Antidepressant use in relation to dementia risk, cognitive decline, and brain atrophy

Abstract

Introduction: We aimed to assess the effect of antidepressant use on dementia risk, cognitive decline, and brain atrophy.

Methods: In this prospective cohort study, we included 5511 dementia-free participants (Mini-Mental State Examination [MMSE] > 25) of the Rotterdam study (57.5% women, mean age 70.6 years). Antidepressant use was extracted from pharmacy records from 1991 until baseline (2002-2008). Incident dementia was monitored from baseline until 2018, with repeated cognitive assessment and magnetic resonance imaging (MRI) every 4 years.

Results: Compared to never use, any antidepressant use was not associated with dementia risk (hazard ratio [HR] 1.14, 95% confidence interval [CI] 0.92-1.41), or with accelerated cognitive decline or atrophy of white and gray matter. Compared to never use, dementia risk was somewhat higher with tricyclic antidepressants (HR 1.36, 95% CI 1.01-1.83) than with selective serotonin reuptake inhibitors (HR 1.12, 95% CI 0.81-1.54), but without dose-response relationships, accelerated cognitive decline, or atrophy in either group.

Discussion: Antidepressant medication in adults without indication of cognitive impairment was not consistently associated with long-term adverse cognitive effects.

Highlights: Antidepressant medications are frequently prescribed, especially among older adults. In this study, antidepressant use was not associated with long-term dementia risk. Antidepressant use was not associated with cognitive decline or brain atrophy. Our results support safe prescription in an older, cognitively healthy population.

Keywords: MRI; antidepressant use; cognitive decline; depression; population‐based.

FIGURE 1

Antidepressant medication use and risk of all‐cause dementia. Estimates shown are from the fully adjusted Model 2, which is adjusted for age, sex, education, smoking status, alcohol use, body mass index, estimated glomerular filtration rate, Center for Epidemiologic Studies Depression scale score, benzodiazepine use, antipsychotic medication use, as well as prevalence of diabetes, hypertension, stroke, parkinsonism, atrial fibrillation, congestive heart failure, coronary heart disease, cancer, and chronic obstructive pulmonary disease. The median daily dose in tricyclic antidepressant users was 39.6 DDD, and in selective serotonin reuptake inhibitor users 315.0 DDD. N_dem , number of dementia cases; N_total, total number of participants in group; CI, confidence interval; DDD, defined daily dose.

FIGURE 2

Antidepressant medication use and change in global cognition (g‐factor) over time. Trajectories show change in global cognition per year for ever users and never users. Global cognition is defined as the standardized compound score (g‐factor) using the first factor of a principal component analysis, including the letter‐digit substitution task, verbal fluency test, Stroop test, 15‐word learning test, and Purdue Pegboard Test. For graphical representation, trajectories are depicted for mean age, sex, education, and cohort; corresponding effect estimates in the main text refer to the fully adjusted models.

FIGURE 3

Antidepressant medication use and change in brain volumes in milliliters (mL) on MRI. Trajectories show change in mL brain volume per year for ever users and never users. For graphical representation, trajectories are depicted for mean age, sex, education, and cohort. Corresponding effect estimates in the main text refer to the fully adjusted models.