TOPLINE:

In predominantly older patients with dementia and epilepsy, starting treatment with valproate as the first antiseizure medication (ASM) was associated with a higher risk for all‑cause death than starting treatment with carbamazepine, largely due to cardiovascular deaths; however, lamotrigine was associated with a lower risk.

METHODOLOGY:

• Researchers conducted a longitudinal cohort study to assess whether survival outcomes differ according to the first ASM prescribed in patients with both dementia and epilepsy.
• Using data from three Swedish health registers, they identified 5764 patients who initiated ASM treatment after a diagnosis of epilepsy and dementia (between 2006 and 2023); more than 86% of patients were older than 65 years, and 51% were women.
• Participants were categorised on the basis of the first ASM prescribed: those who received carbamazepine (n = 1578), valproate (n = 746), lamotrigine (n = 922), levetiracetam (n = 2098), and other ASMs (n = 420).
• All registry data were cross-referenced and anonymised by the respective register authorities prior to analysis.

TAKEAWAY:

• Carbamazepine was the most commonly initiated ASM before 2015, after which levetiracetam became the dominant first‑line option. Valproate was the second most frequently used ASM before 2008, but its usage declined subsequently.
• After adjustments, patients receiving valproate had a 34% higher risk for all-cause death than those receiving carbamazepine; however, those receiving lamotrigine had a 16% lower risk. These associations were most pronounced during the first 5 years of follow-up (P < .05 for all).
• Compared with carbamazepine, valproate was associated with a 30% higher risk for cardiovascular death (95% CI, 1.11-1.52), whereas lamotrigine was linked to a 21% lower risk (95% CI, 0.66-0.94).
• The risk for all-cause and cardiovascular deaths did not differ significantly between levetiracetam and carbamazepine.

IN PRACTICE:

"[The study] findings emphasize the importance of informed ASM selection in this vulnerable population and demonstrate the value of large-scale registry data in guiding safer therapeutic strategies," the authors wrote.

SOURCE:

This study was led by Johan Zelano, Institute of Neuroscience and Physiology, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden. It was published online on April 17, 2026, in Neurology.

LIMITATIONS:

The shorter follow ‑ up period and fewer observed deaths for lamotrigine and levetiracetam may have limited long-term comparison with older ASMs. The study included older patients, potentially limiting the generalisability of the findings. Additionally, registers lacked information on ethnicity or sex.

DISCLOSURES:

This study was funded by the Swedish Research Council, Swedish State through the ALF agreement, and Knut and Ragnvi Jacobsson Family Foundation. One author declared receiving speaker or advisory board honoraria and research grants from various pharmaceutical companies.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.