1 Department of Cardiology, The Affiliated Traditional Chinese Medicine Hospital, Southwest Medical University, 646000 Luzhou, Sichuan, China
†These authors contributed equally.
Abstract
The relationship between negative emotions and atrial fibrillation (AF) has garnered significant attention, yet observational studies have yielded contradictory findings regarding the causal associations between the two. Our study sought to provide genetic evidence for a causal relationship between negative emotions and AF through Mendelian randomization (MR) study.
Utilizing genetic variations associated with negative emotions and AF as instrumental variables (IVs), a two-sample MR study was implemented. The potential causality between the two was initially assessed by using negative emotions as exposure and AF as outcome. Subsequently, potential reverse causality was evaluated by using AF as exposure and negative emotions as outcome. The inverse variance weighted (IVW) method served as the primary analysis for the two-sample MR, supplemented by weighted median method, MR-Egger regression, Simple mode method, and Weighted mode method. Sensitivity analyses were performed using the MR pleiotropy residual sum and outlier test (MR-PRESSO), Cochran Q test, and leave-one-out analysis to ensure the robustness of the results.
The two-sample MR analyses revealed that genetic susceptibility to AF had no potential causal effect on negative emotions (p > 0.05). Conversely, genetic susceptibility to negative emotions was positively correlated with an increased relative risk of AF [odds ratio (OR), 1.173, 95% confidence interval (CI): 1.115–1.235, p = 8.475 × 10-10]. Furthermore, neither horizontal pleiotropy nor heterogeneity was detected in the analysis.
Genetic evidence from the study supports a potential causal link between negative emotions and AF. The study suggests that negative emotions may elevate the risk of AF, and the escalation of negative emotions in AF patients is more likely attributable to modifiable factors rather than genetically related factors.
Graphical Abstract
Keywords
- atrial fibrillation
- negative emotions
- Mendelian randomization
Atrial fibrillation (AF), the most prevalent rapid cardiac arrhythmia encountered in clinical settings, can readily result in heart failure, stroke, and even life-threatening conditions. In the last five decades, the incidence of AF has tripled [1]. It is projected that by 2050, at least 72 million people in Asia are expected to be diagnosed with AF as medical technology advances and average life expectancy increases [2]. The symptoms associated with AF, coupled with the increasing prevalence of the AF population, have led to a growing number of individuals seeking healthcare services, thereby creating a substantial socioeconomic burden. Therefore, early detection and prevention strategies are essential for the effective management of AF.
Negative emotions, a pervasive health issue worldwide, encompass depression, anxiety, anger, and sadness, the most common of which are anxiety and depression. Anxiety is the most prevalent negative emotion, with an estimated global prevalence of approximately 7.3% [3]. Depression ranks among the top three leading causes of non-fatal health loss, affecting over 280 million individuals worldwide [4]. Several meta-analyses indicate that anxiety and depression are independent risk factors for cardiovascular disease, with a specific correlation observed between their occurrence and the development as well as adverse outcomes of cardiovascular disease [5, 6, 7]. There exists a complex relationship between negative emotions and AF. Studies suggested that negative emotions may foster an environment conducive to the initiation and perpetuation of AF. Simultaneously, AF can induce varying degrees of negative emotions in patients [8, 9]. Although an increasing number of observational studies are investigating the link between negative emotions and the development of AF, the causal impact of negative emotions on AF remains contradictory [10, 11].
The majority of studies in this field are currently observational studies, primarily due to objective factors such as ethics and research methods. However, these studies can be impacted by confounding factors and reverse causation, which limits their ability to provide strong evidence for a causal relationship between negative emotions and AF. Genome-wide association studies (GWAS) offer a new approach to investigating the genetic basis of diseases by analyzing genetic variants in large populations to determine their association with diseases [12]. Mendelian randomization (MR), a method utilizing genetic variations as instrumental variables (IVs), allows for the assessment of causal links between exposures and outcomes [13]. In MR studies, allelic genes are randomly allocated at conception following Mendel’s second law, and mutations at specific loci occur before the onset of the disease. By effectively reducing biases that may arise in observational studies, MR studies have been widely used to explore potential connections between diseases, to identify new treatment methods and prevention strategies. In light of these considerations, our study uses GWAS data on negative emotions and AF to perform a two-sample MR analysis. This analysis aims to evaluate the association between negative emotions and AF, thereby overcoming the limitations of traditional observational studies and providing clarity on the causal relationship between negative emotions and AF.
The two-sample MR study must adhere to three core assumptions [14]: ① the relevance assumption: genetic variations must be strongly correlated with exposure; ② the exclusivity assumption: genetic variations have no direct association with outcome and can solely impact the outcome through exposure; and ③ the independence assumption: genetic variations are not related to confounding factors.
Our study utilized published GWAS data on negative emotions and AF and screened single nucleotide polymorphisms (SNPs) that meet the three core assumptions above as IVs for a two-sample MR analysis to uncover the causal link between negative emotions and AF. Considering the potential effect of reverse causation, we initially conducted an MR analysis with AF as the outcome and negative emotions as exposure. Subsequently, by treating AF as the exposure and negative emotions as the consequence, the causal link between AF and negative emotions was evaluated. The study employed the inverse variance weighted (IVW) method as the primary analysis method and simultaneously used Weighted median method, MR-Egger regression, Simple mode method, and Weighted mode method for analysis. Sensitivity analyses were carried out using various methods to ensure the reliability and robustness of the results (Fig. 1).
Fig. 1.
The flowchart illustrating the analyses for the two-sample Mendelian randomization (MR) study. Assumption ②, the exclusivity assumption; Assumption ③, the independence assumption; SNP, single nucleotide polymorphism.
The exposure and outcome data were extracted from the IEU Open GWAS databases (https://gwas.mrcieu.ac.uk/), and to minimize bias, both samples in this study were sourced from European populations. Specifically, the data for AF genetic variants were obtained from the study conducted by Nielsen et al. [15]. After examining genetic variants in 60,620 cases and 970,216 controls from six cohorts, they identified 111 genomic regions significantly associated with AF. This dataset comprises 33,519,037 SNPs and is openly accessible in the IEU Open GWAS database. Similarly, the data for the genetic variants associated with negative emotions were sourced from the IEU Open GWAS database, comprising 459,560 samples and 9,851,867 SNPs. Among these samples, there were 158,565 cases and 300,995 controls. Unfortunately, the negative emotions mentioned here mainly include anxiety and depression, as we did not find useful data regarding other negative emotions such as sadness and anger. All genetic association data for the study are presented in the Supplementary Table 1.
The best IVs were chosen in accordance with the two-sample MR core assumption to
guarantee the authenticity and precision of the causal connection between
negative emotions and AF. First, IVs were chosen from SNPs associated with the
exposure, and these SNPs needed to meet the following criteria: (i) SNPs were
significant loci at the genome-wide level (p
We performed a positive control MR analysis to illustrate the anticipated effect on outcome that has established a causal association with the exposure, thereby supporting the validity of the genetic instruments. Anxiety and depression have been linked to hypertension, so we included hypertension as an additional positive control outcome [18]. We selected ischemic stroke as a positive control outcome for AF, given that AF is the primary underlying cause of this type of stroke [19]. Only exposures showing anticipated associations with these favorable control outcomes were subjected to the primary MR analysis. Genetic association information for these positive control outcomes is available in the Supplementary Table 1.
The two-sample MR analysis of the screened SNPs was employed using R software (version 4.2.3; The R Foundation for Statistical Computing, Vienna, Austria) with the two-sample MR package and the MR-PRESSO package [20]. The IVW method was predominantly employed in this study to evaluate the causal correlation between AF and negative emotions. In cases where only one SNP was available, the Wald ratio method was utilized [21]. Additionally, as supplements, Weighted mode method, Simple mode method, MR-Egger regression approach, and Weighted median method were applied. The results were presented as odds ratio (OR), serving as the effect measure to portray the strength of the link between anticipated genetic susceptibility to exposure and the likelihood of outcome.
Several sensitivity analyses were conducted to ensure the robustness of the
findings. First, the heterogeneity of each SNP estimate was assessed using
Cochrane’s Q test. When the Q value was significant
(p
Initially, 44 SNPs associated with negative emotion at a genome-wide
significance (p
According to the result of the positive control analysis, genetic variations in
negative emotions were linked to a higher risk of hypertension (OR 2.327, 95%
confidence interval (CI) 1.046–5.176, p = 0.038). Additionally, when
functional variants of AF were utilized as instruments, the results showed that
genetic variations in AF were linked to an increased risk of ischemic stroke (OR
1.003, 95% CI 1.002–1.004, p = 2.173
The MR estimates of genetically predicted negative emotion-related features and
the risk of AF are presented in Table 1 and Fig. 2. Utilizing the IVW approach,
we observed evidence supporting a potential causal association between negative
emotions and AF (OR 1.173, 95% CI 1.115–1.235, p
= 8.475
Fig. 2.
Forest plot of results from forward and reverse Mendelian randomization (MR) analysis. Abbreviations: AF, atrial fibrillation; SNP, single-nucleotide polymorphism; OR, odds ratio; CI, confidence interval.
Fig. 3.
MR leave-one-out sensitivity analysis for negative emotions on AF. Notice: The MR estimates for negative emotions on AF using the inverse-variance-weighted (IVW) approach are depicted by circles, while the bars represent the 95% confidence range of the MR values. Abbreviations: AF, atrial fibrillation; MR, Mendelian randomization.
| Outcome | SNP | Methods | OR (95% CI) | p value | Heterogeneity | Pleiotropy |
| AF | 34 | IVW | 1.173 (1.115,1.235) | 8.475 |
0.417 | |
| Weighted median | 1.188 (1.104,1.278) | 3.841 |
||||
| MR-Egger | 1.036 (0.786,1.365) | 8.030 |
0.408 | 0.375 | ||
| Simple mode | 1.198 (1.029,1.394) | 2.6189 |
||||
| Weighted mode | 1.193 (1.035,1.374) | 2.017 |
||||
| Negative emotions | 102 | IVW | 1.002 (0.999,1.006) | 0.156 | 0.074 | |
| Weighted median | 1.002 (0.997,1.008) | 0.306 | ||||
| MR-Egger | 1.000 (0.994,1.006) | 0.990 | 0.073 | 0.389 | ||
| Simple mode | 1.002 (0.991,1.013) | 0.747 | ||||
| Weighted mode | 1.001 (0.995,1.007) | 0.688 |
Abbreviations: AF, atrial fibrillation; SNP, single-nucleotide polymorphism; OR, odds ratio; CI, confidence interval; IVW, inverse variance weighted method; MR, Mendelian randomization.
A reverse two-sample MR analysis was carried out using the same methodology, with AF as exposure and negative emotions as the outcome, to determine whether AF had a causal effect on negative emotions. Using the SNP screening process mentioned above, 102 SNPs highly related to AF were ultimately comprised in the two-sample MR analysis (Supplementary Table 4). The results of the IVW method (OR 1.002, 95% CI 0.999–1.006, p = 0.156) indicated no proof of a genetic susceptibility to AF being associated with the occurrence of negative emotions. This finding was consistent across other MR approaches, including MR-Egger regression, Weighted median method, Simple mode, and Weighted mode, which were also negative results, indicating that there is probably no causal relationship between genetically forecast characteristics related to AF and the risk of negative emotions. Furthermore, sensitivity analysis did not find significant heterogeneity or horizontal pleiotropy (Table 1, Fig. 2).
Our study assessed the causal link between negative emotions and AF by identifying SNPs from publicly available GWAS data for a two-sample MR analysis. The results indicated that an inherited susceptibility to negative emotions was linked to a heightened probability of developing AF. Interestingly, despite numerous observational studies demonstrating that patients with AF are more likely to experience anxiety and depression [25, 26], our study did not find an inverse relationship. Genetic variants that increased the risk of AF did not correspond to an increased risk of negative emotions. This finding gives us a new perspective, suggesting that modifiable factors rather than hereditary ones may have a greater impact on the increase of negative emotions in patients with AF.
Although there are few researches on whether negative emotions can cause AF, and
the evidence is inconsistent, negative emotions like anxiety, depression, and
anger have been demonstrated to significantly affect the heart, being considered
both independent risk factors and triggers for AF episodes [8, 27, 28]. A
meta-analysis showed that anxiety and depression, two frequent negative emotions,
were associated with a 10% and 25% increase in the risk of AF, respectively,
while anger was linked to a 15% increase in AF incidence [29]. In a 13-year
follow-up study involving 6644 participants with baseline data on depressive
symptoms and no history of AF, Garg et al. [30] found that a Centre for
Epidemiological Studies Depression Scale score of
Although anxiety, depression, and other negative emotions are prevalent in AF patients, the findings of the current study suggest that genetic variants linked to an increased risk of AF do not correspond to an increased risk of negative emotions. The findings may offer hope, suggesting that the rise in depression among patients with AF could be influenced more by modifiable factors than genetically related ones. Numerous studies have demonstrated that interventions such as patient education, physical exercise, and comprehensive disease management approaches significantly reduce the incidence of anxiety and depression in AF patients [33, 34, 35]. In a randomized controlled trial, Hendriks et al. [36] discovered that a novel nurse-led integrated chronic care approach notably improved the quality of life, including reductions in anxiety and depression, among patients with AF compared to conventional care, along with enhancing AF-related knowledge in the nurse-led care group. Similar findings were observed in another study where AF patients were provided with “mobile AF”, a mobile phone app integrating clinical decision support tools, AF-related knowledge, and patient involvement in decision-making to enhance treatment and care for AF patients [37]. AF patients may experience anxiety and depression due to various factors, including fear of surgery, potential medication side effects (such as stomach bleeding or brain hemorrhage), and the condition of AF and its complications. Patients with prolonged negative emotions are prone to autonomic nervous system dysfunction, neuroendocrine disorders, inflammatory responses, and overactivation of the hypothalamic-pituitary-adrenal (HPA) axis [8, 27]. These pathophysiological changes create an environment conducive to triggering and perpetuating AF. Patients with AF complicated by anxiety, depression, and other negative emotions face an increased risk of recurrent AF after radiofrequency ablation, and their quality of life is significantly diminished [38, 39]. Therefore, identifying negative emotions in AF patients and implementing strategies to reduce them may improve clinical outcomes, and patient quality of life, and alleviate the economic burden associated with AF. This necessitates healthcare professionals’ attention to assessing patients’ psychological status, improving the diagnosis and treatment rate of negative emotional disorders such as anxiety or depression, and adopting a more holistic approach to managing AF patients, rather than solely focusing on medications and surgical treatments, as has been the practice in the past.
Despite efforts to minimize the effects of reverse causality and confounding variables, there are additional limitations in our study. Unfortunately, the study cannot elucidate the association between anger/sadness and AF due to the unavailability of GWAS data for these emotions in the database. Furthermore, the applicability of the study’s findings to other demographics remains uncertain, as the GWAS data analyzed were derived from a European population.
In summary, our study suggests that genetic susceptibility to negative emotions correlates with a heightened risk of developing AF. However, genetic variants linked to increased AF risk do not correspond to an increased risk of negative emotions. The escalation of negative emotions in patients with AF is more likely attributable to the relationship between how patients perceive their disease and how their disease is managed.
The authors confirm that this study analyzed publicly available datasets. These data can be found here: IEU OpenGWAS project (https://gwas.mrcieu.ac.uk/, accessed on 5 August 2023).
XTS and LZZ designed the research study. WSLH and CT provided help and advice on collecting data and explaining results. XTS, YQP, and HL analyzed the data. XTS and LZZ wrote the manuscript. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.
Not applicable.
We sincerely thank the IEU OpenGWAS database and Jonas B Nielsen et al. and other related researchers for providing GWASs summary data.
This work was funded by Technological Research Special Project of Sichuan Provincial Administration of Traditional Chinese Medicine (2023MS146).
The authors declare no conflict of interest.
Supplementary material associated with this article can be found, in the online version, at https://doi.org/10.31083/j.rcm2510356.
References
Publisher’s Note: IMR Press stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.