- Email: [email protected]
Optimizing therapies for neurobehavioral comorbidities of epilepsy using chronic ambulatory electrocorticography
Epilepsy & Behavior 102 (2020) 106814
Contents lists available at ScienceDirect
Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh
Brief Communication
Optimizing therapies for neurobehavioral comorbidities of epilepsy using chronic ambulatory electrocorticography Alexandra T. Issa Roach a, Ganne Chaitanya a, Kristen O. Riley b, Wolfgang Muhlhofer a, Sandipan Pati a,⁎ a b
Department of Neurology, University of Alabama at Birmingham, AL, United States of America Department of Neurosurgery, University of Alabama at Birmingham, AL, United States of America
a r t i c l e
i n f o
Article history: Received 29 August 2019 Revised 18 November 2019 Accepted 19 November 2019 Available online xxxx Keywords: Depression Anxiety Focal epilepsy Psychogenic nonepileptic seizures Neuromodulation
a b s t r a c t There is an unmet need to improve therapy for neuropsychiatric comorbidities that are highly prevalent in persons with epilepsy (PWE). However, diagnosing and monitoring the neurobehavioral symptoms is challenging as their presentation can overlap with seizures. In this retrospective study, we report the advantage of chronic ambulatory electrocorticography (ECoG) from implanted Responsive Neurostimulator System (RNS®) in characterizing these psychosomatic paroxysms as a possible ictal, postictal, or interictal phenomenon and how the diagnosis guided the therapy choices. Five out of 21 patients with RNS had neuropsychiatric symptoms (panic attack, psychosis, conversion, and somatization disorders) that overlapped with their seizure semiology and were found to benefit from the use of RNS ECoG data by timely diagnosing and titrating targeted therapies. The cases illustrate the use of RNS ECoG data in diagnosing and improving the management of comorbidities in PWE. The ability to access RNS ECoG data and correlate it with patient symptoms is unique among available therapeutic options for PWE. © 2019 Elsevier Inc. All rights reserved.
1. Introduction There is an unmet need to improve therapy for various neuropsychiatric comorbidities and cognitive dysfunctions that have a high coincidence in persons with epilepsy (PWE) [1,2]. Anxiety, depression, psychosis, and impaired memory are some of the highly prevalent comorbidities in PWE [1,3,4]. Beyond seizure control, effective diagnosis, treatment, and monitoring of these comorbidities are critical as they impact the quality of life [5]. However, diagnosing these comorbidities remains challenging because a) seizures can masquerade as paroxysms of increased anxiety or panic attacks [6]; b) psychosis can be temporally related to increased epileptiform activities including seizure and postictal state [7]; and c) patient may fail to report symptoms due to seizure-related memory loss or social stigma. Additionally, PWE can report newer spells that are psychogenic nonepileptic seizures (PNES). The coexistence of PNES in PWE can pose a diagnostic challenge and can confound the treatment outcome [8]. Characterizing these events utilizing video-electroencephalography (vEEG) as the gold standard helps to discriminate these psychosomatic events from seizures. However, vEEG can be only performed for a limited amount of time (typically 3– 5 days, max 1–2 weeks), and is often restricted to the inpatient setting, ⁎ Corresponding author at: Epilepsy and Cognitive Neurophysiology Laboratory, Department of Neurology, University of Alabama at Birmingham, CIRC 312, 1719 6th Avenue South, Birmingham, AL 35294, United States of America. E-mail address: [email protected] (S. Pati).
https://doi.org/10.1016/j.yebeh.2019.106814 1525-5050/© 2019 Elsevier Inc. All rights reserved.
which limits access to this diagnostic tool. Chronic ambulatory electrocorticography (ECoG) from implanted Responsive Neurostimulator System (RNS®) can overcome these limitations by providing EEG that can be used to characterize and classify these psychosomatic paroxysms as a possible ictal, postictal, or interictal phenomenon and help with treatment decisions accordingly [9]. The RNS includes a cranially seated programmable neurostimulator that is connected to two leads, which are surgically placed over the suspected seizure onset zone [9]. From the two leads (4 channels), up to 12 min of electrocorticographic activity can be stored in the neurostimulator at any one time. Patients are provided with a customized telemetry unit (a wand that connects to a password-protected laptop) that can be used to download and archive the ECoG for any event of interest. Further, the patients can guide the storage of ECoG data by swiping a magnet of the neurostimulator itself whenever they have an event of concern. Here, we demonstrate the utility of the RNS in the management of psychiatric and neurological comorbidities in patients with drug-resistant focal epilepsy. 2. Methods A single-center, retrospective study was conducted at a level IV epilepsy center. All patients with implanted RNS at the University of Alabama at Birmingham (UAB) epilepsy neuromodulation clinic were included in the study. Observations from patient visits and management, as documented in the electronic medical record, were reviewed. Since no
2
A.T. Issa Roach et al. / Epilepsy & Behavior 102 (2020) 106814
additional appointments or treatment adjustments were based on this retrospective chart review, no formal consent was obtained as part of the study. The UAB Institutional Review Board approved the study. 3. Results Among 21 patients with implanted RNS, there were five patients (23%) with significant neurobehavioral comorbidities whose presentation overlapped with their seizures and, hence, benefitted from the use of the ECoG data stored by the RNS. This group included 3 females and 2 males with 27–46 years of age. All patients were diagnosed with intractable multifocal epilepsy, and the epileptogenic cortex was not amenable to resection. 3.1. Case #1: amygdala-hippocampal seizures mimicking panic attacks This is a 43-year-old right-handed woman with previous right anterior temporal lobectomy (ATL) who was initially referred for evaluation of possible seizures mimicking panic attacks and worsening of comorbid generalized anxiety disorder and depression that was refractory to medications. She underwent stereoelectroencephalography (sEEG) evaluation that confirmed that some of her panic-like attacks were indeed seizures (focal aware seizures (FAS)) emanating from the left
amygdala and hippocampus. She had RNS with a depth lead implanted in the left amygdala–hippocampus and a subtemporal strip lead. She was asked to swipe the magnet whenever she had a panic attack, and ECoG data were used to distinguish seizures that mimic panic attacks from independent panic attacks (Fig. 1). The use of the ECoG data stored by the RNS allowed her medical team to distinguish the incidence of an ictal panic attack from a panic attack related to her underlying generalized anxiety disorder. Based on the ECoG findings, the psychiatrist optimized the dosing of her antidepressant (sertraline) and as a needed benzodiazepine (lorazepam). Her cognitive-behavioral therapy (CBT) sessions were adjusted accordingly. She now endorses significantly fewer panic attacks, and this may be secondary to improved seizure control and management of anxiety disorder. 3.2. Case #2: acute postictal psychosis in a person with borderline personality disorder (BPD) This patient is a 33-year-old right-handed man with a history of drug-resistant focal epilepsy with sEEG confirming bilateral mesial temporal lobe epilepsy (TLE) that was treated with RNS. He was diagnosed with comorbid borderline personality disorder (BPD) that was associated with psychosis and was treated with olanzapine. He would frequently report periods of worsening of psychiatric symptoms of
Fig. 1. Sample electrocorticography and detected events in subject #1 (ictal panic attacks). The Patient Data Management System (PDMS) from the RNS contains data uploaded from the RNS programmer laptop. A — Number of long episodes (defined as events lasting N30 s; green bar plot) from August 2018–Jan. 2019. In this patient, long episodes were characteristic of electrographic seizures as seen in the 4-channel electrocorticography in C. Magnet swipe by the patient is indicated by “M”. Electrocorticography helped to discriminate magnet swipe events (in B) as panic attacks (D) and panic attacks due to seizure (E). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
A.T. Issa Roach et al. / Epilepsy & Behavior 102 (2020) 106814
paranoia, hyperreligious delusions, and emotional lability, which was primarily attributed to his underlying psychiatric conditions. However, there was also a concern that the psychosis could be part of a seizure or postictal phenomenon resultant from seizures during sleep. The patient was encouraged to swipe the RNS magnet during these episodes. A review of ECoG demonstrated that some of these events preceded long epileptiform activity (also called long episodes) (Fig. 2). These data obtained from RNS were used to optimize therapy. Instead of increasing his olanzapine, as needed lorazepam was used to treat the postictal psychosis, and ongoing adjustments of his antiseizure medications, RNS detection, and stimulation settings resulted in self-reported improvement in underlying psychiatric symptoms. 3.3. Cases #3–4: new-onset PNES in chronic epilepsy The third patient is a 35-year-old, right-handed woman with drug-resistant bilateral mesial TLE that was confirmed with sEEG evaluation. The RNS was placed with leads in bilateral hippocampi, and over a year, the patient had a decrease in her seizure frequency. Despite such improvement, the patient subsequently reported a new semiology of her seizure that was not detected by the RNS. She was admitted to the epilepsy monitoring unit (EMU) where her new spell type was confirmed as PNES. Her PNES was managed with CBT, and the event-related ECoG data stored by the RNS were used to monitor the progress of her therapy. She continued
3
to swipe her RNS magnet for all events, and we were able to ascertain which ones of them were PNES versus epileptic seizures. The fourth patient is a 46-year-old right-handed woman who, after undergoing sEEG evaluation, was found to have bilateral mesial TLE. The RNS was placed with leads in the bilateral hippocampi. The ECoG successfully detected and treated right and left hippocampal seizures. Several months after the initiation of therapy, the patient reported a new spell type characterized by nausea followed by left-sided body jerking and loss of consciousness. The ECoG did not detect any of these events. The patient was admitted to EMU, where a working diagnosis of PNES was confirmed although a brief seizure from insular cortex was in the differential diagnosis. The patient is undergoing CBT, and the ECoG data provided by the RNS are used to monitor the progress of her therapy. Within three months after starting CBT, the magnet swipe decreased from 59/month to 26/month (45% decrease), while detected seizures (long episodes confirmed with visual inspection) during the same period were 2 and 3/month. In both these cases, video-EEG was used to confirm the diagnosis of PNES, while the RNS ECoG was used to quantify the seizures in the ambulatory setting. 3.4. Case #5: somatic symptom disorder (SSD) mimicking seizures A 27-year-old right-handed man with a large right parietotemporal focal cortical dysplasia (FCD) had sEEG evaluation that confirmed seizures
Fig. 2. Sample electrocorticography and detected events in subject #2 (postictal psychosis). A — Number of long episodes (defined as events lasting N30 s; green bar plot) between Dec. 5 and 11, 2018. B — Example of a long episode representing electrographic seizure recorded at 2:45 AM in the 4-channel electrocorticography. C — Electrocorticography showing seizure termination and transition to the postictal state. Patient's caregiver swiped the magnet multiple times (see D also) as the patient was reporting worsening of psychosis that lasted for days. D — Electrocorticography showing persistent epileptiform activities. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
4
A.T. Issa Roach et al. / Epilepsy & Behavior 102 (2020) 106814
with the RNS can be challenging, given the limitations of the device stated earlier in the discussion. Some programming approaches that can help mitigate the chances of “false negative” ECoG (i.e., ECoG that did not show seizure even though a seizure actually occurred around that time) are as follows: a) encouraging frequent data downloads immediately after clinical events to minimize overwriting; b) encourage patient or caregiver to magnet swipe within a minute after the clinical event; c) increasing ECoG length; and d) shortening Long Episode duration. The long-term outcome of RNS neurostimulation therapy showed stable seizure reduction in patients with neocortical and mesiotemporal epilepsies [18]. Additionally, data on quality of life and cognition are favorable in patients with RNS [19,20]. The availability of ambulatory ECoG provides the opportunity to manage comorbidities in epilepsy that can mimic seizures and contributes to the overall poor quality of life.
localized to the right mesial temporal and parietal regions. He underwent right ATL and limited resection of the right mesial parietal region followed by RNS implanted with strip leads placed over the lateral parietal regions. He also had a significant comorbid anxiety disorder. In the first few months postimplantation, he reported a substantial increase in paroxysmal dizziness, impending doom, and fear of falling that limited his mobility; he ended up bedbound. The patient was convinced that these prolonged dizzy spells (lasting minutes to up to an hour), described as a feeling of being on a rocking boat, were seizures. He underwent ambulatory EEG and, subsequently, an EMU admission that identified these events as PNES. The ECoG and the scalp EEG recorded simultaneously did not show any electrographic changes during these spells. Multiple specialists, including ear, nose and throat (ENT) and neurohospitalist, ruled out any organic pathology. Antiseizure drugs were adjusted without much success. Finally, psychiatry was consulted for the treatment of his generalized anxiety disorder. The patient was counseled appropriately and encouraged to swipe his RNS magnet with every dizzy spell. The epileptologist reviewed the ECoG and provided consistent and reassuring feedback that the events were nonepileptic. Over two months, the patient improved significantly (magnet swipe decreased from 12/month to one/ month), and his dizziness eventually resolved completely.
SP has served as a paid consultant for NeuroPace, Inc. but declares no targeted funding or compensation for this study. None of the authors share any competing interests.
4. Discussion and conclusion
Acknowledgment
The five cases described above are great examples of the additional use of readily available ECoG data provided by the RNS in the management of comorbidities that are highly prevalent in PWE. The ability to access ECoG data and correlate them with patient magnet swipes during any paroxysmal event is unique compared with other therapeutic options in the field [9]. However, the RNS has its limitations. (1) The spatial sampling is restricted to only two four-contact cortical strip and/or depth leads. Hence, the presence of any electrical activity outside the sensors will not be detected. (2) Although the neurostimulator continuously senses and monitors electrographic activity through the leads, the memory capacity is limited to storage of 12 min of ECoG daily with the newer RNS. (3) The magnet swipe triggers the storage of ECoG for a limited period prior to the swipe (typically 60 s). Therefore, a late magnet swipe (i.e., more than 60 s after a clinical event) will miss the activity that is temporally correlated with symptoms, resulting in a ‘false negative’ ECoG. (4) Electrographic seizures stored for the ‘Long Episode’ trigger must be detected by the neurostimulator and persist above the detection threshold for a minimum duration of time (typically 30 s). Despite these limitations, the RNS has been used to lateralize and localize seizure foci, monitor medication effects, and characterize temporal patterns of seizures and the running-down phenomenon [10–14]. Seizures can present as panic attacks, worsening anxiety, dizziness, acute psychosis, agitation, and hallucinations [15]. All these symptoms are also prevalent in chronic epilepsy and can be attributed to comorbid depression, anxiety, medication side effects, and other neuropsychiatric disorders [2,4,6]. The availability of EEG correlates of these paroxysms is critical in differentiating epileptic from nonepileptic events. Epilepsy-related neurobehavioral symptoms have been first reported in the 1800s by Gowers and Jackson [16,17]. These symptoms are often classified in their timely relationship to the actual seizure (i.e., preictal, ictal, or postictal). Affective symptoms like fear, anxiety, and panic attacks can represent ictal phenomena commonly associated with limbic seizures [3,4,6]. Psychosis can be an ictal, postictal, or an interictal phenomenon. Interictal psychosis is frequently associated with other mood disorders (so-called as interictal dysphoric disorder) [7]. Postictal psychosis is managed with as needed benzodiazepines, while interictal dysphoric disorder is treated effectively with antidepressants [3,7]. The availability of ECoG can help to categorize these neurobehavioral symptoms, guide their management appropriately, and monitor treatment response. Similarly, the ECoG also offers the advantage to diagnose PNES early in PWE and avoid further mistreatments by not adding antiseizure medications or prevent multiple ER visits. However, characterizing paroxysmal events
SP and WM would like to thank NeuroPace representatives Ms. Kaitlyn Wilmer-Fierro and Mr. Mark Griffin and UAB study coordinator Ms. Kathleen Hernando.
Declaration of competing interest
References [1] Lin JJ, Mula M, Hermann BP. Uncovering the neurobehavioural comorbidities of epilepsy over the lifespan. Lancet 2012;380:1180–92. [2] Strine TW, Kobau R, Chapman DP, Thurman DJ, Price P, Balluz LS. Psychological distress, comorbidities, and health behaviors among U.S. adults with seizures: results from the 2002 National Health Interview Survey. Epilepsia 2005;46:1133–9. [3] Mula M, Monaco F. Ictal and peri-ictal psychopathology. Behav Neurol 2011;24:21–5. [4] Kanner AM. Obstacles in the treatment of common psychiatric comorbidities in patients with epilepsy: what is wrong with this picture? Epilepsy Behav 2019;98:291–2. [5] Devinsky O. Therapy for neurobehavioral disorders in epilepsy. Epilepsia 2004;45 (Suppl. 2):34–40. [6] Mula M. Epilepsy-induced behavioral changes during the ictal phase. Epilepsy Behav 2014;30:14–6. [7] Maguire M, Singh J, Marson A. Epilepsy and psychosis: a practical approach. Pract Neurol 2018;18:106–14. [8] Benbadis SR, Agrawal V, WOt Tatum. How many patients with psychogenic nonepileptic seizures also have epilepsy? Neurology 2001;57:915–7. [9] Skarpaas TL, Jarosiewicz B, Morrell MJ. Brain-responsive neurostimulation for epilepsy (RNS((R)) system). Epilepsy Res 2019;153:68–70. [10] King-Stephens D, Mirro E, Weber PB, Laxer KD, Van Ness PC, Salanova V, et al. Lateralization of mesial temporal lobe epilepsy with chronic ambulatory electrocorticography. Epilepsia 2015;56:959–67. [11] Chan AY, Knowlton RC, Chang EF, Rao VR. Seizure localization by chronic ambulatory electrocorticography. Clin Neurophysiol Pract 2018;3:174–6. [12] Skarpaas TL, Tcheng TK, Morrell MJ. Clinical and electrocorticographic response to antiepileptic drugs in patients treated with responsive stimulation. Epilepsy Behav 2018;83:192–200. [13] Baud MO, Kleen JK, Mirro EA, Andrechak JC, King-Stephens D, Chang EF, et al. Multiday rhythms modulate seizure risk in epilepsy. Nat Commun 2018;9:88. [14] Geller AS, Friedman D, Fang M, Doyle WK, Devinsky O, Dugan P. Running-down phenomenon captured with chronic electrocorticography. Epilepsia Open 2018;3: 528–34. [15] Berg AT, Altalib HH, Devinsky O. Psychiatric and behavioral comorbidities in epilepsy: a critical reappraisal. Epilepsia 2017;58:1123–30. [16] Gowers WR. Epilepsy and other chronic convulsive diseases: their causes, symptoms, and treatment. Old Hickory Bookshop; 1901. [17] Hughlings-Jackson J. On a particular variety of epilepsy (“intellectual aura”), one case with symptoms of organic brain disease. Brain 1888;11:179–207. [18] Bergey GK, Morrell MJ, Mizrahi EM, Goldman A, King-Stephens D, Nair D, et al. Longterm treatment with responsive brain stimulation in adults with refractory partial seizures. Neurology 2015;84:810–7. [19] Heck CN, King-Stephens D, Massey AD, Nair DR, Jobst BC, Barkley GL, et al. Two-year seizure reduction in adults with medically intractable partial onset epilepsy treated with responsive neurostimulation: final results of the RNS System pivotal trial. Epilepsia 2014;55:432–41. [20] Loring DW, Kapur R, Meador KJ, Morrell MJ. Differential neuropsychological outcomes following targeted responsive neurostimulation for partial-onset epilepsy. Epilepsia 2015;56:1836–44.
Contents lists available at ScienceDirect
Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh
Brief Communication
Optimizing therapies for neurobehavioral comorbidities of epilepsy using chronic ambulatory electrocorticography Alexandra T. Issa Roach a, Ganne Chaitanya a, Kristen O. Riley b, Wolfgang Muhlhofer a, Sandipan Pati a,⁎ a b
Department of Neurology, University of Alabama at Birmingham, AL, United States of America Department of Neurosurgery, University of Alabama at Birmingham, AL, United States of America
a r t i c l e
i n f o
Article history: Received 29 August 2019 Revised 18 November 2019 Accepted 19 November 2019 Available online xxxx Keywords: Depression Anxiety Focal epilepsy Psychogenic nonepileptic seizures Neuromodulation
a b s t r a c t There is an unmet need to improve therapy for neuropsychiatric comorbidities that are highly prevalent in persons with epilepsy (PWE). However, diagnosing and monitoring the neurobehavioral symptoms is challenging as their presentation can overlap with seizures. In this retrospective study, we report the advantage of chronic ambulatory electrocorticography (ECoG) from implanted Responsive Neurostimulator System (RNS®) in characterizing these psychosomatic paroxysms as a possible ictal, postictal, or interictal phenomenon and how the diagnosis guided the therapy choices. Five out of 21 patients with RNS had neuropsychiatric symptoms (panic attack, psychosis, conversion, and somatization disorders) that overlapped with their seizure semiology and were found to benefit from the use of RNS ECoG data by timely diagnosing and titrating targeted therapies. The cases illustrate the use of RNS ECoG data in diagnosing and improving the management of comorbidities in PWE. The ability to access RNS ECoG data and correlate it with patient symptoms is unique among available therapeutic options for PWE. © 2019 Elsevier Inc. All rights reserved.
1. Introduction There is an unmet need to improve therapy for various neuropsychiatric comorbidities and cognitive dysfunctions that have a high coincidence in persons with epilepsy (PWE) [1,2]. Anxiety, depression, psychosis, and impaired memory are some of the highly prevalent comorbidities in PWE [1,3,4]. Beyond seizure control, effective diagnosis, treatment, and monitoring of these comorbidities are critical as they impact the quality of life [5]. However, diagnosing these comorbidities remains challenging because a) seizures can masquerade as paroxysms of increased anxiety or panic attacks [6]; b) psychosis can be temporally related to increased epileptiform activities including seizure and postictal state [7]; and c) patient may fail to report symptoms due to seizure-related memory loss or social stigma. Additionally, PWE can report newer spells that are psychogenic nonepileptic seizures (PNES). The coexistence of PNES in PWE can pose a diagnostic challenge and can confound the treatment outcome [8]. Characterizing these events utilizing video-electroencephalography (vEEG) as the gold standard helps to discriminate these psychosomatic events from seizures. However, vEEG can be only performed for a limited amount of time (typically 3– 5 days, max 1–2 weeks), and is often restricted to the inpatient setting, ⁎ Corresponding author at: Epilepsy and Cognitive Neurophysiology Laboratory, Department of Neurology, University of Alabama at Birmingham, CIRC 312, 1719 6th Avenue South, Birmingham, AL 35294, United States of America. E-mail address: [email protected] (S. Pati).
https://doi.org/10.1016/j.yebeh.2019.106814 1525-5050/© 2019 Elsevier Inc. All rights reserved.
which limits access to this diagnostic tool. Chronic ambulatory electrocorticography (ECoG) from implanted Responsive Neurostimulator System (RNS®) can overcome these limitations by providing EEG that can be used to characterize and classify these psychosomatic paroxysms as a possible ictal, postictal, or interictal phenomenon and help with treatment decisions accordingly [9]. The RNS includes a cranially seated programmable neurostimulator that is connected to two leads, which are surgically placed over the suspected seizure onset zone [9]. From the two leads (4 channels), up to 12 min of electrocorticographic activity can be stored in the neurostimulator at any one time. Patients are provided with a customized telemetry unit (a wand that connects to a password-protected laptop) that can be used to download and archive the ECoG for any event of interest. Further, the patients can guide the storage of ECoG data by swiping a magnet of the neurostimulator itself whenever they have an event of concern. Here, we demonstrate the utility of the RNS in the management of psychiatric and neurological comorbidities in patients with drug-resistant focal epilepsy. 2. Methods A single-center, retrospective study was conducted at a level IV epilepsy center. All patients with implanted RNS at the University of Alabama at Birmingham (UAB) epilepsy neuromodulation clinic were included in the study. Observations from patient visits and management, as documented in the electronic medical record, were reviewed. Since no
2
A.T. Issa Roach et al. / Epilepsy & Behavior 102 (2020) 106814
additional appointments or treatment adjustments were based on this retrospective chart review, no formal consent was obtained as part of the study. The UAB Institutional Review Board approved the study. 3. Results Among 21 patients with implanted RNS, there were five patients (23%) with significant neurobehavioral comorbidities whose presentation overlapped with their seizures and, hence, benefitted from the use of the ECoG data stored by the RNS. This group included 3 females and 2 males with 27–46 years of age. All patients were diagnosed with intractable multifocal epilepsy, and the epileptogenic cortex was not amenable to resection. 3.1. Case #1: amygdala-hippocampal seizures mimicking panic attacks This is a 43-year-old right-handed woman with previous right anterior temporal lobectomy (ATL) who was initially referred for evaluation of possible seizures mimicking panic attacks and worsening of comorbid generalized anxiety disorder and depression that was refractory to medications. She underwent stereoelectroencephalography (sEEG) evaluation that confirmed that some of her panic-like attacks were indeed seizures (focal aware seizures (FAS)) emanating from the left
amygdala and hippocampus. She had RNS with a depth lead implanted in the left amygdala–hippocampus and a subtemporal strip lead. She was asked to swipe the magnet whenever she had a panic attack, and ECoG data were used to distinguish seizures that mimic panic attacks from independent panic attacks (Fig. 1). The use of the ECoG data stored by the RNS allowed her medical team to distinguish the incidence of an ictal panic attack from a panic attack related to her underlying generalized anxiety disorder. Based on the ECoG findings, the psychiatrist optimized the dosing of her antidepressant (sertraline) and as a needed benzodiazepine (lorazepam). Her cognitive-behavioral therapy (CBT) sessions were adjusted accordingly. She now endorses significantly fewer panic attacks, and this may be secondary to improved seizure control and management of anxiety disorder. 3.2. Case #2: acute postictal psychosis in a person with borderline personality disorder (BPD) This patient is a 33-year-old right-handed man with a history of drug-resistant focal epilepsy with sEEG confirming bilateral mesial temporal lobe epilepsy (TLE) that was treated with RNS. He was diagnosed with comorbid borderline personality disorder (BPD) that was associated with psychosis and was treated with olanzapine. He would frequently report periods of worsening of psychiatric symptoms of
Fig. 1. Sample electrocorticography and detected events in subject #1 (ictal panic attacks). The Patient Data Management System (PDMS) from the RNS contains data uploaded from the RNS programmer laptop. A — Number of long episodes (defined as events lasting N30 s; green bar plot) from August 2018–Jan. 2019. In this patient, long episodes were characteristic of electrographic seizures as seen in the 4-channel electrocorticography in C. Magnet swipe by the patient is indicated by “M”. Electrocorticography helped to discriminate magnet swipe events (in B) as panic attacks (D) and panic attacks due to seizure (E). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
A.T. Issa Roach et al. / Epilepsy & Behavior 102 (2020) 106814
paranoia, hyperreligious delusions, and emotional lability, which was primarily attributed to his underlying psychiatric conditions. However, there was also a concern that the psychosis could be part of a seizure or postictal phenomenon resultant from seizures during sleep. The patient was encouraged to swipe the RNS magnet during these episodes. A review of ECoG demonstrated that some of these events preceded long epileptiform activity (also called long episodes) (Fig. 2). These data obtained from RNS were used to optimize therapy. Instead of increasing his olanzapine, as needed lorazepam was used to treat the postictal psychosis, and ongoing adjustments of his antiseizure medications, RNS detection, and stimulation settings resulted in self-reported improvement in underlying psychiatric symptoms. 3.3. Cases #3–4: new-onset PNES in chronic epilepsy The third patient is a 35-year-old, right-handed woman with drug-resistant bilateral mesial TLE that was confirmed with sEEG evaluation. The RNS was placed with leads in bilateral hippocampi, and over a year, the patient had a decrease in her seizure frequency. Despite such improvement, the patient subsequently reported a new semiology of her seizure that was not detected by the RNS. She was admitted to the epilepsy monitoring unit (EMU) where her new spell type was confirmed as PNES. Her PNES was managed with CBT, and the event-related ECoG data stored by the RNS were used to monitor the progress of her therapy. She continued
3
to swipe her RNS magnet for all events, and we were able to ascertain which ones of them were PNES versus epileptic seizures. The fourth patient is a 46-year-old right-handed woman who, after undergoing sEEG evaluation, was found to have bilateral mesial TLE. The RNS was placed with leads in the bilateral hippocampi. The ECoG successfully detected and treated right and left hippocampal seizures. Several months after the initiation of therapy, the patient reported a new spell type characterized by nausea followed by left-sided body jerking and loss of consciousness. The ECoG did not detect any of these events. The patient was admitted to EMU, where a working diagnosis of PNES was confirmed although a brief seizure from insular cortex was in the differential diagnosis. The patient is undergoing CBT, and the ECoG data provided by the RNS are used to monitor the progress of her therapy. Within three months after starting CBT, the magnet swipe decreased from 59/month to 26/month (45% decrease), while detected seizures (long episodes confirmed with visual inspection) during the same period were 2 and 3/month. In both these cases, video-EEG was used to confirm the diagnosis of PNES, while the RNS ECoG was used to quantify the seizures in the ambulatory setting. 3.4. Case #5: somatic symptom disorder (SSD) mimicking seizures A 27-year-old right-handed man with a large right parietotemporal focal cortical dysplasia (FCD) had sEEG evaluation that confirmed seizures
Fig. 2. Sample electrocorticography and detected events in subject #2 (postictal psychosis). A — Number of long episodes (defined as events lasting N30 s; green bar plot) between Dec. 5 and 11, 2018. B — Example of a long episode representing electrographic seizure recorded at 2:45 AM in the 4-channel electrocorticography. C — Electrocorticography showing seizure termination and transition to the postictal state. Patient's caregiver swiped the magnet multiple times (see D also) as the patient was reporting worsening of psychosis that lasted for days. D — Electrocorticography showing persistent epileptiform activities. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
4
A.T. Issa Roach et al. / Epilepsy & Behavior 102 (2020) 106814
with the RNS can be challenging, given the limitations of the device stated earlier in the discussion. Some programming approaches that can help mitigate the chances of “false negative” ECoG (i.e., ECoG that did not show seizure even though a seizure actually occurred around that time) are as follows: a) encouraging frequent data downloads immediately after clinical events to minimize overwriting; b) encourage patient or caregiver to magnet swipe within a minute after the clinical event; c) increasing ECoG length; and d) shortening Long Episode duration. The long-term outcome of RNS neurostimulation therapy showed stable seizure reduction in patients with neocortical and mesiotemporal epilepsies [18]. Additionally, data on quality of life and cognition are favorable in patients with RNS [19,20]. The availability of ambulatory ECoG provides the opportunity to manage comorbidities in epilepsy that can mimic seizures and contributes to the overall poor quality of life.
localized to the right mesial temporal and parietal regions. He underwent right ATL and limited resection of the right mesial parietal region followed by RNS implanted with strip leads placed over the lateral parietal regions. He also had a significant comorbid anxiety disorder. In the first few months postimplantation, he reported a substantial increase in paroxysmal dizziness, impending doom, and fear of falling that limited his mobility; he ended up bedbound. The patient was convinced that these prolonged dizzy spells (lasting minutes to up to an hour), described as a feeling of being on a rocking boat, were seizures. He underwent ambulatory EEG and, subsequently, an EMU admission that identified these events as PNES. The ECoG and the scalp EEG recorded simultaneously did not show any electrographic changes during these spells. Multiple specialists, including ear, nose and throat (ENT) and neurohospitalist, ruled out any organic pathology. Antiseizure drugs were adjusted without much success. Finally, psychiatry was consulted for the treatment of his generalized anxiety disorder. The patient was counseled appropriately and encouraged to swipe his RNS magnet with every dizzy spell. The epileptologist reviewed the ECoG and provided consistent and reassuring feedback that the events were nonepileptic. Over two months, the patient improved significantly (magnet swipe decreased from 12/month to one/ month), and his dizziness eventually resolved completely.
SP has served as a paid consultant for NeuroPace, Inc. but declares no targeted funding or compensation for this study. None of the authors share any competing interests.
4. Discussion and conclusion
Acknowledgment
The five cases described above are great examples of the additional use of readily available ECoG data provided by the RNS in the management of comorbidities that are highly prevalent in PWE. The ability to access ECoG data and correlate them with patient magnet swipes during any paroxysmal event is unique compared with other therapeutic options in the field [9]. However, the RNS has its limitations. (1) The spatial sampling is restricted to only two four-contact cortical strip and/or depth leads. Hence, the presence of any electrical activity outside the sensors will not be detected. (2) Although the neurostimulator continuously senses and monitors electrographic activity through the leads, the memory capacity is limited to storage of 12 min of ECoG daily with the newer RNS. (3) The magnet swipe triggers the storage of ECoG for a limited period prior to the swipe (typically 60 s). Therefore, a late magnet swipe (i.e., more than 60 s after a clinical event) will miss the activity that is temporally correlated with symptoms, resulting in a ‘false negative’ ECoG. (4) Electrographic seizures stored for the ‘Long Episode’ trigger must be detected by the neurostimulator and persist above the detection threshold for a minimum duration of time (typically 30 s). Despite these limitations, the RNS has been used to lateralize and localize seizure foci, monitor medication effects, and characterize temporal patterns of seizures and the running-down phenomenon [10–14]. Seizures can present as panic attacks, worsening anxiety, dizziness, acute psychosis, agitation, and hallucinations [15]. All these symptoms are also prevalent in chronic epilepsy and can be attributed to comorbid depression, anxiety, medication side effects, and other neuropsychiatric disorders [2,4,6]. The availability of EEG correlates of these paroxysms is critical in differentiating epileptic from nonepileptic events. Epilepsy-related neurobehavioral symptoms have been first reported in the 1800s by Gowers and Jackson [16,17]. These symptoms are often classified in their timely relationship to the actual seizure (i.e., preictal, ictal, or postictal). Affective symptoms like fear, anxiety, and panic attacks can represent ictal phenomena commonly associated with limbic seizures [3,4,6]. Psychosis can be an ictal, postictal, or an interictal phenomenon. Interictal psychosis is frequently associated with other mood disorders (so-called as interictal dysphoric disorder) [7]. Postictal psychosis is managed with as needed benzodiazepines, while interictal dysphoric disorder is treated effectively with antidepressants [3,7]. The availability of ECoG can help to categorize these neurobehavioral symptoms, guide their management appropriately, and monitor treatment response. Similarly, the ECoG also offers the advantage to diagnose PNES early in PWE and avoid further mistreatments by not adding antiseizure medications or prevent multiple ER visits. However, characterizing paroxysmal events
SP and WM would like to thank NeuroPace representatives Ms. Kaitlyn Wilmer-Fierro and Mr. Mark Griffin and UAB study coordinator Ms. Kathleen Hernando.
Declaration of competing interest
References [1] Lin JJ, Mula M, Hermann BP. Uncovering the neurobehavioural comorbidities of epilepsy over the lifespan. Lancet 2012;380:1180–92. [2] Strine TW, Kobau R, Chapman DP, Thurman DJ, Price P, Balluz LS. Psychological distress, comorbidities, and health behaviors among U.S. adults with seizures: results from the 2002 National Health Interview Survey. Epilepsia 2005;46:1133–9. [3] Mula M, Monaco F. Ictal and peri-ictal psychopathology. Behav Neurol 2011;24:21–5. [4] Kanner AM. Obstacles in the treatment of common psychiatric comorbidities in patients with epilepsy: what is wrong with this picture? Epilepsy Behav 2019;98:291–2. [5] Devinsky O. Therapy for neurobehavioral disorders in epilepsy. Epilepsia 2004;45 (Suppl. 2):34–40. [6] Mula M. Epilepsy-induced behavioral changes during the ictal phase. Epilepsy Behav 2014;30:14–6. [7] Maguire M, Singh J, Marson A. Epilepsy and psychosis: a practical approach. Pract Neurol 2018;18:106–14. [8] Benbadis SR, Agrawal V, WOt Tatum. How many patients with psychogenic nonepileptic seizures also have epilepsy? Neurology 2001;57:915–7. [9] Skarpaas TL, Jarosiewicz B, Morrell MJ. Brain-responsive neurostimulation for epilepsy (RNS((R)) system). Epilepsy Res 2019;153:68–70. [10] King-Stephens D, Mirro E, Weber PB, Laxer KD, Van Ness PC, Salanova V, et al. Lateralization of mesial temporal lobe epilepsy with chronic ambulatory electrocorticography. Epilepsia 2015;56:959–67. [11] Chan AY, Knowlton RC, Chang EF, Rao VR. Seizure localization by chronic ambulatory electrocorticography. Clin Neurophysiol Pract 2018;3:174–6. [12] Skarpaas TL, Tcheng TK, Morrell MJ. Clinical and electrocorticographic response to antiepileptic drugs in patients treated with responsive stimulation. Epilepsy Behav 2018;83:192–200. [13] Baud MO, Kleen JK, Mirro EA, Andrechak JC, King-Stephens D, Chang EF, et al. Multiday rhythms modulate seizure risk in epilepsy. Nat Commun 2018;9:88. [14] Geller AS, Friedman D, Fang M, Doyle WK, Devinsky O, Dugan P. Running-down phenomenon captured with chronic electrocorticography. Epilepsia Open 2018;3: 528–34. [15] Berg AT, Altalib HH, Devinsky O. Psychiatric and behavioral comorbidities in epilepsy: a critical reappraisal. Epilepsia 2017;58:1123–30. [16] Gowers WR. Epilepsy and other chronic convulsive diseases: their causes, symptoms, and treatment. Old Hickory Bookshop; 1901. [17] Hughlings-Jackson J. On a particular variety of epilepsy (“intellectual aura”), one case with symptoms of organic brain disease. Brain 1888;11:179–207. [18] Bergey GK, Morrell MJ, Mizrahi EM, Goldman A, King-Stephens D, Nair D, et al. Longterm treatment with responsive brain stimulation in adults with refractory partial seizures. Neurology 2015;84:810–7. [19] Heck CN, King-Stephens D, Massey AD, Nair DR, Jobst BC, Barkley GL, et al. Two-year seizure reduction in adults with medically intractable partial onset epilepsy treated with responsive neurostimulation: final results of the RNS System pivotal trial. Epilepsia 2014;55:432–41. [20] Loring DW, Kapur R, Meador KJ, Morrell MJ. Differential neuropsychological outcomes following targeted responsive neurostimulation for partial-onset epilepsy. Epilepsia 2015;56:1836–44.