Valproate Linked to Higher Mortality in Patients With Dementia and Epilepsy

Navigating the clinical intersection of cognitive decline and seizure management

The clinical management of patients with neurocognitive disorders is increasingly complicated by the bidirectional relationship between dementia and epilepsy, where each condition significantly elevates the risk of developing the other [1, 2]. Late-onset epilepsy often serves as a harbinger of accelerated cognitive decline or underlying cerebrovascular disease, yet diagnosing seizures in the elderly remains difficult due to atypical presentations such as impaired awareness or confusion rather than motor convulsions [3, 4]. While the International League Against Epilepsy has standardized the definition of the disease, selecting an appropriate treatment regimen for patients with existing dementia is fraught with concerns regarding drug-drug interactions and the exacerbation of cognitive symptoms [5, 6]. High-quality evidence comparing the long-term safety and survival outcomes of different antiseizure medications in this specific population has historically been limited, leaving clinicians to rely on expert consensus rather than robust comparative data [6]. A new population-wide cohort study now addresses this gap by examining how the initial choice of therapy influences mortality risk in thousands of patients facing these dual diagnoses.

Longitudinal analysis of Swedish registry data

To evaluate the impact of antiseizure medication selection on survival, researchers conducted a population-wide cohort study utilizing data from Swedish national registers. The study population consisted of 5,764 individuals who had been diagnosed with dementia and subsequently initiated treatment with an antiseizure medication following a diagnosis of epilepsy after January 1, 2006. This cohort included 2,811 men (48.8%) and 2,953 women (51.2%), with the analysis of their clinical data extending through December 2023. To identify specific treatments, the authors tracked dispensed medications categorized under the Anatomical Therapeutic Code N03, which is a standardized classification system for drugs acting on the nervous system. This systematic tracking allowed the researchers to monitor real-world prescribing habits across the entire Swedish population, providing a high level of external validity to the findings.

Quantifying the mortality risk across medication classes

The researchers utilized Cox proportional hazards regression (a statistical method for investigating the association between the survival time of patients and specific predictor variables) to determine the relationship between specific antiseizure medications and all-cause mortality. To ensure the findings reflected the impact of the medications rather than baseline health differences, the Cox regression models adjusted for age, sex, year of antiseizure medication start, and comorbidities. This rigorous adjustment was critical given the complex medical profiles of patients living with both dementia and epilepsy, where polypharmacy and frailty often confound survival data. The analysis revealed that use of valproate was associated with the highest risk of death in persons with epilepsy and dementia. Specifically, valproate (n = 746) was associated with an increased adjusted hazard ratio (aHR) of death of 1.34 (95% CI 1.20-1.48). In contrast, other common treatments showed more favorable survival profiles. Lamotrigine (n = 922) was associated with a reduced aHR of death of 0.84 (95% CI 0.75-0.93), representing a 16 percent reduction in mortality risk compared to the reference group. Levetiracetam (n = 2,098) was associated with an aHR of death of 0.93 (95% CI 0.85-1.03), suggesting a neutral to slightly positive survival trend. When comparing the different therapeutic options, the study found that lamotrigine and, in some models, levetiracetam were associated with better survival than both valproate and carbamazepine. These findings provide quantitative evidence that the choice of antiseizure medication significantly influences long-term outcomes in this vulnerable population, suggesting that the increased mortality risk associated with valproate is a distinct clinical consideration when managing seizures in the context of neurocognitive decline.

Cardiovascular outcomes and statistical robustness

Beyond all-cause mortality, the researchers specifically assessed causes of death and the risk of cardiovascular death to identify potential drivers of the observed survival differences. They found that cardiovascular causes of death were more common among users of valproate or carbamazepine than among users of lamotrigine and levetiracetam. When using carbamazepine as the reference point, valproate was associated with an increased risk of cardiovascular death (aHR: 1.30; 95% CI 1.11-1.52). Conversely, lamotrigine was associated with a reduced risk of cardiovascular death (aHR 0.79; 95% CI 0.66-0.94) compared with carbamazepine. These findings suggest that the cardiovascular safety profile of these agents, particularly the potential for metabolic or vascular side effects, may be a significant factor in the long-term management of patients with comorbid dementia and epilepsy. To ensure the reliability of these associations, the authors conducted several sensitivity analyses. The risks remained similar in analyses using restricted mean survival time (a statistical measure of the average time until an event occurs within a specific follow-up window, which provides a more intuitive sense of life expectancy). The results were also consistent when using propensity score-matched sets of participants (a method used to balance groups by their likelihood of receiving a specific treatment to mimic a randomized trial) and when applying balancing weights in regression models. Furthermore, the increased mortality risk associated with valproate persisted in cases where epilepsy onset occurred in the context of existing dementia. These findings provide real-world support for existing expert guidelines that recommend newer antiseizure medications for older adults and those with cognitive impairment. For the practicing physician, these data emphasize that the choice of antiseizure therapy in patients with dementia is not merely a matter of seizure control but a critical factor in optimizing overall survival and cardiovascular health.

References

1. Tan Z, Wang F, Wu W, Yu L, Wu J, Wang L. Bidirectional relationship between late-onset epilepsy (LOE) and dementia: A systematic review and meta-analysis of cohort studies.. Epilepsy & behavior : E&B. 2024. doi:10.1016/j.yebeh.2024.109723

2. Chen L, Yang W, Yang F, et al. The crosstalk between epilepsy and dementia: A systematic review and meta-analysis.. Epilepsy & behavior : E&B. 2024. doi:10.1016/j.yebeh.2024.109640

3. Pirgit ML, Beniczky S. EEG and semiology in the elderly: A systematic review.. Seizure. 2025. doi:10.1016/j.seizure.2024.09.003

4. Wall J, Knight J, Emsley HCA. Late-onset epilepsy predicts stroke: Systematic review and meta-analysis.. Epilepsy & behavior : E&B. 2021. doi:10.1016/j.yebeh.2020.107634

5. Fisher RS, Acevedo CA, Arzimanoglou A, et al. ILAE Official Report: A practical clinical definition of epilepsy. Epilepsia. 2014. doi:10.1111/epi.12550

6. Musaeus CS, Nilsson C, Cooper C, et al. Pharmacological Medical Treatment of Epilepsy in Patients with Dementia: A Systematic Review.. Current Alzheimer research. 2021. doi:10.2174/1567205018666211126121529