for low- and high-pass filters. The algorithms correspond to infinite impulse-response filters that have been designed by applying the bilinear transformation to the transfer functions of the corresponding analog filters, resulting in a recursive digital filter with seven real coefficients. Expressions for filter coefficients as a function of the cut-off frequency and the sampling period are derived. Filter performance is evaluated and discussed. As in the case of their analog counterparts, their transfer function shows marked flattening over the pass band and gradually higher attenuation can be seen at frequencies above or below the cut-off frequency, with a slope of around 60 dB/decade. There is a 3 dB attenuation at the cut-off frequency and a gradual increase in phase shift over one decade above or below the cut-off frequency. Low-pass filters show a maximum overshoot of 8% and high-pass filters show a maximum downwards overshoot of approximately 35%. The filter is mildly under-damped, with a damping factor of 0.5. On an IBM 300GL personal computer at 600 MH with 128 MB RAM, filtering time with MATLAB 5.2 running under Windows 98 is of the order of 50 ms for 60000 samples. This will be adequate for on-line electroencephalography (EEG) applications. The simplicity of the algorithm to calculate filter coefficients for an arbitrary cut-off frequency can be useful to modern EEG laboratories and software designers for electrophysiological applications.

Modern electroencephalography: its role in epilepsy management.

Epilepsy is the area in which electroencephalography is probably of greatest potential clinical value, and yet is most abused. In general, the sensitivity of the waking interictal EEG for detecting epilepsy and its specificity for distinguishing epilepsy from other episodic disorders are both very limited, and routine examination for diagnostic screening or follow up is of little value. However, as this review attempts to demonstrate, EEG is of crucial importance for answering specific, clearly defined questions which commonly arise in the management of seizure disorders, especially so, when non-routine and if necessary complex investigations are undertaken, to address the problems of individual patients.

Psychomotor EEG variant of Gibbs: an association with underlying structural pathology.

OBJECTIVE: To describe the association between a unilateral mid-temporal rhythmic theta discharge ("psychomotor variant of Gibbs") with neuroimaging-demonstrated underlying mass lesion. METHODS: Standard routine awake and scalp electroencephalography, continuous video-EEG monitoring and magnetic resonance brain imaging were employed in the diagnostic work-up of a 9-year-old boy with a severe behavioural disturbance and episodic outbursts of aggression. RESULTS: EEG showed a unilateral mid-temporal rhythmic discharge which was continuous in drowsiness and which remained confined to the right hemisphere. MRI showed a lesion in the temporal horn of the right lateral ventricle displacing superiorly the white matter stem of the right temporal lobe. CONCLUSION: A rhythmic mid-temporal theta discharge, commonly regarded as a benign EEG variant, may, in some patients, reflect underlying structural pathology. Neuroimaging should be considered particularly when this EEG pattern remains confined to one hemisphere.

Continuous bitemporal rhythmic subclinical epileptiform activity in an adult without epileptic seizures.

We report the occurrence of an unusually sustained, non-reactive, bilateral, rhythmic, subclinical discharge in a young adult without epileptic seizures

Sensitivity of recordings at sphenoidal electrode site for detecting seizure onset: evidence from scalp, superficial and deep foramen ovale recordings.

OBJECTIVES: Some authors have recently stressed that the position of the tip of sphenoidal electrodes plays a crucial role in their efficacy in detecting ictal onset. An opportunity to test this hypothesis is provided by recordings from the most superficial contacts of foramen ovale (FO) electrode bundles because these contacts are located at the FO, in a position equivalent to that of optimally located sphenoidal electrodes. To simplify wording, recordings obtained by superficial FO electrodes will hereafter be called sphenoidal recordings, although they have not been obtained with standard sphenoidal electrodes. The sensitivities of simultaneous scalp and sphenoidal recordings for detecting ictal onset have been compared with each other, and with a 'gold standard' provided by simultaneous deep intracranial FO recordings from the mesial aspect of the temporal lobe. METHODS: Three hundred and fourteen seizures obtained from 110 patients under telemetric presurgical assessment for temporal lobe epilepsy have been studied. Scalp electrodes included anterior temporal placements. All scalp electrodes were considered when identifying seizure onset but the anterior temporal electrodes were most frequently involved. RESULTS: Ictal onset time at sphenoidal and scalp recordings: initial ictal changes appeared simultaneously in scalp and sphenoidal recordings in 123 seizures (39.2%). Initial changes occurred earlier in sphenoidal recordings in 63 seizures (20.1%), whereas they were seen earlier on the scalp in 76 seizures (24.2%). Artefacts prevented the comparison between sphenoidal and scalp recordings in 16 seizures (5.1%) and no ictal changes were seen on the scalp and/or sphenoidal recordings in 36 seizures (11.5%). In most of the 63 seizures where ictal changes appeared earlier in sphenoidal recordings, a delayed ipsilateral scalp onset was seen as the signal amplitude increased or scalp changes could be identified retrospectively on the scalp with an onset which appeared simultaneous and ipsilateral to the initial sphenoidal changes. Sphenoidal recordings supplied additional information when compared to scalp recordings in only 22 seizures (7%): in 5 seizures with artefacts on the scalp, in 6 seizures with no changes on the scalp and in 11 seizures with discrepant laterality at onset. Congruence in laterality with respect to deep intracraneal FO recordings: of the 61 seizures with unilateral onset on the scalp, onsets at sphenoidal recordings and deep FO electrodes were ipsilateral in most cases. In only 3 of these 61 seizures (4.9%), sphenoidal recordings lateralized ipsilateral to the deep FO electrodes in the presence of a contralateral onset on the scalp. In 14 among the 122 seizures (11.5%) with bilateral asymmetrical onset on the scalp, sphenoidal recordings lateralized seizure onset ipsilateral to the deep FO electrodes in the presence of a contralateral scalp onset. Thus, when compared with scalp EEG, sphenoidal recordings increased laterality congruence with respect to deep FO electrodes in 17 seizures (5.4%). CONCLUSIONS: Extracranial electrodes located next to the FO at the sphenoidal electrode site yield an improvement over suitable surface electrodes in the identification of ictal onset in only 5.4-7% of seizures. Such improvement derives from the fact that the low amplitude signals often seen at seizure onset may show higher amplitude on sphenoidal than on scalp recordings.

Early serial EEG in hypoxic ischaemic encephalopathy.

OBJECTIVES: To perform early serial EEGs in infants with hypoxic ischaemic encephalopathy (HIE) and compare the findings with neurodevelopmental outcome. METHODS: Nine full-term neonates with HIE had simultaneous video-EEG polygraphic studies within 8 h of birth. The EEG was repeated at 12-24 h intervals. All surviving infants had a neurodevelopmental assessment at 1 year. RESULTS: Two infants had a normal or mildly abnormal EEG within 8 h of birth and neurodevelopmental outcome was normal. Seven infants had severely depressed background activity in the first 8 h of life. In 3 infants the EEG activity recovered within 12-24 h showing continuous activity with no or only minor abnormalities. All these infants had a normal outcome. The remaining 4 infants, who also had an initially inactive recording, subsequently developed severe background abnormalities. At follow-up, two infants had died and the remainder developed major neurological sequelae. CONCLUSIONS: Early EEG is an excellent prognostic indicator for a favourable outcome if normal within the first 8 h of life and for a poor outcome if the background activity continues to be inactive or grossly abnormal beyond 8-12 h of life. However, an inactive or very depressed EEG within the first 8 h of life can be associated with good outcome if the EEG activity recovers within 12 h.

Comparison of sphenoidal, foramen ovale and anterior temporal placements for detecting interictal epileptiform discharges in presurgical assessment for temporal lobe epilepsy.

OBJECTIVES: Some authors have recently stressed that the position of the tip of the sphenoidal electrode plays a crucial role in its efficacy to detect mesio-basal spikes. We have tested this hypothesis by comparing the sensitivity of a contact of a foramen ovale bundle located at the foramen ovale (CFO) with scalp electrodes in detecting interictal epileptiform discharges. We have also compared deep and superficial foramen ovale contacts in the same bundle in order to establish whether deeper contacts can detect epileptiform discharges not seen at the foramen ovale or on the scalp. METHODS: The sensitivity for detecting epileptiform discharges of simultaneous intracranial and scalp EEG recordings from 20 patients under telemetric presurgical assessment for temporal lobe epilepsy were compared. RESULTS: Out of 2280 epileptiform discharges evaluated, about 70% were seen only at the deep foramen ovale contacts. Out of the 722 discharges recorded by CFO and/or scalp electrodes, 698 were seen at the CFO and 690 at the scalp anterior temporal electrode. Only on 29 occasions (4.15%) were discharges recorded at the CFO and not at the anterior temporal electrode. On 21 occasions (3.04%) CFO failed to detect discharges seen at the anterior temporal electrode. CONCLUSIONS: Our findings confirm previous results suggesting that sphenoidal electrodes, however accurately positioned, offer no significant increase in detection sensitivity compared with anterior temporal scalp electrodes. In addition, these results confirm that a large proportion of discharges seen at the deepest foramen ovale contacts are not seen either on the scalp nor at the superficial foramen ovale contacts.

Combined use of subdural and intracerebral electrodes in preoperative evaluation of epilepsy.

For intracranial recording of partial seizures considered to originate from one of the temporal or frontal lobes, the team in the Utrecht Academic Hospital has used subdural, multicontact, flexible electrodes since 1972. These are introduced through bilateral, frontocentral trephine holes and are manipulated under fluoroscopy to cover most of the cerebral convexity. It became evident that in many patients, additional placements to record from intracerebral structures were indispensable. Therefore, using the same trephine holes, an additional 2 to 4 depth electrodes were stereotactically implanted in the mesial temporal and/or frontal structures, as appropriate. An extensive intra- and extracerebral spatial representation of the epileptogenic zone was thus obtained. We report here the methods for manufacturing and applying these electrodes and our clinical experience with 28 patients. The results obtained so far stress the value of combining subdural and depth electroencephalographic monitoring in the presurgical selection of patients suffering from medically refractory complex partial seizures. By miniaturizing the electrodes, extensive areas of the brain can be investigated without craniotomy or multiple burr holes.