Credible Mendelian Randomization Studies in the Presence of Selection Bias Using Control Exposures

Association between trans fatty acids and COVID-19: A multivariate Mendelian randomization study

Traditional observational studies have suggested a potential association between trans fatty acids (TFAs), which are considered to be health-damaging fatty acids, and coronavirus disease 2019 (COVID-19). However, whether there is a causal relationship between them is currently unclear. We aimed to investigate the causal link between genetically determined TFAs and COVID-19. We performed univariate and multivariate Mendelian randomization (MR) studies using summary statistics from the European Pedigree TFAs (n = 8013), COVID-19 susceptibility (n = 159 840), COVID-19 hospitalization (n = 44 986), and COVID-19 severity (n = 18 152) genome-wide association studies (GWAS). The inverse variance weighted (IVW) method was used as the primary MR analysis, and several other methods were used as supplements. In univariate MR analysis, higher levels of circulating trans, cis-18:2 TFAs were positively associated with a higher COVID-19 hospitalization rate (p < 0.0033; odds ratio [OR] = 1.637; 95% confidence interval [CI]: 1.116-2.401) and COVID-19 severity (p < 0.0033; OR = 2.575; 95% CI: 1.412-4.698). Furthermore, in multivariate MR analysis, trans, cis-18:2 had an independent and significant causal association with a higher COVID-19 hospitalization rate (p = 0.00044; OR = 1.862; 95% CI = 1.316-2.636) and COVID-19 severity (p = 0.0016; OR = 2.268; 95% CI = 1.361-3.779) after the five TFAs were adjusted for each other. Together, our findings provide evidence that trans, cis-18:2 TFAs have an independent and robust causal effect on COVID-19 hospitalization and severity.

Roles of 21 Genera of Human Gut Microbiota in Barrett's Esophagus Risk: A Mendelian Randomization Study

Background: Lack of definitive evidence supports the putative hypothesis that gut microbiota dysbiosis is associated with Barrett’s esophagus (BE). We conducted a two-sample Mendelian randomization study to assess the associations of 21 genera of human gut microbiota with BE.

Methods: We identified independent genetic instruments for 21 genera of gut microbiota (including nine dominant genera, four core genera among individuals of European ancestry, and eight esophagus-specific genera of gut microbiota) from MiBioGen (up to 18,340 participants). We applied them to summary statistics from the largest publicly available genome-wide association study on BE (9,680 cases and 31,211 controls). We obtained the causal estimates of genetically predicted higher genera of gut microbiota and BE using the inverse variance weighting method. Sensitivity analyses included weighted median, MR-Egger, MR-RAPS, and MR-PRESSO.

Results: We found that genetically predicted higher Actinomyces (OR: 0.76 per unit increase in log odds of having BE, 95% CI: 0.70–0.83) and higher Ruminiclostridium (OR: 0.75, 95% CI: 0.63–0.90) were significantly associated with a lower risk of BE. No associations of other genera of gut microbiota with BE were noted, apart from suggestive associations of higher Alistipes (OR: 0.77; 95% CI: 0.61–0.99), higher Eubacterium (OR: 0.89; 95% CI: 0.80–0.99), and higher Veillonella (OR: 0.76; 95% CI: 0.56–1.02) with a lower risk of BE, and higher Faecalibacterium (OR: 1.15; 95% CI: 0.99–1.33) with a higher risk of BE.

Conclusion: This study suggests that higher Actinomyces and higher Ruminiclostridium might protect against BE.