Article Text

Mendelian randomisation shows diverticular disease and irritable bowel syndrome increase the risk of haemorrhoidal disease
  1. Simonas Juzenas1,2,
  2. David Ellinghaus2,
  3. Serena Sanna3,4,
  4. Andre Franke2,5,
  5. Mauro D'Amato6,7,8
  1. 1Institute of Biotechnology, Life Science Centre, Vilnius University, Vilnius, Lithuania
  2. 2Institute of Clinical Molecular Biology, Kiel University, Kiel, Germany
  3. 3Institute for Genetic and Biomedical Research (IGBR), Consiglio Nazionale delle Ricerche (CNR), Cagliari, Italy
  4. 4University of Groningen, Department of Genetics, University Medical Center Groningen, Groningen, The Netherlands
  5. 5University Hospital Schleswig-Holstein (UKSH), Kiel, Germany
  6. 6Department of Medicine and Surgery, LUM University, Casamassima, Italy
  7. 7Ikerbasque, Basque Foundation for Science, Bilbao, Spain
  8. 8Gastrointestinal Genetics Lab, CIC bioGUNE – BRTA, Derio, Spain
  1. Correspondence to Professor Mauro D'Amato, Department of Medicine and Surgery, Università LUM Giuseppe Degennaro, Casamassima, Puglia, Italy; damato{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Recently in Gut, Zhu and colleagues1 investigated causality of genetic risk effects across three genetically correlated conditions: haemorrhoidal disease (HEM), diverticular disease (DIV) and IBS. Using genome-wide association study (GWAS) summary statistics from our previous HEM GWAS2 and other studies,3–5 they conducted unidirectional two-sample Mendelian randomisation (MR) and concluded ‘HEM might not be a trigger for developing diverticular disease or IBS, providing novel insights into a modest genetic correlation in conditions. Simultaneously, the observed causal (protective, ed.) effects of HEM on diverticular disease and IBS requires further exploration’. Although we did not originally make any claims of causality from the observed genetic correlations, the findings of Zhu et al are surprising and not in line with previous results,2 hence we set out to investigate further the relationship between HEM, DIV and IBS. Here, we briefly report an analysis that differs from Zhu et al in that: (1) we used best practice two-sample MR analysis,6 (2) causality was studied bi-directionally (testing HEM both as exposure (cause) and outcome (effect) versus DIV and IBS risk) and (3) summary statistics from the largest, recent GWAS meta-analysis including >53 000 IBS cases were used to test genetic risk effects of HEM relative to IBS (online supplemental material).7 By these means, we obtained results different from Zhu et al and new evidence of causative effects of DIV and (to a lesser extent) IBS towards the risk of HEM. Figure 1, online supplemental figure S1 and S2 illustrate the outcome of our analyses, outlined below.

Supplemental material

Figure 1

Bi-directional Mendelian randomisation (MR) analyses indicate a causal relationship between the effects of SNPs associated with DIV and IBS, respectively, on the development of HEM. The graph illustrates the associations between (A) IBS and HEM as well as (B) DIV and HEM, with ORs and their corresponding 95% CIs shown on the x-axis. The y-axis represents HEM as either an exposure or an outcome variable. MR analyses were conducted by using inverse variance weighted (IVW), MR Egger, weighted median, simple mode, weighted mode and CAUSE methods. Significance levels: *p<0.05; **p<0.01; ***p<0.001; else not significant. DIV, diverticular disease; HEM, haemorrhoidal disease.

We first optimised the instrument variable (IV) selection for MR by using consensus thresholds and methods6 to select genetic variants that are: associated with the exposure (p<5×10−8 or p<5×10−5); not in linkage disequilibrium (LD) (r2<0.001 across a 1 Mb window); and not horizontally pleiotropic (having no direct association with the outcome) (pMR-PRESSO<0.05). Then, using the selected variants as IVs, we performed bi-directional two-sample MR analyses8 9 using inverse variance weighted (IVW), MR Egger, weighted median, simple mode, weighted mode and CAUSE methods. Additionally, to ensure the robustness of the causal effects, we performed leave-one-out IVW regression analysis, which shows that causal effects are not driven by a few outlying genetic variants (online supplemental material).

Hence, this pipeline was first applied to testing HEM as exposure towards the risk of developing DIV or IBS (outcome) as in Zhu et al. In contrast to Zhu et al, no significant findings were obtained in our analysis in this direction (see HEM exposure in figure 1). However, when either DIV or IBS were tested as exposure on HEM as outcome, significant observations were made using multiple MR models (see HEM outcome in figure 1). In particular, MR based on the IVW method resulted in strongest and most significant findings, both for IBS and for DIV showing predisposing effects on HEM. Remarkably, DIV was associated with nearly five times increased HEM risk (OR=4.8; p=2×10−4). While their results were limited to testing HEM as exposure, discrepancies with Zhu et al are most likely due to the fact that they used IVs variants that are in LD and therefore not independent (as required for MR analyses; a detailed analysis of methodological issues is reported in the online supplemental material).

In summary, we show that DIV and IBS significantly predispose to HEM. This is in line with real-world data, as a most recent colonoscopy study also demonstrated: not only were patients with DIV more likely to have internal HEM (OR=2.21) but also those with more than 10 diverticula had the highest risk of HEM (OR=3.33).10 Indeed, DIV was deemed by the authors even more important than diet or constipation for internal HEM development.

Ethics statements

Patient consent for publication

Ethics approval

Not applicable.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Twitter @SimonasJuzenas, @damato_mauro

  • Contributors SS, AF and MD designed and supervised the study; SJ and DE contributed to data acquisition, statistical and computational analyses; SJ, DE, SS, AF and MD analysed and interpreted data; SJ and MD drafted the manuscript, with input and critical revision from all other authors. All authors approved the final draft of the manuscript.

  • Funding The study received funds through AF’s grantFR 2821/19-1 from the German Research Foundation (DFG). SS is supported by Next Generation EU grant Project Age-It (DM MUR 1557 11.10.2022), ERC Starting Grant 2022 (101075624) and NutrAGE grant (DM MUR 844 16.07.2021). SJ is supported by MSCA-IF-2020 grant (101030265). The study received infrastructure support from the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) Cluster of Excellence 2167 'Precision Medicine in Chronic Inflammation (PMI)' (EXC 2167-390884018).

  • Competing interests MD has received unrestricted research grants and consulting fees from QOL Medical LLC.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.