Abstract
Purnell BS, Braun A, Fedele D, Murugan M, Boison D. Brain Commun. 2022;4(5):fcac232. doi:10.1093/braincomms/fcac232. PMID: 36196086; PMCID: PMC9525001 Sudden unexpected death in epilepsy is the leading cause of epilepsy related death. Currently, there are no reliable methods for preventing sudden unexpected death in epilepsy. The precise pathophysiology of sudden unexpected death in epilepsy is unclear; however, convergent lines of evidence suggest that seizure-induced respiratory arrest plays a central role. It is generally agreed that sudden unexpected death in epilepsy could be averted if the patient could be rapidly ventilated following the seizure. The diaphragm is a muscle in the chest which contracts to draw air into the lungs. Diaphragmatic pacing is a surgical intervention which facilitates normal ventilation in situations, such as spinal cord injury and sleep apnoea, in which endogenous respiration would be inadequate or non-existent. In diaphragmatic pacing, electrodes are implanted directly onto diaphragm or adjacent to the phrenic nerves which innervate the diaphragm. These electrodes are then rhythmically stimulated, thereby eliciting contractions of the diaphragm which emulate endogenous breathing. The goal of this study was to test the hypothesis that seizure-induced respiratory arrest and death can be prevented with diaphragmatic pacing. Our approach was to induce respiratory arrest using maximal electroshock seizures in adult, male, C57BL6 mice outfitted with EEG and diaphragmatic electrodes (n = 8 mice). In the experimental group, the diaphragm was stimulated to exogenously induce breathing. In the control group, no stimulation was applied. Breathing and cortical electrographic activity were monitored using whole body plethysmography and EEG, respectively. A majority of the animals that did not receive the diaphragmatic pacing intervention died of seizure-induced respiratory arrest. Conversely, none of the animals that received the diaphragmatic pacing intervention died. Diaphragmatic pacing improved postictal respiratory outcomes (two-way ANOVA, P < 0.001) and reduced the likelyhood of seizure-induced death (Fisher’s exact test, P = 0.026). Unexpectedly, diaphragmatic pacing did not instantly restore breathing during the postictal period, potentially indicating peripheral airway occlusion by laryngospasm. All diaphragmatically paced animals breathed at some point during the pacing stimulation. Two animals took their first breath prior to the onset of pacing and some animals had significant apnoeas after the pacing stimulation. Sudden unexpected death in epilepsy results in more years of potential life lost than any other neurological condition with the exception of stroke. By demonstrating that seizure-induced respiratory arrest can be prevented by transient diaphragmatic pacing in animal models we hope to inform the development of closed-loop systems capable of detecting and preventing sudden unexpected death in epilepsy.
Commentary
“The diaphragm is the only organ which only and all mammals have and without which no mammals can live.” 1 Mammals breathe by active inhalation that occurs due to diaphragm muscle contraction. Without oscillatory diaphragm contraction and relaxation, death is imminent for any mammal, including humans. Although other expiratory and postinspiratory muscle contraction can shape and heavily influence breathing, rhythmic diaphragm contraction is all that is needed to sustain life. A recent paper by Purnell and colleagues, from Rutgers University, demonstrates this most exquisitely by exogenously stimulating the diaphragm to prevent fatality in a mouse model of sudden unexpected death in epilepsy (SUDEP). 2
SUDEP is the sudden, unexpected, nontraumatic, and nondrowning death of a person with epilepsy for which postmortem examination does not reveal another cause of death. Sudden unexpected death in epilepsy accounts for up to 17% of all deaths of persons with epilepsy. 3 For those with epilepsy under 50 years of age, SUDEP is the second most frequent cause of death after cancer 4 and across all ages SUDEP is second only to stroke in potential life years lost. 5 Treatments for the prevention of SUDEP are nearly nonexistent, with the usual goal being to reduce or eliminate the occurrence of seizures, particular of the tonic–clonic variety. However, many epilepsy patients are refractory to standard anti-seizure drugs, and consequently at high risk of SUDEP. In addition, it is clear that SUDEP can occur after achieving apparent seizure freedom and even when very few seizures have been observed. A most notable example would be the actor Cameron Boyce, who reportedly only had 5 seizures over as many years prior to succumbing to SUDEP in 2019. Clearly, there is a great need for the development of SUDEP-specific treatments.
Underlying mechanisms of SUDEP are obscure. Considering it is mostly a diagnosis of exclusion, there are likely multiple mechanisms of SUDEP, both across all cases and even within individual cases. One put forward, however, has gained significant traction over the past several years: that of seizure-induced apnea. 6 In the clinic, the MORTEMUS study has provided 9 recorded cases of SUDEP events. In all cases, patients had tonic–clonic seizures that produced apnea minutes prior to terminal asystole. 7 This cascade of events has been demonstrated in a number of mouse models now. Furthermore, mechanical ventilation has had success preventing seizure-induced death in some of these models, suggesting that temporary exogenous stimulation of breathing could be sufficient to prevent death in cases of SUDEP. Indeed, the MORTEMUS study found 3 nonfatal cases that appeared nearly identical to the SUDEP cases, but in which cardiopulmonary resuscitation was instigated within minutes of apnea initiation. 7
Mechanical ventilation is a fine method to demonstrate the validity of rescue breathing as a potential therapeutic target; however, in most cases of SUDEP, patients are alone in bed at night 3 and it is unlikely to be effective in practice. Diaphragmatic pacing is an already FDA approved surgical intervention that facilitates breathing in a number of conditions where neural drive to breath is absent or impaired. 8
Purnell and colleagues perform intricate, yet straightforward experiments to demonstrate that diaphragm pacing can prevent seizure-induced death in mice (aka SUDEP). 2 Mice had EEG headsets implanted that included electrodes extended to both sides of the diaphragm, near the 2 innervation points of the phrenic nerve. Their method allowed for confirmation of electrode placement during the surgery, facilitating optimal diaphragm stimulation. Once recovered, mice had breathing recorded by whole-body plethysmography in tandem with EEG. The authors used the maximal electroshock (MES) method via electrodes clipped to the ears, which induced tonic–clonic seizures that were fatal in more than half of cases, as has been described for this method before. In mice where diaphragm pacing was enacted via electrical stimulation 10 seconds after MES induction at 4 Hz, approximating a normal mouse breathing rate, fatality from MES significantly decreased to 0%. Diaphragm stimulation had no impact on the usual MES seizure severity score, the flexion/extension ratio, indicating increased survival was not due to anticonvulsant effects, but likely due to overcoming the apnea that is secondary to the seizure. Finally, the authors demonstrated that postictal breathing was significantly improved by diaphragm pacing.
As mentioned above, SUDEP is most likely multifactorial, and MES may only accurately model a fraction of SUDEP cases. Future studies examining diaphragm pacing in other models of SUDEP will be informative. In addition, even in the MES model, the results had some surprises. For instance, diaphragm stimulation did not immediately produce inspiration, suggesting that other mechanisms of respiratory arrest (e.g., laryngeal or diaphragm contraction) could also be occurring at certain times during apnea. Eliminating seizure-induced apnea in entirety may require multiple or different approaches, but this does not detract from the life-saving possibility of diaphragm pacing.
No therapy is for all patients. Diaphragm pacing is certainly invasive and would likely be targeted only to those that are at high risk of SUDEP. While reliable biomarkers for SUDEP are currently rudimentary at best, this would include patients that experience regular tonic–clonic seizures, nocturnal seizures, or seizures that present with apnea. 3,6 Further work that can help predict SUDEP occurrence will be helpful. In addition, diaphragm pacing can be less effective for obese patients 9 and over stimulation of the diaphragm can lead to muscle fatigue. 10 The latter point should not be of concern when seizures and apnea seldomly occur. 10 Diaphragm pacing has already been used effectively for spinal cord injury, central hypoventilation syndrome, and central sleep apnea, demonstrating its effectiveness and safety. 8 For spinal cord injury, it has been reported that diaphragm pacing decreases treatment cost and improves quality of life compared to mechanical ventilation. 11 One can envision a device which detects respiratory arrest during or after a seizure and then automatically enacts electrical stimulation of the diaphragm—and in fact, such a device is already FDA approved for central sleep apnea: the Remedē system by Zoll Respicardia, Inc. Thus, these exciting results by Purnell and colleagues represent a novel, yet clearly translatable intervention for SUDEP that would allow patients and loved ones alike to breathe easier.
