Gut microbiota associated with low anterior resection syndrome after rectal cancer surgery

Subgroups of LARS-related symptoms

Of the 109 patients, 91 were classified as severe LARS and 18 as mild. Patient baseline characteristics were shown in the Supplementary Table. S1. The mean age was 62.2 years, and 101 (92.7%) of the patients were male. The mean tumor location from the anal verge was 7 cm, and 58.7% of his patients had ileostomy reversal after a reorienting ileostomy. Preoperative or postoperative radiation therapy was administered to 55 (50.5%) patients, and 3 (2.8%) patients experienced anastomotic leakage.

Questionnaire items for frequency, clustering, and urgency were grouped and contributed to the first PC variable (PC1LARS) in PCA. Other symptoms, such as liquid fecal incontinence and gas incontinence, accounted for a larger proportion of the second PC variable (PC2LARS) (fig. 1).

Principal component analysis (PCA) using low anterior resection syndrome (LARS) scores from 109 patients. (a) Biplot of PCA scores, (B.) PCA loading plot for LARS questionnaire items, (C.) factor loadings for each principal component (PC). This explains the presence of 55.3% of the total data variance in the two principal components.

In addition, the sum of the scores for items related to frequency, clustering, and urgency is Sub 1LARSthe sum of flatus and liquid stool incontinence scores is Sub 2 Ra. Therefore, similar to previous studies,³³, PC1LARS and Sub 1LARS Supports frequency-dominant LARS, PC2LARS and Sub 2 Ra Corresponds to LARS with predominance of incontinence.

table 1 summarizes the descriptions of these LARS-related variables and the ranges for each.in order to Total Lars, a predefined metric of the LARS questionnaire was applied. To further interpret the new symptom-based his LARS subgroups, we introduced grouping criteria for each LARS-related variable. For subgrouping items in the LARS questionnaire, a binary grouping variable derived from a new continuous variable was devised. The criteria that apply to each level are indicated in the “Level” column of this table.Supplementary table S2 A summary of the distributions of LARS-related variables.

Cut-off values ​​for LARS-related dichotomous variables were determined considering patient distribution.in the Sub 1LARS, the ‘severe’ classification was determined if the frequency of bowel movements was greater than 4 times a day, with cluster symptoms and a sense of urgency at least once a week.in the Sub 2 Ra“Severe” refers to symptoms of incontinence, both flatus and liquid stools occurring at least once a week (Supplementary Table) S3). The two group factors were significantly independent of each other (chi-square = 0.410, P = 0.522).

microbial diversity

To compare LARS groups, microbial alpha diversity was measured using different methods in terms of abundance and diversity (Table 2).No significant correlation was observed between Total Lars Significant correlations were found between scores and all alpha diversity indices. Sub 1LARS score Sub 2 Ra And the richness of alpha diversity. Briefly, the number of observed features and indicators of microbial abundance, including Chao1, were monitored and revealed a significant negative correlation. Sub 1LARS [Spearman correlation coefficient (ρ) = −0.217, P = 0.023 with Observed feature number; ρ = −0.222, P = 0.020 with Chao1]While showing a significant positive correlation at Sub 2 Ra (ρ = 0.221, P = 0.021 for observed features, ρ = 0.221, P = 0.021 for Chao1).

Additionally, a mean comparative analysis of the alpha diversity index for each LARS-related binary variable was performed (Figure 1). 2). Microbial abundance assessed using Chao1 was significantly lower at severe levels of bacteria. sub1LARS_ group [median 115 (SD 59.1) vs. 146 (57.7), severe vs. mild, respectively, P = 0.0487] But on a more serious level, Sub 2LARS_ Group [132 (60.1) vs. 112 (55.5), P = 0.0419]. Beta diversity showed no difference between LARS groups (Supplementary Figure). S1).

Comparison of microbial alpha diversity among low anterior resection syndrome (LARS) groups. (a) total LARS_groups(B.) sub1LARS_ group(C.) Sub 2LARS_ Group. Wilcoxon rank sum tests were performed for each comparison and nominal P-values ​​are shown.

Severity of LARS according to gut microbiota enterotype

The gut microbiome has been classified into three enterotypes and the main bacterial taxa contributing to it are: Prevotellaceae, Ruminococcaceaeand Bacteroidae at the family level (Fig. 3a). A different composition was identified. sub1LARS_ group compared to Sub 2LARS_group and total LARS_groups (fig. 3b).in serious situations sub1LARS_ group, ratio of Bacteroidae higher enterotypes, Prevotellaceae Enterotypes tended to be lower than mild cases sub1LARS_ group (P = 0.692). in contrast, Sub 2LARS_ Group and total LARS_groups showed higher Prevotellaceae Enterotypes were higher in the severe group than in the mild group (P = 0.625 and P = 0.311, respectively).

Enterotypes of gut microbiota according to severity and group of low anterior resection syndrome (LARS). (a) the relative abundance of major contributing family-level taxa for the three enterotypes of the gut microbiota; (B.) organization of three enterotypes in the LARS group. The values ​​displayed in the bar chart (B.) represents the proportion of each enterotype within a column (level) of the LARS group. Enterotype proportions for each column in each LARS group are shown as percentage values ​​in bar graphs. There was no significant difference in enterotypic composition between mild/mild and severe/severe disease in each LARS group (chi-square test, P > 0.05).

Abundant taxa that differ among LARS groups

Twenty of a total of 123 genera showed a significant relationship with at least one of the LARS-related continuous variables (absolute Spearman rank coefficient ≥ 0.2, nominal P < 0.05). Four).

Heatmap of the relationship between bacterial classification by LARS group and severity of low anterior resection syndrome (LARS). Twenty taxa with significant results (nominal P-values) corresponding to LARS severity are listed at the genus level. The signed log-transformed P-values ​​of the coefficients of each MaAsLin2 linear model are displayed on a heatmap. Genera containing zeros in more than 90% of the samples were excluded from the analysis, resulting in a total of 123 genera. Each taxa positively associated with LARS severity is marked in red and those negatively associated in blue.

As Total Lars score increases, granuli catella (linear model coefficient β = 0.27; 95% CI, 0.02 to 0.52; P = 0.037) and Catenibacterium (β = 0.26; 95% CI, 0–0.51; P = 0.049) increased and butyric acid (β = −0.55; 95% CI, −0.97 to −0.12; P = 0.013) decreased and the direction of the relationship was consistent with other LARS-related variables.

PC1LARS and Sub 1LARS again PC2LARS and Sub 2ALRS showed similar patterns of correlation for each genus, but the respective different symptom groups of frequency-dominant vs. incontinence-dominant differed greatly in their taxonomic relationships. The main genera of lactic acid bacteria, bifidobacteria (β = −0.76; 95% CI, −1.37 to −0.15; P = 0.016 PC1LARS β = −0.79. 95% CI, -1.40 to -0.18. P = 0.013 Sub 1LARS), Lactic acid bacteria (β = −0.58; 95% CI, −1.00 to −0.15; P = 0.009 PC1LARS β = −0.42; 95% CI, −0.85 to 0.01; P = 0.060 Sub 1LARS), and Weissella (β = −0.33; 95% CI, −0.60 to −0.06; P = 0.017 PC1LARS) showed a significant negative relationship with the severity of the frequency-dominant pattern, but a weakly positive relationship with the severity of the incontinence-dominant pattern (PC2LARS and Sub 2 Ra).

A differential abundance analysis with binary variables was performed. The difference in richness is butyric acid of total LARS_groups (β = −1.34; 95% CI, −2.48 to −0.2; P = 0.023), subdori granules of sub1LARS_ group (β = −2.52; 95% CI, −4.07 to −0.96; P = 0.002), and Ocillibacter of Sub 2LARS_ Group (β = 1.74; 95% CI, 0.38 to 3.10; P = 0.014) was significant.Rich Lactic acid bacteria (β = –1.28; 95% CI, -2.21 to -0.34; P = 0.009) and bifidobacteria (β = −1.36; 95% CI, −2.73 to 0; P = 0.052) was lower in critically ill patients sub1LARS_ group than the mild group (Fig. 2) Five).

Abundant differences in the lactate-producing microbiome among the low anterior resection syndrome (LARS) groups. (a) Spearman correlation between LARS severity and relative abundance of the microbiome in the LARS group (*P < 0.05). (B.C.) comparison of relative abundance of genera Lactic acid bacteria and bifidobacteria in minor (mild) and major (severe) LARS groups (MaAsLin2 linear model P-values ​​are shown).

Probiotic strain abundance in LARS severity

lactic acid bacteria, streptococci, enterococci, bacilli, and bifidobacteria Used as a probiotic³⁸. PC1LARS and Sub 1LARS was significantly negatively correlated with Lactic acid bacteria [Spearman correlation coefficient [ρ]= −0.223, P = 0.020; respectively, ρ = −0.205, P = 0.033, Figures 2, 3, 4, 5, 6 Fivea].of sub1LARS_ groupthe severe group showed a statistically lower Lactic acid bacteria (P = 0.009, Fig. Fiveb) below bifidobacteria (P = 0.052, Figure Fivec).