Anderson Matheus Pereira da Silva, Gabriel Caruso Novaes Tudella, Alice Mi Lee, Diogo Haddad Santos, Cristine Mella Cukiert, George Perry
Epilepsy was associated with higher subsequent recording of dementia and mortality over 2 years. No biomarker, neuroimaging, or neuropathological measure was available; the estimates describe the order in which diagnoses were documented and do not establish that epilepsy marks the underlying neurodegenerative process. Differential ascertainment and residual confounding remain plausible contributors.
PURPOSE: Epilepsy and early-onset Alzheimer disease (AD) are both common in adults with Down syndrome (DS), and their temporal relationship remains unclear. We examined whether a recorded diagnosis of epilepsy was associated with higher subsequent recording of all-cause dementia, AD, non-Alzheimer dementia, and mortality among adults with DS aged 40 years or older.
METHODS: Retrospective cohort study using deidentified electronic health records from the TriNetX Global Collaborative Network. Adults with DS (ICD-10-CM Q90; age ≥40 years; 2010-2025) were classified as having epilepsy (G40 recorded before any dementia code) or not; secondary seizures were excluded. Cohorts were balanced by 1:1 propensity score matching. Outcomes were assessed from 90 to 730 days after the index date using hazard ratios (HRs) and Kaplan-Meier analyses.
RESULTS: After matching, 1812 patients per cohort were retained (mean [SD] age, 54.3 [7.0] vs 54.2 [7.4] years). Epilepsy was associated with higher subsequent recording of all-cause dementia (12.9%vs 6.5%; HR, 2.03 [95% CI, 1.59-2.60]), AD (6.1%vs 3.2%; HR, 1.97 [1.41-2.76]), non-Alzheimer dementia (12.9%vs 6.3%; HR, 2.08 [1.63-2.65]), and mortality (12.8%vs 5.1%; HR, 2.55 [2.00-3.24]). Estimates were similar by age and sex, and when exposure required an antiseizure medication prescription alongside the diagnostic code.
CONCLUSION: Epilepsy was associated with higher subsequent recording of dementia and mortality over 2 years. No biomarker, neuroimaging, or neuropathological measure was available; the estimates describe the order in which diagnoses were documented and do not establish that epilepsy marks the underlying neurodegenerative process. Differential ascertainment and residual confounding remain plausible contributors.