Carotid body stimulation as a potential intervention in sudden death in epilepsy

Abstract

Objective: To investigate carotid body (CB) mechanisms related to sudden death during seizure. Ictal activation of oxygen-conserving reflexes (OCRs) can trigger fatal cardiorespiratory collapse in seizing rats, which presents like human sudden unexpected death in epilepsy (SUDEP). The CB is strongly implicated in OCR pathways; we hypothesize that modulating CB activity will provide insight into these mechanisms of death.

Methods: Long-Evans rats were anesthetized with urethane. Recordings included: electrocorticography, electrocardiography, respiration via nasal thermocouple, and blood pressure (BP). The mammalian diving reflex (MDR) was activated by cold water delivered through a nasal cannula. Reflex and stimulation trials were repeated up to 16 times (4 pre-intervention, 12 post-intervention) or until death. In some animals, one or both carotid bodies were denervated. In some animals, the CB was electrically stimulated, both with and without MDR. Seizures were induced with kainic acid (KA).

Results: Animals without seizure and with no CB modulation survived all reflexes. Non-seizing animals with CB denervation survived 7.1 ± 5.4 reflexes before death, and only 1 of 7 survived past the 12-trial threshold. Electrical CB stimulation without seizure and without reflex caused significant tachypnea and hypotension. Electrical CB stimulation with seizure and without reflex required higher amplitudes to replicate the physiological responses seen outside seizure. Seizing animals without CB intervention survived 3.2 ± 3.6 trials (per-reflex survival rate 42.0% ± 44.4%), and 0 of 7 survived past the 12-trial threshold. Seizing animals with electrical CB stimulation survived 10.5 ± 4.7 ictal trials (per-reflex survival rate 86.3% ± 35.0%), and 6 of 8 survived past the 12-trial threshold.

Significance: These results suggest that, during seizure, the ability of the CB to stimulate a restart of respiration is impaired. The CB and its afferents may be relevant to fatal ictal apnea and SUDEP in humans, and CB stimulation may be a relevant intervention technique in these deaths.

Keywords: Carotid sinus nerve; Diving reflex; Ictal apnea; SUDEP; Seizure.

Fig. 1.

Left: Microhook electrode fabricated from a 50-μm wire. Right: photograph in a female LE rat of electrode placement around the carotid sinus nerve (CSN), which branches off the glossopharyngeal nerve (IX) and passes under the hypoglossal nerve (XII) before entering the carotid body. The external carotid artery (ECA) and occipital artery (OA) serve as important surgical landmarks and so are also labeled. This procedure was performed on the left side of the animal. The top of the photo is the rostral direction, and photo left and right are medial and lateral, respectively.

Fig. 2.

Mean arterial pressure (A), heart rate (B), and respiratory rate (C) measurements during mammalian diving reflex challenges under various experimental parameters. Thick lines indicate the average value over time for all trials, with thin dashed lines indicating the 95% confidence intervals for the values over time across trials. t = 0 seconds corresponds to the time of reflex challenge initiation. All x-axes are equivalent. Intrahippocampal and systemic kainic acid data sets have been combined in both ictal categories to facilitate comparison to carotid body denervation.

Fig. 3.

Average heart rate, mean arterial pressure, and respiratory rate changes resulting from 10 s of carotid sinus nerve (CSN) stimulation in an anesthetized rat. All changes are normalized to the average baseline amplitude in the 10-s window before stimulation began. Measurements for each amplitude were taken at the point of greatest difference from the baseline amplitude in the 20 s following start of stimulation. Asterisks indicate statistically significant difference between the measurement following stimulation and the pre-stimulation baseline (Wilcoxon Signed Rank test). Data were averaged from the total 20 stimuli across 4 animals at each amplitude. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Fig. 4.

Average heart rate, mean arterial pressure, and respiratory rate changes resulting from 10 s of carotid sinus nerve (CSN) stimulation in an anesthetized rat following seizures induced by administration of kainic acid. As in Fig. 3, all changes are normalized to the average baseline amplitude in the 10-s window before stimulation began. Measurements for each amplitude were taken at the point of greatest difference from the baseline amplitude in the 20 s following start of stimulation. Asterisks indicate statistically significant difference between the measurement following stimulation and the pre-stimulation baseline (Wilcoxon Signed Rank test) and are displayed vertically to reduce overlap between data points. Data were averaged from the total 180 stimuli across 11 animals. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Fig. 5.

Example physiological traces of nonfatal and fatal diving reflex challenges. a) Data from a diving reflex challenge in a non-seizing animal which is survived. b) Data from a fatal diving reflex challenge from an animal with seizures induced with kainic acid. ECoG = electrocorticography, ECG = electrocardiography, Resp. = respiration via nasal thermocouple, MDR = mammalian diving reflex. The vertical dashed bar indicates the start of the reflex stimulus, which was applied for 20 seconds.

Fig. 6.

Animals with carotid sinus nerve (CSN) stimulation survive more ictal mammalian diving reflex (MDR) challenges than animals without CSN stimulation. Experiments were terminated upon animal death or after 12 ictal MDR challenges, whichever was fewer, except one which went to 16 ictal challenges. Data represent n = 10 animals without CSN stimulation (left boxes) and n = 8 animals with CSN stimulation (right boxes). “Reflexes survived” indicates number of reflexes with full recovery prior to fatal reflex or experiment termination. Central line of boxes indicates the data median, with upper and lower quartiles represented by the thick end lines of the boxes. Non-overlapping shaded regions indicate statistically significant differences for p < 0.05.