Update of use of dietary indices to control for diet in human gut microbiota studies

Abstract Background In our recent publication (DOI: https://doi.org/10.1186/s40168-018-0455-y ) we concluded that of the three dietary indices we studied the Healthy Eating Index (HEI) was the index of choice where researchers wish to account for the role of diet in microbiome association studies. Following correspondence from its creators, we replicated our initial study with an additional index, the Dietary inflammation index (DII ® ) using an updated data analysis pipeline for microbiota comparisons that incorporate the use of Amplicon Sequence Variants (ASVs) rather than the previously utilised Operational Taxonomic Units (OTUs). The energy-adjusted DII (E-DII) reflects the inflammatory potential of an individual’s diet, after controlling for total energy intake, which is of interest given the known association between the microbiota and inflammatory disease. Results Using data from 5047 participants of the TwinsUK cohort, we observed the E-DII to be a valid dietary measure within this cohort. The E-DII had a stronger association with frailty, a measure of health deficit, than the HEI (E-DII: β = 0.12, p < 0.001; HEI: β = −0.09, p < 0.001) But not with BMI (E-DII: β = 0.04, p <0.01, HEI: β = −0.07, p <0.001). In a subset of individuals (n = 1853) with existing microbiota data, the E-DII had a stronger association with measures of microbiota species (alpha) diversity; the HEI had a higher number of associations with differences of taxa abundance. Of interest are the differences where the two indexes do not overlap). For example, the genus with the strongest association with the E-DII but not significantly associated with the HEI was assigned as Escherichia/Shigella (logFC=0.692, q<0.001). Both indexes were associated with PCoAs of weighted UniFrac distances. Conclusion Both the E-DII and HEI are strong candidates for use as covariates in microbiota association studies where measures of the microbiota are captured using ASVs. The E-DII is an interesting alternative to the HEI particularly where study designs are looking to assess causality of the microbiota in driving inflammation and inflammatory disease.