Wan et al.

BMC Psychiatry (2025) 25:416 BMC Psychiatry

https://doi.org/10.1186/s12888-025-06843-y

RESEARCH Open Access

Efficacy and safety of antidepressant in post-

myocardial infarction associated depression:
a meta-analysis and systematic review
Hongquan Wan1, He Li2, Shuxin Luan1* and Chunguo Zhang2*

Abstract
Objective This study aimed to summarize the available data and assess whether antidepressants are effective and
well-tolerated in the treatment of post-myocardial infarction (MI)-associated depression.
Materials and methods A comprehensive search of public databases (PubMed, Embase, Web of Science, Ovid,
EBSCO, and the Cochrane Library) was conducted for publications on interventions for post-MI depression before
October 2024. Keywords included post-myocardial infarction depression, antidepressants, myocardial infarction, and
depression. Pooled data were analyzed using Stata software.
Results A total of twelve studies were included. At baseline, no significant difference was observed in depression
severity between the antidepressant treatment and control groups (pooled SMD = -0.022, 95% CI: -0.087–0.044).
Antidepressant treatment significantly reduced depression scores after long-term follow-up (pooled SMD = -1.023,
95% CI: -1.671– -0.375). The incidence of adverse cardiac events was not significantly higher in the treatment group
compared to the control group (pooled HR = 0.893, 95% CI: 0.793–1.005). Antidepressants did not increase the risk
of all-cause mortality (pooled HR = 0.957, 95% CI: 0.699–1.311), and there was no significant difference in the risk of
rehospitalization for heart disease (pooled HR = 1.070, 95% CI: 0.820–1.398). Antidepressant treatment was associated
with a reduced risk of MI recurrence (pooled HR = 0.787, 95% CI: 0.693–0.894) and revascularization procedures
(pooled HR = 0.858, 95% CI: 0.755–0.975). Moderate-certainty evidence (GRADE assessment) supports antidepressant
efficacy in improving depressive symptoms, while low-certainty evidence suggests potential cardiac risk reduction.
Conclusion This meta-analysis demonstrates that antidepressants are effective and well-tolerated in the treatment
of post-MI depression. Antidepressants can improve depressive symptoms without adversely affecting long-
term prognosis. The clinical application of these findings should consider the moderate certainty for symptom
improvement and low certainty for MI recurrence benefits.
Keywords Post-myocardial infarction depression, Antidepressant, Myocardial infarction, Depression, Meta-analysis

*Correspondence:
Shuxin Luan
wanhq@jlu.edu.cn
Chunguo Zhang
qp8m2l@163.com
1
Department of Mental Health, The First Hospital of Jilin University, No.1
Xinmin Road, Changchun 130021, China
2
Department of Pain Medicine, The First Hospital of Jilin University, No.1
Xinmin Road, Changchun 130021, China

© The Author(s) 2025. Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0
International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you
give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the
licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or
other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or
exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit ​h​t​t​p​​:​/​/​​c​r​e​a​​t​i​​
v​e​c​​o​m​m​​o​n​s​.​​o​r​​g​/​l​​i​c​e​​n​s​e​s​​/​b​​y​-​n​c​-​n​d​/​4​.​0​/.
Wan et al. BMC Psychiatry (2025) 25:416 Page 2 of 14

Introduction clinical trials. Details of the search strategy are provided

Depression and coronary heart disease (CHD) are lead- in the supplementary methods section.
ing causes of global health burden and disability, as high-
lighted by the World Health Organization [1, 2]. A strong Study selection and data extraction
association between depression and myocardial infarc- Clear diagnostic criteria for depression were based on
tion (MI), a key factor in the etiology of CHD, has been well-established instruments, including the Beck Depres-
well documented [3, 4]. While MI survivors frequently sion Inventory (BDI) [18–20], the Hamilton Depression
experience depression, there is a notable gender dispar- Rating Scale (HAMD) [21], or other validated diagnostic
ity in both its prevalence and its impact on cardiac prog- tools. A structured psychiatric interview, based on the
nosis and all-cause mortality [5]. Evidence suggests that Diagnostic and Statistical Manual of Mental Disorders,
screening for and treating depression can modestly alle- Fifth Edition (DSM-V) criteria, was used to diagnose
viate depressive symptoms in post-MI patients [6]. More- major or minor depressive disorder [22]. Studies were
over, subclinical depressive symptoms in MI patients may included only if depression diagnosis was confirmed
also contribute to increased morbidity and mortality [7]. through (1) DSM-IV/ICD-10 criteria via structured inter-
A growing body of evidence has indicated that post-MI view (e.g., SCID), or (2) validated scales (PHQ-9 ≥ 10,
depression is not only detrimental to the quality of life HAM-D ≥ 18).
but also correlates with worsened clinical outcomes, Two independent investigators conducted the data-
including increased morbidity and mortality [8, 9]. base search and screened the results according to the
The reported prevalence of post-MI depression varies predefined inclusion and exclusion criteria. A third
widely. Almamari et al. estimated that major depression investigator was consulted as a referee to resolve any dis-
affects 5% of post-MI patients depending on diagnostic agreements between the initial two reviewers. The inclu-
criteria, with higher rates in those with prior psychiat- sion criteria were as follows:
ric history [10], while Wilkowska et al. reported a much Population: Eligible participants were adults (≥ 18
higher rate of 34.4% using the Beck Depression Inventory years) with (1) confirmed MI diagnosis and (2) comor-
(BDI) with a threshold of ≥ 10 [11]. Akhtar et al. provided bid major depressive disorder (MDD) was diagnosed
an intermediate estimate of 14% [12]. Meta-analyses indi- according to Diagnostic and Statistical Manual of Mental
cate that post-MI depression is associated with a 1.6- to Disorders, Fifth Edition (DSM-5) criteria for depressive
2.7-fold increased risk of mortality and cardiovascular disorders, specifically requiring the presence of at least
events during long-term follow-up [13], highlighting the one core symptom (depressed mood or loss of interest)
critical need for effective interventions [14]. and five or more additional symptoms (e.g., sleep dis-
Antidepressants are frequently prescribed to post-MI turbances, fatigue) persisting for ≥2 weeks. Studies were
patients, with usage rates around 6% [15]. Recent data included only if depression diagnosis was confirmed
from Fehr et al. corroborate this figure, reporting a pre- through: Structured Clinical Interview for DSM-5 (SCID-
scription rate of 6.3% for antidepressants in patients 5), or ICD-10 criteria validated by the Mini-International
diagnosed with post-MI depression at discharge [16], Neuropsychiatric Interview (MINI), and (3) post-MI
suggesting a consistent need for antidepressant interven- patients with regular follow-up;
tion. Banankhah et al. emphasized that effective treat- Intervention: Studies investigating SSRIs or TCAs
ment of depression can reduce mortality in post-MI were prioritized; other classes (e.g., SNRIs, MAOIs) were
patients, underscoring the importance of monitoring and excluded post hoc due to insufficient data (< 3 studies per
optimizing treatment [17]. However, concerns remain subclass).
regarding the potential adverse cardiac effects of anti- Control: Conventional treatment excluding
depressants [16]. To address these ongoing debates, this antidepressants;
meta-analysis aims to evaluate the efficacy and safety of Outcomes: Evaluation of depression severity, all-cause
antidepressant therapy for post-MI depression. mortality, rehospitalization due to heart disease, repeated
percutaneous coronary intervention, and MI recurrence.
Materials and methods Data extraction fields included study design (RCT,
Search strategy cohort, case-control), sample size, follow-up duration,
This study adhered to PRISMA guidelines and conducted and outcome ascertainment methods.
a comprehensive search of several databases (PubMed, Studies not designed as randomized controlled trials
Embase, Web of Science, Ovid, EBSCO, and the (RCTs) or prospective cohort studies were excluded.
Cochrane Library) for articles published up until October
2024. Final searches across database were conducted on Data synthesis and analysis
31 October 2024. Keywords used in the search included Reference citations were created and duplicate records
myocardial infarction, depression, antidepressants, and were screened using Noteexpress Bibliography Software
Wan et al. BMC Psychiatry (2025) 25:416 Page 3 of 14

(version 3.2.0; Beijing Aegean Software Co., Ltd., China). Results

Relevant data from each study, including the first author’s Database search and baseline demographic characteristics
name, year of publication, country, sample size, baseline of the included studies
demographic characteristics, and outcome measures, A total of 925 articles were initially screened, and 12
were extracted and compiled by two independent inves- studies [16, 23–33] met the inclusion criteria for fur-
tigators. The quality of RCTs was assessed using the ther meta-analysis. A flowchart detailing the publication
Jadad Scale, while non-RCTs were evaluated using the filtration process is shown in Fig. 1. The characteristics
Newcastle-Ottawa Scale (NOS). The GRADE framework of the included studies are summarized in Table 1, with
was applied to assess the certainty of evidence for each baseline patient demographics presented in Supplemen-
outcome (depression response, cardiac events, mortal- tary Table 1. A quality assessment of the included studies
ity), evaluating risk of bias, inconsistency, indirectness, is provided in Table 2. Additionally, the GRADE frame-
imprecision, and publication bias. work was applied to assess the certainty of the evidence,
and the findings are presented in Table 3.
Statistics
Stata statistical software (version 12.0, University of Comparison of depression severity between
Texas Stata Corporation) was used to analyze and inte- antidepressant treatment and control groups at baseline
grate the extracted data. The statistical heterogeneity of and long-term follow-up
the data was assessed using the Inverse Variance (I-V) Data on depression severity at baseline, as measured by
method. Heterogeneity was quantified using the Q test, the BDI or HAMD scores, were available from six stud-
calculated as: SIGMA_i{ (1/variance_i)*(effect_i - effect_ ies [23, 26, 28, 31–33]. These data showed no significant
pooled)^2 } where variance_i = ((upper limit - lower heterogeneity (p = 0.840, I² = 0.0%). The baseline depres-
limit)/(2*z))^2. The I² statistic was used to describe the sion levels were comparable between the antidepressant
heterogeneity among studies, with values > 50% or a treatment and control groups (pooled SMD = -0.022, 95%
p-value < 0.05 indicating high heterogeneity. When high CI: -0.087–0.044), as shown in Fig. 2A. At long-term fol-
heterogeneity was present, a random-effects model was low-up, eight studies [24–33] reported that antidepres-
used to calculate pooled estimates; otherwise, a fixed- sant treatment significantly reduced depression scores
effects model was applied. For continuous variables, the (pooled SMD = -1.023, 95% CI: -1.671– -0.375), although
standardized mean difference (SMD) was calculated. For significant heterogeneity was observed (Fig. 2B). Sensitiv-
binary variables, results were expressed as odds ratios ity analysis confirmed the robustness of these findings,
(OR) with 95% confidence intervals (95% CI). Publication with antidepressant treatment showing a consistent effect
bias was assessed using Egger’s test and visualized using in reducing depression severity after excluding individual
a funnel plot. A two-tailed p-value < 0.05 was consid- studies (Fig. 2C).
ered statistically significant. Sensitivity analysis was per-
formed to evaluate the robustness of the results obtained Evaluation of the safety of antidepressant treatment in
from the random-effects model. Publication bias refers to patients with post-MI depression
the selective dissemination of research findings based on Adverse cardiac events were reported in six studies [24–
their statistical significance or direction, which may dis- 32], and no significant heterogeneity was detected among
tort the pooled effect estimates. Two widely used statis- them (Fig. 3A). The overall incidence of adverse cardiac
tical methods to detect publication bias are Begg’s rank events was not higher in the treatment group compared
correlation test and Egger’s linear regression test, both to the control group (pooled HR = 0.893, 95% CI: 0.793–
of which are often accompanied by graphical tools for 1.005). All-cause mortality data were available from
visual assessment. Below is an academic interpretation seven studies [16, 23–25, 29, 31, 32], and no significant
of these tests and their graphical representations. Begg’s heterogeneity was found (Fig. 3B and C). Antidepressant
and Egger’s tests, supported by their graphical counter- treatment did not increase the risk of all-cause mortality
parts, are essential tools for diagnosing publication bias (pooled HR = 0.957, 95% CI: 0.699–1.311).
in meta-analyses. While Begg’s test provides a non-para- The risk of rehospitalization due to heart disease was
metric rank-based assessment, Egger’s regression offers a assessed in six studies [16, 26, 27, 31, 33], with significant
parametric framework to quantify asymmetry. Research- heterogeneity detected (p < 0.001, I² = 96.6%) (Fig. 4A).
ers must acknowledge the limitations of these tests and In the primary analysis, the risk of rehospitalization
integrate them into a broader strategy for evaluating the did not differ significantly between the antidepressant
robustness and validity of meta-analytic findings. and control groups (pooled HR = 1.070, 95% CI: 0.820–
1.398) (Fig. 4A). After sensitivity analysis and exclud-
ing one study [16] (Fig. 4B), the risk of rehospitalization
Wan et al. BMC Psychiatry (2025) 25:416 Page 4 of 14

Fig. 1 Study retrieval and selection

Table 1 The characteristics of included studies

Author Year Region Number of patients Study design AD Mean follow-up duration
Fehr et al. 2019 Switzerland 8911 prospective cohort not specified 1 year
Kim et al. 2018 Korea 300 RCT escitalopram 8.4 years
Smolderen et al. 2017 United States 759 prospective cohort not specified 1 year
Zuidersma et al. 2013 Netherlands 311 RCT mirtazapine and citalopram 5.2 years
Honig et al. 2007 Netherlands 91 RCT mirtazapine 0.5 year
van Melle et al. 2007 Netherlands 331 RCT not specified 1.5 years
Strik et al. 2006 Netherlands 54 RCT fluoxetine 1 year
O’Connor et al. 2005 United States 369 RCT sertraline 0.5 year
Taylor et al. 2005 United States 1689 RCT SSRIs 0.5 year
Berkman et al. 2003 United States 2481 RCT sertraline 2.4 years
Glassman et al. 2002 United States 369 RCT Sertraline 1 year
Strik et al. 2000 Netherlands 54 RCT fluoxetine 1 year

was not increased by antidepressant treatment (pooled study [16], which largely contributed to the heterogeneity
HR = 0.977, 95% CI: 0.913–1.045) (Fig. 4C). (Fig. 5B), the re-analysis indicated that antidepressants
significantly reduced the risk of MI recurrence (pooled
Cardiac safety outcomes HR = 0.787, 95% CI: 0.693–0.894) (Fig. 5C). The effect of
Six studies [16, 23, 26, 29, 31, 32] assessed the effect of antidepressant t treatment on the risk of repeated percu-
antidepressants on the risk of recurrent MI. Signifi- taneous coronary intervention (re-PCI) was evaluated in
cant heterogeneity was observed among these studies four studies (Fehr et al. 2019; Kim et al. 2018; van Melle et
(Fig. 5A), and the pooled data suggested that antidepres- al. 2007; Berkman et al. 2003) (Fig. 6A). Antidepressants
sant treatment did not significantly reduce the risk of did not reduce the risk of re-PCI (pooled HR = 1.011,
recurrent MI (pooled HR = 0.931, 95% CI: 0.374–2.316). 95% CI: 0.540–1.892), with high heterogeneity observed
However, after sensitivity analysis and removing one (p < 0.001, I² = 98.2%). However, after excluding one study
Wan et al. BMC Psychiatry (2025) 25:416 Page 5 of 14

Table 2 Quality assement of eligible literatures

Quality Assesment
RCT study (Jadad scale) Author Prospective Clear defini- (1) (2) (3) (4) (5)
design tion of study
population
Kim et al. 2018 Yes Yes Yes Yes Yes Yes Yes
Zuidersma 2013 Yes Yes Yes Yes Yes Yes Yes
et al.
Honig et al. 2007 Yes Yes Yes Yes Yes Yes Yes
van Melle 2007 Yes Yes Yes Yes Yes Yes Yes
et al.
Strik et al. 2006 Yes Yes Yes Yes Yes Yes Yes
O’Connor 2005 Yes Yes Yes Yes Yes Yes Yes
et al.
Taylor et al. 2005 Yes Yes Yes Yes Yes Yes Yes
Berkman 2003 Yes Yes Yes Yes Yes Yes Yes
et al.
Glassman 2002 Yes Yes Yes Yes Yes Yes Yes
et al.
Strik et al. 2000 Yes Yes Yes Yes Yes Yes Yes
non-RCT study (Newcastle- Author Study Clear defini- Clear Clear Blindness Representa- Compa-
Ottawa scale) design tion of study defini- definition to AD or tivenessof rability
population tion of of related placebo the study between
different endpoints population inter-
type of vention
AD and
control
groups
Fehr et al. 2019 243 Yes Yes Yes No Yes Yes
Smolderen 2017 543 Yes No Yes No Yes Yes
et al.
(1) Was the study described as randomized?
(2) Was the randomization method appropriate?
(3) Was the study described as double-blind?
(4) Was the blinding method appropriate?
(5) Was there a description of withdrawals and dropouts?

[16], which was the primary contributor to the heteroge- post-myocardial infarction (MI) depression. Although
neity (Fig. 6B), the re-analysis showed that antidepres- baseline depression severity did not significantly dif-
sants had a positive effect on reducing the risk of re-PCI fer between the treatment and control groups (pooled
(pooled HR = 0.858, 95% CI: 0.755–0.975) (Fig. 6C). SMD = -0.022, 95% CI: -0.087–0.044), antidepressant
While antidepressants improved depressive symptoms, treatment led to a significant reduction in depression
their association with reduced MI recurrence should be scores following long-term follow-up (pooled SMD
interpreted cautiously given confounding in observa- = -1.023, 95% CI: -1.671– -0.375). The analysis also
tional studies. demonstrated that antidepressant use did not increase
the risk of adverse cardiac events (pooled HR = 0.893,
Publication bias analysis 95% CI: 0.793–1.005), all-cause mortality (pooled
Publication bias was assessed using both Begg’s (Fig. 7A) HR = 0.957, 95% CI: 0.699–1.311), or rehospitalization
and Egger’s tests (Fig. 7B). Both tests indicated a sym- for heart disease (pooled HR = 1.070, 95% CI: 0.820–
metrical distribution of the included studies (Begg’s test: 1.398). Additionally, antidepressants were linked to a
p = 0.138, Egger’s test: p = 0.309), suggesting no significant decreased risk of MI recurrence (pooled HR = 0.787,
publication bias among the included articles. 95% CI: 0.693–0.894) and revascularization pro-
cedures (pooled HR = 0.858, 95% CI: 0.755–0.975).
Discussion Post-MI depression is an established independent
In this meta-analysis, antidepressant treatment was risk factor for poor cardiac prognosis [34]. This study
associated with improvements in depressive symp- underscores the clinical value of antidepressant treat-
toms and a favorable safety profile for patients with ment in improving outcomes for patients with post-MI
Wan et al. BMC Psychiatry (2025) 25:416 Page 6 of 14

Table 3 GRADE assessment

Outcome Study (Au- Study Design Risk of Inconsistency Indirectness Imprecision Cer- Comments
thor, Year) Bias tainty of
Evidence
1-Year Fehr et al. Observational Serious None None None Low Confounding likely;
Mortality (2019) large sample size sup-
ports findings.
Smolderen et Observational Moderate None None None Moderate Adjusted for confound-
al. (2017) ers; consistent findings.
Glassman et RCT Low None None None High Robust design;
al. (2002) well-powered.
Berkman et al. RCT Low None None None High Large RCT with strong
(2003) methodology.
Long-Term Kim et al. RCT Low None None None High Direct evidence; small
Cardiac (2018) but well-conducted
Events study.
Zuidersma et RCT Moderate None None Serious Moderate Small sample size leads
al. (2013) to imprecision.
Taylor et al. RCT Low None None None High Large, well-conducted
(2005) study.
van Melle et RCT Moderate None None Serious Moderate Small sample size lim-
al. (2007) its generalizability.
Depressive Honig et al. RCT Moderate None None Serious Moderate Effective treatment but
Symptom (2007) small sample size.
Reduction
Strik et al. RCT Moderate None Serious Serious Low Limited generalizability
(2006) and small sample.
Strik et al. RCT Moderate None None Serious Moderate Small sample size but
(2000) consistent efficacy.
Glassman et RCT Low None None None High Strong evidence for
al. (2002) symptom reduction in
post-MI patients.

depression. While some reports have raised concerns associated with a poorer prognosis at 1-year follow-
about the potential cardiac risks of antidepressants up. However, the study had limited data on depres-
[16], the overall evidence suggests that they are safe sion diagnosis and antidepressant classification,
and effective when appropriately prescribed. Berkman which reduced its quality. The benefits of antidepres-
et al. observed improvements in depressive symptoms sant treatment may also be constrained by heteroge-
and social isolation with interventions for post-MI neous response rates across different populations. For
depression, even though these interventions did not instance, de Jonge et al. reported that non-response
significantly affect event-free survival [31]. In contrast, to treatment for post-MI depression might be linked
van Melle et al. found no difference in mean Beck to increased cardiac events [35]. Additionally, Tay-
Depression Inventory (BDI) scores between patients lor et al., using data from the ENRICHD trial, found
treated with antidepressants and those who were not, that patients who received either selective serotonin
and reported that antidepressant treatment neither reuptake inhibitors (SSRIs) or tricyclic antidepres-
altered long-term depression status nor improved car- sants had lower mortality compared to those who
diac prognosis [26]. However, specific antidepressants did not, though there was no effect on the incidence
such as escitalopram and sertraline have shown partic- of non-fatal MI [29]. However, it is important to
ular promise in reducing major adverse cardiac events note that ENRICHD was a randomized trial of cog-
and managing recurrent post-MI depression, respec- nitive behavioral therapy (CBT) with some partici-
tively [23, 32]. pants receiving antidepressants, rather than a direct
To evaluate the effect of antidepressants on 1-year comparison of antidepressants to other treatments.
prognosis in patients with acute MI, logistic regres- A follow-up study by Glassman from the SADHART
sion analysis compared baseline characteristics and trial (sertraline vs. placebo) found that while sertra-
outcomes between patients who received antidepres- line did not affect 7-year major adverse cardiac events
sant medication at discharge and those who did not (MACE), the short-term depression response, regard-
[16]. This study suggested that antidepressants were less of treatment, was associated with lower 7-year
Wan et al. BMC Psychiatry (2025) 25:416 Page 7 of 14

Fig. 2 Comparison of the depressive degree between the antidepressant-treated group and the control group at baseline and long-term follow-up
duration: before sensitivity analysis (A); sensitivity analysis (B); after sensitivity analysis (the x-axis indicates the standardized mean difference (SMD), and
fixed-effect model is applied). A unity line should be SMD = 0. (C)

mortality compared to subjects with persistent depres- depression after MI, emphasizing the need for close
sion [36]. Other studies have also demonstrated that monitoring and, when necessary, adjusting antide-
non-response to treatment or persistent depression is pressant treatment to improve patient prognosis [17].
linked to increased long-term morbidity and mortal- Wilkowska et al. suggested that fluctuations in morn-
ity [37, 38]. Emerging research on biomarkers such as ing and afternoon serum cortisol levels in patients with
interleukin-6, tumor necrosis factor-α, and C-reactive post-MI depression could help identify those with lon-
protein shows promise in predicting responses to anti- ger-lasting depressive symptoms. A flattened diurnal
depressant treatment and elucidating the relationship serum cortisol profile was easily detectable [11]. SSRIs
between inflammation and depression [39, 40]. More- may be recommended as the first-line treatment for
over, soluble cytokine receptors and cytokines are post-MI depression [44, 45], though further research
closely associated with MI during both the acute and is needed to confirm their long-term benefits and to
chronic phases [41]. This opens the possibility of com- compare them with newer antidepressants. Post-MI
bining anti-inflammatory treatments with antidepres- patients often suffer from depression and anxiety,
sants to improve post-MI depression outcomes. For which negatively impact treatment adherence, recov-
example, mirtazapine, a dual-acting antidepressant, ery, and overall quality of life. However, due to the
has been found effective in treating post-MI depres- atypical nature of post-MI depression symptoms, rec-
sion without increasing adverse events [27]. ognition rates remain low among healthcare providers,
The Patient Health Questionnaire-9 (PHQ-9) and potentially delaying diagnosis and treatment. Training
the Beck Depression Inventory (BDI) are key tools clinicians to identify and treat post-MI depression is
for diagnosing post-MI depression and can help iden- crucial for improving patient outcomes and reducing
tify patients at higher risk for poor outcomes [42, 43]. the incidence of missed diagnoses.
These tools were therefore introduced as diagnostic
criteria for post-MI depression. Effective treatment of
depression can reduce the mortality of patients with
Wan et al. BMC Psychiatry (2025) 25:416 Page 8 of 14

Fig. 3 Evaluation of the adverse cardiac event risks of antidepressant treatment in patients with post-MI depression: before sensitivity analysis (A); sen-
sitivity analysis (B); after sensitivity analysis (the x-axis indicates the standardized mean difference (SMD), and fixed-effect model is applied). A unity line
should be SMD = 0. (C)

Limitations examining the impact of treatment adherence on post-

This meta-analysis has several limitations. First, 21 MI depression.
articles were excluded due to lack of access to full-text
publications, which could introduce selection bias. Conclusion
Additionally, the study did not have sufficient data on Antidepressants are probably effective and safe for
the specific types and dosages of antidepressants used, managing post-MI depression. Their use probably
preventing subgroup analyses that might reveal poten- improves depressive symptoms and reduces MI recur-
tial differences in treatment effects. Only SSRIs and rence risk, highlighting their clinical relevance. Further
mirtazapine were included in this analysis, and other studies are needed to validate these findings. Although
classes of antidepressants were excluded, limiting the antidepressants demonstrate moderate-certainty effi-
generalizability of the findings. Another limitation is cacy in alleviating post-MI depression without wors-
the reliance on standardized mean differences (SMD) ening cardiac prognosis, observed MI risk reduction
to combine depression scores. While this approach in sensitivity analyses requires validation in RCTs, as
is consistent, it may exaggerate effect sizes and lacks high-quality trials showed no significant benefit.
clinically intuitive interpretation. Furthermore, the
study did not assess treatment acceptability or adher- Relevance for clinical practice
ence beyond the adverse event profiles. The focus on In the present study, we were aimed to summarize those
long-term outcomes also meant that the potential accessible data and testify whether the antidepressant
short-term effects of acute depressive episodes were is effective and tolerable in post-myocardial infarction
not explored. Future studies should address these associated depression. Our study indicated that Efficacy
gaps by evaluating novel antidepressant treatments, and tolerability of antidepressant in post-MI associated
exploring the role of anti-inflammatory therapies, and depression was proved. Antidepressant could improve
Wan et al. BMC Psychiatry (2025) 25:416 Page 9 of 14

Fig. 4 Evaluation of the re-hospitalization risks of antidepressant treatment in patients with post-MI depression: before sensitivity analysis (A); sensitivity
analysis (B); after sensitivity analysis (sensitivity analysis excluding Fehr et al. 2019 (Non-RCT), the x-axis indicates the standardized mean difference (SMD),
and fixed-effect model is applied). A unity line should be SMD = 0. (C)

symptom of post-MI depression without weakening the

long-term prognosis.
Wan et al. BMC Psychiatry (2025) 25:416 Page 10 of 14

Fig. 5 Assessment of the recurrent myocardial infarction risks of anti-depressive treatment in patients with post-myocardial infarction depression: before
sensitivity analysis (A); sensitivity analysis (B); after sensitivity analysis (C)
Wan et al. BMC Psychiatry (2025) 25:416 Page 11 of 14

Fig. 6 Assessment of the risk of repeated percutaneous coronary intervention of anti-depressive treatment in patients with post-myocardial infarction
depression: before sensitivity analysis (A); sensitivity analysis (B); after sensitivity analysis (C)
Wan et al. BMC Psychiatry (2025) 25:416 Page 12 of 14

Fig. 7 Publication bias analysis: Begg’s test: X-axis: Rank of effect sizes (ascending order); Y-axis: Rank of variances or standard errors (descending order)
(A); Egger’s test: X-axis: Standard error (SE) of effect sizes; Y-axis: Effect size (e.g., odds ratio, standardized mean difference; regression line: a best-fit line with
slope (β) and intercept (α)) (B). X-axis: Standard error (SE) of effect sizes; Y-axis: Effect size (e.g., odds ratio, standardized mean difference)

Abbreviations
MACE Major adverse cardiovascular events Supplementary Material 1
CHD Coronary heart disease
Supplementary Material 2
BDI Beck depression inventory
MI Myocardial infarction Supplementary Material 3
PRISMA Preferred reporting items for systematic reviews and
meta-analyses
HAMD Hamilton depression rating scale Acknowledgements
NOS Newcastle-ottawa scale None.
SMD Standardized mean difference
OR Odds ratios Author contributions
CI Confidence interval HQ W, SX L, and CG Z contributed to the study conception and design. HQ W,
HR Hazard ratio SX L, Q H, and CG Z collected the data and performed data analysis. HQ W, SX
PCI Percutaneous coronary intervention L, Q H, and CG Z contributed to the interpretation of the data and completion
of figures and tables. HQ W, SX L, Q H, and CG Z contributed to the drafting of
the manuscript and final approval of the submitted version.
Supplementary Information
The online version contains supplementary material available at ​h​t​t​p​s​:​​​/​​/​d​o​​i​.​​o​r​​ Funding
g​​/​​1​0​​.​1​1​​​8​6​​/​s​1​2​​8​8​8​-​​0​2​5​-​0​​6​8​4​3​-​y. None.
Wan et al. BMC Psychiatry (2025) 25:416 Page 13 of 14

Data availability of secondary prevention measures–a longitudinal study. J Psychosom Res.

No datasets were generated or analysed during the current study. 2012;72(1):5–10.
15. Benazon NR, Mamdani MM, Coyne JC. Trends in the prescribing of antide-
pressants following acute myocardial infarction, 1993–2002. Psychosom Med.
Declarations 2005;67(6):916–20.
16. Fehr N, Witassek F, Radovanovic D, Erne P, Puhan M, Rickli H. Antidepressant
Ethics approval and informed consent prescription in acute myocardial infarction is associated with increased
Since meta-analyses compile and synthesize findings from existing literature, mortality 1 year after discharge. Eur J Intern Med. 2019;61:75–80.
they do not require additional ethics approval or informed consent from 17. Banankhah SK, Friedmann E, Thomas S. Effective treatment of depres-
participants. sion improves post-myocardial infarction survival. World J Cardiol.
2015;7(4):215–23.
Consent for publication 18. Wang YP, Gorenstein C. Psychometric properties of the Beck Depression
Not applicable. Inventory-II: a comprehensive review. Revista brasileira de psiquiatria (Sao
Paulo, Brazil: 1999). 2013;35(4):416– 31.
Competing interests 19. Richter P, Werner J, Heerlein A, Kraus A, Sauer H. On the validity of the Beck
The authors declare no competing interests. depression inventory. A review. Psychopathology. 1998;31(3):160–8.
20. Lovibond PF, Lovibond SH. The structure of negative emotional States:
Received: 12 June 2024 / Accepted: 9 April 2025 comparison of the depression anxiety stress scales (DASS) with the Beck
depression and anxiety inventories. Behav Res Ther. 1995;33(3):335–43.
21. Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry.
1960;23(1):56–62.
22. Malhi GS, Mann JJ, London. England). 2018;392(10161):2299–312.
References 23. Kim JM, Stewart R, Lee YS, Lee HJ, Kim MC, Kim JW, et al. Effect of Escitalo-
1. Hasnain M, Vieweg WV, Lesnefsky EJ, Pandurangi AK. Depression screen- pram vs placebo treatment for depression on Long-term cardiac outcomes
ing in patients with coronary heart disease: a critical evaluation of the AHA in patients with acute coronary syndrome: A randomized clinical trial. JAMA.
guidelines. J Psychosom Res. 2011;71(1):6–12. 2018;320(4):350–8.
2. Lichtman JH, Bigger JT Jr., Blumenthal JA, Frasure-Smith N, Kaufmann PG, Les- 24. Smolderen KG, Buchanan DM, Gosch K, Whooley M, Chan PS, Vaccarino V, et
pérance F, et al. Depression and coronary heart disease: recommendations al. Depression treatment and 1-Year mortality after acute myocardial infarc-
for screening, referral, and treatment: a science advisory from the American tion: insights from the TRIUMPH registry (Translational research investigating
heart association prevention committee of the Council on cardiovascu- underlying disparities in acute myocardial infarction patients’ health Status).
lar nursing, Council on clinical cardiology, Council on epidemiology and Circulation. 2017;135(18):1681–9.
prevention, and interdisciplinary Council on quality of care and outcomes 25. Zuidersma M, Conradi HJ, van Melle JP, Ormel J, de Jonge P. Depression
research: endorsed by the American psychiatric association. Circulation. treatment after myocardial infarction and long-term risk of subsequent car-
2008;118(17):1768–75. diovascular events and mortality: a randomized controlled trial. J Psychosom
3. Headrick JP, Peart JN, Budiono BP, Shum DHK, Neumann DL, Stapelberg NJC. Res. 2013;74(1):25–30.
The heartbreak of depression: ‘Psycho-cardiac’ coupling in myocardial infarc- 26. van Melle JP, de Jonge P, Honig A, Schene AH, Kuyper AM, Crijns HJ, et al.
tion. J Mol Cell Cardiol. 2017;106:14–28. Effects of antidepressant treatment following myocardial infarction. Br J
4. Echols MR, O’Connor CM. Depression after myocardial infarction. Curr Heart Psychiatry: J Mental Sci. 2007;190:460–6.
Fail Rep. 2010;7(4):185–93. 27. Honig A, Kuyper AM, Schene AH, van Melle JP, de Jonge P, Tulner DM, et
5. Doyle F, McGee H, Conroy R, Conradi HJ, Meijer A, Steeds R, et al. Systematic al. Treatment of post-myocardial infarction depressive disorder: a ran-
review and individual patient data Meta-Analysis of sex differences in depres- domized, placebo-controlled trial with Mirtazapine. Psychosom Med.
sion and prognosis in persons with myocardial infarction: A MINDMAPS 2007;69(7):606–13.
study. Psychosom Med. 2015;77(4):419–28. 28. Strik JJ, Honig A, Klinkenberg E, Dijkstra J, Jolles J. Cognitive performance
6. Thombs BD, Roseman M, Coyne JC, de Jonge P, Delisle VC, Arthurs E, et al. following Fluoxetine treatment in depressed patients post myocardial infarc-
Does evidence support the American heart association’s recommendation to tion. Acta Neuropsychiatrica. 2006;18(1):1–6.
screen patients for depression in cardiovascular care? An updated systematic 29. Taylor CB, Youngblood ME, Catellier D, Veith RC, Carney RM, Burg MM,
review. PLoS ONE. 2013;8(1):e52654. et al. Effects of antidepressant medication on morbidity and mortality
7. García Vicente E, Del Villar Sordo V, García YGEL. [Post-myocardial infarc- in depressed patients after myocardial infarction. Arch Gen Psychiatry.
tion depression]. Anales de medicina interna (Madrid, Spain: 1984). 2005;62(7):792–8.
2007;24(7):346– 51. 30. O’Connor CM, Glassman AH, Harrison DJ. Pharmacoeconomic analysis of Ser-
8. Zuidersma M, Conradi HJ, van Melle JP, Ormel J, de Jonge P. Self-reported traline treatment of depression in patients with unstable angina or a recent
depressive symptoms, diagnosed clinical depression and cardiac morbidity myocardial infarction. J Clin Psychiatry. 2005;66(3):346–52.
and mortality after myocardial infarction. Int J Cardiol. 2013;167(6):2775–80. 31. Berkman LF, Blumenthal J, Burg M, Carney RM, Catellier D, Cowan MJ, et al.
9. Bigotte Vieira M, Baptista RB, Costa J, Vaz-Carneiro A. [Analysis of the Effects of treating depression and low perceived social support on clinical
Cochrane Review: Pharmacotherapy for Hyperuricemia in Hypertensive events after myocardial infarction: the enhancing recovery in coronary heart
Patients. Cochrane Database Syst Rev. 2017;4:CD008652.]. Acta medica portu- disease patients (ENRICHD) randomized trial. JAMA. 2003;289(23):3106–16.
guesa. 2017;30(5):356– 60. 32. Glassman AH, O’Connor CM, Califf RM, Swedberg K, Schwartz P, Bigger JT Jr.,
10. Almamari RSS, Muliira JK, Lazarus ER. Self-reported sleep quality and depres- et al. Sertraline treatment of major depression in patients with acute MI or
sion in post myocardial infarction patients attending cardiology outpatient unstable angina. JAMA. 2002;288(6):701–9.
clinics in Oman. Int J Nurs Sci. 2019;6(4):371–7. 33. Strik JJ, Honig A, Lousberg R, Lousberg AH, Cheriex EC, Tuynman-Qua HG, et
11. Wilkowska A, Rynkiewicz A, Wdowczyk J, Landowski J. Morning and after- al. Efficacy and safety of Fluoxetine in the treatment of patients with major
noon serum cortisol level in patients with post-myocardial infarction depres- depression after first myocardial infarction: findings from a double-blind,
sion. Cardiol J. 2019;26(5):550–4. placebo-controlled trial. Psychosom Med. 2000;62(6):783–9.
12. Akhtar MS, Malik SB, Ahmed MM. Symptoms of depression and anxiety in 34. Bhatt DL, Lopes RD, Harrington RA. Diagnosis and treatment of acute coro-
post-myocardial infarction patients. J Coll Physicians Surgeons–Pakistan: nary syndromes: A review. JAMA. 2022;327(7):662–75.
JCPSP. 2004;14(10):615–8. 35. de Jonge P, Honig A, van Melle JP, Schene AH, Kuyper AM, Tulner D,
13. Meijer A, Conradi HJ, Bos EH, Thombs BD, van Melle JP, de Jonge P. Prognostic et al. Nonresponse to treatment for depression following myocardial
association of depression following myocardial infarction with mortality infarction: association with subsequent cardiac events. Am J Psychiatry.
and cardiovascular events: a meta-analysis of 25 years of research. Gen Hosp 2007;164(9):1371–8.
Psychiatry. 2011;33(3):203–16. 36. Glassman AH, Bigger JT Jr., Gaffney M. Psychiatric characteristics associated
14. Myers V, Gerber Y, Benyamini Y, Goldbourt U, Drory Y. Post-myocardial with long-term mortality among 361 patients having an acute coronary
infarction depression: increased hospital admissions and reduced adoption
Wan et al. BMC Psychiatry (2025) 25:416 Page 14 of 14

syndrome and major depression: seven-year follow-up of SADHART partici- 43. Huffman JC, Doughty CT, Januzzi JL, Pirl WF, Smith FA, Fricchione GL. Screen-
pants. Arch Gen Psychiatry. 2009;66(9):1022–9. ing for major depression in post-myocardial infarction patients: operating
37. Carney RM, Freedland KE, Steinmeyer B, Blumenthal JA, de Jonge P, Davidson characteristics of the Beck depression Inventory-II. Int J Psychiatry Med.
KW, et al. History of depression and survival after acute myocardial infarction. 2010;40(2):187–97.
Psychosom Med. 2009;71(3):253–9. 44. Rajeswaran T, Plymen CM, Doherty AM. The effect of antidepressant medica-
38. Carney RM, Blumenthal JA, Freedland KE, Youngblood M, Veith RC, Burg MM, tions in the management of heart failure on outcomes: mortality, cardiovas-
et al. Depression and late mortality after myocardial infarction in the enhanc- cular function and depression - a systematic review. Int J Psychiatry Clin Pract.
ing recovery in coronary heart disease (ENRICHD) study. Psychosom Med. 2018;22(3):164–9.
2004;66(4):466–74. 45. Fernandes N, Prada L, Rosa MM, Ferreira JJ, Costa J, Pinto FJ, et al. The impact
39. Tulner DM, Smith OR, Schins A, de Jonge P, Quere M, Delanghe JR, et al. of SSRIs on mortality and cardiovascular events in patients with coronary
Antidepressive effect of Mirtazapine in post-myocardial infarction depression artery disease and depression: systematic review and meta-analysis. Clin Res
is associated with soluble TNF-R1 increase: data from the MIND-IT. Neuropsy- Cardiol. 2021;110(2):183–93.
chobiology. 2011;63(3):169–76.
40. Diniz BS, Teixeira AL, Talib LL, Mendonça VA, Gattaz WF, Forlenza OV. Increased
soluble TNF receptor 2 in antidepressant-free patients with late-life depres- Publisher’s note
sion. J Psychiatr Res. 2010;44(14):917–20. Springer Nature remains neutral with regard to jurisdictional claims in
41. Abbate A, Toldo S, Marchetti C, Kron J, Van Tassell BW, Dinarello CA. Interleu- published maps and institutional affiliations.
kin-1 and the inflammasome as therapeutic targets in cardiovascular disease.
Circul Res. 2020;126(9):1260–80.
42. DiSante JL, Bires AM, Cline TW, Waterstram-Rich K. An analysis of the
prevalence of depression Post-Myocardial infarction. Crit Care Nurs Q.
2017;40(2):124–36.