Introduction

The global coronavirus disease 2019 (COVID-19) pandemic, caused by the novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) [1], has led to the loss of more than 6.5 million lives worldwide [2]. In addition to these direct human costs, the pandemic has led to extensive lifestyle changes in efforts to curtail the spread of clinical disease [3]. These include disruptions in food and drink consumption patterns [4], exercise habits [5], and altered recreational drug [6] and alcohol [7] use. Preventative lifestyle changes including lockdowns, social distancing, mandatory mask wearing, extensive hygiene practices, and sanitization of surfaces [8] were all adopted to varying degrees. While prudent from a public health perspective, these lifestyle changes may have had unintended consequences on other aspects of human health [9, 10].

The human gut microbiome plays extensive roles in human health and disease, and contributes to host metabolism, physiology, and immune function [11]. Alterations in gut microbial profiles have been demonstrated in gastrointestinal conditions including inflammatory bowel disease (IBD) and irritable bowel syndrome (IBS) [12], as well as diseases such as obesity [13], type 2 diabetes [14], and more [15]. Within the gut, the small bowel microbiome is critical to nutrient absorption and host adaptation to dietary changes [16]. Given these important influences of the gut microbiome on human health, and the fact the gut microbiome is highly influenced by lifestyle practices [9, 17], it is critical to consider how the gut microbiome may have been impacted by COVID-19 prevention measures. In fact, it has been suggested that the COVID-19 pandemic may compound the existing decline in microbial diversity due to increased sanitization techniques, antibiotic use, and urban living (the hygiene hypothesis) [9].

While previous studies [18] have explored the impact of COVID-19 infection on the gut microbiome, lifestyle changes during the COVID-19 pandemic may have important, independent effects on the gut microbiome. In addition, these previous studies used stool samples [18]. Although stool sampling is non-invasive and cost effective, there are significant differences between the microbial populations of the small intestine and stool [19], and stool samples are not representative of the entire gastrointestinal tract. Recognizing this, the REIMAGINE (Revealing the Entire Intestinal Microbiota and its Associations with the Genetic, Immunologic, and Neuroendocrine Ecosystem) study was initiated to explore the interrelationships between the microbial populations of the small bowel and human health and disease [20].

In this study, we performed an initial, retrospective analysis to determine the potential effects of lifestyle changes during the COVID-19 pandemic on the composition of the duodenal microbiome, using samples and data obtained from REIMAGINE subjects who underwent upper endoscopy before the pandemic (from February 2019 to March 2020 [pre-pandemic group]) and subject who underwent upper endoscopy during the pandemic but after restrictions such as lockdowns began to ease and elective surgeries and procedures were allowed to resume (April 2021–September 2021 [intra-pandemic group]). For the intra-pandemic group, only individuals who tested negative for COVID-19 were included. As a secondary analysis, we also compared levels of circulating cytokines and chemokines in these two groups, in order to determine if any microbiome differences seen were associated with changes in inflammatory profiles.

Methods

Study Subjects

Potential study subjects were identified from the REIMAGINE study. Subjects aged between 18 and 85 years old undergoing a standard of care upper endoscopy (esophagogastroduodenoscopy [EGD]) without colon preparation are eligible for participation. REIMAGINE subjects complete a detailed questionnaire that includes self-documented family and medical history. All medical information provided by subjects is verified by the previous medical records. Based on their procedure date, subjects were divided into two groups: pre-pandemic, from February 2019 to March 2020; and intra-pandemic, from April 2021 to September 2021. Study enrollment was halted from April 2020 through March 2021 due to pandemic-related restrictions on non-elective surgeries and procedures. The pre-pandemic and intra-pandemic groups were then matched by age (± 5 years), body mass index (BMI, (± 3 kg/m2)), and sex. All data were de-identified prior to analysis. Of note, all subjects from the intra-pandemic period underwent polymerase chain reaction (PCR) testing for COVID-19, and only COVID-19-negative subjects with no recorded history of COVID-19 were included in this study.

The Institutional Review Board at Cedars-Sinai approved the study protocol (Pro00035192), and all subjects provide informed written consent prior to participation.

Study Objectives

The primary objective of this study was to identify differences in the composition of the duodenal microbiome in subjects before and during the COVID-19 pandemic that were potentially associated with pandemic-related lifestyle changes.

The secondary objective was to identify differences in circulating levels of a panel of cytokines and chemokines, in order to determine if any microbiome differences seen were associated with changes in inflammatory profiles.

Duodenal Luminal Samples

Samples of luminal fluid were obtained from the duodenum during the EGD, using a sterile double-lumen aspiration catheter (Hobbs Medical, Inc.) as described previously [20].

Blood Collection and Analysis

After completing the study questionnaire, fasting blood samples were collected. Sera were analyzed using a Luminex FlexMap 3D (Luminex Corporation, Austin, TX, USA) with a magnetic bead-based multiplex panel to assess circulating cytokines and chemokines. Two immunoassay kits, together totaling 29 analytes, were run from a single aliquot to limit freeze thaws on aliquots. The first kit was the Th17 Magnetic Bead Panel (EMD Millipore Corp., Billerica, MA, USA, cat. #HT17MG-14 K-PX25), a 25-plex assay for the following analytes: chemokine macrophage inflammatory protein 3 α (MIP3α, also known as C–C Motif Chemokine Ligand 20 (CCL20)), granulocyte–macrophage colony-stimulating factor (GM-CSF), interferon gamma (IFNγ), interleukin (IL) 1β, IL2, IL4, IL5, IL6, IL9, IL10, IL13, IL12p70, IL15, IL17A, IL17F, IL21, IL22, IL23,IL 25(IL17E), IL27, IL28A, IL31, IL33, and tumor necrosis factor alpha (TNFα) and beta (TNFβ). A second, custom 4-bead panel, the Milliplex Human Cytokine, Chemokine, Growth Factor Panel A panel (cat #HCYTA-60 K), included monocyte chemoattractant protein-1 (MCP-1), IL3, IL8, and IL18. The kits were run using protocol-specified equipment settings after an overnight incubation.

Aspirate Processing

Equal volumes of sterile 1X dithiothreitol (DTT) and duodenal aspirates were mixed and vortexed for approximately 30 s until thoroughly blended as described previously [20]. Following this samples were centrifuged at maximum speed (> 13,000 RPM) for 5 min, and the supernatant was removed. Then 1 mL of sterile All Protect reagent (Qiagen, Hilden, Germany) was added to the microbial pellet. The pellets under All Protect were then stored at − 80 °C until DNA extraction was performed [20].

DNA Extraction and Isolation

Duodenal aspirate pellets were thawed on ice and processed as described previously [20]. Subsequentially, microbial DNAs were isolated from each sample using the MagAttract PowerSoil DNA KF Kit (Qiagen) on a KingFisher Duo (Thermo Fisher Scientific, Waltham, MA, USA) [20], and quantitated using a Qubit™ ds high sensitivity DNA BR Assay kit (Invitrogen by Thermo Fisher Scientific) on a Qubit™ 4 Fluorometer (Invitrogen, Carlsbad, CA, USA).

Library Preparation and 16S rRNA Sequencing and Analysis

The 16S V3 and V4 regions were amplified with specific primers (S-D-Bact-0341-b-S-17 and S-D-Bact-0785-a-A-21) modified to include Illumina sequencing adapters as described previously [20]. The quality of the library was assessed on an Agilent 2100 Bioanalyzer System. The MiSeq System (Illumina, San Diego, California) was used for paired-end sequencing of amplicons (2 × 301 cycles) with the MiSeq reagent v3 Kit with 600 cycles and 5–10% PhiX (Illumina).

Sequencing and Statistical Analysis

CLC Genomic Workbench v. 10.1.1 and CLC Microbial Genomics Module v. 2.5 software (Qiagen) were used for Operational Taxonomic Unit (OTU) clustering and taxonomic analysis against the SILVA database v132. In instances where species level classifications were not presented in the SILVA database, Greengene Database 2013 v.13_8 was used. Default parameters were used for minimum occurrences and chimera crossover cost, and the creation of new OTUs was not allowed. Low depth samples (< 5000 sequences per sample) were removed from the analysis. MicrobiomeAnalyst tools were used to calculate microbial alpha diversity, using Shannon’s Index and Simpson’s Index. On MicrobiomeAnalyst, Bray–Curtis metric was used for calculation of inter-sample variability (beta diversity) statistical significance using PERMANOVA test. The PCoA ordination plot was originated on MetaboAnalyst after a multivariate dimensionality-reduction analysis was performed (Partial Least Squares Discriminant Analysis [PLS-DA]).

A rarefied OTU table was created and used to identify OTUs with significantly different relative abundances (RA) between the different groups. CLC Microbial Genomics Module v.2.5 (Qiagen) and MicrobiomeAnalyst were used for numerous comparisons and statistical analyses. The OTU table was rarefied to the minimal number of reads assigned to a sample, and a Negative Binomial GLM model was used to obtain maximum likelihood estimates for the fold change (FC) of an OTU between different groups. No filtering of low prevalence organisms was performed prior to determination of RA. Significance was determined using the Wald test, and P values were corrected using False Discovery Rate (FDR). A P value < 0.05 was considered to be statistically significant after controlling for FDR. Two-tailed Spearman r correlations, Mann–Whitney tests, and graph construction were performed using rarefied OTU tables and GraphPad Prism 7.02 (GraphPad Software, La Jolla, CA, USA) and IBM SPSS Statistics Version software. Fisher’s Exact test was used to identify associations between categorical variables between different groups, again using GraphPad Prism 7.02 (GraphPad Software).

Core microbiome analyses and correlation constructions were performed using the Core Microbiome tool and correlation network tool from MicrobiomeAnalyst [21]. Relative abundances were calculated based on the percentage of a given taxon in the total taxa identified in the duodenal luminal microbiome. The core microbiome at the phylum, family, and genus levels was constructed based on the prevalence of a specific feature (phylum, family, genus) in at least 65% of samples with a minimum relative abundance of 0.01%. The rank of each feature was defined based on the prevalence and relative abundance distribution in each group. Correlations and associations between microbial taxa were generated based on Spearman rank correlation, using a correlation threshold of at least 0.3 and P values lower than 0.05.

The datasets generated during this study are available at the National Center for Biotechnology Information (NCBI) BioProject Repository (https://www.ncbi.nlm.nih.gov/bioproject) under BioProject ID PRJNA860162.

Results

Subject Demographics

A total of 508 REIMAGINE study subjects had been recruited at the time of this study. Of these, 56 subjects were recruited during the intra-pandemic period (April 2021–September 2021), and 38 of these subjects’ samples had undergone 16S rRNA gene sequencing at the time of analysis. Fifty-six (56) subjects from the pre-pandemic period (February 2019–March 2020) were matched to the intra-pandemic subjects for age, sex, and BMI, and were also included in the analysis, for a total of 94 subjects. The average age of subjects in both groups was 53 years (± 16) (Table 1). History of smoking and use of PPIs were not significantly different between groups, nor were there significant differences in the racial and ethnic compositions of the groups (Table 1).

Table 1 Subject demographics
Full size table

Duodenal Microbial Alpha and Beta Diversity

Duodenal microbial alpha diversity and richness was not significantly different between the pre-pandemic and intra-pandemic groups, as determined using Shannon’s Index (P = 0.417) and Simpson’s Index (P = 0.507) (Supplemental Fig. 1). However, the degree of change in duodenal microbial community composition (beta diversity), analyzed using PERMANOVA, was significantly different in the intra-pandemic groups when compared to pre-pandemic (P < 0.002) (Fig. 1).

Fig. 1
figure 1

Partial Least Squares Discriminant Analysis (PLS-DA) plot of the beta diversity of the duodenal microbiome in the pre-pandemic and intra-pandemic groups (P < 0.002)

Full size image

Differences in the Duodenal Luminal Microbiome in Pre-pandemic Versus Intra-pandemic Subjects

Significant taxonomic differences were identified in the duodenal microbiome between groups, including changes in the relative abundance (RA) of a total of 2 phyla, 3 classes, 6 orders, 4 families, and 22 genera (Fig. 2). Within the core duodenal luminal microbiome, which was defined as the most widespread microbial components of the microbiome found across all at least 65% of subjects, the most prevalent phylum within the pre-pandemic group was Firmicutes, followed by Proteobacteria, Actinobacteria, Bacteroidetes, and Fusobacteria (Fig. 3A). In the intra-pandemic group, Firmicutes was still the most prevalent phylum, but Fusobacteria moved up in rank to the second most prevalent phylum, followed by Bacteroidetes which moved up in rank to the third most prevalent phylum, and Proteobacteria to the fourth most prevalent (Fig. 3B). In contrast, the relative abundance of phylum Actinobacteria was slightly lower in the intra-pandemic group vs. the pre-pandemic group (FC = − 0.40, P = 9.83E-8), and was now the fifth most prevalent phylum (Figs. 3 and 4, Supplemental Table 1). Lastly, a broader analysis that included all duodenal microbes revealed that phylum Deinococcus-Thermus, which was not part of the core duodenal microbiome and which is known to be highly resistant to extreme conditions [22], was significantly more prevalent in the intra-pandemic group when compared to the pre-pandemic group (FC = 3.42, P = 1.69E-03) (Fig. 4, Supplemental Table 1).

Fig. 2
figure 2

Heat tree of the duodenal microbiome in the intra-pandemic vs. the pre-pandemic groups. Nodes represent taxonomic levels, and greater line thicknesses and font sizes denote higher relative abundances (RA). Taxa shown in red have lower RA in the intra-pandemic group, and taxa shown in blue have higher RA in the intra-pandemic group. Only microbial taxa that are statistically different between groups (P < 0.05) are labeled

Full size image
Fig. 3
figure 3

Core duodenal microbiome profiles at the phylum level in the pre-pandemic group (A) and intra-pandemic group (B)

Full size image
Fig. 4
figure 4

Correlation coefficients for the top 15 phyla comparing the pre-pandemic group to the intra-pandemic group

Full size image

Looking further into the 5 core phyla, the family level was also different in the pre- vs. intra-pandemic groups. Families Streptococcaceae (phylum Firmicutes), Veillonellaceae (phylum Firmicutes), and Prevotellaceae (phylum Bacteroidetes) were the most prevalent in both the pre-pandemic and intra-pandemic groups (Fig. 5). However, the prevalence of family Fusobacteriaceae (phylum Fusobacteria) was higher in intra-pandemic subjects, going from the fifth most prevalent in pre-pandemic to the fourth most prevalent intra-pandemic (Fig. 5). In addition, family Carnobacteriaceae (phylum Proteobacteria) went from sixth most prevalent in the pre-pandemic group to eighth most prevalent in the intra-pandemic group (Fig. 5). Significantly, family Micrococcaceae (phylum Actinobacteria), which was part of the family level core in the pre-pandemic group, decreased in prevalence and was no longer part of the core microbiome in the intra-pandemic group.

Fig. 5
figure 5

Core duodenal microbiome profiles at the family level in the pre-pandemic group (A) and intra-pandemic group (B)

Full size image

We identified several other significant changes at the family level. The RA of several Gram-negative families were significantly lower in the intra-pandemic group vs. the pre-pandemic group, including Enterobacteriaceae (phylum Proteobacteria) (FC = − 1.12, P = 2.02E-2), Bacteroidaceae (phylum Bacteroidetes) (FC = − 3.44, P = 1.58E-2), and Pseudomonadaceae (phylum Proteobacteria) (FC = − 2.42, P = 3.76E-2) (Supplemental Table 1). Moving outside of these core families, the Gram-negative family Thermaceae (phylum Deinococcus-Thermus) (FC = 2.70, P = 1.69E-3) was significantly higher in the intra-pandemic group when compared to pre-pandemic (Supplemental Table 1).

The core microbiome at the genus level was also significantly different in the intra-pandemic group compared to pre-pandemic. The prevalence of genus Rothia (family Micrococcaceae, phylum Actinobacteria) was significantly lower in the intra-pandemic group compared to pre-pandemic, and was no longer part of the core duodenal microbiome at the genus level in these subjects (Fig. 6).

Fig. 6
figure 6

Core duodenal microbiome profiles at the genus level in the pre-pandemic group (A) and the intra-pandemic group (B)

Full size image

In addition to genus Rothia (phylum Actinobacteria) (FC = − 1.62, P = 6.85E-7), the RA of Serratia (phylum Proteobacteria) (FC = − 5.94, P = 4.19E-6), Pseudomonas (phylum Proteobacteria) (FC = − 2.34, P = 0.0376), and Escherichia-Shigella (phylum Proteobacteria) (FC = − 1.66, P = 0.0092) were significantly lower in the intra-pandemic group compared to pre-pandemic (Fig. 2, Supplemental Fig. 2, Supplemental Table 1), whereas the RA of Fusobacterium (phylum Fusobacteriota), Leptotrichia (phylum Fusobacteriota), Granulicatella (phylum Bacillota), and Gemella (phylum Firmicutes) were higher in the intra-pandemic group compared to pre-pandemic (Fig. 6, Supplemental Fig. 2).

Levels of the Inflammatory Marker IL-18 Are Lower in the Intra-pandemic Group

Of the pro- and anti-inflammatory chemokines and cytokines tested, only IL-18 (P = 0.018) was significantly different between groups, with lower circulating levels in the intra-pandemic group (P = 0.017) when compared to pre-pandemic (Supplemental Table 2) (Fig. 7). The difference between groups remained significant even after the removal of outliers (P = 0.022). There were no significant associations between circulating chemokine and cytokine levels and duodenal microbial populations, but there was a trend towards an association between higher circulating IL-18 levels and higher levels of Rothia that did not reach significance (Spearman R = 0.187, P = 0.073).

Fig. 7
figure 7

Circulating levels of IL-18 in pre-pandemic vs. intra-pandemic subjects

Full size image

Discussion

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) caused a global pandemic leading to the loss of millions of lives, with ongoing implications for public health [23, 24]. In this study, we demonstrate that the composition of the duodenal luminal microbiome in intra-pandemic individuals was significantly different from that in pre-pandemic individuals. Of note, none of the intra-pandemic subjects in our cohort tested positive for COVID-19, or had a known history of COVID-19, so these differences were not due to direct COVID-19 effects. We found significant differences in beta diversity between the two groups, and significant taxonomic differences from the phylum to the genus levels. These included lower levels of potential bacterial disruptors of the duodenal microbiome [25] in the intra-pandemic group, such as the genera Escherichia-Shigella (phylum Proteobacteria) and Rothia (phylum Actinobacteria), as well as the Proteobacterial families Enterobacteriaceae and Pseudomonadaceae. Interestingly, we also found that the RA of phylum Deinococcus-Thermus, which is typically a rare phylum within the duodenum and contains microbes that are resistant to changes in heat, pH, and radiation [26] as well as chlorination [27], was higher in the intra-pandemic group than in the pre-pandemic group. Given the critical impact the COVID-19 pandemic has had on lifestyles, combined with the key roles of small bowel microbial populations in digestion [28] and the regulation of the immune system [29], these findings may, when taken together, suggest that the lifestyle modifications adapted during the pandemic have implications for the small bowel microbiome, and consequently gastrointestinal health, beyond those which are currently recognized.

While we saw no significant differences in alpha diversity in intra- vs. pre-pandemic subjects, there were significant differences in beta diversity, suggesting that duodenal microbial community structure was significantly altered during the pandemic. Specifically, both the prevalence and the RA of phylum Actinobacteria, and of genus Rothia, were lower in the intra-pandemic group than in the pre-pandemic group. Similar to our findings, Peng et al. found significantly lower richness within phylum Actinobacteria in the stool microbiome of non-COVID subjects during the pandemic [8]. The authors hypothesized that, as species richness within Actinobacteria in the infant gut was previously found to be higher in infants with greater exposure to the natural environment [30], this lowered Actinobacteria species richness might be reflective of decreased outdoor recreational activities during the pandemic [8], and this may also be the case for our subjects. Interestingly, we also found higher RA of phylum Fusobacteria in the intra-pandemic group. While increases in Fusobacterium are associated with digestive diseases, particularly colorectal cancer [31], Fusobacteria has also been linked to psychiatric conditions such as depression [31]. Given the widespread increase in the prevalence of depression during the pandemic [32], our finding of increases RA of Fusobacteria in the intra-pandemic group may be associated with reduced mental health in these subjects.

We also identified lower prevalence of phylum Proteobacteria in the intra-pandemic vs. pre-pandemic group, as well as a trend towards lower RA. While this trend did not reach significance at the overall phylum level, the RA of several Proteobacterial families and genera were significantly lower in the intra-pandemic vs. pre-pandemic group, including gram-negative families Enterobacteriaceae, and Pseudomonadaceae, and the genera Serratia (family Yersiniaceae), Escherichia-Shigella (family Enterobacteriaceae), and Pseudomonas (family Pseudomonadaceae). Phylum Proteobacteria contains known human pathogens [33] and increased levels of Proteobacteria have been suggested as a microbial signature of disease in the gut [34], particularly Escherichia-Shigella, which is increased in the stool microbiome of subjects with conditions ranging from coronary heart disease [35] and Crohn’s disease [36] to Parkinson’s [37]. Moreover, Gram-negative bacteria have an outer membrane that can prevent antibiotics from penetrating the cell, and thus contributes to antibiotic resistance, which is a subject of increasing clinical concern [38, 39]. Lastly, Proteobacterial taxa, particularly Gram-negative taxa such as Escherichia-Shigella, have recently been identified as disruptors of the small bowel microbiome, i.e., taxa that displace strict anaerobes which are common in the duodenum [25, 40,41,42,43]. The lower RA of Gram-negative Proteobacterial taxa in the duodenal microbiome of intra-pandemic subjects may indicate a shift towards a more balanced microbial composition, although further work is needed to confirm this.

Lastly, the RA of the rare phylum Deinococcus-Thermus, and of family Thermaceae, were significantly higher in the intra-pandemic as compared to pre-pandemic group. As noted above, phylum Deinococcus-Thermus is comprised of extremophilic bacteria [26], and a recent study examining the nasal/oropharyngeal microbiota in 21 COVID-19 patients who were admitted to the ICU, and 10 controls who were negative for COVID-19, found that Deinococcus-Thermus was present only in controls, and not in COVID patients [44], which the authors surmised might be due to pandemic-related lifestyle changes, such as increased hand-washing, use of sanitizers, and surface disinfectants [45].

Circulating levels of interleukin (IL)-18, a proinflammatory cytokine, were also significantly lower in the intra-pandemic group compared to the pre-pandemic group. Previous studies have associated elevated IL-18 production in the pathogenesis of various diseases such as rheumatoid arthritis [46], type-1 diabetes [47], and Crohn’s disease [48]. In pathogenic conditions, following disruption of the epithelial barrier, bacteria enter the lamina propria, which causes local macrophage production of IL-18 and promotes dysbiosis [49]. Our finding of lower levels of IL-18 in the intra-pandemic group may reflect the lower RA of disruptor taxa in duodenal microbiome, although further studies are needed to confirm this.

It is critical to understand that the majority of gut microbiome studies performed to date use fecal samples as a measure of gut microbial content. The cost and difficulty of sampling the small intestine result in stool being more commonly used to study the gut microbiome [50]. However, there are notable differences in acidity, transit time, and microbial load along the length of the gastrointestinal tract, resulting in differences in the microbial populations colonizing the different segments of the gut [51]. Importantly, the small intestine is central to human health, as it is the primary site for nutrient absorption, metabolism, and immune function [28]. As part of the REIMAGINE study to explore the roles of the small bowel microbiome in human health and disease, we developed novel techniques to sample the small bowel, reducing contamination with saliva and optimizing the recovery of DNA from small bowel microbes [20]. Using these validated techniques, we found that the small bowel is more heavily colonized than previously thought [20, 25], and that its microbial populations differ significantly from those in stool [19], illustrating the importance of studying the small bowel microbiota directly.

Our study has some limitations. Subjects were undergoing upper endoscopy for various reasons, including but not limited to screening purposes due to risk factors and determining the cause of intestinal complaints, and may not be fully representative of normal healthy individuals. Further, this is a cross-sectional study with a single sample per subject, and future studies might benefit from a longitudinal study design. Pre-pandemic subjects were matched to intra-pandemic subjects by age, BMI, and sex. However, we could not assess for other potential confounders that could affect the microbiome, such as diet, and the pandemic is known to have brought about significant changes in diet and eating habits [52]. Lastly, while our study included a total of 94 subjects (56 pre-pandemic and 38 intra-pandemic), an even larger sample size could confirm the microbiome trends we identified.

In conclusion, this examination of the small bowel microbiome before and during the pandemic suggests that the duodenal microbiome underwent significant changes from the phylum to the genus levels, including lowered prevalence and RA of potential disruptor taxa from phylum Proteobacteria, particularly Escherichia-Shigella, and of phylum Actinobacteria and genus Rothia, in the intra-pandemic group. Conversely, the intra-pandemic group exhibited increased RA of phylum Deinococcus-Thermus, which includes resistant extremophiles [26], which may reflect increased sanitization practices during the pandemic. Given the key roles of small bowel microbiota in host metabolism and immune function, these changes may have important implications for human health.