Mendelian randomization analysis reveals higher whole body water mass may increase risk of bacterial infections.

Abstract

Background and purpose: The association of water loading with several infections remains unclear. Observational studies are hard to investigate definitively due to potential confounders. In this study, we employed Mendelian randomization (MR) analysis to assess the association between genetically predicted whole body water mass (BWM) and several infections.

Methods: BWM levels were predicted among 331,315 Europeans in UK Biobank using 418 SNPs associated with BWM. For outcomes, we used genome-wide association data from the UK Biobank and FinnGen consortium, including sepsis, pneumonia, intestinal infections, urinary tract infections (UTIs) and skin and soft tissue infections (SSTIs). Inverse-variance weighted MR analyses as well as a series of sensitivity analyses were conducted.

Results: Genetic prediction of BWM is associated with an increased risk of sepsis (OR 1.34; 95% CI 1.19 to 1.51; P = 1.57 × 10^- 6), pneumonia (OR: 1.17; 95% CI 1.08 to 1.29; P = 3.53 × 10^- 4), UTIs (OR: 1.26; 95% CI 1.16 to 1.37; P = 6.29 × 10^- 8), and SSTIs (OR: 1.57; 95% CI 1.25 to 1.96; P = 7.35 × 10^- 5). In the sepsis and pneumonia subgroup analyses, the relationship between BWM and infection was observed in bacterial but not in viral infections. Suggestive evidence suggests that BWM has an effect on viral intestinal infections (OR: 0.86; 95% CI 0.75 to 0.99; P = 0.03). There is limited evidence of an association between BWM levels and bacteria intestinal infections, and genitourinary tract infection (GUI) in pregnancy. In addition, MR analyses supported the risk of BWM for several edematous diseases. However, multivariable MR analysis shows that the associations of BWM with sepsis, pneumonia, UTIs and SSTIs remains unaffected when accounting for these traits.

Conclusions: In this study, the causal relationship between BWM and infectious diseases was systematically investigated. Further prospective studies are necessary to validate these findings.