Sedation mediates part of Citalopram's effect on agitation in Alzheimer's disease

Abstract

Background: We found a benefit of citalopram for agitation in the Citalopram for Agitation in Alzheimer's Disease study (CitAD), and wondered if this was mediated by a sedative effect. CitAD was a randomized, placebo-controlled, double-blind, parallel group trial conducted at 8 academic centers in the United States and Canada from August 2009 to January 2013. One hundred sixty-two participants with probable Alzheimer's disease (AD) and clinically significant agitation were analyzed in this study. Participants received a psychosocial intervention and were randomized to receive either citalopram or placebo (approximately half assigned to each group). Participants were rated on the Neurobehavioral Rating Scale Agitation subscale and measures of sedation (i.e., fatigue and somnolence).

Methods: Using the MacArthur Foundation procedures for documenting a mediator effect, we performed a secondary analysis examining whether sedation mediates the effect of treatment on agitation outcome.

Results: We found a statistically significant mediating effect of sedation on agitation outcomes, but the magnitude of the effect was small, only explaining 11% of the variance in agitation, with a significant, but modest effect size of 0.16 (95% CI: 0.08 to 0.22).

Conclusions: The benefit of citalopram was partly due to sedation but largely due to other mechanisms of action.

Keywords: Agitation; Citalopram; Sedation.

Figure 1

Mean Change in NBRS-A (top), Fatigue (middle), and Somnolence (bottom) across Time for Placebo and Citalopram Groups, with Standard Error bars. Dashed line for Citalopram (n=82), and solid line for Placebo (n=80).

Figure 2

Effect of Change in Sedation versus Change in Agitation Across Placebo and Citalopram Groups. Both Citalopram and Placebo groups show that large increases in sedation are associated with larger improvements is agitation as reflected by the two nearly parallel lines. The differences between the two treatment groups are that as a whole the Citalopram patients are more sedated than the Placebo patients as reflected by the Star’s position offset to the right. The vertical line with arrows represents the raw effect and also includes the mediation effect. The bold portion of line to the arrows represents the effect of treatment had there been no mediating effect of sedation change. Thus the dotted portion of the line represents the mediating effect of sedation change.