Abstract
Depression is associated with an increased risk of death in patients with heart failure (HF); however, the association between the use of antidepressants and HF prognoses remains controversial. Therefore, this meta-analysis aimed to evaluate the effect of antidepressants on the risk of death in HF patients. We retrieved data from the PubMed and EMBASE databases until August 2019 for studies reporting the use of antidepressants in HF patients. Data were extracted from the eligible articles, and a random effects model was used to pool the effect estimates (risk ratios (RRs) and 95% confidence intervals (CIs)). A total of 8 studies were included in this meta-analysis. Overall, the use of antidepressants was associated with increased risks of all-cause death (RR = 1.27; 95% CI, 1.21–1.34) and cardiovascular death (RR = 1.14; 95% CI, 1.08–1.20) in HF patients with or without depression. Specifically, HF patients with depression taking antidepressants had increased risks of all-cause death (RR = 1.21; 95% CI, 1.16–1.27) and cardiovascular death (RR = 1.21; 95% CI, 1.13–1.30). Compared with nonusers, the use of selective serotonin reuptake inhibitors (SSRIs), tricyclics (TCAs), and selective serotonin reuptake inhibitors (SNRIs) significantly increased the rate of all-cause death (SSRIs (RR = 1.26; 95% CI, 1.19–1.32), TCAs (RR = 1.30; 95% CI, 1.16–1.46), and SNRIs (RR = 1.17; 95% CI, 1.08–1.26)) but not cardiovascular death (SSRIs (RR = 1.03; 95% CI, 0.84–1.26), TCAs (RR = 1.02; 95% CI, 0.86–1.21), and SNRIs (RR = 0.92; 95% CI, 0.48–1.78)). Based on current publications, the use of antidepressants could increase the risk of all-cause death in HF patients, regardless of whether they have depression or the type of antidepressants they use. Further study is needed to determine the relationship between antidepressant use and cardiovascular death.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
Introduction
Heart failure (HF) is associated with poor quality of life and frequent hospitalizations. Because of the increased risk of HF-related death, HF represents a heavy burden on both family and social health costs. Psychological disorders are prevalent in patients with HF, and these illnesses could lead to poor outcomes in HF [1, 2]. Prior meta-analyses have reported a 21.5% prevalence of depression among patients with HF [3]; further studies have demonstrated that depression in HF is associated with poorer prognoses and increased risk of death [3,4,5], particularly in elderly patients (age > 65 years) [4]. As such, the prescription of antidepressants is increasing in clinical practice [6].
Antidepressants are categorized as selective serotonin reuptake inhibitors (SSRIs), tricyclics (TCAs), and others (tetracyclics, noradrenergic and specific serotonergic antidepressants (NaSSAs), selective serotonin reuptake inhibitors (SNRIs)). SSRIs and TCAs are the two major classes prescribed for cardiac patients with depression. However, the effect of antidepressant use on the risk of death in HF patients remains controversial. In patients with end-stage HF and depression, taking medications of β-blockers combined with SSRIs is associated with a reduced risk of cardiovascular death compared with the use of β-blockers without SSRIs [7]. Chung et al. [8] found that antidepressant use was not a predictor of cardiovascular hospital admission or all-cause death in HF patients when depressive symptoms remained. In contrast, several studies reported that antidepressants did not significantly reduce the risks of death or hospitalization in HF patients with or without depression [9,10,11]. Moreover, Brouwers et al. [12] and Fosb et al. [13] have proposed increased risks of death associated with antidepressants used in HF patients. Therefore, the role of antidepressants in HF patients should be further evaluated to guide treatment in clinical practice. We conducted a comprehensive meta-analysis to evaluate the effects of antidepressants on the risk of death in HF patients.
Methods
We performed this meta-analysis based on the protocol and reporting of the results from the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) statement [14].
Study search strategy
The PubMed and EMBASE computer-based databases were systematically searched from inception to August 2019 for studies that evaluated the use of antidepressants in HF patients. The following keywords were used and limited to retrieval in the titles and abstracts: heart failure AND (psychological factors OR psychological distress OR emotional distress OR depression OR depressive symptom OR antidepressant) AND (mortality OR death). We did not apply the language restrictions in the search. Non-English studies were translated into English by an author (JY-M) using Google’s automatic-translation software. The literature search strategy is presented in Supplemental Table 1. The reference lists of all relevant studies were also searched to identify additional reports.
Eligibility criteria
Studies were included if they met the following criteria: (1) studies that reported antidepressant use and the consequent risk for all-cause death and cardiovascular death in HF patients; (2) randomized controlled trials (RCTs) or observational studies with comparison and control groups; and (3) studies that reported the unadjusted or adjusted risk ratios (RRs) with 95% confidence intervals (CIs). Reviews, letters to the editor, case reports, comments, editorials, and meeting abstracts were excluded. In the case of multiple publications analyzing the same population, the study with the longest follow-up time or with the largest sample size was selected.
Study selection and data extraction
According to the predefined criteria, study selection and data extraction were performed by two interdependent authors (WF-H and YH-F). All discrepancies were resolved through discussion. The first phase of study selection was based on the titles and abstracts. In the second phase, the full texts were assessed to detect whether the study met the inclusion criteria. For each study, the following information was recorded: the first author, publication year, geographic location, participant information (sex, age, and sample size), categories of antidepressants, follow-up time, and RRs with 95% CIs. When both unadjusted and adjusted RRs were presented in one study, the maximally adjusted RR value was extracted.
Quality assessment
The methodological quality of the RCTs was assessed independently by two authors (WF-H and YH-F) using the Cochrane Risk of Bias Tool [15]. We assessed the quality of observational studies using the Newcastle-Ottawa score [16, 17].
Statistical analysis
The RRs with 95% CIs were regarded as the common risk estimates. The consistency test was evaluated using Cochran’s Q test complemented with the I2 statistic, where P < 0.1 or I2 > 50% indicated high heterogeneity. The effect measures were transformed to their natural logarithms (logRR), and the standard error (SElogRR) was calculated from the corresponding CIs. An inverse-variance weighted random effects model was used to pool these natural logarithms. In the sensitivity analysis, a fixed effects model was used to examine the robustness of the results. We performed a subgroup analysis based on the types of antidepressants (SSRIs, TCAs, and SNRIs). The publication bias was assessed using a funnel plot. All statistical analyses were performed using Review Manager version 5.3 software (Cochrane Collaboration 2014, Nordic Cochrane Center Copenhagen, Denmark).
Results
Study selection
The flowchart of electronic retrievals is shown in Supplemental Figure 1. Ultimately, a total of 8 studies (2 RCTs [9, 10] and 6 observational studies [11,12,13, 18,19,20]) published between 2007 and 2016 were included in this meta-analysis. The basic characteristics of the selected studies are shown in Table 1. Overall, 5 studies were from Europe, and 3 studies were from North America. The sample size of these studies ranged from 372 to 99,335, with a total of 125,763 participants. The mean age of patients ranged from 56.8 to 73.5 years. The follow-up duration varied from 1.4 to 5.0 years. The methodologic quality of all the included studies was acceptable.
Relationship between antidepressant use and death
A total of 8 and 5 studies reported the outcomes of all-cause death and cardiovascular death, respectively. As shown in Fig. 1, the use of antidepressants was associated with the risks of all-cause death (RR = 1.27; 95% CI, 1.21–1.34, P < 0.00001; I2 = 64%) and cardiovascular death (RR = 1.14, 95% CI, 1.08–1.20, P < 0.00001; I2 = 32%) in patients with HF with or without depression.
Sensitivity analysis
In the sensitivity analysis, the aforementioned results remained stable after omitting 1 study at a time. In addition, only two RCTs focused on HF patients with depression [9, 10]. Consistent with the results of the main analyses, HF patients with depression taking antidepressants had increased risks of all-cause death (RR = 1.21; 95% CI, 1.16–1.27; P < 0.00001; I2 = 0%) and cardiovascular death (RR = 1.21; 95% CI, 1.13–1.30; P < 0.00001; I2 = 0%) compared with nonusers (Fig. 2). The results also did not change when we reperformed these analyses with fixed effects models. In addition, the included study of Veien et al. [19] specifically assessed the association between depression and mortality risk, and pharmacologically treated depression was defined as treatment with at least one antidepressant. We performed the sensitivity analysis and excluded this study, and the corresponding results did not change.
Subgroup analysis
The subgroup analysis was performed based on the types of antidepressants (SSRIs, TCAs, SNRIs). As a result, compared with the corresponding nonusers, the use of SSRIs, TCAs, and SNRIs significantly increased the rates of all-cause death (SSRIs (RR = 1.26; 95% CI, 1.19–1.32; P < 0.00001), TCAs (RR = 1.30; 95% CI, 1.16–1.46; P < 0.00001), and SNRIs (RR = 1.17; 95% CI, 1.08–1.26; P < 0.0001); Fig. 3) but not cardiovascular death (SSRIs (RR = 1.03; 95% CI, 0.84–1.26; P = 0.77), TCAs (RR = 1.02; 95% CI, 0.86–1.21; P = 0.85), and SNRIs (RR = 0.92; 95% CI, 0.48–1.78; P = 0.81; Fig. 4).
Publication bias
As shown in Fig. 5, no obvious publication bias was observed by inspecting the funnel plot.
Discussion
To the best of our knowledge, we conducted the first meta-analysis to assess the relationship between antidepressant use and death in HF patients. With the use of data from 8 studies, our current meta-analysis suggested the following: (1) the use of antidepressants (SSRIs, TCAs, SNRIs) was associated with an increased risk of all-cause death in HF patients, regardless of whether the patients had clinical depression; and (2) there was a relatively weaker association (14% increased risk) between antidepressant use and cardiovascular death. However, the use of SSRIs, TCAs, or SNRIs was not associated with an increased rate of cardiovascular death when these drugs were analyzed separately.
Our current data found a significantly elevated risk of all-cause death in HF patients treated with antidepressants. The reasons for this finding are not yet clear, but different hypotheses have been proposed. In patients with HF, the use of antidepressants appears to improve depressive symptoms based on the validated measuring method of depression [21,22,23,24]. However, HF patients taking antidepressants still show high rates of at least mild-to-moderate depressive symptoms [25]. As such, antidepressants might not be capable of sufficiently alleviating depressive symptoms to reduce the risk of death. A prior meta-analysis indicated an increased risk of death in HF patients with major depression but not mild depression [26]. This finding might suggest that the use of antidepressants would not reduce the risk of death in these patients with mild depression. Moreover, drug interactions could increase the risk of death by aggravating the side effects of drugs. For example, a case of bradycardia after the co-administration of paroxetine and metoprolol has been reported before [27]. In addition, β-blockers that are associated with reduced death risk are administered less frequently in HF patients with depression, contributing to the decreased survival rate [28]. Finally, some included studies might have patient selection bias, indicating that HF patients taking antidepressants actually suffered from more serious depression symptoms than nonusers did.
SSRIs were considered the standard pharmacological treatment of depression through the selective inhibition of the reuptake of 5-hydroxytryptamine (5-HT) by the presynaptic membrane in the central nervous system to increase the concentration of 5-HT in the synaptic cleft. Adverse cardiovascular responses during the administration of SSRIs are rare because of their high selectivity for therapeutic targets. TCAs can block the reuptake of noradrenaline and 5-HT by the presynaptic membranes of nerve endings and increase the concentration of monoamines at the synaptic cleft. In addition, TCAs can also retard the acetylcholine receptor, histamine H1 receptor, and noradrenalin ɑ1 receptor, which can trigger adverse nervous system reactions, such as excessive sedation, delirium, extrapyramidal symptoms, and epileptic seizures, as well as an increased heart rate and postural hypotension. Current guidelines on the pharmacological management of depression in patients with cardiac disease also propose that SSRIs are relatively safer than TCAs [29]. TCAs could promote arrhythmias because of increased heart rate and prolonged QT intervals. These pharmacological characteristics could potentially increase the risk of cardiovascular death. However, interestingly, although a 14% increased risk of cardiovascular death was found in HF patients taking any antidepressants, SSRIs, TCAs, and SNRIs did not appear to increase the risk of cardiovascular death when they were analyzed separately. The overall increased risk of cardiovascular death in HF patients might be explained by other antidepressants, such as tetracyclics and NaSSAs that were potentially linked to a higher risk of cardiovascular death [12]. Our data in relation to the association between antidepressant use and cardiovascular death should be interpreted cautiously. Further study should confirm these findings in HF patients.
Since depression could increase the risk of death among HF patients, HF patients with depression symptoms are treated with antidepressants more frequently, even if the patients have no clinical depression. Consequently, antidepressants have been increasingly used by patients in recent years to alleviate depression symptoms. Our findings suggest that HF patients should be cautious when antidepressants are administered because of the increased risk of death. Antidepressant treatments for HF patients should be based on individual patient-level therapeutic strategies, such as the degree of patients’ depression, HF severity, and general conditions.
Several limitations should be considered in this meta-analysis. First, residual confounders might exist given the nature of the observational data used in this study. Second, a higher prevalence of antidepressant use was observed in female patients with HF than in male patients, which raised the possibility that antidepressants might function differently depending on sex. However, the subgroup analysis based on sex could not be performed due to the limited data. Third, the number of included studies in some comparisons was relatively small, thus limiting the validity of these findings. Fourth, the major objective was pooling RCTs and observational studies; however, the results of our sensitivity analysis based on RCTs were consistent with the results of the main analyses. Finally, the methods of clinically validated assessments of depression differed across studies, and the heterogeneous populations might limit the generalizability of the findings to all HF patients in clinical practice.
Conclusions
Based on current publications, the use of antidepressants was associated with an increased risk of all-cause death in HF patients, regardless of whether they have clinical depression or the type of antidepressants they use. Further study should confirm the association between antidepressant use and cardiovascular death in HF patients.
References
Celano CM, Villegas AC, Albanese AM, Gaggin HK, Huffman JC (2018) Depression and anxiety in heart failure: a review. Harv Rev Psychiatry 26(4):175–184
Lin TK, Hsu BC, Li YD, Chen CH, Lin JW, Chien CY, Weng CY (2019) Prognostic value of anxiety between heart failure with reduced ejection fraction and heart failure with preserved ejection fraction. J Am Heart Assoc 8(12)
Rutledge T, Reis VA, Linke SE, Greenberg BH, Mills PJ (2006) Depression in heart failure a meta-analytic review of prevalence, intervention effects, and associations with clinical outcomes. J Am Coll Cardiol 48(8):1527–1537
Gathright EC, Goldstein CM, Josephson RA, Hughes JW (2017) Depression increases the risk of mortality in patients with heart failure: a meta-analysis. J Psychosom Res 94:82–89
Sokoreli I, de Vries J, Pauws SC, Steyerberg EW (2016) Depression and anxiety as predictors of mortality among heart failure patients: systematic review and meta-analysis. Heart Fail Rev 21(1):49–63
Jha MK, Qamar A, Vaduganathan M, Charney DS, Murrough JW (2019) Screening and management of depression in patients with cardiovascular disease. J Am Coll Cardiol 73(14):1827–1845
Tousoulis D, Antoniades C, Drolias A, Stefanadi E, Marinou K, Vasiliadou C, Tsioufis C, Toutouzas K, Latsios G, Stefanadis C (2008) Selective serotonin reuptake inhibitors modify the effect of β-blockers on long-term survival of patients with end-stage heart failure and major depression. J Card Fail 14(6):456–464
Chung ML, Dekker RL, Lennie TA, Moser DK (2013) Antidepressants do not improve event-free survival in patients with heart failure when depressive symptoms remain. Heart Lung 42(2):85–91
Angermann CE, Gelbrich G, Störk S, Gunold H, Edelmann F, Wachter R, Schunkert H, Graf T, Kindermann I, Haass M, Blankenberg S, Pankuweit S, Prettin C, Gottwik M, Böhm M, Faller H, Deckert J, Ertl G (2016) Effect of escitalopram on all-cause mortality and hospitalization in patients with heart failure and depression. JAMA 315(24):2683–2693
O’Connor CM, Jiang W, Kuchibhatla M, Silva SG, Cuffe MS, Callwood DD, Zakhary B, Stough WG, Arias RM, Rivelli SK, Krishnan R (2010) Safety and efficacy of sertraline for depression in patients with heart failure. J Am Coll Cardiol 56(9):692–699
O’Connor CM, Jiang W, Kuchibhatla M, Mehta RH, Clary GL, Cuffe MS, Christopher EJ, Alexander JD, Califf RM, Krishnan RR (2008) Antidepressant use, depression, and survival in patients with heart failure. Arch Intern Med 168(20):2232
Brouwers C, Christensen SB, Damen NL, Denollet J, Torp-Pedersen C, Gislason GH, Pedersen SS (2016) Antidepressant use and risk for mortality in 121,252 heart failure patients with or without a diagnosis of clinical depression. Int J Cardiol 203:867–873
EL Fosb L, Gislason GH, Poulsen HE, Hansen ML, Folke F, Schramm TK, Olesen JB, Bretler D, Abildstrøm SZ, Sørensen R, Hvelplund A, Køber L, Torp-Pedersen C (2009) Prognosis in heart failure and the value of β-blockers are altered by the use of antidepressants and depend on the type of antidepressants used. Circ Heart Fail 2(6):582–590
Moher D, Liberati A, Tetzlaff J, Altman DG (2009) Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 6(7):e1000097
Higgins JP, Altman DG, Gotzsche PC, Juni P, Moher D, Oxman AD, Savovic J, Schulz KF, Weeks L, Sterne JA (2011) The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ 343:d5928
Wells GA, Shea B, O’Connell D, Peterson J, Welch V, Losos M, Tugwell P (2014) The Newcastle Ottawa Scale (NOS) for assessing the quality if nonrandomized studies in meta-analyses. Available at: http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp. Accessed 1 August 2019
Zhu W, Wan R, Liu F, Hu J, Huang L, Li J, Hong K (2016) Relation of body mass index with adverse outcomes among patients with atrial fibrillation: a meta-analysis and systematic review. J Am Heart Assoc 5(9):e4006
Diez-Quevedo C, Lupón J, González B, Urrutia A, Cano L, Cabanes R, Altimir S, Coll R, Pascual T, de Antonio M, Bayes-Genis A (2013) Depression, antidepressants, and long-term mortality in heart failure. Int J Cardiol 167(4):1217–1225
Veien KT, Videbæk L, Schou M, Gustafsson F, Hald-Steffensen F, Hildebrandt PR (2011) High mortality among heart failure patients treated with antidepressants. Int J Cardiol 146(1):64–67
Sherwood A, Blumenthal JA, Trivedi R, Johnson KS, O’Connor CM, Adams KF, Dupree CS, Waugh RA, Bensimhon DR, Gaulden L, Christenson RH, Koch GG, Hinderliter AL (2007) Relationship of depression to death or hospitalization in patients with heart failure. Arch Intern Med 167(4):367–373
Samartzis L, Dimopoulos S, Tziongourou M, Koroboki E, Kyprianou T, Nanas S (2013) SSRIs versus exercise training for depression in chronic heart failure: a meta-analysis of randomized controlled trials. Int J Cardiol 168(5):4956–4958
Lekakis J, Ikonomidis I, Papoutsi Z, Moutsatsou P, Nikolaou M, Parissis J, Kremastinos DT (2010) Selective serotonin re-uptake inhibitors decrease the cytokine-induced endothelial adhesion molecule expression, the endothelial adhesiveness to monocytes and the circulating levels of vascular adhesion molecules. Int J Cardiol 139(2):150–158
Fraguas R, Da Silva Telles RM, Alves TCTF, Andrei AM, Rays J, Iosifescu DV, Wajngarten M (2009) A double-blind, placebo-controlled treatment trial of citalopram for major depressive disorder in older patients with heart failure: the relevance of the placebo effect and psychological symptoms. Contemp Clin Trials 30(3):205–211
Gottlieb SS, Kop WJ, Thomas SA, Katzen S, Vesely MR, Greenberg N, Marshall J, Cines M, Minshall S (2007) A double-blind placebo-controlled pilot study of controlled-release paroxetine on depression and quality of life in chronic heart failure. Am Heart J 153(5):868–873
Jimenez JA, Redwine LL, Rutledge TR, Dimsdale JE, Pung MA, Ziegler MG, Greenberg BH, Mills PJ (2012) Depression ratings and antidepressant use among outpatient heart failure patients: implications for the screening and treatment of depression. Int J Psychiatry Med 44(4):315–334
Fan H, Yu W, Zhang Q, Cao H, Li J, Wang J, Shao Y, Hu X (2014) Depression after heart failure and risk of cardiovascular and all-cause mortality: a meta-analysis. Prev Med 63:36–42
Konig F, Hafele M, Hauger B, Loble M, Wossner S, Wolfersdorf M (1996) Bradycardia after beginning therapy with metoprolol and paroxetine. Psychiatr Prax 23(5):244–245
Kim C, Duan L, Phan DQ, Lee M (2019) Frequency of utilization of beta blockers in patients with heart failure and depression and their effect on mortality. Am J Cardiol 124(5), 746–750
Davidson KW, Kupfer DJ, Bigger JT, Califf RM, Carney RM, Coyne JC, Czajkowski SM, Frank E, Frasure-Smith N, Freedland KE, Froelicher ES, Glassman AH, Katon WJ, Kaufmann PG, Kessler RC, Kraemer HC, Krishnan KR, Lesperance F, Rieckmann N, Sheps DS, Suls JM (2006) Assessment and treatment of depression in patients with cardiovascular disease: National Heart, Lung, and Blood Institute working group report. Ann Behav Med 32(2):121–126
Contributors
WF-H and YH-F performed the whole meta-analysis, including literature search, data extraction, statistical analysis, and writing the manuscript. Y-Z, B-W, and JY-M offered help in the process of revision.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
Not required.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Electronic supplementary material
ESM 1
(DOC 71 kb)
Rights and permissions
About this article
Cite this article
He, W., Zhou, Y., Ma, J. et al. Effect of antidepressants on death in patients with heart failure: a systematic review and meta-analysis. Heart Fail Rev 25, 919–926 (2020). https://doi.org/10.1007/s10741-019-09850-w
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10741-019-09850-w