The presence of symptoms of depression and anxiety, as well as a higher abundance of Bacteroides stercoris in stool, were found to be correlated with an elevated risk of FC (OR>1). Furthermore, a potential causal relationship between FC and symptoms of depression and anxiety (OR>1) was established. However, no causal effect of FC on BSAIS was observed. These findings are visually represented in Figure 1b.
1. Depression symptoms (DS) and functional constipation (FC)
The attributes of all SNPs incorporated in our MR analyses to explore potential causal associations between DS and FC are detailed in Additional file 1: Table S1-S2. The scatter plots of the causality and the reverse causation between DS and FC estimated by five MR analysis methods are presented in Figure 1c-d.
The findings from the IVW analysis indicated a statistically significant causal association between DS and FC (OR = 1.122, 95% CI: 1.031-1.221, P = 0.008), as presented in Table 1. These results suggest that individuals with DS may have an elevated risk of developing FC. The four remaining methods, including MR Egger (OR = 1.124, 95% CI: 0.932-1.354, P = 0.233), Weighted median (OR = 1.098, 95% CI: 0.974-1.238, P = 0.126), Simple mode (OR = 1.095, 95% CI: 0.851-1.410, P = 0.487), and Weighted mode (OR = 1.098, 95% CI: 0.855-1.411, P = 0.470), are presented in Table 1. The four remaining methods demonstrated consistent directional effects with the IVW analysis, showing OR greater than 1. The sensitivity analysis revealed that the MR-Egger intercept test did not detect pleiotropy (P = 0.983), while the MR Egger and IVW methods analyses indicated no heterogeneity among the IVs, with corresponding P values of 0.372 and 0.425, as presented in Table 2.
An inverse MR analysis was conducted to investigate the presence of reverse causality. The IVW results indicated a statistically significant causal association between FC and DS (OR = 1.156, 95% CI: 1.048-1.275, P = 0.004), suggesting that individuals exhibiting FC may be at increased risk of developing DS, as illustrated in Table 1. Furthermore, the results indicated that MR Egger (OR = 1.179, 95% CI: 0.945~1.471, P = 0.164), Weighted median (OR = 1.068, 95% CI: 0.943~1.208, P = 0.300), Simple mode (OR = 1.001, 95% CI: 0.783~1.279, P = 0.997), and Weighted mode (OR = 1.001, 95% CI: 0.775~1.292, P = 0.997) yielded directional effects consistent with the IVW estimates, as detailed in Table 1. The analysis of heterogeneity, as assessed by MR Egger (Q=22.978, P= 0.150) and IVW (Q=23.028, P=0.190), as well as pleiotropy, as determined by the MR-egger_intercept test (P= 0.850), revealed no significant evidence of heterogeneity or pleiotropy, as presented in Table 2.
2. Anxiety symptoms (AS) and functional constipation (FC)
A comprehensive listing of SNPs utilized in our MR analyses to explore potential causal associations between anxiety symptoms and functional constipation was provided in Additional file 1: Table S3-S4. The scatter plots of the causality and the reverse causation between AS and FC estimated by five MR analysis methods are presented in Figure 1e-f.
Utilizing the IVW model of MR analysis, a statistically significant causal association was identified between AS and FC (OR = 1.129, 95% CI: 1.037-1.230, P = 0.005). These results indicate that AS may elevate the likelihood of experiencing FC, as illustrated in Table 1. The IVW estimates demonstrated consistent directional effects for MR Egger (OR = 1.047, 95% CI: 0.777~1.410, P =0.766), Weighted median (OR = 1.143, 95% CI: 1.024~1.275, P = 0.017), Simple mode (OR = 1.227, 95% CI: 0.953~1.582, P = 0.126), and Weighted mode (OR = 1.224, 95% CI:0.945~1.585, P = 0.138), as outlined in Table 1. The analysis revealed that heterogeneity was not statistically significant (MR Egger Q=29.511, P= 0.201 and IVW Q=29.848, P=0.230), and pleiotropy was also not significant (MR-egger_intercept test p = 0.606), as presented in Table 2.
In line with the principles of MR, the findings from inverse MR investigations also revealed a statistically significant causal association between FC and AS (OR = 1.164, 95% CI: 1.057~1.281, P = 0.002), as presented in Table 1. Furthermore, in addition to the IVW estimates, the MR Egger (OR = 1.194, 95% CI: 0.966~1.475, P = 0.119), Weighted median (OR = 1.152, 95% CI: 1.005~1.321, P = 0.043), Simple mode (OR = 1.121, 95% CI: 0.858~1.466, P = 0.414), and Weighted mode (OR = 1.132, 95% CI: 0.871~1.471, P = 0.365) all demonstrated consistent effects in the same direction, as detailed in Table 1. Subsequent sensitivity analyses indicated that the presence of heterogeneity (MR Egger Q=14.156, P= 0.656 and IVW Q=14.225, P=0.714) or horizontal pleiotropy (MR-egger_intercept test p = 0.795) did not yield statistically significant results, as presented in Table 2.
3. Bacteroides stercoris abundance in stool (BSAIS) and functional constipation (FC)
The study presents an overview of the SNPs utilized in the MR analysis to explore potential causal associations between BSAIS and FC. Additional file 1: Table S5-S6 provides detailed information on the SNPs. The scatter plots of the causality and the reverse causation between BSAIS and FC estimated by five MR analysis methods are presented in Figure 1g-h.
In the MR analysis examining the association between BSAIS and FC, our primary causal estimate derived from the IVW method indicated a significant causal relationship (OR = 1.137, 95% CI: 1.048~1.233, P = 0.002), as presented in Table 1. Additional MR methods, including MR Egger (OR = 1.125, 95% CI: 0.913~1.386, P = 0.307), Weighted median (OR = 1.099, 95% CI: 0.983~1.229, P = 0.097), Simple mode (OR = 1.092, 95% CI: 0.907~1.314, P = 0.379), and Weighted mode (OR = 1.084, 95% CI: 0.907~1.295, P = 0.400), demonstrated consistent directional effects, as depicted in Table 1. Furthermore, the analysis revealed that there was no evidence of heterogeneity (MR Egger Q=7.537, P=0.375 and IVW Q=7.550, P=0.479) or pleiotropy (MR-egger_intercept test p=0.916) among the IVs during sensitivity analyses, as detailed in Table 2.
An inverse MR analysis was conducted to investigate the presence of reverse causality. The IVW results indicated no statistically significant causal association between FC and BSAIS (OR = 0.962, 95% CI: 0.814~1.136, P = 0.645), as illustrated in Table 1.