The effects of barometric pressure changes and other climate factors on the frequency of seizures

Virginia Thornley, M.D., Neurologist, Epileptologist

June 3, 2018


It is not common for a patient to complain of seizures seeming to increase immediately before a hurricane or a big storm. Do these changes truly correlate with outside environmental factors? This article seeks to review the literature to determine the cause and mechanisms of how weather risk factors might affect epilepsy and frequency of seizures.  There is a paucity of information of barometric effects and weather changes on exacerbation of seizure frequency.

Changes in atmospheric pressure correlated with seizures 

Atmospheric pressure is defined as the weight of the atmosphere. At sea level, it is 101,325 pascals, 14.5969 pounds/square inch or 1013.3 millibars. It is also referred to as barometric pressure.

In one article studying 191 patients, with an increase in atmospheric pressure variability, seizures were noted to increase. The atmospheric pressure was obtained from metropolitan weather stations in Seattle. The maximum, minimum and changes were correlated with the number of seizures being monitored in a telemetry unit over 2005-2006. Patients with known epilepsy had an odds ratio of 2.6 (p=0.02) if the atmospheric pressure varied over 5.5mBar (1).

Higher temperatures correlated with more febrile seizures

In another study of 108,628 pediatric patients from January 2005-December, 2015 were studied regarding the effect of barometric pressure on the frequency of seizures. They were classified as febrile seizures, afebrile, epilepsy or status epilepticus. 53% presented as febrile seizures while 5.9% presented as status epilepticus. Mean atmospheric pressure was 1015.5hPa over the 11 year period. The mean temperature was 14.7 degrees Celsius with a variation of 8.3 degrees Celsius throughout the day.  The study demonstrated febrile seizures were influenced by the temperature. At lower temperatures, the emergency room visits were less while at higher temperatures the visits increased (2).


Low barometric pressure, high air humidity increases seizures, high ambient temperature improved seizures

In another study where temperature, barometric pressure, and humidity were correlated with seizure frequency, 604 patients were studied between 2006-2010. The study showed that with a 10.7hPa lower atmospheric pressure there was an increase in seizures by 14%. Those with less severe seizures had an increase of 36%. Relative humidity of >80% correlated with increased seizures of 48%. A high ambient temperature of more than 20 degrees Celsius reduced seizures by 46% (4).

Cold temperature worsen seizures

In a study of 30 patients ages (19-54), patients with epilepsy appeared to have more active seizures during the seasons of spring, autumn and winter and less during summer of about 7%. During stable weather, it was 43% patients and unstable weather 63% had seizures. EEG’s changes occurred more frequently during winter. During winter seizures increased by 40%, in spring it increased 40% and spring by 43.3% (3).

In summary

While anecdotally, there is a correlation of exacerbation of seizure frequency to weather changes, the literature shows mixed results and some of them are small in number. One study showed a correlation of changes of more than 5.5mBar in barometric pressure leading to increased seizures frequency, another showed that it is the reduction in the atmospheric pressure itself that increased seizures. 1 study showed that high humidity may increase seizures. 2 studies showed that cold temperatures worsened seizures, while 1 study showed that higher ambient temperature worsened febrile seizures.

The data that was demonstrated is not uniform in the acquisition of information and there is a large variety of conditions. One study was primarily taken from ER visits another was information from inpatient video EEG monitoring units where the subset of patients may be completely different. In addition, there is a wide heterogeneity in etiologies of seizures which comes into play. Regardless, patients know their own symptoms, usually, if something is noted to trigger an event is it probably real.





  1. Doherty, et al, “Atmospheric pressure and seizure frequency in the epileptic unit: preliminary observations,” Epilepsia, 2007, Sep., 48 (9):1764-1767.
  2. Kim, et al, “The effects of weather on pediatric seizure; a single -center retrospective study,” Sci. Total Environ. , 2017, Dec., (609):535-540.
  3. Motta, et al, “Seizure frequency and bioelectric brain activity in epileptic patients in stable and unstable atmospheric pressure and temperature in different seasons of the year–a preliminary report,” Neurol. Neurochir. Pol, 2011, Nov.-Dec., 45(6):561-566.
  4. Rakers, et al, “Weather as a risk factor for epileptic seizures: a case-crossover study,” Epilepsia, 2017, Jul., 58(7): 1297-95.
cannabidiol, Epilepsy

Scientific and clinical evidence of cannabidiol (CBD) and seizure control: mechanisms, randomized controlled clinical trials, open label trials and animal models

Virginia Thornley, M.D., Neurologist, Epileptologist 


May 22, 2018


There are numerous scientific studies that have studied the effect of cannabidiol by itself on seizure control encompassing animal models, longitudinal observational studies, case series and currently randomized double-blinded placebo-controlled clinical trials. It is difficult to ignore the wealth of information regarding the medical value of cannabidiol with a significant role in the treatment of epilepsy.

The endocannabinoid pathway and cannabinoids

The endocannabinoid pathway is found naturally within our system, comprising of receptors, transporters, and endocannabinoids. It is responsible for the sense of well-being one gets after running referred to as the “runner’s high,” and not endorphins, serotonin or noradrenergic neurotransmitters as their molecular sizes are too large to pass through the blood-brain barrier. There are 2 types of receptors, CB1 and CB2 receptors. CB1 is found predominantly within the nervous system and is the receptor on which tetrahydrocannabinol works and it is through this binding where psychoactive properties arise. There are two metabolites within the endocannabinoid pathway, anandamide for which tetrahydrocannabinol (THC) is a phytomimetic and 2-arachidonoyl-glycerol for which cannabidiol is a phytomimetic. Cannabidiol (CBD) acts as an inverse agonist on the CB1 receptor, with a weak affinity. 100 times of cannabidiol is needed to get the same psychoactive properties as tetrahydrocannabinol. When CBD is combined with THC the side effects of paranoia, hyperactivity and agitation become less because it is an inverse agonist of the CB1 receptor. In many animal studies, cannabidiol has anti-inflammatory, anti-oxidative and neuroprotective actions within the nervous system (8).

Mechanisms by which cannabidiol works 

It is thought to modulate the neurotransmitter system. Endocannabinoids are increased as a result if hyperexcitability in the nervous system. CBD can regulate intracellular calcium during hyperexcitability states in the hippocampus in the temporal lobe. CBD can regulate NMDA (N-methyl-D-aspartate) receptor transmission and increase serotonergic 5HT-1A (5-hydroxytryptamine)receptor transmission and reduces GABA, 5-HT1A, and norepinephrine synaptic uptake (9). Cannabidiol is thought to be neuroprotective through its role in controlling intracellular calcium. Excess calcium can activate a cascade of neurochemical events leading to cell degeneration and death through lipases, endonucleases, and proteases. In one study in rat models, there was a suggestion that treatment of seizures was not just at the neurotransmitter level but also modulates the oscillatory nature, neuronal loss and post-ictal lethargy of the status epilepticus model.

Scientific evidence in animal models

Animal studies show that the effectiveness of cannabis is at the level of the CB1 receptor. With the deletion of the CB1 receptors in the forebrain excitatory neurons in the mice model, Kainate-induced seizures were more prominent. The presence of CB1 receptors in the hippocampal gyrus seems to protect against Kainate-induced seizures. Viral-induced CB1 overexpression resulted in less Kainate-induced seizures, CA pyramidal cell 3 cell death. This demonstrates that the presence of the CB1 receptor can limit seizures and reduces gliosis and apoptosis (4).



In animal studies, the CB1 receptors increased 1 week after pilocarpine-induced seizures in the CA1-3 striatum oriens and the dentate gyrus. Patients with temporal lobe epilepsy had reduced Anandamide and increased CB1 receptors suggesting an up-regulation of the CB1 receptor as a homeostatic mechanism in the presence of seizures which can reduce excitatory neurotransmitters (4). This compensatory mechanism may be impaired with long-standing seizures and hippocampal sclerosis and refractoriness to pharmacologic measures.

Case series report

In a small study on patients with tumors with seizures, in 3 patients who were medically refractory were started on cannabidiol (Epidiolex) to treat seizures. 2 out of the 3 had improvement in seizures while all 3 had improvement in the severity in the University of Alabama (2).

Evidence in longitudinal observational studies

In one study of 57 patients, ages 1-20 years old, CBD:THC was given at a ratio of 20:1 with the CBD component of 11.4 mg/kg/day. The patients were followed longitudinally for 3 months with a follow-up time of 18 months. 56% or 26 patients had <50% reduction of seizures. No difference was noted between the causes of the seizure and the type of cannabis used. Younger ages of 10 years old and below had a statistically better outcome compared to an older age. Those with higher doses of CBD of >11.4mg/kg/day had a statistically better outcome compared to 11.4mg/kg/day and below. There were side effects in about 46% of patients leading to stopping the protocol. These studies suggest that cannabidiol enriched treatment may be beneficial in seizure control particularly in the pediatric population.  (1).

Open-label studies

In an open-label trial, 214 patients were studied between the ages 1-30, with pharmacoresistant epilepsy. There were 162 in the safety follow-up of 12 weeks, 137 were in the efficacy analysis. For the safety group, 33 had Dravet syndrome and 31 had Lennox-Gastaut syndrome. The rest had medically refractory seizures from different causes. Side effects were mild to moderate including diarrhea, lack of appetite, somnolence, fatigue, and convulsion. 5 had a cessation of treatment related to adverse effects. Serious events were reported in 48 patients with 1 death unrelated to cannabidiol. 20 had severe adverse effect including status epilepticus. The median number of seizures at baseline was 30 which was reduced to 15 per month with a 36.5% reduction of motor seizures (7).

Evidence in randomized controlled clinical trials 

In a multi-country study was performed on Dravet syndrome and effect of cannabidiol in a randomized double-blind trial of cannabidiol versus placebo and in young adults between the ages of 2-18. Dravet syndrome is an epileptic syndrome involving myoclonic epilepsy during childhood which may progress attributed to an SCN1A gene abnormality. There was a 4 week baseline period followed by a 14 week treatment period. The dosages of cannabidiol were increased gradually to 20mg/kg/day. Those in the cannabidiol group was matched to a placebo control. The endpoints were the percentage of change and Caregiver Global Impression of Change (CGIC). In 23 center in the U.S. and in Europe, 120 patients underwent randomization, mean age was 9.8 years old. 108 completed treatment. The median number of drugs was 3 and the most commonly taken were clobazam, valproate, stiripentol, levetiracetam, and topiramate. The most common type of seizures was generalized tonic-clonic followed by secondary generalized tonic-clonic seizures. 114/118 children presented with developmental delay. Adverse reactions were mild to moderate including somnolence, diarrhea and loss of appetite. Elevated liver enzymes were found in those taking valproate likely related to drug-drug interactions. The reduction of seizures was considered meaningful while no change in non-convulsive episodes was noted. In the cannabidiol group, convulsive seizures reduced from 12.4 seizures to 5.9 per month while the placebo control group had a reduction of seizures from 14.9 to 14.1 which was not statistically significant. A reduction of more than 50% of seizures occurred in 43% of patients in the cannabidiol group and 27% in the control cohort. 3 patients in the cannabidiol group and no one in the placebo group became free of seizures. 62% of caregivers thought the condition improved in the cannabidiol group as opposed to 34% in the placebo group (5).


Another randomized placebo-controlled trial in Lennox-Gastaut syndrome was done using cannabidiol versus placebo. Lennox-Gastaut Syndrome is characterized by multiple seizure types with a slow spike and wave of 2.5 Hz or slower on EEG.  This study covered 30 clinical trial centers between the ages 2-55 with 2 or more seizures per week over 28 days. 225 patients were randomized with 76 in the group for cannabidiol at 20mg/kg/day, 73 in the cannabidiol group at 10mg/kg/day and 76 in the placebo cohort. The reduction in median of drop attacks was 41.9% in the 20mg cannabidiol group, 37% in the 10mg cannabidiol group and 17.2% in the placebo group which was statistically significant. Side effects were somnolence, diarrhea and poor appetite which was dose-related. 9% had higher liver function tests. The study concluded that addition of cannabidiol of either 10mg/kg/day or 20mg/kg/day in addition to standard anti-epileptic agents resulted in a significant reduction of seizures(6).

Cannabidiol as an add-on adjunct for refractory seizures

In another study in Slovenia, add-on cannabidiol was given to 66 patients who were deemed medically refractory at a dosage of 8mg/kg/day. 32 or 48% of patients experienced fewer seizures of more than 50% reduction. 14 (21%) were seizure free. No patient had to worsen and 15 or 22.7% there was no effect. Patients reported less robust seizures, less recovery time and less time duration of the seizures as positive outcomes. Adverse effects were seen in 5 patients or 0.07% of patients. They concluded that there are some beneficial effects of cannabidiol as an add-on adjunctive treatment in controlling medically refractory epilepsy(3). However, this study focused on cannabidiol as an adjunctive treatment, not as monotherapy.  Regardless, there are some beneficial aspects as evidenced in this study (3).

In summary

There is growing evidence that cannabidiol which is the non-psychoactive component of the Cannabis sativa plant is effective in treating intractable seizures, from the mouse model to randomized controlled clinical trials, which can no longer be ignored. There are mostly mild to moderate side effects involving the gastointestinal and neuropsychiatric system, although severe adverse outcomes include status epilepticus. There were no fatal outcomes associated with the use of cannabidiol. The real question are the long-term side effects and drug-drug interactions which can be studied once the cannabidiol is well-established as a conventional agent in the future.





  1. Hausman-Kedem, M., et al, “Efficacy of CBD-enriched medical cannabis for treatment of refractory epilepsy in children and adolescents – an observational longitudinal study,” Brain Dev., 2018 Apr., pii:S0387-7604 (18)30112-8 doi: 10.1016/j.braindev2018.03.013. (Epub ahead of print)
  2. Warren, et al, “The use of cannabidiol for seizure management in patients with brain tumor-related epilepsy,” Neurocase, 2017, Oct.-Dec., 23 (5-6):287-291.
  3. Neubauer, D., et al, “Cannabidiol for treatment of refractory childhood epilepsies: experience from a single tertiary epilepsy center in Slovenia,” Epilepsy Behav., 2018 Apr., 81:79-85. doi:10.1016/j.yebeh.2018.02.009. (Epub ahead of print)
  4. Rosenberg, et al, “Cannabinoids and epilepsy,” Neurotherapeutics, 2015, Oct., 12 (4):747-768.
  5. Devinsky, O., et al, “Trial of cannabidiol for drug-resistant seizures in the Dravet Syndrome,” New England Journal of Medicine, 2017, 376: 2011-2020.
  6. Devinsky, et al, “Effect of cannabidiol on drop seizures in the Lennox-Gastaut Syndrome,” NEJM, 2018, May,  378:1888-1897.
  7. Devinsky, et al, “Cannabidiol in patients with treatment-resistant epilepsy: an open label interventional trial,” Lancet Neurology, 2016, Mar., 15 (3):270-8.
  8. Fernandez-Ruiz, et al, “Prospects of cannabinoid therapies in basal ganglia disorder,” British Journal of Pharmacology, 2011, Aug., 163 (7):1365-1378.
  9. Do Val-da-Silva, et al, “Protective effects of cannabidiol against seizures and neuronal death in a rat model of mesial temporal lobe epilepsy,” Front. Pharmacol., 2017, 8:131.

Vagal nerve stimulation device: its role in medically refractory partial epilepsy and reports of weight loss


Virginia Thornley, Neurologist, Epileptologist


April 15, 2018


The vagal nerve stimulation device is an implanted device that exerts its effort by pulses of electrical activity that stimulates the vagal nerve or cranial nerve X. It had initially been found to work in animal studies in the 1990’s then later applied in clinical studies.

Mechanism of action 

For years, the mechanism was unknown and was used rather effectively in the clinical realm. The elucidated mechanisms were thought to be that the vagal nerve stimulator modifies the highly synchronized electrical activity that occurs in epilepsy through desynchronization via the vagal nerve. In addition, there is increased regional cerebral perfusion, and there is increased GABA neurotransmitters which are inhibitory towards electrical activity causing seizures and a decrease in glutamate which is known to increase excitation with the brain. There are GABA-A receptor increases, an increase in locus ceruleus produced noradrenergic substances which are released through the vagal nerve and an increase in serotonin transmissions through the raphe nucleus.

Role in controlling seizures

In the original open-label trial in 5 clinical trials, the vagal nerve stimulation device was found to be effective in reducing seizures by 50%. 454 patients had the implanted device and clinical information was obtained from 440. A cardiac stimulation device was implanted along with a coil in the ipsilateral vagal nerve. At 1 year of implantation, more than 50% of reduction of seizures occurred in 36.8% of patients at year 1, 43.2% year 2, and 42.7% at year 3.  The most common side effect at year 2 was hoarseness of about 9.8% and headache in 4.5% and at 3 years there was shortness of breath in 3% (4).

In one retrospective study from 1997 to 2008, 436 patients were found with implanted vagal nerve stimulation devices from ages 1-76, 220 were women and 216 were men. 33 had poor follow-up and 3 had removal due to infection. The mean frequency of seizures was better at 50% reduction.  There was 90% better control on 90 patients, >75% control in 162 patients and 50% control in 255 patients, <50% control in 145 patients. Permanent damage to the vagal nerve happened in 2.8% or 11 patients out of the 400 patients (after the removal of the ones lost to follow-up and infected) (5).

Long-term value of vagal nerve stimulating device, effectiveness after 5 years

There have been many studies reported that it may be effective short-term. But there was one pediatric study that reported success in seizure control in longer than 5 years. In a study of 56 pediatric patients ages 4-17, >9.8% were seizure free after 9 months, 24% after 2 years, 46.4% after 3 years and 54% after 5 years.11 out of the 56 patients became seizure free. After 5 years 62% of the patients had fewer seizures after 5 years.

What happens from diagnosis to implantation to use

A patient is identified as medically refractory, meaning a patient who has already failed 2 or more agents. Once control is failed after 2 anti-epileptic drugs after an adequate dosage and trial,  the likelihood of being seizure free becomes significantly less.  It is usually applied to patients with partial seizures, the most common being temporal lobe epilepsy. After appropriate identification is done, the patient undergoes a procedure where a cardiac device is implanted under the skin which generates an electrical impulse. A wire or coil is attached to the vagal nerve which reacts to this signal and emits an electrical pulse which inhibits the seizure which is electrical activity in the brain by disrupting this through various mechanisms. The device can be programmed to have a set frequency, amount of power and can be set to automatic with features where the patient can apply a magnet to inhibit the seizure when it is about to occur. The magnet is typically swiped over the cardiac device which was implanted over the left side of the chest. The settings can be changed in the doctor’s office adjusting according to the number and frequency of seizures.

Common side effects

Some of the most common side effects reported include hoarseness, cough, throat irritation, dyspnea, insomnia, dyspepsia, and vomiting. The symptoms are related to the location of the device near the nerve causing local irritation and likely due to the functions subserved by the vagal nerve.

Incidental weight loss effect

Vagal nerve stimulation device was applied to treatment-resistant patients with depression where an incidental effect on weight loss was found. One study in 33 patients showed that the vagal nerve stimulator implanted in patients seemed to alter cravings for sweet food which may play a part in weight loss (2). There have been some conflicting studies proving that there is no weight loss in vagal nerve stimulating device at the settings recommended in epilepsy in 21 patients (3). In a large study of 503 patients from 15 study centers, vagal nerve blockade was applied intrabdominally. 294 patients were randomized to treated (192) and to control groups (102). Therapy involved electrical stimulation through an external power source to the vagal nerves in the subdiaphragm which inhibits afferent and efferent vagal transmission. At 12 months, the excess weight loss in the treated group was 17% and in the control group, it was 16%. There was no statistic difference between the two groups, however, the post-study analysis demonstrated a possible result in weight loss related to the system check of the devices using low charges which may have caused weight loss in the control group (6).

In conclusion

There is strong evidence that the vagal nerve stimulation device is effective at reducing seizures of >50% of the medication-resistant epilepsy patient. It is effective even after 5 years of implantation. There are very little side effects which are mild to moderate. In addition, it can cause weight loss.


  1. Serdaroglu, et al, “Long-term effect of vagus nerve stimulation in pediatric intractable epilepsy: an extended follow-up,” Child’s Nervous System, 2016, 32 (4):641-646.
  2. Bodenlos, “Vagus nerve stimulation acutely alters food craving in adults with depression,” Appetite, 2007, 48: 145-153.
  3. Koren, et al, “Vagus nerve stimulation does not lead to significant changes in body weight in patients with epilepsy,” Epilepsy Behav. 2006;8:246–249.
  4. Morris, et al, “Long-term treatment with vagus nerve stimulation with refractory epilepsy,” Neurology, 1999, 53 (8):1731-1735.
  5. Elliot, et al, “Vagus nerve stimulation in 436 consecutive patients with treatment-resistant epilepsy: long-term outcomes and predictors of response,” Epilepsy Behavior, 2011, Jan., 20(1):57-83.
  6. Sarr, et al, “The EMPOWER study:randomized, prospective, double-blind, multicenter trial of vagal blockade to induce weight loss in morbid obesity,” Obes. Surg., 2012, Nov., 22 (11):1771-82.



Seizure alert dogs: can they really sense seizures of their owners?

Virginia Thornley, M.D., Neurologist, Epileptologist

March 28, 2018


Seizures are a result of recurrent electrical impulses in the brain causing repetitive symptoms pertaining to that area. At times, patients will not know when they occur.

Scientific studies

Seizure alert dogs are used to detect seizures that are undetectable to humans which may be either through olfactory senses or a change in the behavior. In one study, patients utilizing the seizure alert dog were studied. Seizure frequency was monitored for 48 weeks including a baseline of 12 weeks after entry into the study. With this mode, there has been a seizure reduction of about 43% compared to baseline. 9/10 patients had a 34% reduction in seizure frequency (1).

One study suggested that dogs have the innate sense of sensing their owners’ seizures. In 63 patients, 29 had pet dogs, 9 stated their dogs could sense their seizures (3).

In some studies of skeptical value, there is no proven benefit, although the presence of pseudoseizure may be a factor, meaning neurological symptoms that appear as seizures but are psychogenic in etiology may throw the seizure alert dogs off. Although some studies may indicate lack of benefit, mode of training may play an influence in the detection. The seizure alert dog likely takes cues from the heart rate or olfactory cues to detect seizures (2).


Downsides to seizure alert dog services

Recipients of service dog must meet certain criteria. This service is usually not covered by medical insurance and patients may avail of this service through assistance programs for a minimal fee.

The service dogs themselves may suffer from stress related to the work required for service. In addition, most dogs train between 6 months and 2 years after which service may be of value for about 7 years. The patient must also forge a bond with their service animal. Becuase it is often not covered by insurance and it may be cost prohibitive, some patients have started training their own dogs for seizure detection. The different levels of training may not be standardized or adequate.





  1. Strong, et al, “Effect of trained seizure alert dogs on the frequency of tonic-clonic seizures,” Seizure, 2002, Sep., 11(6):402-405.
  2. Brown, et al, “Can seizure-alert dogs predict seizures?” Epilepsy Res., 2011, Dec., 97(3):236-242.
  3. Dalziel, et al, “Seizure-alert dogs: a review and preliminary study,” Seizure, 2003, Mar., 12 (2):115-120.
  4. Strong, et al, “Seizure alert dog-fact or fiction?”Seizures, 1999, Feb., 8 (1):62-65.

Epilepsy surgery in temporal lobe epilepsy due to mesial temporal sclerosis: the timeline in investigative work-up from the neurologist’s office to the O.R.

Virginia Thornley, M.D. Neurologist, Epileptologist

March 27, 2018


Temporal lobe epilepsy is one of the most common types of seizures. The most common cause and one of the most successfully treated causes of temporal lobe epilepsy treated through surgery is mesial temporal sclerosis. This article focuses on mesial temporal sclerosis and does not include discussions of other types of temporal lobe epilepsy due to other causes such as tumors, cystic lesions or head injury or non-lesional temporal lobe epilepsy.  In order to identify a patient, the symptoms are generally stereotypical which suggest localizing towards one focus.  An early age of identification may portend a better outcome since frequent temporal lobe seizures may cause the development of circuitry to the opposite side causing another focus to develop on the opposite temporal lobe. In addition, it is important to control temporal lobe epilepsy because of the location of the seizures are in the hippocampus which is important in memory. Many patients complain of poor memory which will continue to progress should seizures remain poorly controlled. Epilepsy surgery is the definitive treatment for temporal lobe epilepsy in mesial temporal sclerosis.


To identify an appropriate candidate for surgery, the patient should have stereotypical seizures which localize towards one focus. While the focus may cause contralateral clinical symptoms, automatisms of the limb are generally ipsilateral to the focus.  Once a patient has been identified, further diagnostics tests are needed in order to confirm this focus including a routine electroencephalogram and an ambulatory 48-72 hour EEG which can be performed out-patient. The only downfall with an ambulatory EEG is that it is subject to the artifact, since the electrodes may be displaced causing poor adherence of the electrode to the scalp causing resistance manifested as artifact and a poor recording. However, it is still a good screening test to determine whether there may be a single focus versus multiple regions affected. Temporal lobe epilepsy may be seen with high voltage epileptiform spike and wave. It may be accompanied by focal delta slowing within the temporal lobe, suggesting temporal lobe dysfunction due to recurring seizures. If a patient is deemed an appropriate candidate, a referral may be made to an epilepsy center where more in-depth investigations are performed.



Admission to an epilepsy center

Expect to stay at least 1 week or more in order to allow the capture of typical seizures and to obtain an adequate sampling of ictal periods and pinter-ictal periods during wakefulness and sleep. A team of specialists is involved with the work-up including a clinical epileptologist who manages the medications and clinical aspect, a clinical neurophysiologist who interprets the video EEG monitoring and correlates it with the clinical symptoms, a neuropsychologist who performs the WADA testing and a slew of clinical EEG technicians who ensure that the electrodes are properly attached throughout the hospital stay. In-depth conferences are held to review the studies of the patients and evaluate which patients are suitable epilepsy candidates. Sometimes, multiple admissions are necessary before seizures can be captured.


During admission, seizures are captured and correlated with the electroencephalographic recordings to determine the focus. More than one focus correlates with a poor outcome, a single focus is necessary. The clinician may provoke seizures by tapering medications safely in the hospital setting. Other techniques include sleep deprivation and encouraging any triggers. The full spectrum of clinical seizures must be captured in order to ensure adequate localization. Bitemporal foci portend a poor outcome.


A high-quality MRI of the brain using epilepsy protocol with thin cuts through the temporal lobes of 1.5mm to 2mm is essential. Coronal views are the best way to visualize the hippocampi to evaluate for hippocampal sclerosis which characterizes temporal lobe epilepsy. Usually, the hippocampus affected is much smaller than the contralateral one with hyperintensity on T2. As a result of excessive seizures, burning off of the cells in the hippocampus occurs so that is it is now atrophic. Although an MRI of the brain may have already been obtained pre-work-up, a higher resolution and exceptional quality brain MRI is likely to be repeated. This will serve as the visual point on which the neurosurgeon operates. Seeing a sclerotic hippocampus gives a high correlation with mesial temporal sclerosis.



Spectroscopy is obtained in-house, where hexamethylpropylenamine oxime (HMPAO) injection is done 30 minutes before an ictus. When the patient has a seizure, the HMPAO perfuses to the area of interest showing where the seizure localizes. Images are obtained. This test has an added value of further localizing the focus. The drawbacks, however, include not being able to predict when a seizure is about to occur and missing the ictus. It is not unusual for this test to be repeated for it to be meaningful. In addition, it can only be done during office hours so that nocturnal seizure will be missed due to lack of adequate staff.


This is a costly examination which may not be available in some epilepsy centers. It uses a 3-dimensional modality for localizing the focus. The MEG dipoles are superimposed on the MRI images.

WADA testing

A neuropsychologist examines the patient’s memory and language by temporarily putting the opposite side of the brain to sleep through injection of amobarbital into the internal carotid artery. Short-term memory and language are examined. The neuropsychologist must determine that there is adequate memory on the contralateral temporal lobe for temporal lobe surgery to be successful. If both temporal lobes are impaired in terms of memory, the patient will suffer from poor memory following the surgery. Other tests are done by the neuropsychologist to check for cognition, any personality disorders and assess for evidence of mood disorders.



This is one of the final steps in the investigation where the cranium over the temporal lobe of interest is removed and electrodes are placed directly on top of the cerebrum. Depth electrodes are placed in order to capture epileptiform discharges buried deep inside the hippocampus which cannot be adequately detected by electrodes laying on top of the temporal lobe. The seizures are recorded and a more accurate mapping of the seizure focus is obtained.


Once all the appropriate investigations are obtained, if all the data points towards a single focus then the patient is deemed an appropriate candidate. Epilepsy conferences are usually held and reviewed by all the specialists involved in the care. Some patients may proceed directly into surgery after mapping. Others may need to go home and return back for another admission to undergo epilepsy surgery. A patient who is still questionable may need to return for more in-depth recording, this may occur in non-lesional epilepsy where the information is not strong enough to justify surgery. The goal of epilepsy surgery is to resect the dysfunctional epileptogenic zone while preserving the functioning surrounding cortex.

After care

Once the surgery is performed, the patient will need to be on anti-epileptic agents for at least 2 years of seizure freedom. In appropriately investigated patients, a favorable outcome of seizure freedom may reach as high as 60%.





The deleterious effect of caffeine on epilepsy and anti-epileptic agents

Virginia Thornley, M.D., Neurologist, Epileptologist
March 25, 2019
Caffeine (1,3,7-methylxantine) is one of the most commonly ingested stimulants in the world. It is not uncommon for someone to ingest a daily consumption of 200mg of caffeine a day. It is ubiquitously found in soda, coffee, tea, and chocolate. It is the bane of every neurologist who treats migraine and patients with insomnia. It acts as a stimulant and many people use it to counter fatigue induced by lack of sleep. Students consume it to stay up at night for late night studying in order to ace their tests the next day. Millions of people ingest caffeine on a regular basis to get through the full work day.
Caffeine worsen seizures
It has been found in animal models to lower the seizure threshold. At low doses, it reduces the efficacy of anti-epileptic agents. At more than 400mg of caffeine per day, in rodent models it is found to induce seizures. In experimental data, use of caffeine is found to lower the seizure threshold. In mouse models, at lower doses below the seizure-inducing effects, it is found to counter the protective beneficial effects of anti-epileptic agents such as carbamazepine, phenytoin, valproate, and phenobarbital as well as newer agents such as topiramate.  There seems to be no effect of caffeine on newer agents such as tiagabine, oxcarbazepine or lamotrigine. There is clinical data confirming that ingesting high doses of caffeine correlates with greater number of seizures.
Dark cocoa and seizures
Dark chocolate is also found to be a proconvulsant, but little is known about the mechanism of action. Dark chocolate is rich in caffeine. In one mouse study, the effect of high intake of dark chocolate on the susceptibility of hippocampal cells to seizures was examined. Dark cocoa appeared not to affect mood behavior but improved motor coordination.  However, electrophysiologic studies showed enhancement of bursts of epileptogenic potential within the dentate gyrus of the hippocampus. There was a reduction in GABA-alpha receptors suggesting that consumption of dark chocolate may alter the synaptic aspect of epileptogenesis in the temporal lobe.
These findings suggest that high consumption of caffeine especially dark cocoa can increase seizure frequency in animal models and in clinical studies. It seems to act as a proconvulsant and reduces receptors that are necessary for inhibiting seizures.
  1. Chroscinska-Krawzyk, et al, “Caffeine and anticonvulsant potency of anti-epileptic drugs: experimental and clinical data,” Pharmacol. Rep., 2011, 63(1):12-18.
  2. Cicvaric, et al, “Sustained consumption of cocoa-based dark chocolate enhances seizure-like events in the mouse hippocampus,” Food Funct., 2018, Mar., 1, 9(3):1532-1544.