Inflammatory bowel disease and rosacea: causal association analysis using bi-directional Mendelian randomization

doi:10.21203/rs.3.rs-2237564/v1

Download PDF

Article

Inflammatory bowel disease and rosacea: causal association analysis using bi-directional Mendelian randomization

https://doi.org/10.21203/rs.3.rs-2237564/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 09 Sep, 2023

Read the published version in Scientific Reports →

You are reading this latest preprint version

Background:

The association between rosacea and inflammatory bowel disease (IBD) has been studied in previous observational studies. It is unclear, however, whether the association is causal or not.

Methods:

Independent genetic variants for IBD were chosen as instruments from published GWAS studies involving 38155 cases with an IBD diagnosis and 48485 controls in order to investigate the causal effect of IBD on rosacea. Summarized data for rosacea were gathered from various GWAS studies that included 1195 cases and 211139 controls without rosacea. Reverse-direction MR analysis was done to investigate the relationship between genetically proxied rosacea and IBD. With the use of the inverse variance-weighted (IVW), MR-Egger, and weighted median approaches, a 2-sample Mendelian randomization study was carried out. Analysis of heterogeneity and sensitivity was performed to examine the pleiotropy and robustness of effect estimates.

Results: The forward-direction of the MR study was to reveal that genetic predisposition to IBD (OR: 1.1291; 95% CI: 1.0444 to 1.2206), UC (OR: 1.2030; 95% CI: 1.0867 to 1.3318) and CD (OR: 1.1291; 95% CI: 1.0444 to 1.2206; p=0.0023) was associated with an increased risk of rosacea. The reverse-direction MR analyses did not demonstrate that a genetic predisposition to rosacea was not associated with total IBD (OR: 0.9683; 95% CI: 0.9112 to 1.0291), UC (OR: 0.9714; 95% CI: 0.8949 to 1.0545) and CD (WM: OR: 0.9751; 95% CI: 0.8971 to 1.0598; p=0.5525).

Conclusion: Our findings provided evidence for a causal impact of IBD, UC, and CD on rosacea, but not vice versa. The elevated incidence of rosacea in patients with IBD should be recognized by doctors to make an early diagnosis and initiate specialized therapy.

Health sciences/Diseases/Gastrointestinal diseases

Health sciences/Diseases/Skin diseases

Biological sciences/Computational biology and bioinformatics

Biological sciences/Genetics

Health sciences/Medical research

inflammatory bowel disease

Crohn’s disease

ulcerative colitis

rosacea

Mendelian randomization

Rosacea is a chronic inflammatory skin condition that produces erythema of the central face region that can be transitory or permanent(1). The prevalence of rosacea was estimated at 5.46% in adults, especially those between ages 45 and 60(2). Rosacea can negatively impact self-esteem, confidence, and quality of life in the case of persistent facial erythema(3). Rosacea has been linked to cardiometabolic illness, Helicobacter pylori infection, Demodex infection, inflammatory bowel disease, migraine, depression, and anxiety in several studies(3–9).

Inflammatory bowel disease (IBD) including Crohn’s disease (CD) and ulcerative colitis (UC), is a chronic and recurring inflammatory disorder of the intestine. Meta-analyses and observational studies support an underlying bidirectional relationship between rosacea and IBD, although mechanisms remain unclear(8, 10–13). Many common genetic, environmental, and pathophysiologic factors are shared by rosacea and IBD(14). Nevertheless, basic and observational studies are susceptible to confounding factors like demographics or environmental exposure including age, gender, smoking, alcohol consumption, and body mass index (BMI)(14–17). Hence, rosacea and IBD remain controversial in terms of any causal bidirectional role.

The causal relationship between exposure and outcome can be assessed by using Mendelian randomization (MR) with genetic variations as instrumental variables(18). This approach simulates the randomization process of a randomized controlled experiment by exploiting the random distribution of genetic variation to eliminate confounding factors and reverse causality(18).In the present study, MR was used to evaluate the causal relationship between rosacea and IBD, along with its strength and direction of causality.

Multiple single-nucleotide polymorphisms (SNPs) representing genetic variation were selected as instrumental variables and analyzed using two-sample MR. The following three hypotheses were adopted (Fig. 1): 1. Instrumental variables (IVs) are directly and robustly related to exposure; 2. IVs are independent of any confounding factors; 3. IVs only affected outcomes via exposure(19). MR analysis was conducted to assess bidirectional causal relationships between IBD and rosacea.

2.1. Data source

Summary statistics for rosacea were obtained from the FinnGen Biobank Analysis Consortium 2021 which contained 1195 cases and 211139 individuals without rosacea (FinnGen Biobank Analysis Consortium; GWAS ID: finn-b-L12_ROSACEA; n = 212334 (1,195 cases and 211,139 controls); https://gwas.mrcieu.ac.uk/datasets/finn-b-L12_ROSACEA/). Summary statistics for IBD were obtained from The IIBDGC Consortium which contained a total sample size of 86640 participants of predominantly European ancestry (cases/controls for IBD: 38155/48485; UC: m17647/47179; CD: 20550/41642; https://www.ibdgenetics.org/#downloads)(20).

The SNPs and corresponding summary statistics used in our analyses were selected from studies enrolling only individuals with European ancestry for both IBD and Rosacea to avoid population bias. Because all of the data utilized is already in the public domain, no further ethical approval is required.

A series of quality control steps were implemented to select eligible IVs. The IVs for IBD were selected from the SNPs associated with IBD at genome-wide significance (p < 5×10^− 8) in the dataset. The SNPs were clumped to achieve independent loci with a threshold linkage disequilibrium (LD) r² = 0.001 and 10,000 kb distance(21). Then, the effect estimates of the selected IVs from the rosacea dataset were extracted. The SNPs for rosacea were selected that were genome-wide significant (p < 5 × 10^− 6) and independently inherited (r2 < 0.001, distance > 10000 kb) without LD from finn-b-L12_ROSACEA GWAS dataset. Then, the corresponding effect estimates of the selected IVs from the IBD GWAS dataset were extracted. In the context of IBD-Rosacea relationship, Rosacea-relevant traits such as cardiometabolic disease, Helicobacter pylori, Demodex infection, inflammatory bowel disease, migraine, depression, and anxiety(4–9) and IBD-relevant traits such as Helicobacter pylori and obesity(22, 23) are most likely to be potential and major confounders. To fulfill the second MR assumption, the PhenoScannerV2 database was used to search for IVs and their proxied traits and the IVs arrogating these traits at a threshold of p < 5×10⁸ and r2 > 0.80 were removed(24, 25).To harmonize the IBD and Rosacea data, all palindromic SNPs with intermediate allele frequencies were removed from the selected instrumental SNPs above(26). These rigorously selected SNPs were used as the final instruments for the MR analyses. Detailed information on IVs is shown in Supplementary Tables 1–5.

2.2. Estimation of causal effect

After the list of SNPs according to the selection criteria above was determined, a forward MR analysis was conducted to estimate the overall effects of the selected SNPs for IBD on Rosacea using multiple methods, including inverse variance weighted (IVW), MR-Egger regression, weighted median (WM), simple mode and weighted mode. Since the results could be biased by horizontal pleiotropy of the IVs, the validity of the results was checked by comparing the effect estimates across the five MR methods.

The reverse-direction MR analysis was then conducted using the same MR methods as above. Effect estimates were reported in odds ratios (OR) with 95% confidential intervals (CI).

2.3. Heterogeneity and sensitivity analysis

SNP heterogeneity was quantified using Cochran Q-statistics and I²-values(27). "Leave-one-out" analysis was used to assess the robustness of MR estimates by eliminating a different SNP in each iteration to quantify the causal influence of outlying SNPs and to ensure that deleting SNPs did not affect the MR estimates.(28). Pleiotropy RESidualSum and Outlier (MR-PRESSO) analysis was applied to outliers of instrumental variables with pleiotropic effects(29).

2.4 instrumental strength and statistical power

An estimation of variance was performed using the formula R² = 2×MAF×(1 − MAF)×(beta/SD)², where MAF denotes the effect allele frequency and beta is the allele effect estimate of the genetic variant(30). The F statistic was calculated using the formula F = [(N - k − 1)/k] × [R²/ (1 - R²)], where k denotes the number of SNPs, N is the sample size, and R² denotes the proportion of exposure variability explained by genotype(31). The F-statistic > 10 indicates a sufficiently strong instrument to explain phenotypic variations, whereas the F-statistic ≤ 10 indicates a weak instrument(32).

2.5 Statistical analysis

Statistical analyses were performed using the TwoSampleMR (version 0.5.6) and MR-PRESSO (version 1.0) packages in R (version 4.2.1), and plots were created by the ggplot2 package. All statistical tests were performed on both sides, and the findings of the MR analysis and sensitivity analysis regarding exposures and outcomes were judged statistically significant at p < 0.05. The concept and processes for this MR study are illustrated in Fig. 1.

3.1. The causal effect of IBD and its subtypes on Rosacea

To investigate the causative influence of IBD on Rosacea risk, significant at the genome-wide level (p < 5 × 10 ^− 8) and independently inherited (r² < 0.001 and distance > 10,000 kb) from the 12,716,084 SNPs, 63 SNPs were tentatively selected as IVs for IBD. By enquiring about the PhenoScanner V2 database, three SNPs related with at least one Rosacea-relevant trait, such as rs6062496, were ruled out. 10 SNPs were not available in the summary-level datasets of Rosacea. 6 potential SNPs were found to be palindromic. Finally, 44 SNPs were approved for MR analysis of IBD's causative influence on rosacea risk. Similarly, according to the above processes and standards, 25 and 34 SNPs were screened as genetic instruments for UC and CD respectively. Based on F statistics and the proportion of variance explained (R²), all genetic instruments were suitable for MR analysis.Supplementary Tables 1–5 provide a summary and full descriptions of the IVs for each exposure.

Cochran's Q test revealed no significant heterogeneity in the analysis of total IBD on rosacea. (Q = 54.1571; p = 0.1184). Thus, fixed-effect models were applied in the IVW analysis. There was no directional pleiotropy bias found in the MR-Egger test (Intercept = 0.0018; p = 0.9408). No SNP outliers were found in the MR-PRESSO global test (RSSobs = 56.5382; p = 0.1400) and leave-one-out MR analysis. There was a strong association discovered between overall IBD and rosacea (IVW: OR: 1.1291; 95% CI: 1.0444 to 1.2206; p = 2.30E-03). WM (WM: OR: 1.1455; 95% CI: 1.0226 to 1.2832; p = 0.019) and Weighted mode (Weighted mode: OR: 1.1408; 95% CI: 1.0072 to 1.2921; p = 0.0461) confirmed the IBD-rosacea association.

The two main subtypes of IBD, UC, and CD, were also analyzed. Cochran's Q test revealed no significant heterogeneity in the analysis of UC on rosacea (Q = 20.8205; p = 0.6493). No directional pleiotropy bias was detected (Intercept=-0.0239; p = 0.4728). No SNP outliers were found in the MR-PRESSO global test (RSSobs = 22.9609; p = 0.6440) and leave-one-out MR analysis. UC was causally associated with rosacea (IVW: OR: 1.2030; 95% CI: 1.0867 to 1.3318; p = 0.0004). WM (WM: OR: 1.2135; 95% CI: 1.0466 to 1.4070; p = 0.0104) confirmed the UC-rosacea association.

Cochran's Q test revealed no significant heterogeneity in the analysis of CD on rosacea (Q = 28.6333; p = 0.6844). No directional pleiotropy bias was detected (Intercept = 0.0027; p = 0.8912). No SNP outliers were found in the MR-PRESSO global test (RSSobs = 29.9833; p = 0.7020) and leave-one-out analysis. CD was also causally associated with rosacea (IVW: OR: 1.1291; 95% CI: 1.0444 to 1.2206; p = 0.0023). WM (WM: OR: 1.1455; 95% CI: 1.0227 to 1.2830; p = 0.0188) confirmed the CD-rosacea association.

3.2 The causal effect of rosacea on IBD and its subtypes

In the reverse-direction MR analysis, significant at the genome-wide level (p < 5 × 10 ^− 6) and independently inherited (r² < 0.001 and distance > 10,000 kb) from the 16,380,452 SNPs, 13 SNPs were tentatively selected as IVs for rosacea. No SNP associated with IBD-relevant traits was found by inquiring about the PhenoScanner V2 database. One SNP rs2298897 was excluded from the analysis because of being palindromic. Finally, 12 SNPs were accepted for MR analysis evaluating the causative influence of rosacea on IBD, UC, and CD risk.

Based on F statistics (F = 4.63) and the proportion of variance explained (R² = 0.03%), our reverse-direction MR analyses would be affected by weak instrument bias. Supplementary Tables 1–5 provide a summary and full descriptions of the IVs for each exposure.

Cochran's Q test revealed no significant heterogeneity in the analysis of rosacea on total IBD (Q = 6.4603; p = 0.1184). In the MR-Egger test, no directional pleiotropy bias was detected (Intercept=-0.0004; p = 0.9761). No SNP outliers were found in the MR-PRESSO global test (RSSobs = 7.2150; p = 0.8560) and leave-one-out MR analysis. However, no direct causal effect of rosacea on total IBD was found (IVW: OR: 0.9818; 95% CI: 0.9390 to 1.0267; p = 0.4209). The results from the WM (WM: OR: 0.9683; 95% CI: 0.9112 to 1.0291; p = 0.2999) and MR Egger (MR Egger: OR: 0.9828; 95% CI: 0.9086 to 1.0631; p = 0.6745) were consistent.

The two main subtypes of IBD, UC, and CD, were also analyzed. Cochran's Q test revealed no significant heterogeneity in the analysis of UC on rosacea (Q = 7.9962; p = 0.7136). No directional pleiotropy bias was detected (Intercept = 0.0062; p = 0.7300). No SNP outliers were found in the MR-PRESSO global test (RSSobs = 9.3873; p = 0.7590) and leave-one-out MR analysis. No evidence of a causal effect UC on rosacea was found (IVW: OR: 0.9941; 95% CI: 0.9392 to 1.0523; p = 0.8390).WM (WM: OR: 0.9714; 95% CI: 0.8949 to 1.0545; p = 0.4882) and MR Egger (MR Egger: OR: 0.9793; 95% CI: 0.8857 to 1.0828; p = 0.6921) confirmed these findings.

Cochran's Q test revealed no significant heterogeneity in the analysis of CD on rosacea (Q = 10.5728; p = 0.4797). No directional pleiotropy bias was detected (Intercept=-0.0123; p = 0.5194). No SNP outliers were found in the MR-PRESSO global test (RSSobs = 12.2924; p = 0.5280) and leave-one-out MR analysis. There was no causal association between CD and rosacea (IVW: OR: 1.0078; 95% CI:0.9497 to 1.0694; p = 0.7982). The results from the WM (WM: OR: 0.9751; 95% CI: 0.8971 to 1.0598; p = 0.5525) and MR Egger (MR Egger: OR: 1.0373; 95% CI: 0.9352 to 1.1505; p = 0.5046) were consistent.

The sensitivity analyses of MR are shown in Table 1. The scatter plots of causal relationships of MR analysis are shown in Fig. 2. The causal relationships are shown in Fig. 3–4. The funnel plot, leave-one-out plot, and forest plot of MR analyses are shown in Figure S1-S3.

We employed MR analysis to investigate the bidirectional association between IBD and rosacea in this investigation. We found evidence that genetic predisposition to IBD, UC, and CD was linked to an elevated risk of rosacea. Besides, we did not find any evidence that the genetic predisposition to rosacea was linked to an elevated risk of IBD, UC, or CD.

Previous observational studies have evaluated the possible link between IBD and rosacea. Contradictory results were, however, reported. Although a positive association was found in rosacea with CD(33, 34), this has not been verified by other research in UC and rosacea(10, 11). A meta-analysis of 13 separate studies involving 5,051,356 participants conducted by Jing Han et al (2019) concluded that the relative risks for IBD, UC, and CD in patients with rosacea were 1.32 (95% CI, 1.18–1.49), 1.19 (95% CI, 1.02–1.38), and 1.52 (95% CI, 1.25–1.84) respectively, and the overall RR (95% CI) for rosacea in total IBD, UC, and CD patients were 1.66 (1.50–1.84), 1.69 (1.48–1.93) and 2.08 (1.26–3.46) respectively (P < 0.05)(8). Another meta-analysis by Fang-Ying Wang et al (2019) consisting of three case-control and three cohort studies, concluded that the meta-analysis of case-control studies found a marginally increased risk of CD(OR: 1.30; 95% CI:0.99 to 1.69) and a considerably increased risk of UC(OR: 1.64; 95% CI 1.43 to 1.89) in rosacea patients, and the meta-analysis of cohort studies found a considerably increased odds of CD(hazard ratio (HR):1.58; 95% CI 1.14 to 2.20) and UC (HR 1.18, 95% CI 1.01 to 1.37) in rosacea patients(17). These reported associations and our findings differ to some extent. We assume that the difference is due to the two distinct analytical methods themselves. Unavoidable clinical confounding variables may alter exposure and outcome, reducing the capacity to make reliable causal inferences in observational research. As a result, even if observational studies discovered a strong association, it was unable to identify a direct causal connection. By using genetic instrumental variables, Mendelian randomization can avoid the impact of these confounding factors and produce a reasonably accurate causal evaluation.

Our findings of MR analysis provided evidence for a causal impact of IBD, UC, and CD on rosacea, but not vice versa. Additionally, our subanalyses verified that patients with UC and CD have a higher risk of developing rosacea. The underlying mechanisms of the link between IBD and rosacea have not yet been fully elucidated. IBD and rosacea are both chronic inflammatory conditions of the skin and gastrointestinal organs involving the interplay between genetic and immunological elements, this raises the possibility that they are related in some way. IBD and rosacea both have abnormalities in innate and adaptive immunity. and they share some common risk factors such as smoking status(15, 33), obesity(16, 22), and small intestinal bacterial overgrowth (SIBO)(35, 36).

We would want to highlight some of our study's merits while simultaneously addressing some of its limitations. The main merit of our work is that, to the greatest of our knowledge, we employed a 2-sample MR technique for the first time to explore the bidirectional relationship between rosacea and IBD. In comparison to observational studies, This strategy is less susceptible to confounding confounding factors, reverse causation, and exposures that do not differ across groups. Furthermore, the subtypes of inflammatory bowel disease are strictly defined to eliminate the effect of disease coexistence on outcomes. Third, a sensitivity analysis was carried out to make sure that causal estimates were consistent and that the results were reliable. There are certain limits as well. Because IVs were few, a lower p(p < 10^− 6) threshold was defined when rosacea was used as an instrumental variable for exposure. Moreover, the F-statistic test indicated that the instrument was weak. Thus, the negative result of our reverse-direction MR analyses should be interpreted with caution. Second, individuals of other ethnicities cannot be confidently predicted based on the current research because it was conducted on individuals of European ancestry. Finally, Although we matched every chosen SNP to the pheWAS database to identify potential confounding variables and related horizontal polymorphisms, this technique does not eradicate the influence of horizontal polymorphisms because many genetic variations' specific biological function is unclear. When higher-quality GWAS research becomes available, future studies will overcome these shortcomings.

Our findings of MR analysis provided evidence for a causal impact of IBD, UC, and CD on rosacea, but not vice versa. The primary IBD subtypes UC and CD appear to have a causal association with rosacea as well.

The elevated incidence of rosacea in patients with IBD should be recognized by doctors to make an early diagnosis and initiate specialized therapy. To further understand the pathophysiological processes behind this association, more research is required.

Data availabilityAll data that supporting the findings of this study were available in the public domain through FinnGen and IIBDGC Consortium.

Funding and Conflicts of interests

This work was financially supported by no funding. All authors declare no conflicts of interest.

Ethics statement

Ethical approval and informed consent were obtained from the original GWAS, no separate ethics statement was required for this study.

Acknowledgements

The authors would like to thank the FinnGen Biobank Analysis Consortium and the IIBDGC Consortium for making their GWAS summary-level information public. This manuscript has never been published in any publication and is not currently under reviewed by any other journal.

Tan J, Almeida LM, Bewley A, Cribier B, Dlova NC, Gallo R, et al. Updating the diagnosis, classification and assessment of rosacea: recommendations from the global ROSacea COnsensus (ROSCO) panel. The British journal of dermatology. 2017;176(2):431-8.
Gether L, Overgaard LK, Egeberg A, Thyssen JP. Incidence and prevalence of rosacea: a systematic review and meta-analysis. The British journal of dermatology. 2018;179(2):282-9.
Baldwin HE, Harper J, Baradaran S, Patel V. Erythema of Rosacea Affects Health-Related Quality of Life: Results of a Survey Conducted in Collaboration with the National Rosacea Society. Dermatol Ther (Heidelb). 2019;9(4):725 − 34.
Zhao YE, Wu LP, Peng Y, Cheng H. Retrospective analysis of the association between Demodex infestation and rosacea. Archives of dermatology. 2010;146(8):896–902.
Christensen CE, Andersen FS, Wienholtz N, Egeberg A, Thyssen JP, Ashina M. The relationship between migraine and rosacea: Systematic review and meta-analysis. Cephalalgia : an international journal of headache. 2018;38(7):1387-98.
Chen Q, Shi X, Tang Y, Wang B, Xie HF, Shi W, et al. Association between rosacea and cardiometabolic disease: A systematic review and meta-analysis. J Am Acad Dermatol. 2020;83(5):1331-40.
Jorgensen AR, Egeberg A, Gideonsson R, Weinstock LB, Thyssen EP, Thyssen JP. Rosacea is associated with Helicobacter pylori: a systematic review and meta-analysis. J Eur Acad Dermatol Venereol. 2017;31(12):2010-5.
Han J, Liu T, Zhang M, Wang A. The relationship between inflammatory bowel disease and rosacea over the lifespan: A meta-analysis. Clinics and research in hepatology and gastroenterology. 2019;43(4):497–502.
Chang HC, Huang YC, Lien YJ, Chang YS. Association of rosacea with depression and anxiety: A systematic review and meta-analysis. Journal of affective disorders. 2022;299:239 − 45.
Li WQ, Cho E, Khalili H, Wu S, Chan AT, Qureshi AA. Rosacea, Use of Tetracycline, and Risk of Incident Inflammatory Bowel Disease in Women. Clinical gastroenterology and hepatology : the official clinical practice journal of the American Gastroenterological Association. 2016;14(2):220-5 e1-3.
Kim M, Choi KH, Hwang SW, Lee YB, Park HJ, Bae JM. Inflammatory bowel disease is associated with an increased risk of inflammatory skin diseases: A population-based cross-sectional study. J Am Acad Dermatol. 2017;76(1):40 − 8.
Wu CY, Chang YT, Juan CK, Shieh JJ, Lin YP, Liu HN, et al. Risk of inflammatory bowel disease in patients with rosacea: Results from a nationwide cohort study in Taiwan. J Am Acad Dermatol. 2017;76(5):911-7.
Meier CR. Rosacea, inflammatory bowel disease and the value of big data and of epidemiological studies. The British journal of dermatology. 2017;176(1):9–10.
Holmes AD, Spoendlin J, Chien AL, Baldwin H, Chang ALS. Evidence-based update on rosacea comorbidities and their common physiologic pathways. J Am Acad Dermatol. 2018;78(1):156 − 66.
Abram K, Silm H, Maaroos HI, Oona M. Risk factors associated with rosacea. J Eur Acad Dermatol Venereol. 2010;24(5):565 − 71.
Aldrich N, Gerstenblith M, Fu P, Tuttle MS, Varma P, Gotow E, et al. Genetic vs Environmental Factors That Correlate With Rosacea: A Cohort-Based Survey of Twins. JAMA dermatology. 2015;151(11):1213-9.
Wang FY, Chi CC. Association of rosacea with inflammatory bowel disease: A MOOSE-compliant meta-analysis. Medicine. 2019;98(41):e16448.
Skrivankova VW, Richmond RC, Woolf BAR, Yarmolinsky J, Davies NM, Swanson SA, et al. Strengthening the Reporting of Observational Studies in Epidemiology Using Mendelian Randomization: The STROBE-MR Statement. Jama. 2021;326(16):1614-21.
Davies NM, Holmes MV, Davey Smith G. Reading Mendelian randomisation studies: a guide, glossary, and checklist for clinicians. Bmj. 2018;362:k601.
Liu JZ, van Sommeren S, Huang H, Ng SC, Alberts R, Takahashi A, et al. Association analyses identify 38 susceptibility loci for inflammatory bowel disease and highlight shared genetic risk across populations. Nature genetics. 2015;47(9):979 − 86.
Shen J, Zhou H, Liu J, Zhang Y, Zhou T, Yang Y, et al. A modifiable risk factors atlas of lung cancer: A Mendelian randomization study. Cancer medicine. 2021;10(13):4587 − 603.
Long MD, Crandall WV, Leibowitz IH, Duffy L, del Rosario F, Kim SC, et al. Prevalence and epidemiology of overweight and obesity in children with inflammatory bowel disease. Inflammatory bowel diseases. 2011;17(10):2162-8.
Tepler A, Narula N, Peek RM, Jr., Patel A, Edelson C, Colombel JF, et al. Systematic review with meta-analysis: association between Helicobacter pylori CagA seropositivity and odds of inflammatory bowel disease. Alimentary pharmacology & therapeutics. 2019;50(2):121 − 31.
Kamat MA, Blackshaw JA, Young R, Surendran P, Burgess S, Danesh J, et al. PhenoScanner V2: an expanded tool for searching human genotype-phenotype associations. Bioinformatics. 2019;35(22):4851-3.
Staley JR, Blackshaw J, Kamat MA, Ellis S, Surendran P, Sun BB, et al. PhenoScanner: a database of human genotype-phenotype associations. Bioinformatics. 2016;32(20):3207-9.
Lukkunaprasit T, Rattanasiri S, Ongphiphadhanakul B, McKay GJ, Attia J, Thakkinstian A. Causal Associations of Urate With Cardiovascular Risk Factors: Two-Sample Mendelian Randomization. Frontiers in genetics. 2021;12:687279.
Greco MF, Minelli C, Sheehan NA, Thompson JR. Detecting pleiotropy in Mendelian randomisation studies with summary data and a continuous outcome. Statistics in medicine. 2015;34(21):2926-40.
Burgess S, Thompson SG. Erratum to: Interpreting findings from Mendelian randomization using the MR-Egger method. European journal of epidemiology. 2017;32(5):391-2.
Verbanck M, Chen CY, Neale B, Do R. Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nature genetics. 2018;50(5):693-8.
Burgess S, Dudbridge F, Thompson SG. Combining information on multiple instrumental variables in Mendelian randomization: comparison of allele score and summarized data methods. Statistics in medicine. 2016;35(11):1880 − 906.
Xu L, Hao YT. Effect of handgrip on coronary artery disease and myocardial infarction: a Mendelian randomization study. Scientific reports. 2017;7(1):954.
Burgess S, Thompson SG, Collaboration CCG. Avoiding bias from weak instruments in Mendelian randomization studies. International journal of epidemiology. 2011;40(3):755 − 64.
Egeberg A, Weinstock LB, Thyssen EP, Gislason GH, Thyssen JP. Rosacea and gastrointestinal disorders: a population-based cohort study. The British journal of dermatology. 2017;176(1):100-6.
Spoendlin J, Karatas G, Furlano RI, Jick SS, Meier CR. Rosacea in Patients with Ulcerative Colitis and Crohn's Disease: A Population-based Case-control Study. Inflammatory bowel diseases. 2016;22(3):680-7.
Parodi A, Paolino S, Greco A, Drago F, Mansi C, Rebora A, et al. Small intestinal bacterial overgrowth in rosacea: clinical effectiveness of its eradication. Clinical gastroenterology and hepatology : the official clinical practice journal of the American Gastroenterological Association. 2008;6(7):759 − 64.
Rana SV, Sharma S, Malik A, Kaur J, Prasad KK, Sinha SK, et al. Small intestinal bacterial overgrowth and orocecal transit time in patients of inflammatory bowel disease. Digestive diseases and sciences. 2013;58(9):2594-8.

TABLE 1 | Sensitivity analyses of MR.

Exposure	Outcome	Number of Ivs	Heterogeneity test		MR-Egger pleiotropy test		MR-PRESSO results
			Q	P-value	Intercept	P-value	RSSobs	P-value	Outlier
IBD	Rosacea	44	54.1571	0.1184	0.0018	0.9408	56.5382	0.1400	None
UC	Rosacea	25	20.8205	0.6493	-0.0239	0.4728	22.9609	0.6440	None
CD	Rosacea	34	28.6333	0.6844	0.0027	0.8912	29.9833	0.7020	None
Rosacea	IBD	12	6.4603	0.8410	-0.0004	0.9761	7.2150	0.8560	None
Rosacea	UC	12	7.9962	0.7136	0.0062	0.7300	9.3873	0.7590	None
Rosacea	CD	12	10.5728	0.4797	-0.0123	0.5194	12.2924	0.5280	None
CD: Crohn’s Disease;IBD: MR: Mendelian randomization;Inflammatory Bowel Disease; UC: Ulcerative Colitis.

No competing interests reported.

Supplementarymaterial.docx

Download PDF

Journal Publication

published 09 Sep, 2023

Read the published version in Scientific Reports →

Editorial decision: Major revision
14 Dec, 2022
Reviews received at journal
12 Dec, 2022
Reviewers agreed at journal
01 Dec, 2022
Reviewers invited by journal
09 Nov, 2022
Editor assigned by journal
09 Nov, 2022
Editor invited by journal
09 Nov, 2022
Submission checks completed at journal
09 Nov, 2022
First submitted to journal
04 Nov, 2022

You are reading this latest preprint version

Inflammatory bowel disease and rosacea: causal association analysis using bi-directional Mendelian randomization

Status:

Journal Publication

Version 1

Abstract

Figures

1. Introduction

2. Materials And Methods

2.1. Data source

2.2. Estimation of causal effect

2.3. Heterogeneity and sensitivity analysis

2.4 instrumental strength and statistical power

2.5 Statistical analysis

3. Results

3.1. The causal effect of IBD and its subtypes on Rosacea

3.2 The causal effect of rosacea on IBD and its subtypes

4. Discussion

5. Conclusions

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1