|
Depression and 24 gastrointestinal diseases: a Mendelian randomization study
Translational Psychiatry volume 13, Article number: 146 (2023) Cite this article
Abstract
The causality of the association between depression and gastrointestinal diseases is undetermined. We conducted Mendelian randomization (MR) analyses to systematically explore the associations of depression with 24 gastrointestinal diseases. Independent genetic variants associated with depression at the genome-wide significance level were selected as instrumental variables. Genetic associations with 24 gastrointestinal diseases were obtained from the UK Biobank study, the FinnGen study, and large consortia. Multivariable MR analysis was conducted to explore the mediation effects of body mass index, cigarette smoking, and type 2 diabetes. After multiple-testing corrections, genetic liability to depression was associated with an increased risk of irritable bowel syndrome, non-alcohol fatty liver disease, alcoholic liver disease, gastroesophageal reflux, chronic pancreatitis, duodenal ulcer, chronic gastritis, gastric ulcer, diverticular disease, cholelithiasis, acute pancreatitis, and ulcerative colitis. For the causal effect of genetic liability to depression on non-alcoholic fatty liver disease, a substantial proportion was mediated by body mass index. Genetic predisposition to smoking initiation mediated half of effect of depression on acute pancreatitis. This MR study suggests that depression may play a causal role in many gastrointestinal diseases.
우울증과 위장 질환의 연관성에 대한 인과관계는 아직 밝혀지지 않았습니다. 우리는 24가지 위장 질환과 우울증의 연관성을 체계적으로 조사하기 위해 멘델 무작위 배정(MR) 분석을 실시했습니다. 게놈 전체 유의성 수준에서 우울증과 관련된 독립적인 유전자 변이를 도구 변수로 선정했습니다. 24개 위장 질환과의 유전적 연관성은 영국 바이오뱅크 연구, 핀젠 연구 및 대규모 컨소시엄에서 얻었습니다. 체질량 지수, 흡연, 제2형 당뇨병의 매개 효과를 탐색하기 위해 다변량 MR 분석을 실시했습니다.
다중 보정 후 우울증에 대한 유전적 책임은
과민성 대장 증후군,
비알코올 지방간 질환,
알코올성 간 질환,
위식도 역류,
만성 췌장염,
십이지장 궤양,
만성 위염,
위궤양,
게실 질환,
담석증,
급성 췌장염 및
궤양성 대장염의 위험 증가와 관련이 있는 것으로 나타났습니다.
비알코올성 지방간 질환에 대한 우울증에 대한 유전적 책임의 인과적 영향의 경우 상당 부분이 체질량 지수에 의해 매개되었습니다. 흡연 시작에 대한 유전적 소인은 급성 췌장염에 대한 우울증 효과의 절반을 매개했습니다. 이 MR 연구는 우울증이 많은 위장 질환에서 인과적인 역할을 할 수 있음을 시사합니다.
Similar content being viewed by others
Genomic data in the All of Us Research Program
Article Open access19 February 2024
The serotonin theory of depression: a systematic umbrella review of the evidence
Article Open access20 July 2022
Article Open access19 February 2024
Introduction
Depression is a common and serious mental illness that limits psychosocial functioning and compromises life quality [1]. The prevalence of digestive system disease has been found to be higher in depressive patients compared to the general population [2, 3]. Most observational studies have investigated the role of gastrointestinal disorder in development of depression [4, 5], but limited on the reverse impact. Previous cohort studies found that depression was associated with an increased risk of irritable bowel syndrome [6], gastroesophageal reflux [7], and peptic ulcer [8]. Evidence from the Nurses’ Health Studies also found that self-reported depressive symptoms were associated with an increased risk of Crohn’s disease but not ulcerative colitis [9]; however, an association between new-onset depression and ulcerative colitis was revealed in another study [10]. The inconsistent findings as well as the limitations of observational studies, like residual confounding and reverse causation, hinder the causal assessment of the associations between depression and gastrointestinal diseases.
우울증은 흔하고 심각한 정신 질환으로 심리사회적 기능을 제한하고 삶의 질을 떨어뜨립니다[1]. 소화기계 질환의 유병률은 일반 인구에 비해 우울증 환자에서 더 높은 것으로 밝혀졌습니다 [2, 3]. 대부분의 관찰 연구는 우울증 발병에 있어 위장 장애의 역할을 조사했지만[4, 5], 그 반대의 영향에 대해서는 제한적이었습니다.
이전의 코호트 연구에 따르면 우울증은
과민성 대장 증후군[6],
위식도 역류[7],
소화성 궤양[8]의 위험 증가와 관련이 있는 것으로 나타났습니다.
간호사 건강 연구에서도 자가 보고된 우울 증상이 궤양성 대장염은 아니지만 크론병의 위험 증가와 관련이 있는 것으로 나타났지만[9], 다른 연구에서는 새로 발병한 우울증과 궤양성 대장염 사이의 연관성이 밝혀졌습니다[10]. 일관되지 않은 연구 결과와 잔존 혼동 및 역 인과 관계와 같은 관찰 연구의 한계는 우울증과 위장 질환 사이의 인과 관계를 평가하는 데 방해가 됩니다.
Mendelian randomization (MR) is a method that employs genetic variants as instrumental variables for the exposure to infer the causality of an exposure-outcome association [11]. Compared to conventional observational studies, MR is by nature less prone to confounding since genetic variants are randomly assorted at conception and, therefore, unrelated to environmental factors. In addition, this method can minimize reverse causation as germline phenotypes cannot be modified by disease status. Although a phenome-wide MR study found some associations of depression with inflammatory and hemorrhagic gastrointestinal diseases [12], the effects of depression on a broad range of gastrointestinal outcomes have not been investigated. Here, we performed an MR study to examine the associations of genetic liability to major depressive disorder with 24 gastrointestinal diseases. To reveal possible mechanistic pathway, we further conducted multivariable MR analysis to examine the mediations of body mass index, tobacco smoking, and type 2 diabetes mellitus.
멘델 무작위 배정(MR)은 유전적 변이를 노출의 도구 변수로 사용하여 노출-결과 연관성의 인과관계를 추론하는 방법입니다[11]. 기존의 관찰 연구와 비교했을 때, MR은 유전적 변이가 임신 시 무작위로 분류되어 환경적 요인과 무관하기 때문에 본질적으로 혼동 가능성이 적습니다. 또한 이 방법은 질병 상태에 따라 생식세포 표현형이 변형될 수 없기 때문에 역 인과관계를 최소화할 수 있습니다. 표현형 전체에 걸친 MR 연구에서 우울증과 염증성 및 출혈성 위장 질환의 연관성이 일부 발견되었지만[12], 광범위한 위장 질환에 대한 우울증의 영향은 조사되지 않았습니다. 이에 저희는 주요 우울장애와 24가지 위장 질환에 대한 유전적 책임의 연관성을 조사하기 위해 MR 연구를 수행했습니다. 가능한 기계적인 경로를 밝히기 위해 체질량 지수, 흡연, 제2형 당뇨병의 매개 변수를 조사하기 위해 다변량 MR 분석을 추가로 수행했습니다.
Method
Figure 1 shows the overview design of the study. This MR investigation was based on publicly available genome-wide association study (GWAS) consortia (Table S1). All MR analyses were performed separately in each dataset, including the UK Biobank study [13], the FinnGen study [14], and other large consortia if available. Individual MR estimates for each gastrointestinal endpoint were pooled. Included studies had been approved by corresponding institutional review boards and ethical committees.
Fig. 1: Study design.
BMI body mass index, GERA Genetic Epidemiology Research on Aging, IIBDGC International Inflammatory Bowel Disease Genetics Consortium, MR Mendelian randomization, MR-PRESSO Mendelian randomization pleiotropy residual sum and outlier, SNP single nucleotide polymorphisms.
Instrumental variable selection
Genetic instrumental variables for major depressive disorder were extracted from the latest GWAS, which meta-analyzed data in 807,553 individuals (246,363 depressive cases and 561,190 controls predominantly of European ancestry) from the UK Biobank study, 23andMe, and Major Depressive Disorder Working Group of the Psychiatric Genomics Consortium (PGC) [15]. In total, 98 single nucleotide polymorphisms (SNPs) associated with depression at the genome-wide significance threshold (P < 5 × 10−8) and without linkage disequilibrium (defined as r2 > 0.01) were identified. Detailed information on used SNPs is presented in Table S2. Odds ratio (ORs) and CIs of outcome were scale to per one-unit increase in log odds of liability to depression.
Gastrointestinal disease data sources
Genetic associations with 24 gastrointestinal diseases were obtained from the UK Biobank study [13], the FinnGen study [14], and two large consortia, including the International Inflammatory Bowel Disease Genetics Consortium (IIBDGC) [16] and Genetic Epidemiology Research on Aging (GERA) [17]. The UK Biobank is a large multicenter cohort study comprising half a million individuals recruited between 2006 and 2010 in the United Kingdom [13]. The database collects information on a wide range of health-related variables, including self-reported information, clinically validated data, and register-based data. Summary statistics of European ancestry in UK Biobank were obtained from GWAS conducted by the Lee lab where the gastrointestinal outcomes were defined by codes of the International Classification of Diseases 9th Revision (ICD-9) and ICD-10. The genetic associations were adjusted for sex, birth year, and the first four genetic principal components. As for the FinnGen study, the latest summary-level genetic data (R7 release) on gastrointestinal diseases were obtained [14]. The FinnGen study involves the collection and analysis of genetic data from over 500,000 participants from the Finnish biobanks, along with their digital health record data from the Care Register for Health Care, and information from the cancer, cause of death, and medication reimbursement registries. The gastrointestinal endpoints were defined by ICD-8, ICD-9, and ICD-10 codes. Genome-wide association analyses were adjusted for sex, age, genetic components, and genotyping batch in FinnGen. Detailed diagnostic codes in UK Biobank and FinnGen are listed in Table S3. We also obtained summary-level data from the IIBDGC [16] for Crohn’s disease (5956 cases and 14,927 controls) and ulcerative colitis (6968 cases and 20,464 controls) and from the GERA for irritable bowel syndrome (3117 cases and 53,520 controls) [17]. Diagnosis of IBD in IIBDGC was based on accepted radiologic, endoscopic, and histopathologic evaluations. GERA used longitudinal electronic health records to obtain clinical information of individuals.
Data sources for possible mediators
Depression has been associated with body mass index (BMI) [18], cigarette smoking [19], and the risk of type 2 diabetes mellitus [20]. In addition, BMI, cigarette smoking and type 2 diabetes mellitus have been associated with a wide range of gastrointestinal diseases in our previous MR studies [21,22,23]. Thus, we considered these three factors potential mediators. The genetic instrumental variables of BMI, smoking initiation, and type 2 diabetes mellitus were respectively extracted from publicly available GWASs [24,25,26], and the detailed information can be found in Table S1. Independent (linkage disequilibrium r2 > 0.01) SNPs associated with BMI, smoking initiation, and type 2 diabetes at genome-wide significance threshold (P < 5 × 10−8) were selected as instrumental variables.
Statistical analysis
SNPs were excluded if unavailable in outcome datasets or defined as ambiguous (i.e, palindromic SNPs with minor allele frequencies >0.42 and <0.58). The primary MR analysis was conducted by the inverse-variance weighted (IVW) method under a multiplicative random effects model. Assuming that all SNPs are valid instruments, the IVW method provides the most precise estimates. Estimates for each outcome from different sources were combined using the fixed-effects meta-analysis. Heterogeneity among estimates of SNPs was evaluated by Cochran’s Q value. To detect potential horizontal pleiotropy and examine the consistency of the associations, three sensitivity analyses including the weighted median [27], MR-Egger [28], and Mendelian randomization pleiotropy residual sum and outlier (MR-PRESSO) [29] analyses were performed. The weighted median method can provide consistent estimates if more than 50% of the weight in the analysis comes from valid genetic instruments [27]. MR-Egger regression provides an MR estimate with adjustment for horizontal pleiotropy detected by its intercept test [28]. MR-PRESSO method can detect SNP outliers with pleiotropic effects and provide an estimate identical to that from IVW after removal of these outliers [29]. The Benjamini-Hochberg method that controls the false discovery rate (FDR) was applied to correct for multiple testing. The association with a nominal P < 0.05 but Benjamini–Hochberg adjusted P > 0.05 was regarded suggestive and the association with a Benjamini–Hochberg adjusted P < 0.05 were deemed significant. All analyses were two-sided and performed using the TwoSampleMR [30], MendelianRandomization [27], and MRPRESSO [29] R packages in R software 4.1.2.
To investigate possible pathways linking depression to gastrointestinal diseases, we conducted a two-step MR analysis [31] to explore the mediation effects of BMI, cigarette smoking, and type 2 diabetes mellitus using multivariable MR analysis. The two-step MR analysis was only performed for significant MR associations in primary analysis. In detail, we first obtained the MR effect estimates for depression on each mediator using the IVW method. Then the multivariable MR was performed to estimate the effect of three mediators on risk of gastrointestinal diseases with adjustment for depression. These two estimates for each gastrointestinal disease were multiplied together to estimate the indirect effect of depression. Finally, the proportion of the total effect explained by the mediators was calculated through dividing the mediated effect by the total effect. We also performed the same multivariable MR analysis for outcomes with the nonsignificant associations to reveal the potential association of depression with gastrointestinal disease independent of the mediators.
Results
Genetic liability to depression was positively associated with 12 of the 24 studied gastrointestinal diseases and these associations remained after multiple comparison corrections (Fig. 2 and Table S4). In detail, genetic predisposition to depression was associated with higher odds of irritable bowel syndrome (OR 1.58; 95% CI: 1.42–1.76; P = 1.42 × 10−16), non-alcohol fatty liver disease (OR 1.46; 95% CI: 1.15–1.85; P = 0.002), alcoholic liver disease (OR 1.44; 95% CI: 1.12–1.85; P = 0.004), gastroesophageal reflux (OR 1.40; 95% CI: 1.30–1.52; P = 2.43 × 10−16), chronic pancreatitis (OR 1.38; 95% CI: 1.08–1.77; P = 0.01), duodenal ulcer (OR 1.37; 95% CI: 1.15–1.64; P = 4.89 × 10−4), chronic gastritis (OR 1.37; 95% CI: 1.15–1.63; P = 3.30 × 10−4), gastric ulcer (OR 1.34; 95% CI: 1.17–1.54; P = 2.97 × 10−5), diverticular disease (OR 1.31; 95% CI: 1.21–1.43; P = 2.89 × 10−10), cholelithiasis (OR 1.25; 95% CI: 1.14–1.37; P = 6.67 × 10−7), acute pancreatitis (OR 1.25; 95% CI: 1.05–1.48; P = 0.011, and ulcerative colitis (OR 1.20; 95% CI: 1.06–1.36; P = 0.003). The results of the sensitivity analysis were generally consistent (Table S5). The MR-Egger intercept tests found the indication of horizontal pleiotropy for diverticular disease in the UK Biobank study (P for MR-Egger intercept <0.05, Table S5) but not for any other outcomes in neither of sources. MR-PRESSO detected 1 to 2 outliers in the analysis for cholelithiasis; however, the association persisted after removal of the SNPs (Table S5).
Fig. 2: Associations of genetic liability to depression with 24 gastrointestinal diseases.
*Significant association after multiple testing. The estimate of irritable bowel syndrome was meta-analysis by combining estimates from the UK Biobank study, the FinnGen study and the Genetic Epidemiology Research on Aging consortium; the estimates of Crohn’s disease and ulcerative colitis were meta-analysis by combining estimates from the UK Biobank study, the FinnGen study and the International Inflammatory Bowel Disease Genetics Consortium; the estimates of other gastrointestinal disease were meta-analysis by combining estimates from the UK Biobank study and the FinnGen study.
Genetic liability to depression was significantly associated with higher levels of BMI and the higher risk of smoking initiation and type 2 diabetes mellitus (Table S6). Table 1 display the results of multivariable MR analyses and mediation effect of individual mediators and combinations of three mediators. We notice that BMI (49.28%) and combination of three mediators (53.54%) mediated quite a proportion of effect of depression on non-alcoholic fatty liver disease. For the causal effect of depression on acute pancreatitis, the proportion mediated by smoking initiation was up to 45.02%.
Table 1 Estimates of depression on gastrointestinal diseases mediated by potential mediators.
In multivariable MR analysis adjusting for genetically predicted BMI, genetic liability to depression was significant associated with increased risk of acute gastritis (OR 1.49; 95% CI: 1.22–1.81; P = 7.35 × 10−5) and cirrhosis (OR 1.28; 95% CI: 1.10–1.48; P = 0.001). When adjusting for genetically predicted smoking initiation (OR 1.35; 95% CI: 1.11–1.65; P = 0.003) and type 2 diabetes mellitus (OR 1.40; 95% CI: 1.11–1.77; P = 0.005) separately, genetic liability to depression was positively associated with cirrhosis (Table S7).
Discussion
We performed a comprehensive MR investigation on the associations of genetic liability to depression with 24 gastrointestinal diseases. We found that genetic liability to depression was associated with the increased risk of 12 gastrointestinal diseases. Multivariable MR analyses indicated that association between depression and non-alcoholic fatty liver disease were substantially mediated by BMI. Genetically prediction to smoking initiation mediated half of effect of depression on acute pancreatitis.
The current MR investigation corroborated previous epidemiological studies’ findings that depression was associated with an increased risk of irritable bowel syndrome [6], non-alcoholic fatty liver disease [32], gastroesophageal reflux [7], gastric ulcer and duodenal ulcer [8]. However, previous evidence on the association between depression and alcoholic liver disease is inconclusive. A cross-sectional study including 398 patients with alcoholic liver disease found that depression was not associated with alcoholic liver disease [33]; however, another study found that the prevalence and incidence of alcoholic liver disease were higher in patients with depression [34]. Our MR analysis found a positive association of genetic liability to depression with alcoholic liver disease. The unmeasured confounding and relatively small sample size might account for the discrepancy. An MR study found that genetic liability depression proxied by 19 SNPs associate with depression at P < 5 × 10−6 was positively associated with both Crohn’s disease and ulcerative colitis risk [35].
저희는 우울증에 대한 유전적 책임과 24가지 위장 질환의 연관성에 대해 포괄적인 MR 조사를 실시했습니다. 그 결과, 우울증에 대한 유전적 책임이 12가지 위장 질환의 위험 증가와 관련이 있는 것으로 나타났습니다. 다변량 MR 분석 결과, 우울증과 비알코올성 지방간 질환 사이의 연관성은 BMI에 의해 상당 부분 매개되는 것으로 나타났습니다. 흡연 시작에 대한 유전적 예측은 우울증이 급성 췌장염에 미치는 영향의 절반을 매개했습니다.
이번 MR 조사는
우울증이
과민성 대장 증후군[6],
비알코올성 지방간 질환[32],
위식도 역류[7],
위궤양 및 십이지장 궤양의 위험 증가와 관련이 있다는
이전 역학 연구 결과를 확증했습니다[8].
그러나
우울증과 알코올성 간 질환의 연관성에 대한 이전의 증거는 결정적이지 않습니다. 알코올성 간질환 환자 398명을 대상으로 한 단면 연구에서는 우울증이 알코올성 간질환과 관련이 없는 것으로 나타났지만[33], 다른 연구에서는 우울증 환자에서 알코올성 간질환의 유병률과 발병률이 더 높다는 사실이 밝혀졌습니다[34]. 저희의 MR 분석에서는 우울증과 알코올성 간 질환에 대한 유전적 책임이 긍정적인 연관성이 있음을 발견했습니다. 측정되지 않은 혼동 요인과 상대적으로 작은 표본 크기가 이러한 불일치를 설명할 수 있습니다. MR 연구에 따르면 19개의 SNP로 대리되는 유전적 책임 우울증은 P < 5 × 10-6에서 우울증과 관련이 있으며 크론병 및 궤양성 대장염 위험과 모두 양의 상관관계가 있는 것으로 나타났습니다 [35].
Our study replicated the positive association between depression and ulcerative colitis. However, a neutral association for Crohn’s disease was identified in this study with a larger sample size and updated instruments. The discrepancy may be caused by undetected horizontal pleiotropy introduced by using SNPs associated with the exposure at a relaxed threshold. Besides, the different definitions of depression in the original GWAS where the instruments were extracted may also attribute to this discrepancy. A phenome-wide MR study in the UK Biobank revealed that major depressive disorder was associated with increased risks of gastroesophageal reflux disease, non-infectious gastroenteritis, and gastrointestinal hemorrhage [12], which supports our findings. Our MR investigation refined the gastrointestinal classification and provided novel findings for gastric ulcer, duodenal ulcer, and chronic gastritis. The associations of depression with acute and chronic pancreatitis were also novel findings, which need to be verified.
본 연구는
우울증과 궤양성 대장염 사이의 긍정적인 연관성을 재현했습니다.
그러나 이번 연구에서는 더 큰 표본 규모와 업데이트된 도구를 사용하여 크론병에 대한 중립적인 연관성이 확인되었습니다. 이러한 불일치는 느슨한 임계값에서 노출과 관련된 SNP를 사용함으로써 발견되지 않은 수평적 다형성 때문에 발생할 수 있습니다. 또한, 기기가 추출된 원래 GWAS에서 우울증에 대한 정의가 다른 것도 이러한 불일치의 원인이 될 수 있습니다. 영국 바이오뱅크의 표현형 전반에 걸친 MR 연구에 따르면 주요 우울 장애는 위식도 역류 질환, 비감염성 위장염, 위장관 출혈의 위험 증가와 관련이 있는 것으로 나타났는데[12], 이는 이번 연구 결과를 뒷받침합니다. MR 조사를 통해 위장관 분류를 세분화하고 위궤양, 십이지장궤양, 만성 위염에 대한 새로운 발견을 했습니다. 우울증과 급성 및 만성 췌장염의 연관성 또한 새로운 발견으로, 이에 대한 검증이 필요합니다.
Previous studies have suggested that nicotine in tobacco may have certain beneficial effects on patients with depression, which included relief of stress and depressive affect, and feeling pleasurable sensations [36]. Besides, nicotine cessation may result in withdrawal symptoms such as anhedonia and depression [37]. However, the health benefits of quitting smoking are immediate and long-lasting and evidence from current study showed that smoking mediated part of effect of depression on gastrointestinal diseases. Considering the complicated relationship between smoking and depression, the risks and benefit assessment of quitting was required for depression patients. The current study also uncovered that genetic liability to depression was associated increased risks of acute gastritis and cirrhosis. This suggested that the depression may have no effects on these diseases if the depression does not result in BMI increase.
이전 연구에 따르면 담배의 니코틴은 우울증 환자에게 스트레스와 우울한 감정을 완화하고 즐거운 감각을 느끼는 등 특정 유익한 효과가 있을 수 있다고 합니다[36]. 또한 니코틴을 끊으면 무감각증과 우울증과 같은 금단 증상이 나타날 수 있습니다[37]. 그러나 금연의 건강상의 이점은 즉각적이고 오래 지속되며, 최근 연구에 따르면 흡연이 우울증이 위장 질환에 미치는 영향의 일부를 매개한다는 증거가 있습니다. 흡연과 우울증 사이의 복잡한 관계를 고려할 때 우울증 환자에 대한 금연의 위험 및 이익 평가가 필요했습니다. 이번 연구에서는 또한 우울증에 대한 유전적 책임이 급성 위염과 간경변의 위험 증가와 관련이 있다는 사실도 밝혀졌습니다. 이는 우울증이 BMI 증가를 초래하지 않는다면 우울증이 이러한 질병에 영향을 미치지 않을 수 있음을 시사합니다.
The current study quantified the mediation effects of BMI, smoking initiation, and type 2 diabetes in the associations between genetic liability to depression and gastrointestinal disease risk. Our findings suggest that the prevention strategies on these three mediators might partly counteract the detrimental effects of depression on many gastrointestinal diseases. The findings of our MR investigation have implications for public health policy that psychologists should pay more attention to gastrointestinal disease screening and prevention for patients with depressive disorder.
이번 연구는 우울증과 위장병 위험에 대한 유전적 책임의 연관성에서 BMI, 흡연 시작, 제2형 당뇨병의 매개 효과를 정량화했습니다. 우리의 연구 결과는 이 세 가지 매개체에 대한 예방 전략이 많은 위장 질환에 대한 우울증의 해로운 영향을 부분적으로 상쇄할 수 있음을 시사합니다. MR 조사 결과는 심리학자들이 우울 장애 환자의 위장 질환 검사와 예방에 더 많은 관심을 기울여야 한다는 공중 보건 정책에 영향을 미칩니다.
In addition to mediated pathways by BMI, smoking, and diabetes, several biological mechanisms might explain the direct effect of depression on gastrointestinal disease. Autonomic dysfunction in depression results in dysfunction of gastric acid secretion [38], which leads to gastroesophageal reflux and peptic ulcer disease [39]. Chronic stress activates the neuroendocrine response to produce cortisol, which disturbs the balance of gut microflora and leads to bowel inflammation [40]. Imbalance in the gut microbiota composition may contribute to intestinal as well as extraintestinal diseases via perturbed microbiota that produces multiple substances including neuropeptides, hormones, and short-chain fatty acids [41].
BMI, 흡연, 당뇨병에 의한 매개 경로 외에도 여러 가지 생물학적 메커니즘이 우울증이 위장 질환에 미치는 직접적인 영향을 설명할 수 있습니다.
우울증의 자율 기능 장애는
위산 분비 기능 장애를 초래하고[38],
이는 위식도 역류와 소화성 궤양 질환으로 이어집니다[39].
만성 스트레스는
신경 내분비 반응을 활성화하여
코르티솔을 생성하고,
이는 장내 미생물의 균형을 방해하고 장 염증을 유발합니다 [40].
장내 미생물 구성의 불균형은
신경 펩타이드,
호르몬,
단쇄 지방산 등
여러 물질을 생성하는 미생물의 교란을 통해
장뿐만 아니라 장외 질환의 원인이 될 수 있습니다 [41].
The major strength of the current study is MR design, which minimizes biases caused by residual confounding and reverse causality. In addition, we examined the associations in two or more independent sources. The results from these data sources were generally consistent, which makes it unlikely that the observed associations were caused by chance. We explored the mediating pathways by conducting multivariable MR analysis, which deepened the mechanistic understandings and provided evidence supports for prevention strategies.
이번 연구의 가장 큰 강점은 잔존 혼동 및 역인과관계로 인한 편향을 최소화하는 MR 설계입니다. 또한 두 개 이상의 독립적인 출처에서 연관성을 조사했습니다. 이러한 데이터 소스의 결과는 대체로 일관성이 있었기 때문에 관찰된 연관성이 우연에 의한 것일 가능성은 낮았습니다. 다변량 MR 분석을 수행하여 매개 경로를 탐색함으로써 기계론적 이해를 심화시키고 예방 전략에 대한 증거를 제공했습니다.
이 연구에는 인정해야 할 한계가 있습니다. 가장 큰 한계는 수평적 플레오트로피, 즉 유전적 변이가 노출뿐만 아니라 결과에 영향을 미칠 수 있다는 것을 완전히 배제할 수 없다는 것입니다. 하지만 몇 가지 민감도 분석을 수행한 결과 연관성이 안정적이라는 사실을 발견했습니다. 또한 MR-Egger 및 MR-PRESSO 분석에서 검출된 플레오트로피의 데이터는 제한적으로 관찰되었습니다. MR은 노출을 대리하기 위해 수정할 수 없는 유전적 변이를 활용한다는 점을 감안할 때, 이 설계는 평생 누적 효과를 반영합니다. 그럼에도 불구하고 우울증은 치료 없이도 시간이 지남에 따라 증상이 부분적으로 완화되는 역동적인 자연 경과를 보입니다. 따라서 이 결과는 관찰 결과와 직접 비교하여 임상 진료에 적용할 수 없습니다. 또 다른 한계는 모든 위장 질환에 대한 도구 변수가 부족하여 우울증과 위장 질환 사이의 양방향 연관성을 조사하지 않았다는 점입니다. 또한 영국 바이오뱅크 연구는 노출 및 결과 데이터 세트에 모두 포함되었기 때문에 관찰 연관성에 대한 MR 추정치가 편향될 수 있습니다. 그러나 대부분의 연관성은 FinnGen 연구에서 복제되었기 때문에 샘플 중복으로 인한 편향은 최소화될 것으로 보입니다. 마지막으로, 본 연구는 유럽 혈통의 개인을 대상으로 진행되었기 때문에 다른 인구집단에 대한 일반화 가능성을 제한할 수 있다는 점에 유의할 필요가 있습니다.
결론적으로, 저희의 MR 조사는 우울증에 대한 유전적 책임이 12가지 위장 질환 결과의 위험 증가와 관련이 있다는 증거를 제공했습니다. BMI, 흡연, 제2형 당뇨병은 이러한 많은 연관성을 매개하는 것으로 나타났습니다.
This study has limitations that warrant acknowledgment. A major limitation is that we could not completely rule out horizontal pleiotropy, which means genetic variants influence the outcome not or not only through the exposure. However, we performed several sensitivity analyses and found the associations were stable. In addition, we observed limited data of pleiotropy detected by MR-Egger and MR-PRESSO analyses. Given that MR utilizes genetic variants that cannot be modified to proxy the exposure, the design reflects the cumulative lifelong effect. Nevertheless, depression has a dynamic natural course in which symptoms partially remit over time even without treatment. Thus, the results might not be directly compared with observational findings and applicable to clinical practice. Another limitation is that we did not examine the bidirectional associations between depression and gastrointestinal diseases due to the lack of instrumental variables for all gastrointestinal diseases. In addition, the UK Biobank study was included in both the exposure and outcome datasets, which might bias MR estimates toward the observational associations. However, most associations were replicated in the FinnGen study, which indicated that the bias caused by sample overlap should be minimal. Finally, it’s important to note that our study was conducted in the individuals of European ancestry, which might limit the generalizability of our findings to other populations.
In conclusion, our MR investigation provided evidence that genetic liability to depression was associated with increased risks of 12 gastrointestinal disease outcomes. BMI, smoking, and type 2 diabetes mellitus appeared to mediate many of these associations.
Data availability
All data analyzed in this study can be obtained by a reasonable request to corresponding authors.
Code availability
Codes used in current study are available from the corresponding author on reasonable request.
References
|