Prevalence of Depression in Coronary Artery Bypass Surgery: A Systematic Review and Meta-Analysis.

Coronary artery bypass graft surgery (CABG) might adversely affect the health status of the patients, producing cognitive deterioration, with depression being the most common symptom. The aim of this study is to analyse the prevalence of depression in patients before and after coronary artery bypass surgery. A systematic review and meta-analysis was carried out, involving a study of the past 10 years of the following databases: CINAHL, LILACS, MEDLINE, PsycINFO, SciELO, Scopus, and Web of Science. The total sample comprised n = 16,501 patients. The total number of items was n = 65, with n = 29 included in the meta-analysis. Based on the different measurement tools used, the prevalence of depression pre-CABG ranges from 19-37%, and post-CABG from 15-33%. There is a considerable presence of depression in this type of patient, but this varies according to the measurement tool used and the quality of the study. Systematically detecting depression prior to cardiac surgery could identify patients at potential risk.


Introduction
Coronary artery disease (CAD) is one of the leading causes of death in developed countries, and it is associated with deteriorated quality of life, disability, and premature death [1]. The usual surgical treatment involves coronary artery bypass graft surgery (CABG). This technique is based on revascularisation by diverting blood flow to other arteries to increase the blood supply to the heart muscle [2].
Although CABG surgery increases life expectancy [3], it is associated with multiple physical complications, including myocardial infarction, stroke, and even kidney failure [4]; in addition to psychological consequences, such as mood disorders, fatigue, weakness, stress, anxiety, and depression [5]. levels prior to or after CABG; (6) the use of a validated scale; (7) written in English, Portuguese, Spanish, or French; and, (8) published in the last 10 years.
The exclusion criteria were the following: (1) paediatric population; (2) a different type of cardiac surgery that was not exclusively CABG (CABG with valve replacement); (3) measurement of depression in relatives; (4) patients with an active treatment deriving from a psychiatric disorder; (5) data from duplicate articles in previous studies; and, (6) no depression data extracted using a validated scale.

Selection of Articles and Information Analysis
Firstly, two authors checked the title and abstract, and, secondly, the full text of the article. A third author was consulted in the case of discrepancy.
For the meta-analysis, we selected the data from those studies that used the same measurement tool, since the inclusion of several measurement tools would not permit the results to be integrated, due to different scores.

Data Extraction
The following variables were recorded: (1) data on the study (author, year, country); (2) type of CABG (first time, elective or emergency); (3) study characteristics (sample, type of study, sex, and follow-up time); (4) measurement tool; and, (5) mean, standard deviation, prevalence of depression. For clinical trials or quasi-experimental studies, we selected only the levels of depression prior to the programme intervention (baseline) or those relating to the control group.
We used the intraclass correlation coefficient to analyse coding reliability, obtaining an average value of 0.97 (minimum = 0.93; maximum = 1), and the Cohen's kappa coefficient with a mean value of 0.94 (minimum = 0.92; maximum = 1).

Assessment of Quality and Measurement of Bias
Two independently authors assessed the quality of the studies, consulting with a third party in the event of a disagreement.
For observational studies (cohort and cross-sectional), we followed the guidelines in "Strengthening the Reporting of Observational Studies in Epidemiology" (STROBE) [19]. We followed the standards in the Cochrane Collaboration Risk of Bias tool for clinical trials [20].
We used a second quality assessment tool to analyse the level of evidence in accordance with the recommendations of the Oxford Centre for Evidence-Based Medicine [21] (Table 1).

Data Synthesis and Statistical Analysis
The meta-analysis included those studies that used the same tool for measuring depression. We performed six meta-analyses using a random-effects model and two meta-analyses using a fixed-effect model, for prevalence levels and confidence intervals, through the statistical package StatsDirect (version 3, StatsDirect Ltd., Cambridge, UK).

Results
The search yielded a total of n = 1874 articles. After reading the title and abstract, 662 were excluded. Figure 1 shows the study selection process.

Results
The search yielded a total of n = 1874 articles. After reading the title and abstract, 662 were excluded. Figure 1 shows the study selection process.

Characteristics of Included Studies
The total sample comprised 16,501 patients, predominantly male (n = 54). Most of the studies were cohort studies (n = 34), followed by cross-sectional studies (n = 12). Thirteen studies evaluated the levels prior to surgery, 23 after surgery, and 29 both before and after. Most of the studies were carried out in the USA (n = 17), followed by Germany (n = 7), Iran (n = 7), and Australia (n = 6) ( Table  1). The depression follow-up ranged from a month prior to surgery (since the pre-assessment clinic appointment) [23] up to six years after surgery [24,25].

Characteristics of Included Studies
The total sample comprised 16,501 patients, predominantly male (n = 54). Most of the studies were cohort studies (n = 34), followed by cross-sectional studies (n = 12). Thirteen studies evaluated the levels prior to surgery, 23 after surgery, and 29 both before and after. Most of the studies were carried out in the USA (n = 17), followed by Germany (n = 7), Iran (n = 7), and Australia (n = 6) ( Table 1). The depression follow-up ranged from a month prior to surgery (since the pre-assessment clinic appointment) [23] up to six years after surgery [24,25].

Meta-Analysis
A total of 1217 patients were included in the meta-analysis prior to CABG surgery, and 596 patients after the operation. Egger's test showed no publication bias in any case.

Meta-Analysis
A total of 1217 patients were included in the meta-analysis prior to CABG surgery, and 596 patients after the operation. Egger's test showed no publication bias in any case.

Levels of Depression Before and After CABG Surgery and Follow Up
Prior to CABG surgery, most of the authors report depression levels within the normal range, although others found mild [36,55,58,83,86] and moderate levels [27,35,66,69] (Table 1).
The majority of authors observed a positive impact on depression prevalence and levels after surgery, as well as in the short and medium term, although others found that these levels increased after surgery [28,32,33,48,49,55,57,89].

Discussion
The prevalence of depression obtained in this study varied between 19% and 37% prior to surgery, and between 15% and 33% after surgery, depending on the type of measurement tool used. Other studies that combine CABG with valve replacement have reported similar percentages, with depression prevalence ranging from 15% pre-CABG [90] to 37.7% post-CABG [51,91], associated with the development of the disease, worse quality of life, longer hospital stays, and high rates of hospital readmissions [8].
Normal levels of pre-CABG depression are observed, although other studies have indicated higher levels, from moderate to severe [92]. However, more than 25% of patients with normal levels are at risk of worsening, for which reason continuous reassessment can identify patients with transient symptoms of depression [93].
High levels of depression prior to the operation predict a worse quality of life [94,95], worse survival after a CABG [12,96], and more symptoms up to six months after surgery [97].
We have observed that depression levels did not go to remission, but they tend to improve in depressive symptoms, which is probably due to an improvement in the patient's quality of life [98], and even due to greater optimism that facilitates commitment to adaptation [99]. Some authors have found a positive impact on patients from eight weeks [100], while others report a slight improvement from the first month post-CABG surgery [101]. For the majority of patients, depression persists after the surgery. Recent meta-analyses demonstrated that patients undergoing heart valve surgery are at risk of cognitive dysfunction up to six months after surgery [102,103].
Although there is a relationship between depression and CABG, its temporal onset is not clear. Depression can be a pre-existing condition, which increases the risk of cardiovascular disease that is related to behavioural alterations in diet, physical activity level, toxic habits, or poor adherence to treatment and recommendations [45]; or, can appear as a consequence of multiple postoperative complications, such as longer hospital stays [23], readmissions [104,105], general pain [104], or even when facing a series of lifestyle changes [12].
Without evaluation, it is unlikely that depression is being treated correctly. Some authors report that more than 50% of patients were receiving medical treatment for depression, even though they had no symptoms of depression [106]. For this reason, the use of measurement tools to confirm the presence and levels of depression makes it possible to identify the at-risk patients, and therefore carry out a more in-depth post-CABG follow-up, of at least nine months [93].
The current study highlights the importance of depression measures before and after CABG in assessing clinically meaningful mood disturbance, in order to provide early intervention. Systematic screening for depression in the period both before and after this procedure is crucial. Planned coaching combined with counselling can reduce these levels [36]. Cardiac rehabilitation programmes [107,108] and cognitive-behavioural therapies are also available, which reduce the levels of depression and even decrease the length of hospital stays [109]. However, further studies are needed to understand the potential prognostic implication of depression and investigate the best ways to approach the treatment of depression in this patient group.
Depression counselling prior to surgery can influence the post-surgical depression levels by positively improving a patient's perception of illness control and management [13]. Planning is therefore an essential part of the healthcare process as it has the potential to promote self-care [36].
From a clinical perspective, these results suggest that strategies that are aimed to improve depression as a disorder, such as the application of policies and depression assessment protocols prior to CABG by health care providers, are essential, because the depression level might help risk stratification in patients undergoing CABG identifying the high-risk groups and the trajectory of recovery experienced [11].
This study has several limitations. Firstly, the heterogeneity in terms of prevalence is due to different estimation methods over time, differences in the timing of assessment and demographic differences between samples, different uses of cut-offs on questionnaire measures, as well as the use of various tools for assessing the symptoms of depression. Secondly, the measuring tools assess the severity of depression symptoms, but they do not replace a formal clinical diagnosis of depression.

Conclusions
There is a high presence of depression both before and after CABG surgery. While this study found an overall improvement in depressive symptoms after CABG, depression persists after the surgery for the majority of patients. The depression levels present prior to the operation may affect postoperative recovery.
Given the prevalence of depression and its impact, early detection is crucial, since it enables the identification of at-risk patients, through a clinical interview that uses validated measurement tools. This enables the medical team to implement preventive strategies as well as monitor the development of the depression.