Abstract
Whitney R, Keller A, Li SA, Datta AN, MacDonald M, Nabavi Nouri M, Pohl D, Sell E, Ronen GM, Sidhu M, Simard-Tremblay E, Pollanen MS, Donner EJ. Epilepsia. 2025 Jun;66(6):1988-2000. doi: 10.1111/epi.18339. Epub 2025 Apr 5. PMID: 40186506. Objective: This study was undertaken to understand the circumstances surrounding pediatric sudden unexpected death in epilepsy (SUDEP) and identify clinical factors that may be associated with SUDEP in childhood. Methods: A retrospective case series was conducted. Pediatric SUDEP cases were collected across Canada from the Ontario Forensic Pathology Service, Canadian Pediatric Surveillance Program, and Canadian Pediatric Epilepsy Network. Demographics, epilepsy history, comorbidities, and circumstances surrounding death were analyzed. Results: Forty-nine children with pediatric SUDEP were analyzed; 25 (51%) were females, and the median age at death was 8 years. Six children (12%) were <2 years of age at the time of death. Information on seizure types 6 months before death was known in 35 children. Twenty-two had tonic–clonic seizures within the last 6 months prior to death (63%). Seven children (18%) had no tonic–clonic seizures in their lifetime. Two thirds of children were treated with ≥2 antiseizure medications. Genetic etiologies were most common (55%). Data on global developmental delay (GDD) was known in 46 children: 12 children (26%) had no impairment, and 34 were globally delayed (74%). Children with GDD had earlier age at seizure onset (P < .001); however, epilepsy duration was similar to those without GDD (P = .170). Similar to adult cohorts, death was often unwitnessed (n = 41/46, 89%). Information on recent infection before death was known in 37 children. Seventeen children (46%) had a recent infection. Significance: Our study represents the largest pediatric SUDEP case series to date. SUDEP occurred in children of all ages, including infants, with a spectrum of epilepsies with and without neurodevelopmental impairment. The circumstances around death (ie, timing of death, witnessed/unwitnessed) were similar to previous SUDEP cohorts. A recent infection was often observed, which could decrease seizure threshold and trigger a terminal seizure and may suggest that times of increased seizure risk could warrant heightened surveillance for SUDEP. However, further research is needed to determine the significance of this finding.
Commentary
The American Epilepsy Society and American Academy of Neurology recommend discussing Sudden Unexpected Death in Epilepsy (SUDEP) with every patient as soon as possible—at or after the initial visit—as a best practice recommendation. 1 However, most child neurologists do not follow this practice despite the fact that the risk of SUDEP in pediatric patients is no different than the adult population at approximately 1.2 per 1000 person-years. 2 This deviation from best practice is mostly attributed to lack of sufficient knowledge and lack of a feeling of ethical and moral obligation to be discussing SUDEP with families. 2
While most epileptologists at tertiary care centers have either had experience with SUDEP or are knowledgeable about it, let us review what is already known about SUDEP:
Childhood Epilepsies are Different From Adult Epilepsies but the Risk of SUDEP is Similar! What are the Specific Risk Factors and How Does Each Increase Risk for SUDEP?
Previous small case series had identified SUDEP risk factors including young age at epilepsy diagnosis, long duration of epilepsy, focal epilepsy, intellectual disability as examples. 3 Subsequently, a systematic review published by the American Academy of Neurology in 2017 1 clarified risk factors. While high frequency of generalized tonic-clonic seizures (GTC) was the only risk factor identified with a high level of confidence; lack of nighttime supervision and absence of nocturnal listening device were identified risk factors with moderate confidence. In general, methodologic limitations due to small numbers, highly selected populations and differences in definitions of potential risk factors hampered such data pooling. The need for large, high-quality studies to assess risk of SUDEP was unmet till 2020 when Sveinsson et al published their study. These results were obtained from a Swedish population-based case control study of SUDEP-related deaths between July 1, 2006, and December 31, 2011. SUDEP incidence under the age of 16 years was 1.11 per 1000 person-years. Key risk factors included experiencing any GTCs during the preceding year, which contributed to a 27-fold increased risk. Further modeling and logistical regression analysis for the presence of exclusively nocturnal GTCs in the preceding year contributed to a 15-fold increased risk while living alone contributed to a five-fold increased risk. Moreover, the interaction of the combination of living alone and having at least one GTC increased the risk by 67-fold. 4 Previously identified associations of young age at epilepsy onset, longer duration of epilepsy, structural etiology became nonsignificant after controlling for GTC frequency in the above study. While Sveinsson's data identified GTCs as the main culprit, the North American SUDEP Registry data published in 2019 5 brought attention to the fact that, of 143 patients with known information, 60% had focal epilepsy. Additionally, treatment responsive or benign epilepsy (what would now be classified as self-limited epilepsies) contributed to a sizable minority of SUDEP deaths.
Another systematic analysis of pediatric SUDEP published in 2019 concluded that there is currently insufficient evidence to determine etiology of pediatric SUDEP due to methodological issues of noncomparative studies with a high risk of bias. 6 Moreover, GTCs are not a common feature of pediatric epilepsy—especially in the younger age group. Genetic factors have also been posited as additional risk factors in SUDEP. This hypothesis is supported by the increased incidence of SUDEP in children with epilepsies caused by pathogenic variants in genes including SCN1A, SCN8A, DEPDC5. 7
Are Circumstances Around Pediatric SUDEP Different?
This would be a logical question to ask because lack of medication adherence, lack of supervision, living alone, as SUDEP-associated risk factors, are likely less applicable in pediatric epilepsy when compared to adult epilepsy. Authors Whitney and Donner 8 investigated the circumstances surrounding SUDEP in Canadian children using SUDEP-related data from the Ontario forensic pathology service (collected over 3 years between December 2014 and 2017), Canadian pediatric surveillance program (pediatricians were surveyed monthly over 2 years between January 2014 and December 2015) and Canadian pediatric epilepsy network (members completed case report form regarding SUDEP between January 1, 2000 and December 31, 2017). This forms the largest retrospective pediatric case series with 49 definite or probable cases of SUDEP to date.
Key Findings in this Paper Related to Epilepsy
Median age at death was 8 years (range 6 months to 17.9 years)
In the six months prior to death, 63% [22/35] children had either primary generalized or focal to bilateral tonic-clonic seizures.
Seven children [18%] had no documented lifetime history of tonic-clonic seizures.
Three children were seizure-free in the 6 months prior to death [one had juvenile myoclonic epilepsy].
Twenty-seven children [55%] had a genetic or presumed genetic etiology while 12 children [25%] had an unknown etiology.
91% of the children were on antiseizure medications (ASM) at the time of death and only 24% children [5/21] had a history of poor compliance.
Seven children had a dosage reduction [7/34, 47%] or had missed a dose in the 24 h preceding death.
Relevant Medical History
Medical history was extracted from the chart and thus further details regarding severity or classification of comorbidities is not known.
Majority of the children had global developmental delay [46/49, 74%]. Seven children were noted to have a cardiac comorbidity [7/43, 16%]. Nine children had a respiratory comorbidity [9 /36, 25%].
Other Notable Variables
Nine children [9/23, 39%] shared a room with someone.
In seven children [7/17, 41%] regular nocturnal checks were performed and in 2 of the 7 children, listening devices were also used.
93% of the patients were asleep [42/45] and death was unwitnessed in 89% [41/46].
Seventeen children [17/37, 46%] were reported to have recent infection prior to death [“recent” is defined as within the same day or few days—per communication with authors].
Findings in Children <2 years of age
Most patients had global developmental delays, were diagnosed with developmental and epileptic encephalopathy, and were on polytherapy with ASMs at the time of death.
Conclusion and Take-Home Message
Authors who are well-respected in the field have put together the largest case series of pediatric SUDEP. However, this data does not add new knowledge that will help to quantify risk associated with above factors for SUDEP in the next pediatric epilepsy patient that you see. Authors published a retrospective case series built on case report forms submitted across several time spans without a population-based control necessary for comparison or quantification of risk factors associated with pediatric SUDEP.
The circumstances around pediatric SUDEP are not vastly different from those of adult SUDEP in this retrospective sample. One valuable and useful study finding adds justification to the practice of giving patients bridge doses of increased ASM around times of illness/infection—especially in patients with other medical comorbidities. Although granularity regarding significance of medical comorbidities, details regarding infection, type of infection, severity of infection related to baseline infection for that child are lacking in the study, bridge dosing may not be an “overkill.” Close to 50% of those lost to SUDEP in the above case series had a prior infection which in turn could have increased chances of terminal seizures.
By capturing the spectrum of pediatric SUDEP, authors have succeeded in highlighting the shocking reality that no child with epilepsy is risk free when it comes to SUDEP. While much more work is needed to individualize and quantify the risk to the next patient I see in clinic; I must continue to talk about SUDEP to ALL my patients with epilepsy. Without being an alarmist, I must counsel patients that NO seizure is risk-free while trying to reduce the frequency of GTCs as best as I can.
