Highlights
– A prospective, nested case–control analysis within the REPO2MSE cohort found four independent SUDEP risk factors: extratemporal epileptogenic zone, BMI ≥30, male sex, and predominantly nocturnal seizures.
– Peri‑ictal peripheral oxygen saturation (SpO2) <80%, focal‑to‑bilateral tonic‑clonic seizure frequency, heart‑rate variability measures, number of antiseizure medications, and history of depression were not independently associated with SUDEP in this cohort.
– Results underscore the importance of better characterizing extratemporal epilepsies and considering targeted diagnostic and surgical pathways to mitigate SUDEP risk in drug‑resistant focal epilepsy.
Background and clinical context
Sudden unexpected death in epilepsy (SUDEP) is the leading direct epilepsy‑related cause of premature mortality in people with epilepsy, particularly those with drug‑resistant disease. Identifying robust, clinically actionable biomarkers to stratify individual SUDEP risk remains an unmet need. Historically, generalized tonic‑clonic seizure (GTCS) burden and poor seizure control have been among the most consistently reported epidemiological risk factors, but peri‑ictal physiological signatures (for example profound hypoxaemia, cardiac arrhythmia, or autonomic instability) and anatomical seizure onset sites have been less well characterized in prospective, multicentre cohorts with standardized in‑hospital monitoring.
Study design
The REPO2MSE study is a nationwide, prospective cohort in France that enrolled adults with drug‑resistant focal epilepsy undergoing inpatient video‑EEG monitoring at 16 epilepsy monitoring units between May 18, 2010, and Aug 23, 2015. Vital status was followed through the end of 2018. SUDEP cases were identified through linkage to national records and adjudicated using medical records and structured interviews regarding circumstances of death.
Nested case–control approach
Within the cohort of 1,074 participants and 6,828 patient‑years of follow‑up, 42 deaths occurred, including 18 adjudicated definite or probable SUDEP cases. Each SUDEP case was matched to four controls on study centre and date of inclusion. The investigators collected detailed clinical information, presurgical investigation results, and raw inpatient recordings including video‑EEG, ECG, and SpO2 traces. SUDEP risk factors were evaluated with LASSO‑penalized conditional logistic regression to select variables associated with SUDEP while controlling for potential overfitting.
Key findings
The study reported a SUDEP incidence of 2.64 per 1,000 patient‑years (95% CI 1.36–3.92) in this drug‑resistant focal epilepsy cohort. Four variables emerged as independent risk factors for SUDEP after multivariable LASSO selection:
- Extratemporal epileptogenic zone (compared with temporal lobe): OR 37.8, 95% CI 3.21–446.2, p=0.0039.
- Body mass index (BMI) ≥30 kg/m2: OR 26.0, 95% CI 2.0–339.6, p=0.013.
- Male sex: OR 12.6, 95% CI 1.5–106.8, p=0.0201.
- Predominantly nocturnal seizures: OR 6.0, 95% CI 1.2–28.7, p=0.026.
Notably, several factors that have biological plausibility or prior supportive evidence were not significantly associated with SUDEP in this analysis. These include the presence of peri‑ictal SpO2 <80% during focal seizures, frequency of focal‑to‑bilateral tonic‑clonic seizures (FTBTC), heart‑rate variability indices, age at epilepsy onset, the number of antiseizure medications, and history of depression.
Interpretation of effect sizes and precision
The observed point estimates are large, particularly for extratemporal onset and obesity, but confidence intervals are wide—reflecting the modest number of SUDEP events (n=18). The magnitude of the odds ratios implies potentially strong associations but also highlights uncertainty; the results are hypothesis‑generating and warrant confirmation in larger cohorts or meta‑analyses.
Biological plausibility and mechanistic considerations
The link between extratemporal epilepsies—especially those involving the perisylvian region and frontal lobes—and SUDEP has several plausible mechanistic pathways. Frontal and insular/perisylvian networks have dense tectic and autonomic connections and are implicated in cardiorespiratory control and arousal. Seizure propagation through these regions might more readily disrupt airway patency, respiratory drive, or autonomic regulation, increasing the likelihood of terminal respiratory or cardiac events. Predominantly nocturnal seizures may heighten risk because of delayed detection and reduced likelihood of rapid external intervention. Obesity may contribute via impaired baseline respiratory function (including sleep‑disordered breathing) and cardiovascular comorbidity, potentially lowering the threshold for fatal peri‑ictal respiratory compromise.
Clinical implications and translational opportunities
The REPO2MSE findings have several potential implications for clinicians caring for people with drug‑resistant focal epilepsy:
- Risk stratification: Incorporating seizure‑onset localization (temporal vs extratemporal), sex, BMI, and nocturnality into individualized SUDEP risk discussions could refine counseling and shared decision‑making, while acknowledging current evidence limitations.
- Diagnostic focus on extratemporal epilepsy: Given the strong association observed, there is an argument for more intensive efforts to localize extratemporal epileptogenic zones using multimodal presurgical assessment (high‑resolution MRI, PET, magnetoencephalography, stereo‑EEG as appropriate) with the intent of offering surgical or neuromodulatory therapies when feasible.
- Weight management and comorbidity optimization: Identification of obesity as a modifiable risk factor suggests that targeted weight loss interventions and screening for sleep‑disordered breathing could form part of SUDEP prevention strategies.
- Nocturnal risk mitigation: For patients with predominantly nocturnal seizures, low‑cost interventions (seizure detection devices, nocturnal supervision where acceptable, counselling on sleeping position and airway safety) could be emphasized while recognizing variable evidence for efficacy in SUDEP prevention.
Strengths and limitations
The study’s strengths include its prospective national design, adjudicated SUDEP outcomes, access to raw physiological recordings during monitored seizures, and the use of matched controls with LASSO regression to reduce overfitting. However, important limitations temper interpretation:
- Small event count: With only 18 SUDEP cases, statistical power is limited, and effect estimates are imprecise (wide confidence intervals).
- Selection bias: The cohort comprises adults with drug‑resistant focal epilepsy undergoing inpatient monitoring at tertiary centres; findings may not generalize to all people with epilepsy, particularly those with generalized epilepsies or well‑controlled seizures.
- Residual confounding: Although matching and penalized regression were used, unmeasured confounders (socioeconomic factors, adherence, sleep‑related variables outside the monitoring setting) could influence associations.
- Negative findings: The absence of association for peri‑ictal SpO2 <80% and FTBTC frequency does not definitively rule out their importance; measurement variability, timing of recordings relative to fatal events, and the inpatient setting may have influenced these null results.
Expert commentary and how this fits with prior literature
The REPO2MSE dataset adds a novel, prospective perspective on seizure‑related biomarkers for SUDEP by focusing on anatomical onset and peri‑ictal physiological recordings. It complements epidemiological literature that has emphasized GTCS frequency and uncontrolled seizures as major risk determinants, while suggesting that anatomical and patient‑level characteristics (extratemporal focus, male sex, obesity, nocturnality) deserve greater attention. Because prior studies have variably implicated cardiac and respiratory peri‑ictal events, the lack of a robust association with peri‑ictal SpO2 here raises important questions about the timing and transferability of monitoring data to real‑world SUDEP events and underscores the need for larger collaborative efforts pooling monitoring datasets.
Practical takeaways for clinicians
Clinicians should incorporate these findings into risk discussions while making clear the preliminary nature of the associations. Key practical points:
- Actively pursue comprehensive lateralization/localization in patients with suspected extratemporal epilepsy; timely referral to epilepsy surgery programs is appropriate when localization is achievable.
- Screen for and manage obesity and sleep‑disordered breathing as part of holistic epilepsy care.
- Consider intensified safety planning for patients with predominantly nocturnal seizures, including discussion of seizure detection technologies and caregiver notification strategies.
- Continue to prioritize seizure control—particularly prevention of GTCS—because of its well‑documented association with SUDEP in the broader literature, even though FTBTC frequency did not emerge as an independent predictor in this specific analysis.
Future research directions
Confirmatory studies in larger, international cohorts are needed to validate extratemporal onset and obesity as SUDEP risk biomarkers and to refine absolute risk estimates. Integration of long‑term ambulatory physiological monitoring (wearable SpO2 and ECG), standardized assessments of sleep‑disordered breathing, and more granular mapping of seizure propagation pathways could clarify mechanistic links. Collaborative meta‑analytic efforts and prospective registries that harmonize presurgical and monitoring data will be critical to move from association to causation and actionable prevention strategies.
Conclusion
The REPO2MSE prospective nested case–control study identifies extratemporal epileptogenic zones, obesity (BMI ≥30), male sex, and predominantly nocturnal seizures as independent predictors of SUDEP in adults with drug‑resistant focal epilepsy. These findings point to underappreciated clinical characteristics that may help refine risk stratification and prevention, especially through improved diagnosis and management of extratemporal epilepsies. Given the small number of SUDEP events and wide confidence intervals, these results should be considered hypothesis‑generating and require replication in larger, diverse cohorts.
Funding
REPO2MSE was funded by the French Ministry of Health (Programme Hospitalier de Recherche Clinique National 2009).
Reference
Ryvlin P, Huot M, Valton L, Maillard L, Bartolomei F, Derambure P, Hirsch E, Michel V, Chassoux F, Petit J, Crespel A, Biraben A, Navarro V, Kahane P, De Toffol B, Thomas P, Rosenberg S, Bernini A, Charlois AL, Craciun L, Chorfa F, Ducouret P, Ferreira A, Leclercq M, Marty M, Mercedes Alvarez B, Sampaio M, Spahr A, Timestit‑Kurland N, Touya M, Roy P, Rheims S; REPO2MSE study group. Seizure‑related biomarkers of sudden unexpected death in epilepsy (SUDEP) in drug‑resistant focal epilepsy (REPO2MSE): a prospective, multicentre case‑control study. Lancet Neurol. 2025 Nov 21:S1474‑4422(25)00379‑5. doi: 10.1016/S1474‑4422(25)00379‑5. Epub ahead of print. PMID: 41285145.