Evaluation of Antidepressant use in Patients with Heart Failure with Preserved Ejection Fraction Enrolled in the America's Subgroup of the TOPCAT trial

Abstract

Background Depression affects at least 20% of the HF population. Selective serotonin reuptake inhibitors (SSRI) are the primary treatment for depression. Although SSRIs do not improve symptoms of depression or affect cardiovascular outcomes in patients with HFrEF, their utility in patients with HFpEF is unknown. Objectives Conduct a post-hoc analysis to compare CV outcomes for HFpEF patients taking SSRIs compared to other antidepressants at baseline. Methods Patient level data from TOPCAT was obtained from the NHLBI via BioLINCC. We included patients enrolled in North or South America and stratified them according to baseline SSRI use. Patients in the SSRI non-user group were taking serotonin-norepinephrine reuptake inhibitors (SNRIs), tricyclic antidepressants (TCAs), atypical antidepressants (AA), and serotonin modulators (SM). Adjusted Cox-proportional models were constructed to estimate hazard ratios for CV outcomes between SSRI Users and Non-Users (covariates included allocation to spironolactone or placebo, age, sex, race, eGFR, LVEF, BNP tertile, DM, A.fib, and PAD). Results Of the 1767 patients enrolled in the America's, 368 patients (20.8%) were prescribed an antidepressant at baseline: 254 (69%) SSRIs, 55 (14.9%) SNRIs, 42 (11.4%) TCAs, 40 (10.9%) AAs, 37 (10.1%) SMs, and 60 (16.3%) patients were prescribed some combination. Patient baseline demographics were well matched for the sub-groups with no significant differences in age, gender, race category, NYHA classification, eGFR, BMI, LVEF, country of origin, or study reported comorbidities. Using cox-proportional models, SSRI use at baseline was associated with an increased risk of the composite endpoint of HF hospitalization, CV death and resuscitated cardiac arrest (adjusted HR, 2.14 [95% CI, 1.18 -3.87] P=0.011) compared to use of other antidepressants ( Figure 1 ). There was an increased risk of HF hospitalizations (adjusted HR, 2.37 [95% CI, 1.20 - 4.67] P=0.013) but no differences for CV death or all-cause death (adjusted HRs, 1.24 [95% CI, 0.64 - 2.37] and 1.15 [95% CI, 0.69 - 1.90], respectively). Conclusions Baseline SSRI use in a sub-group of patients prescribed antidepressants in the TOPCAT trial predicted a composite of CV outcomes and favored SSRI Non-Users. This difference between groups was driven by an increase in HF hospitalization for SSRI users. Further investigations are warranted into the optimal treatment of depression for patients with HFpEF.