A two-way, two-sample MR Analysis was conducted to explore the connection between Asthma-related pneumonia and HF. The MR Results indicated that individuals with a genetic susceptibility to Asthma-related pneumonia have a heightened risk of developing HF. However, this study did not establish a causal link between the effects of HF and Asthma-related pneumonia risk.
3.1Causal effect of Asthma-related pneumonia on HF via forward MR
After conducting various analyses such as excluding linkage disequilibrium, F-value screening, and allele adjustment, we identified 4 SNPs that are associated with heart failure. We created forest plots, wherein instrumental variables were used as the ordinate and the effect value of each instrumental variable was used as the abscissa, as illustrated in Figure 2. Our study indicates that there is a positive causal relationship between pneumonia and heart failure based on the IVW results. Pneumonia can increase the risk of heart failure (OR=1.113,95%CI:1.027~1.206, P=0.009). The weighted median method also reveals a significant causal relationship (OR=1.153,95%CI:1.046~1.271, P=0.004) (table 1, table 3). However, the MR-Egger method (OR=1.482, 95% CI: 0.678~3.237, P = 0.428) and the simple model (OR=1.170, 95% CI: 1.006 ~1.361, P=0.134) showed a positive causal relationship with the same direction but no statistical significance. This may be attributed to the low efficiency caused by the large confidence interval of MR-Egger. Simple models are less accurate as they do not take into account variance. Our main findings are based on the IVW method, which is widely considered to be the strongest and most reliable among the existing methods. This method utilizes all available genetic tools, thereby improving statistical power and providing the most accurate and efficient results without the presence of heterogeneity and horizontal pleiotropy. Figure 3 displays the scatter plot of the causal effect value for each genetic instrumental variable. Each black dot represents a SNP, and the line on each dot represents the 95%CI. The vertical coordinate shows the effect of the SNP on the outcome of heart failure, while the horizontal coordinate shows the effect of the SNP on the exposure factor of pneumonia. The colored straight line represents the MR Fitting results.
To assess the differences between all IVs, we conducted a heterogeneity test using the Q statistic. The results for the MR-Egger and IVW tests were Cochran's Q=2.820, P= 0.244, and Cochran's Q=3.556, P=0.314, respectively. These results indicate no heterogeneity in the causal relationship between pneumonia and HF patients. Additionally, genetic polymorphisms can affect the results, as shown in Table 5. The MR-Egger intercept P=0.545, indicates that there is no causal relationship between the two.
Through one-by-one exclusion analysis, we found that the inclusion of SNPs had no significant effect on the causal relationship between the two groups. We did not find any large fluctuation points, and all P values were greater than 0.05, as shown in Figure 4. The funnel plot is symmetrical, indicating that a single SNP as an instrumental variable is less likely to be affected by bias in the analysis of causality between the two. Please refer to Figure 5 for more information.
3.2 Causal association of HF with Asthma-related pneumonia via reverse MR
We conducted a meticulous study to determine if HF has a causal effect on the risk of Asthma-related pneumonia. Our approach involved a two-sample MR Analysis with HF as the exposure factor and pneumonia as the outcome. We utilized the same GWAS database, IV selection method, analysis method, and test method as our previous research. We selected 12 genome-wide SNPs that were closely related to HF as instrumental variables and F > 10 was included to ensure that there was no possibility of weak instrumental variable bias and that the findings were dependable (Table 2) (Figure 6). However, all of the methods we used, namely MR-Egger, IVW, simple mode, and weighted median, failed to demonstrate any evidence of a causal relationship between HF and the risk of Asthma-related pneumonia (P>0.05) (Table 4) (Figure 7). The Cochran’s Q test showed no heterogeneity among the IVs, and MR-Egger's result was Cochran's Q=7.856, P= 0.345, while IVW's was Cochran's Q=7.910, P=0.442. The pleiotropy test using MR-Egger intercept analysis also showed no abnormalities, P=0.832 (P>0.05) (Table 5) (Figure 8, Figure 9).
Table 1. Detailed information of instrumental variables utilized in the MR analysis of the causal effects of Asthma-related pneumonia on Heart failure.
SNP
|
Effect allele
|
Other allele
|
Se
|
Beta
|
P
|
rs11693204
|
A
|
G
|
0.0225
|
0.13
|
7.95E-09
|
rs12603332
|
C
|
T
|
0.0204
|
0.1263
|
5.62E-10
|
rs142807069
|
G
|
A
|
0.0272
|
0.1646
|
1.51E-09
|
rs28688825
|
G
|
A
|
0.0295
|
-0.1607
|
4.93E-08
|
MR, Mendelian Randomization; SNP, single nucleotide polymorphism; SE, standard error.
Table 2. Detailed information of instrumental variables utilized in the MR analysis of the causal effects of Heart failure on Asthma-related pneumonia.
SNP
|
Effect allele
|
Other allele
|
Se
|
Beta
|
P
|
rs11745324
|
A
|
G
|
0.0095
|
-0.0528
|
2.34E-08
|
rs1510226
|
C
|
T
|
0.0285
|
0.162
|
1.27E-08
|
rs1556516
|
C
|
G
|
0.0078
|
0.0622
|
1.57E-15
|
rs17042102
|
A
|
G
|
0.0121
|
0.1103
|
5.71E-20
|
rs17617337
|
T
|
C
|
0.0095
|
-0.0561
|
3.65E-09
|
rs4135240
|
C
|
T
|
0.0084
|
-0.0486
|
6.84E-09
|
rs4746140
|
C
|
G
|
0.0109
|
-0.0666
|
1.10E-09
|
rs4766578
|
A
|
T
|
0.0079
|
-0.0433
|
4.90E-08
|
rs55730499
|
T
|
C
|
0.0157
|
0.1058
|
1.83E-11
|
rs56094641
|
G
|
A
|
0.008
|
0.0454
|
1.21E-08
|
rs600038
|
C
|
T
|
0.0096
|
0.0569
|
3.68E-09
|
rs660240
|
C
|
T
|
0.0097
|
0.0611
|
3.25E-10
|
MR, Mendelian Randomization; SNP, single nucleotide polymorphism; SE, standard error.
Table 3 Mendelian randomization estimates for Asthma-related pneumonia on HF
Exposure
|
Outcome
|
No. of IVs
|
Methods
|
Beta
|
SE
|
OR(95%CI)
|
P
|
Asthma-related pneumonia
|
HF
|
4
|
MR Egger
|
0.393
|
0.399
|
1.482(0.678~3.237)
|
0.428
|
|
|
|
IVW
|
0.107
|
0.041
|
1.113(1.027 ~1.206)
|
0.009
|
|
|
|
Simple mode
|
0.157
|
0.077
|
1.170(1.006 ~1.361)
|
0.134
|
|
|
|
Weighted median
|
0.142
|
0.049
|
1.153(1.046~1.271)
|
0.004
|
HF, Heart failure; IVs, instrumental variables; IVW, inverse variance weighting; SE, standard error; OR, odds ratio: CI, confidence interval.
P < 0.05 was considered statistically significant.
Table 4 Heterogeneity and horizontal pleiotropy for Mendelian randomization analysis
Exposure
|
Outcome
|
Heterogeneity
|
Horizontal pleiotropy
|
Method
|
Cochran’s Q
|
P
|
Egger intercept
|
SE
|
P
|
Asthma-related pneumonia
|
HF
|
MR Egger
|
2.820
|
0.244
|
-0.040
|
0.056
|
0.545
|
|
|
IVW
|
3.556
|
0.314
|
|
|
|
HF
|
Asthma-related pneumonia
|
MR Egger
|
7.856
|
0.345
|
0.006
|
0.028
|
0.832
|
|
|
IVW
|
7.910
|
0.442
|
|
|
|
HF, Heart failure; IVW, inverse variance weighting; SE, standard error;
P < 0.05 was considered statistically significant.
Table 5 Mendelian randomization estimates for HF on Asthma-related pneumonia.
Exposure
|
Outcome
|
No. of IVs
|
Methods
|
Beta
|
SE
|
OR(95%CI)
|
P
|
HF
|
Asthma-related pneumonia
|
12
|
MR Egger
|
-0.070
|
0.427
|
0.932(0.403~ 2.154)
|
0.874
|
|
|
|
IVW
|
0.018
|
0.136
|
1.018(0.780~ 1.328)
|
0.893
|
|
|
|
Simple mode
|
-0.213
|
0.338
|
0.808(0.416~ 1.567)
|
0.547
|
|
|
|
Weighted median
|
-0.037
|
0.182
|
0.963(0.673~ 1.377)
|
0.838
|
HF, Heart failure; IVs, instrumental variables; IVW, inverse variance weighting; SE, standard error; OR, odds ratio: CI, confidence interval.
P < 0.05 was considered statistically significant.