- Email: [email protected]
COMMON ANTIDEPRESSANT MEDICATIONS AND RISK OF DEMENTIA: A PROSPECTIVE COHORT STUDY
Podium Presentations: Wednesday, July 27, 2016 DT-01-03
COMMON ANTIDEPRESSANT MEDICATIONS AND RISK OF DEMENTIA: A PROSPECTIVE COHORT STUDY
Laura M. Heath1, Shelly L. Gray1, Denise Boudreau2, Karen L. Edwards3, Stephanie M. Fullerton1, Ken Thummel1, Eric B. Larson2, 1University of Washington, Seattle, WA, USA; 2Group Health Research Institute, Seattle, WA, USA; 3University of California, Irvine, Irvine, CA, USA. Contact e-mail: [email protected] Background: The purpose of this study was to determine if antidepressant type was associated with dementia risk. Effects may differ by antidepressant class, as some have known anticholinergic activity (tricyclic antidepressants (TCAs) and paroxetine), which may increase risk, while long-term effects of selective serotonin reuptake inhibitors (SSRIs) are unknown. Methods: Adult Changes in Thought (ACT) is a prospective population-based cohort in an integrated health care delivery system. Initial recruitment occurred from 1994-1996 and 2000-2003, followed by continuous enrollment since 2004. For these analyses eligible participants had at least 10 years of enrollment in the health care system at baseline (N¼3342, data through December 31, 2013). Primary exposures were SSRIs (paroxetine vs. others) and TCAs. All other antidepressant classes were grouped together and adjusted for in the models. Computerized pharmacy dispensing data were used to ascertain cumulative medication exposure, defined as total number of standardized daily doses (TSDDs) dispensed over a 10-year period (a rolling window from ACT entry and moving forward in time throughout follow-up). The most recent year was excluded to avoid use
Table 1 Risk of dementia according to Antidepressant 10-year cumulative total standardized daily dose (TSDDs) TSDDs SSRIs Paroxetine 0 1-90 91-365 366-1095 1095+ Other SSRIs 0 1-90 91-365 366-1095 1095+ TCAs 0 1-90 91-365 366-1095 1095+
Adjusted HR* (95% CI)
1 1.68 (1.20, 2.37) 1.46 (0.95, 2.24) 2.06 (1.30, 3.28) 1.49 (0.89, 2.49) 1 0.81 (0.56, 1.16) 1.01 (0.72, 1.42) 0.88 (0.56, 1.40) 1.07 (0.78, 1.47) 1 0.93 (0.72, 1.20) 1.09 (0.84, 1.41) 1.13 (0.81, 1.56) 1.16 (0.90, 1.51)
*Adjusted for all other time-varying antidepressant class use (all antidepressant classes analyzed simultaneously), age (via the time-axis), cohort, gender, depression (time-varying), race, education (some college vs none), and baseline variables including: age, dichotomized CASI score, comorbid vascular disease (including cardiovascular disease, hypertension, and diabetes), body mass index category, history of cigarette smoking, self-rated health (fair/poor vs better), and regular exercise (15 minutes at least 3 times per week). Individuals with bipolar disorder excluded.
P1135
related to prodromal symptoms. Dementia risk was analyzed using Cox proportional hazards models adjusted for age via the time axis, and for demographic characteristics, health behaviors, and health status. Results: During a mean follow-up of 7.9 years, 864 participants (25.9%) developed dementia (699 Alzheimer’s disease). An antidepressant was used by 48.6% during the study period. Hazard ratios (HR) and 95% confidence intervals for dementia risk are summarized in Table 1. Notably, paroxetine was associated with increased risk of dementia at all TSDD categories (HRs 1.46-2.06), although the confidence interval crossed 1 for some categories. Other SSRIs and TCAs were not significantly associated with increased risk of dementia. Results were similar for Alzheimer’s disease. Conclusions: Other SSRI and TCA use were not associated with dementia risk. However, paroxetine use was associated with higher risk of dementia and Alzheimer’s disease, even in participants with 90 or fewer TSDDs. Although this association may not be causal due to confounding by indications associated with premorbid symptoms, it is plausible given that other anticholinergic medications have been associated with dementia risk.
DT-01-04
EFFECTS OF FORTASYN CONNECT (SOUVENAID) ON LONGITUDINAL BRAIN ATROPHY MEASURES IN PRODROMAL ALZHEIMER’S DISEASE: RESULTS OF THE DOUBLE-BLIND RANDOMISED CONTROLLED LIPIDIDIET TRIAL
Pieter Jelle Visser1,2, Hilkka Soininen3, Mara ten Kate2, Miia Kivipelto3,4, Tobias Hartmann5,6, the LipiDiDiet Study Group, 1School for Mental Health and Neuroscience, Maastricht University, Maastricht, Netherlands; 2 Alzheimer Centre, VU University Medical Centre, Amsterdam, Netherlands; 3University of Eastern Finland and Kuopio University Hospital, Kuopio, Finland; 4Karolinska Institutet and Stockholm Gerontology Research Center, Stockholm, Sweden; 5Deutsches Institut f€ur Demenz Pr€avention (DIDP), Saarland University, Homburg, Germany; 6 Department of Experimental Neurology, Saarland University, Homburg, Germany. Contact e-mail: [email protected] Background: Synaptic and neuronal loss contribute to atrophy on structural Magnetic Resonance Imaging (MRI) and are all correlated with cognitive impairment observed already early in Alzheimer’s disease (AD) (Jack & Holtzman, 2013; Terry et al., 1991). Results from mechanistic and clinical studies suggested an augmented therapeutic potential through intake of specific nutrients in early, especially prodromal, AD. Here a multi-nutrient combination ‘Fortasyn Connect’, as present in Souvenaid, was used for the long-term randomised controlled LipiDiDiet clinical trial on prodromal AD1. This nutrient combination has been designed to target several pathologies linked to AD (Van Wijk et al., 2014). In mild AD dementia patients, it previously showed beneficial effects on functional brain network organisation (De Waal et al., 2014) and episodic memory (Scheltens et al., 2010, 2012). Methods: The LipiDiDiet clinical trial (NTR1705) is a 6-year, 1:1 randomised, double-blind, parallel-group, multi-centre, multi-country study in 311 subjects with prodromal AD (criteria Dubois et al., 2007). Subjects receive active (Fortasyn Connect in Souvenaid) or iso-caloric control product once daily. Main outcomes include brain volumes based on MRI (3D T1-weighted anatomical scans; hippocampal, whole-brain and ventricular volumes), assessed annually and analysed centrally using MAPS-HBSI for hippocampal volume, BSI for whole-brain volume and longitudinal Freesurfer for ventricular
COMMON ANTIDEPRESSANT MEDICATIONS AND RISK OF DEMENTIA: A PROSPECTIVE COHORT STUDY
Laura M. Heath1, Shelly L. Gray1, Denise Boudreau2, Karen L. Edwards3, Stephanie M. Fullerton1, Ken Thummel1, Eric B. Larson2, 1University of Washington, Seattle, WA, USA; 2Group Health Research Institute, Seattle, WA, USA; 3University of California, Irvine, Irvine, CA, USA. Contact e-mail: [email protected] Background: The purpose of this study was to determine if antidepressant type was associated with dementia risk. Effects may differ by antidepressant class, as some have known anticholinergic activity (tricyclic antidepressants (TCAs) and paroxetine), which may increase risk, while long-term effects of selective serotonin reuptake inhibitors (SSRIs) are unknown. Methods: Adult Changes in Thought (ACT) is a prospective population-based cohort in an integrated health care delivery system. Initial recruitment occurred from 1994-1996 and 2000-2003, followed by continuous enrollment since 2004. For these analyses eligible participants had at least 10 years of enrollment in the health care system at baseline (N¼3342, data through December 31, 2013). Primary exposures were SSRIs (paroxetine vs. others) and TCAs. All other antidepressant classes were grouped together and adjusted for in the models. Computerized pharmacy dispensing data were used to ascertain cumulative medication exposure, defined as total number of standardized daily doses (TSDDs) dispensed over a 10-year period (a rolling window from ACT entry and moving forward in time throughout follow-up). The most recent year was excluded to avoid use
Table 1 Risk of dementia according to Antidepressant 10-year cumulative total standardized daily dose (TSDDs) TSDDs SSRIs Paroxetine 0 1-90 91-365 366-1095 1095+ Other SSRIs 0 1-90 91-365 366-1095 1095+ TCAs 0 1-90 91-365 366-1095 1095+
Adjusted HR* (95% CI)
1 1.68 (1.20, 2.37) 1.46 (0.95, 2.24) 2.06 (1.30, 3.28) 1.49 (0.89, 2.49) 1 0.81 (0.56, 1.16) 1.01 (0.72, 1.42) 0.88 (0.56, 1.40) 1.07 (0.78, 1.47) 1 0.93 (0.72, 1.20) 1.09 (0.84, 1.41) 1.13 (0.81, 1.56) 1.16 (0.90, 1.51)
*Adjusted for all other time-varying antidepressant class use (all antidepressant classes analyzed simultaneously), age (via the time-axis), cohort, gender, depression (time-varying), race, education (some college vs none), and baseline variables including: age, dichotomized CASI score, comorbid vascular disease (including cardiovascular disease, hypertension, and diabetes), body mass index category, history of cigarette smoking, self-rated health (fair/poor vs better), and regular exercise (15 minutes at least 3 times per week). Individuals with bipolar disorder excluded.
P1135
related to prodromal symptoms. Dementia risk was analyzed using Cox proportional hazards models adjusted for age via the time axis, and for demographic characteristics, health behaviors, and health status. Results: During a mean follow-up of 7.9 years, 864 participants (25.9%) developed dementia (699 Alzheimer’s disease). An antidepressant was used by 48.6% during the study period. Hazard ratios (HR) and 95% confidence intervals for dementia risk are summarized in Table 1. Notably, paroxetine was associated with increased risk of dementia at all TSDD categories (HRs 1.46-2.06), although the confidence interval crossed 1 for some categories. Other SSRIs and TCAs were not significantly associated with increased risk of dementia. Results were similar for Alzheimer’s disease. Conclusions: Other SSRI and TCA use were not associated with dementia risk. However, paroxetine use was associated with higher risk of dementia and Alzheimer’s disease, even in participants with 90 or fewer TSDDs. Although this association may not be causal due to confounding by indications associated with premorbid symptoms, it is plausible given that other anticholinergic medications have been associated with dementia risk.
DT-01-04
EFFECTS OF FORTASYN CONNECT (SOUVENAID) ON LONGITUDINAL BRAIN ATROPHY MEASURES IN PRODROMAL ALZHEIMER’S DISEASE: RESULTS OF THE DOUBLE-BLIND RANDOMISED CONTROLLED LIPIDIDIET TRIAL
Pieter Jelle Visser1,2, Hilkka Soininen3, Mara ten Kate2, Miia Kivipelto3,4, Tobias Hartmann5,6, the LipiDiDiet Study Group, 1School for Mental Health and Neuroscience, Maastricht University, Maastricht, Netherlands; 2 Alzheimer Centre, VU University Medical Centre, Amsterdam, Netherlands; 3University of Eastern Finland and Kuopio University Hospital, Kuopio, Finland; 4Karolinska Institutet and Stockholm Gerontology Research Center, Stockholm, Sweden; 5Deutsches Institut f€ur Demenz Pr€avention (DIDP), Saarland University, Homburg, Germany; 6 Department of Experimental Neurology, Saarland University, Homburg, Germany. Contact e-mail: [email protected] Background: Synaptic and neuronal loss contribute to atrophy on structural Magnetic Resonance Imaging (MRI) and are all correlated with cognitive impairment observed already early in Alzheimer’s disease (AD) (Jack & Holtzman, 2013; Terry et al., 1991). Results from mechanistic and clinical studies suggested an augmented therapeutic potential through intake of specific nutrients in early, especially prodromal, AD. Here a multi-nutrient combination ‘Fortasyn Connect’, as present in Souvenaid, was used for the long-term randomised controlled LipiDiDiet clinical trial on prodromal AD1. This nutrient combination has been designed to target several pathologies linked to AD (Van Wijk et al., 2014). In mild AD dementia patients, it previously showed beneficial effects on functional brain network organisation (De Waal et al., 2014) and episodic memory (Scheltens et al., 2010, 2012). Methods: The LipiDiDiet clinical trial (NTR1705) is a 6-year, 1:1 randomised, double-blind, parallel-group, multi-centre, multi-country study in 311 subjects with prodromal AD (criteria Dubois et al., 2007). Subjects receive active (Fortasyn Connect in Souvenaid) or iso-caloric control product once daily. Main outcomes include brain volumes based on MRI (3D T1-weighted anatomical scans; hippocampal, whole-brain and ventricular volumes), assessed annually and analysed centrally using MAPS-HBSI for hippocampal volume, BSI for whole-brain volume and longitudinal Freesurfer for ventricular