Predictors and reasons for epilepsy patients to decline surgery: a prospective study.

BACKGROUND: In patients with drug-resistant focal epilepsy, resective surgery is the most successful treatment option to achieve seizure freedom. However, a surprisingly high rate of patients declines their physicians' recommendation to undergo removal of the seizure focus or-if necessary-further video-EEG monitoring (VEM). METHODS: In this prospective study, consecutive patients in presurgical assessment with at least one scalp VEM between 2016 and 2018 were included. We assessed both epilepsy-related and psychosocial variables as well as decision-making of physicians and patients, including reasons for decline in the latter. RESULTS: Out of 116 patients with a total of 165 VEM, 20 patients were eventually found to be ineligible for resection, 51 declined, and 45 agreed on recommendations for resection or further VEM diagnostics. Patients most frequently declined due to general fear of brain surgery (n = 30, 59%) and currently lower seizure frequency (n = 11, 22%). An independent predictor of patients' decline was less epilepsy-related fear (OR 0.43; p = 0.02) assessed in a standardised questionnaire. CONCLUSION: Half of the patients potentially eligible for resective surgery decline the operation or further VEM procedures. Patients who decline are more fearful of brain surgery than of ongoing disabling seizures. More insight is needed to improve counselling of patients.

Presurgical video-EEG monitoring with foramen ovale and epidural peg electrodes: a 25-year perspective.

BACKGROUND: Epilepsy surgery cases are becoming more complex and increasingly require invasive video-EEG monitoring (VEM) with intracranial subdural or intracerebral electrodes, exposing patients to substantial risks. We assessed the utility and safety of using foramen ovale (FO) and epidural peg electrodes (FOP) as a next step diagnostic approach following scalp VEM. METHODS: We analyzed clinical, electrophysiological, and imaging characteristics of 180 consecutive patients that underwent FOP VEM between 1996 and 2021. Multivariate logistic regression was used to assess predictors of clinical and electrophysiological outcomes. RESULTS: FOP VEM allowed for immediate resection recommendation in 36 patients (20.0%) and excluded this option in 85 (47.2%). Fifty-nine (32.8%) patients required additional invasive EEG investigations; however, only eight with bilateral recordings. FOP VEM identified the ictal onset in 137 patients, compared to 96 during prior scalp VEM, p = .004. Predictors for determination of ictal onset were temporal lobe epilepsy (OR 2.9, p = .03) and lesional imaging (OR 3.1, p = .01). Predictors for surgery recommendation were temporal lobe epilepsy (OR 6.8, p < .001), FO seizure onset (OR 6.1, p = .002), and unilateral interictal epileptic activity (OR 3.8, p = .02). One-year postsurgical seizure freedom (53.3% of patients) was predicted by FO ictal onset (OR 5.8, p = .01). Two patients experienced intracerebral bleeding without persisting neurologic sequelae. CONCLUSION: FOP VEM adds clinically significant electrophysiological information leading to treatment decisions in two-thirds of cases with a good benefit-risk profile. Predictors identified for electrophysiological and clinical outcome can assist in optimally selecting patients for this safe diagnostic approach.

Surgery in intractable epilepsy-physicians' recommendations and patients' decisions.

OBJECTIVES: To identify demographic and clinical variables independently associated with patients' decisions against their physicians' recommendations for resective epilepsy surgery or further scalp video-EEG monitoring (sca-VEM), semi-invasive (sem-)VEM with foramen ovale and/or peg electrodes, and invasive (in-)VEM. METHODS: Consecutive patients, who underwent presurgical assessment with at least one sca-VEM between 2010 and 2014, were included into this retrospective analysis. Multivariate analysis was used to identify independent variables associated with patients' decisions. RESULTS: Within the study period, 352 patients underwent 544 VEM sessions comprising 451 sca-, 36 sem-, and 57 in-VEMs. Eventually, 96 patients were recommended resective surgery, and 106 were ineligible candidates; 149 patients denied further necessary VEMs; thus, no decision could be made. After sca- or additional sem-VEM, nine out of 51 eligible patients (17.6%) rejected resection. One hundred and ten patients were recommended in-VEM, 52 of those (47.2%) declined. Variables independently associated with rejection of in-VEM comprised intellectual disability (OR 4.721, 95% CI 1.047-21.284), extratemporal focal aware non-motor seizures ("aura") vs. no "aura" (OR 0.338, 95% CI 0.124-0.923), and unilateral or bilateral vs. no MRI lesion (OR 0.248, 95% CI 0.100-0.614 and 0.149, 95% CI 0.027-0.829, respectively). CONCLUSIONS: During and after presurgical evaluation, patients with intractable focal epilepsy declined resections and intracranial EEGs, as recommended by their epileptologists, in almost 20% and 50% of cases. This calls for early and thorough counseling of patients on risks and benefits of epilepsy surgery. Future prospective studies should ask patients in depth for specific reasons why they decline their physicians' recommendations.

Craniotomy Size for Subdural Grid Electrode Placement in Invasive Epilepsy Diagnostics.

BACKGROUND: Traditionally, for subdural grid electrode placement, large craniotomies have been applied for optimal electrode placement. Nowadays, microneurosurgeons prefer patient-tailored minimally invasive approaches. Absolute figures on craniotomy size have never been reported. To elucidate the craniotomy size necessary for successful diagnostics, we reviewed our single-center experience. METHODS: Within 3 years, 58 patients with focal epilepsies underwent subdural grid implantation using patient-tailored navigation-based craniotomies. Craniotomy sizes were measured retrospectively. The number of electrodes and the feasibility of the resection were evaluated. Sixteen historical patients served as controls. RESULTS: In all 58 patients, subdural electrodes were implanted as planned through tailored craniotomies. The mean craniotomy size was 28 ± 15 cm2 via which 55 ± 16 electrodes were implanted. In temporal lobe diagnostics, even smaller craniotomies were applied (21 ± 11 cm2). Craniotomies were significantly smaller than in historical controls (65 ± 23 cm2, p < 0.05), while the mean number of electrodes was comparable. The mean operation time was shorter and complications were reduced in tailored craniotomies. CONCLUSION: Craniotomy size for subdural electrode implantation is controversial. Some surgeons favor large craniotomies, while others strive for minimally invasive approaches. For the first time, we measured the actual craniotomy size for subdural grid electrode implantation. All procedures were straightforward. We therefore advocate for patient-tailored minimally invasive approaches - standard in modern microneurosurgery - in epilepsy surgery as well.

The anticonvulsant lamotrigine enhances Ih in layer 2/3 neocortical pyramidal neurons of patients with pharmacoresistant epilepsy.

Alterations of the hyperpolarization activated nonselective cation current (Ih) are associated with epileptogenesis. Accordingly, the second-generation antiepileptic drug lamotrigine (LTG) enhances Ih in rodent hippocampus. We directly evaluated here whether LTG fails to enhance Ih in neocortical slices from patients with pharmacoresistant epilepsy. With somatic current clamp recordings we observed that LTG depolarized the membrane potential, decreased the input resistance and increased the "sag" in human layer 2/3 neocortical pyramidal neurons when confounding IKir was blocked. In subsequent voltage clamp recordings we confirmed a LTG induced increase of Ih that was qualitatively similar to the one we found in rat neocortical and hippocampal pyramidal neurons. This increase is sufficient to curtail single excitatory postsynaptic potentials (EPSPs) and reduces their temporal summation in human neocortical pyramidal neurons under physiological conditions, i.e. without blocking any other currents, as estimated by sharp microelectrode recordings. Taken together LTG increases Ih and thereby alters neuronal excitability, even in neurons of pharmacoresistant patients. However, whether this increase fully countervails the deficits of Ih in epileptic patients, remains elusive.

Components of neuronal chloride transport in rat and human neocortex.

Considerable evidence indicates disturbances in the ionic gradient of GABAA receptor-mediated inhibition of neurones in human epileptogenic tissues. Two contending mechanisms have been proposed, reduced outward and increased inward Cl⁻ transporters. We investigated the properties of Cl⁻ transport in human and rat neocortical neurones (layer II/III) using intracellular recordings in slices of cortical tissue. We measured the alterations in reversal potential of the pharmacologically isolated inhibitory postsynaptic potential mediated by GABAA receptors (IPSPA) to estimate the ionic gradient and kinetics of Cl⁻ efflux after Cl⁻ injections before and during application of selected blockers of Cl⁻ routes (furosemide, bumetanide, 9-anthracene carboxylic acid and Cs+). Neurones from human epileptogenic cortex exhibited a fairly depolarized reversal potential of GABAA receptor-mediated inhibition (EIPSP-A) of -61.9 ± 8.3 mV. In about half of the neurones, the EIPSP-A averaged -55.2 ± 5.7 mV, in the other half, 68.6 ± 2.3 mV, similar to rat neurones (-68.9 ± 2.6 mV). After injections of Cl⁻, IPSPA recovered in human neurones with an average time constant (τ) of 19.0 ± 9.6 s (rat neurones: 7.2 ± 2.4 s). We calculated Cl⁻ extrusion rates (1/τ) via individual routes from the τ values obtained in different experimental conditions, revealing that, for example, the K+-coupled Cl⁻ transporter KCC2 comprises 45.3% of the total rate in rat neurones. In human neurones, the total rate of Cl⁻ extrusion was 63.9% smaller, and rates via KCC2, the Na+-K+-2Cl⁻ transporter NKCC1 and the voltage-gatedCl− channelClCwere smaller than in rat neurones by 80.0%, 61.7% and 79.9%, respectively. The rate via anion exchangers conversely was 14.4% larger in human than in rat neurones. We propose that (i) KCC2 is the major route of Cl⁻ extrusion in cortical neurones, (ii) reduced KCC2 is the initial step of disturbed Cl⁻ regulation and (iii) reductions in KCC2 contribute to depolarizing EIPSP-A of neurones in human epileptogenic neocortex.

Irregular RNA splicing curtails postsynaptic gephyrin in the cornu ammonis of patients with epilepsy.

Anomalous hippocampal inhibition is involved in temporal lobe epilepsy, and reduced gephyrin immunoreactivity in the temporal lobe epilepsy hippocampus has been reported recently. However, the mechanisms responsible for curtailing postsynaptic gephyrin scaffolds are poorly understood. Here, we have investigated gephyrin expression in the hippocampus of patients with intractable temporal lobe epilepsy. Immunohistochemical and western blot analyses revealed irregular gephyrin expression in the cornu ammonis of patients with temporal lobe epilepsy and four abnormally spliced gephyrins lacking several exons in their G-domains were isolated. Identified temporal lobe epilepsy gephyrins have oligomerization deficits and they curtail hippocampal postsynaptic gephyrin and GABA(A) receptor α2 while interacting with regularly spliced gephyrins. We found that cellular stress (alkalosis and hyperthermia) induces exon skipping in gephyrin messenger RNA, which is responsible for curtailed postsynaptic gephyrin and GABA(A) receptor α2 scaffolds. Accordingly, we did not obtain evidence for gephyrin gene mutations in patients with temporal lobe epilepsy. Cellular stress such as alkalosis, for example arising from seizure activity, could thus facilitate the development of temporal lobe epilepsy by reducing GABA(A) receptor α2-mediated hippocampal synaptic transmission selectively in the cornu ammonis.

Evaluating reference genes to normalize gene expression in human epileptogenic brain tissues.

Several reference genes have been used to quantify gene expression in human epilepsy surgery tissue. However, their reliability has not been validated in detail, although this is crucial in interpreting epilepsy-related changes of gene expression. We evaluated 12 potential reference genes in neocortical tissues resected from patients with temporal lobe epilepsy (TLE) with either few or many seizures (n=6 each) and post mortem controls (n=6) using geNorm and NormFinder algorithms. For all candidate reference genes threshold cycle (C(T)) values were measured. geNorm analysis revealed that the expression of e.g. glyceraldehyde-3-phosphate-dehydrogenase (GAPDH) and hypoxanthine phosphoribosyl-transferase (HPRT) is unstable, whereas synaptophysin (SYP) and neuron-specific enolase (NSE)/mitochondrial 39S ribosomal protein L28 (MRPL) are most stably expressed. The geometric mean of SYP, NSE and MRPL levels is recommended as normalization factor (NF). NormFinder analysis, in contrast, indicated HPRT as the most stable single gene and recommended the geometric mean of TATA-box binding protein (TBP) and NSE levels as NF. Different values of upregulation of glial fibrillary protein (GFAP) expression were found in TLE tissue compared to control tissue depending on the NF used: 4.5-fold (geNorm-NF), 4.7-fold (NormFinder-NF), 4.2-fold (vs. GAPDH) and 7.8-fold (vs. HPRT). The expression of GABA(A) receptor subunit α5 (GARα5) was unaltered in the TLE groups compared to controls (geNorm-NF, NormFinder-NF, vs. GAPDH). However, normalization to HPRT suggests an apparent increase of GARα5 expression. In conclusion, the geNorm-NF (SYP/NSE/MRPL) and the NormFinder-NF (TBP/NSE) are equally suitable for normalization of gene expression in the human epileptogenic neocortex. In contrast, normalization to single and probably less stably expressed genes may not deliver accurate results.

Verapamil attenuates the malignant treatment course in recurrent status epilepticus.

In the scenario of refractory status epilepticus, the recommended approach of intensive care treatment is limited with respect to the available pharmacodynamic variability and its impeding, severe side effects. Alternative treatment options are therefore urgently needed. In the case described, a patient with nonlesional frontal lobe epilepsy had a high-frequency series of tonic seizures, which evolved into a malignant form of status epilepticus. Co-administration of verapamil, a potent multidrug transporter inhibitor, was followed by significant reduction in seizure frequency. We discuss the putative role of verapamil and the specific risk factors for this malignant treatment course.

Hyperpolarization-activated cation currents in human epileptogenic neocortex.

PURPOSE: Hyperpolarization-activated cation currents (I(H)) play a pivotal role in the control of neuronal excitability. In animal models of epilepsy both increases and decreases of I(H) have been reported. We, therefore, characterized properties of I(H) in human epileptogenic neocortex. METHODS: Layer II/III neurons in slices from epilepsy surgery tissues and rat cortex were investigated with whole-cell patch-clamp recordings. RESULTS: A total of 484 neurons from 96 temporal lobe epilepsy (TLE) tissues and 32 neurons from 8 frontal lobe epilepsy (FLE) tissues were recorded. Voltage-clamp recordings revealed on hyperpolarizing command steps two time- and voltage-dependent inward currents, namely a fast, Ba(2+)-sensitive current (K(IR)) and a slowly activating current, namely consisting of two kinetically distinct components sensitive to the established I(H) blocker ZD7288. Only, the fast component (I(H)(fast)) of TLE neurons was on average smaller and activated more slowly (density 2.7 +/- 1.6 pA/pF; tau 38.4 +/- 34.0 ms) than in FLE neurons (4.7 +/- 2.3 pA/pF; 16.6 +/- 7.9 ms; p < 0.001 for both). Within the TLE tissues the I(H)(fast) density (averaged per patient) was smaller in cases with numerous annual grand mal seizures (GM; 2.2 +/- 0.6 pA/pF) compared to those with few GM (2.8 +/- 1.0 pA/pF; p = 0.0184). A similar difference was obtained in the case of complex partial seizures (CPS; many CPS 2.2 +/- 0.6 pA/pF; few CPS 2.9 +/- 1.0 pA/pF, p = 0.0037). DISCUSSION: The biophysical properties of I(H) in cortices from TLE, FLE, and rat tissue suggest a deficit of HCN1 subunits in the human epileptogenic neocortex, which in turn may increase excitability and probability of seizure activity.