Hashimoto’s thyroiditis (HT) and Graves’ disease (GD) are the main types of autoimmune thyroid disease (AITD). AITDs are the most common organ-specific autoimmune disorders. Although the clinical manifestations of GD and HT are different, such as hyperthyroidism and hypothyroidism, respectively, GD and HT share similar immune-mediated mechanisms of disease, even alternating from one to the other (1, 2). Many studies have revealed the possible causes of AITD, such as genetic susceptibility factors, dysregulation of the immune system, inflammation, stress, and other environmental factors; however, its etiology remains unclear (3, 4).
Emerging evidence suggests that the alterations of the gut microbiota play a key role in the development and progress of AITD in individuals. From the embryology aspect, the thyroid and gut share a common embryological origin, explaining some morphological and functional similarities between the gut and thyroid follicular cells (5). The association between autoimmune thyroid disorders and gut autoimmune disease atrophic gastritis was first described in the early 1960s (6). More recently, owing to the development of the 16S ribosomal RNA (16SrRNA) gene sequencing technique, the gut microbiota, which comprises trillions of microorganisms, has been proposed to be involved in the pathogenesis of many autoimmune diseases, such as type 1 diabetes, lupus nephritis, Rheumatoid arthritis, Celiac Disease, and AITD (7–10). Although there is no direct evidence that AITD and gut microbiota have a cause-effect relationship, several studies have suggested that the thyroid-gut axis has beneficial or detrimental effects on thyroid function (11). The gut microbiota shapes the thyroid mainly through the following possible microbial-related mechanisms. First, dysbiosis leads to the damaged intestinal barrier and increased intestinal permeability, allowing the antigens to pass into the circulation and activate the immune system (12). Second, the antibodies in the circulation may react with the bacterial antigen and enhance the activation of the inflammasome in the thyroid gland (13). Guo et al. (14) has found that the expression of the inflammasome, including the NOD-like receptor (NLR) family pyrin domain containing 3 (NLRP3), AIM2, caspase-1, and IL-1β mRNA and protein from patients with HT, was significantly increased, which can be regulated by the gut microbiota and its metabolism (15–17). Third, the short-chain fatty acids (SCFAs),metabolites of commensal bacteria fermentation of dietary fiber, are speculated to play a crucial role in the development, functioning, and modulation of the immune system (18, 19). For example, butyrate, a SCFAs, is associated with reduced levels of TNF-α, IL-6 and suppressed activation of the NLRP3 inflammasome via GPR109A (20).
Recently, many researchers have found that AITD patients have reduced α diversity and abundances of certain microbiota compared with healthy controls (21–23). The α diversity mainly contains community diversity (Simpson and Shannon) and community richness indices (ACE and Chao1) (24). Among the AITD patients, those with HT tend to have a higher Chao1 value than healthy volunteers; however, patients with GD have a lower Chao1 value than the controls (25, 26). In addition, the current results revealed the correlation between the clinical parameters of AITD, such as TRAb or TPOAb and the certain microbiota (22, 27, 28). For example, Chen et al. (29) found that the proportion of Synergistetes was negatively correlated with TRAb, and Jiang et al. (30) found that Lactobacillus was positively correlated with TRAb. At the phylum level, Yang et al. (31) found a higher Firmicutes/Bacteroidetes ratio in GD patients than in the control group, which may be relevant to inflammation disease, whereas Hanaa et al. (32) found that the ratio was significantly decreased in patients with AITD. Due to different conflicting data, a further study of the association between the gut microbiota and AITD is needed. To better understand the potential role of gut microbiota in the pathogenesis of AITD,we carried out a meta-analysis to assess the alteration in the microbial population between patients with AITD and healthy controls at different levels.
Materials and Methods
We conducted a systematic literature search in PubMed, Web of Science, Embase, and Cochrane databases up to August 2021 using the following search string: (thyroiditis OR Hashimoto Disease OR Thyroiditis, Autoimmune OR Hashimoto Thyroiditis OR Thyroiditis, and Chronic Lymphocytic OR Chronic Lymphocytic Thyroiditis OR Thyroid Diseases OR Graves’ Disease OR Disease, Graves OR Goiter, Exophthalmic OR Basedow’s Disease OR Hyperthyroidism, Autoimmune) AND (microbiota OR Gut Microbiome OR Microbial Community OR Microbial Community OR Gastrointestinal Microbiome OR Microbiome OR Gut Flora OR Gastrointestinal Microbiota OR Microflora, Gastrointestinal OR Gastric Microbiome OR Intestinal Microbiota OR Intestinal Flora).
Inclusion and Exclusion Criteria
Studies were considered eligible if they met the following criteria: 1) investigating the gut microbiota and patients diagnosed with AITD. 2) providing sufficient data on the relationship between AITD and intestinal microbiota and could be extracted to analyze the 95% confidence interval (CI). 3) written in English. 4) full-text availability. However, comments, animal model subjects, conference abstract, and reviews were excluded. We also excluded studies with incomplete outcome data on the percentages or relative abundance of gut microbiota and studies with fewer than 20 participants.
Three reviewers independently extracted the following data from each study: authors, publication year, country of population, population age, clinical parameters of thyroid function, diagnosis of AITD, and microbiology assessment methods.
Two reviewers completed the quality assessment using the Newcastle-Ottawa scale (NOS) to evaluate all the included studies, comprising the trial group selection, comparability, and exposure. The total score ranged from 5−9, where a higher score represents a higher quality of assessment. All the discrepancies or poor agreement were resolved through a consensus discussion with a third author.
Standardized mean differences (SMD) were calculated between the HT and GD groups to assess the bacterial alpha diversity indices (Chao1). SMD >0 indicates that participants with HT have a higher level of richness in the intestinal microflora. A fixed- and random-effects were used to assess the percentages or relative abundances of certain gut microbiota with AITD compared with the healthy controls. We examined the statistical heterogeneity using the I2 statistic. I2 values of 25%, 50%, and 75% were related to low, moderate, and high heterogeneity, respectively. A Random-effects model was used to pool the results when high heterogeneity (I2>50%) existed. Additionally, the fixed-effects model was used if the heterogeneity was low. Furthermore, we performed a sensitivity analysis as well as Begg’s test to assess the potential influences of bias. All statistical analyses were performed using Stata software (version 12.0).
Characteristics of Included Studies
The initial literature searches retrieved 282 records from the online database. Among them,92 studies were excluded for duplication, and after the review of titles and abstracts,175 articles were eliminated because they did not fulfill the inclusion criteria. Then, we evaluated the remaining 15 articles individually; seven were excluded because the studies did not provide quantitative or appropriate data on the gut microbiota relative abundance. Finally, eight studies were included in our meta-analysis (Figure 1). This meta-analysis included 196 patients with AITD and 160 age-matched healthy controls (Table 1). The clinical parameters of AITD and microbiology assessment methods used in the included studies are shown in Table 2. Fecal samples were collected, and microbiota was analyzed by pyrosequencing or high-throughput sequencing of the 16SrRNA gene, real-time PCR, and PCR-DGGE of the 16SrRNA gene (33). The percentage of gut microbiota composition at a different level and relevant abundance were analyzed based on the assessment methods, preventing the potential deviation caused by the detection methods