In light of the current circumstances, this MR study meticulously accounted for the full array of modifiable risk factors for IBD. Gut microbiota abundance serves as a robust indicator of IBD risk. Furthermore, MS, PSC, and AS were significantly linked to elevated IBD risk. Notably, potential associations between body fat percentage, whole body fat mass, primary hypertension, and increased IBD risk. Conversely, the comprehensive analysis indicated that omega-3 fatty acids, asthma, type 2 diabetes, and college or university degree were tentatively associated with reduced IBD risk. Upon adjusting for alcohol intake, no significant association between omega-3 fatty acids, asthma, RA, MS, type 2 diabetes, and IBD risk, hinting that these associations may be influenced by alcohol intake. These illuminating findings provide a clearer understanding of the prevention and management of IBD.
In recent years, several MR studies investigating the risk factors of IBD [26–29], yet they all possess limitations. For instance, Yang et al. found causal association between MS and IBD risk. However, the causal relationship remains uncertain, as false-positive inferences due to commonly correlated pleiotropy are a prevalent concern[28]. Additionally, Xie Y et al. observed a heightened association between PSC and an increased IBD risk[29]. Notably, their inclusion of a multi-ethnic population may have introduced the potential risk of population stratification, consequently leading to spurious associations between exposure and variation. Another MR study highlighted the imbalance of gut microbiota as IBD risk factor, based on a single inverse variance, potentially biasing the interpretation[30]. Furthermore, a one-sample MR study revealed causal association between asthma onset in childhood and reduced IBD risk in adults, which have less statistical power compared to the two-sample approach[31]. Similarly, investigations into food intakes encountered limitations stemming from sample size issues[32]. In our study, we endeavored to mitigate the aforementioned limitations through specific approaches. Firstly, we exclusively analyzed genetic variants from individuals of European ancestry as data sources to minimize potential bias. Additionally, to bolster conclusiveness, we combined the results from three databases. A MR study revealed positive association between alcohol intake and CD risk. However, sensitivity analyses did not confirm this association, emphasizing the need to address horizontal pleiotropy and heterogeneity. Furthermore, a prospective cohort study uncovered that moderate beer consumption was linked to reduced CD risk, while liquor consumption was associated with increased UC risk[33]. While our study found suggestive association of alcohol consumption with IBD risk in discovery data, no significant association was observed in combined results. Nevertheless, the increase in sample size enhanced statistical power, rendering our integrated results potentially more representative.
One study found no causal association between gut microbiota and IBD risk[34], while another study confirmed the opposite[30]. This discrepancy prompts further investigation. In our research, we sought to reconcile these conflicting findings by integrating results from three databases. The results (FinnGen dataset) revealed causal relationship between gut microbiota and IBD risk, also been found in combined results. This underscores further investigation into the relationship between gut microbiota and IBD risk, delving into the types of intestinal flora, their distribution within the intestine, and the causal relationship between intestinal flora and specific subtypes of IBD such as CD or UC.
We were able to ascertain causal association between MS and IBD risk, a finding reinforced by the FinnGen dataset and our combined dataset. Similar outcomes were observed for PSC and AS, with our results aligning with prior study[29]. However, another MR analysis found no causal association of AS with IBD risk, may stem from database updates and large sample sizes[35]. Additionally, a causal role of RA in IBD risk was found[36], whereas our research indicated association of RA with reduced IBD risk, a trend confirmed in the combined results. Consequently, we firmly believe that the robustness of our findings. Moreover, another study demonstrated that protective effect of asthma in reduced IBD risk[31], align with our findings.
Prior study found that serum omega-6 containing metabolites have causal association with CD risk[9]. Subsequently, MR investigation that delved into causal metabolites linked to multiple autoimmune diseases but did not focus on omega-3 fatty acids[37]. Therefore, our study incorporates omega-3 fatty acids into the analysis. Diminished association between omega-3 fatty acids and IBD risk that was subsequently corroborated in combined results. Omega-3 fatty acids play a multifaceted role in human physiology. Firstly, they promote the excretion of cholesterol from the stool and inhibit the synthesis of lipids and lipoproteins in the liver. Secondly, they are involved in arachidonic acid metabolism. These actions culminate in increased excretion and decreased synthesis of lipoproteins, leading to reduction in adiposity measures. This aligns with our outcomes, which evidenced strong association between body fat percentage and whole-body fat mass and increased IBD risk. However, our study did not find a significant association between obesity and IBD risk. This discrepancy can be attributed to the fact that body fat percentage and whole-body fat mass are correlated with obesity, determined by the degree of obesity or body mass index. Furthermore, obesity can be characterized by abdominal obesity and peripheral obesity. Importantly, our study did not account for the impact of fat distribution, which may have influenced the result.
Our study represents a pioneering effort in illustrating the causal association between primary hypertension, type 2 diabetes, and IBD risk using MR and GWAS. We found that higher causal association between primary hypertension and increased IBD risk. We also found that lower association between type 2 diabetes and decreased IBD risk. The heterogeneous causes of secondary diabetes, the presence of multiple complications, and variations in the duration of onset emphasize the need for more precise prospective studies to thoroughly examine the impact of type 2 diabetes on IBD risk. In conclusion, our study has made valuable contributions to the understanding of intricate relationships between primary hypertension, type 2 diabetes, and IBD risk.
Our study possesses several strengths. Firstly, the comprehensive approach to MR encompassing the discovery, validation, and meta-analysis stages represents a significant strength. This comprehensive methodology enhances credibility and reliability of the results, providing valuable insights into causal relationships between risk factors and IBD. Furthermore, the incorporation of multiple supplementary and sensitivity analyses to validate the instrumental variable assumptions for traits is another strength. By addressing potential issues such as pleiotropy, outliers, and sample overlap, our study demonstrates a high level of statistical rigor and robustness, bolstering the credibility of causal associations identified. The focus on a population of European ancestry also represents a strength, as it helps mitigate potential population stratification biases, thereby enhancing the internal validity of the findings. Additionally, to minimize sample overlap between exposure and outcome data sources stands out as a distinctive strength. By maintaining a low-down rate of sample overlap, our study effectively mitigates the risk of weak instrument bias, thereby enhancing the robustness and reliability of the causal associations identified. These strengths underscore the significance and reliability of our findings.
Our study possesses several limitations. The limitation related to the sample size being restricted to European populations is an important consideration. While this focus enhances the internal validity of the findings within this demographic, it also underscores the need for future research to evaluate modifiable risks of IBD in other racial and ethnic groups. Additionally, the recognition of the need for future studies to identify SNPs associated with IBD severity and analyze the relationship of modifiable risk factors with IBD severity is an important research direction. While the absence of studies on the association of specific SNPs with IBD severity precluded correlation analysis in our study, this limitation highlights an area for potential future research to delve deeper into the nuanced impact of modifiable risk factors on IBD severity. By openly acknowledging these limitations, our study demonstrates a commitment to transparency and rigor, while also providing valuable insight into areas for further investigation.