It remains unclear whether obstructive sleep apnea is causally related to asthma or subtypes. To elucidate this relationship, we examined the bidirectional causality between obstructive sleep apnea and asthma or subtypes using an MR approach with data from the FinnGen genome database. The results showed that obstructive sleep apnea had a causal effect on both allergic and nonallergic asthma. In addition, allergic asthma also increased the risk of obstructive sleep apnea. However, to our surprise, the causal effect of nonallergic asthma on obstructive sleep apnea was not supported in the MR approach. These findings are robust to sensitivity analyses in different MR models.
Our findings are consistent with those of previous studies. For example, a retrospective cohort study showed a significantly higher risk of developing OSA in asthma patients than in the general population[10]. In addition, a recent systematic review and meta-analysis showed that there is also a causal relationship between OSA and asthma[11]. The bidirectional relationship between asthma and obstructive
sleep apnea disease has been increasingly recognized, and there have been a number of attempts to find supportive evidence for a link between the two. According to a cross-sectional study conducted by Althoff et al[12], individuals with both asthma and obstructive sleep apnea had higher levels of L-arginine metabolites in their serum than those with asthma alone. This finding suggests the possibility of a shared underlying mechanism for both conditions. Cisneros et al[13] showed in a retrospective study that patients with obstructive sleep apnea combined with asthma showed a significant effect on the treatment of continuous positive airway pressure ventilation, suggesting that the same therapeutic strategy may exist for both. In addition, Prasad et al[5] systematically described the interactions between asthma and obstructive sleep apnea disease based on epidemiological, clinical, and laboratory findings, providing supportive evidence for a causal relationship between the two. However, this evidence is mainly limited to observational studies, and the results may be affected by confounding factors and reverse causation. Although there have been high-quality prospective findings supporting a causal relationship between obstructive sleep apnea and asthma, it is important to note that the study had a small sample size and was limited in its scope, as it focused solely on one direction of the relationship without considering subcomponents[14]. Currently, there is insufficient strong evidence to support a bidirectional causal relationship between asthma and obstructive sleep apnea. Our analyses utilize large-scale genome association study (GWAS) data, which effectively minimizes confounders and reverse causality bias often present in observational studies. Our findings suggest that the examined causal relationship between asthma and obstructive sleep apnea is predicted based on genetic variants, providing a relatively more reliable result.
Despite the high degree of association between asthma and obstructive sleep apnea, the mechanisms associated with the link remain hypothetical. Some hypotheses linking obstructive sleep apnea to asthma include increased vagal tone during apnea, hypoxia-associated oxidative stress and changes in intrathoracic pressure exacerbating gastroesophageal reflux[15]. Other hypotheses linking asthma to obstructive sleep apnea include chronic inflammation of the airway, vagal excitation, airway remodeling, and the effects of corticosteroids[5]. Based on the results of our in forward MR analyses, it was found that patients with obstructive sleep apnea were more likely to have nonallergic asthma. This difference is believed to be related to obesity-mediated factors. It is well-known that there is often a coexistence between obesity and sleep apnea, and comorbid obesity is also more frequent in patients with nonallergic asthma than in those with allergic asthma. Additionally, obesity is associated with airway resistance, oxidative stress, and gastroesophageal reflux, which are also more severe in this population[16]. Furthermore, in subgroup analyses in reverse MR, we found that allergic asthma was able to increase the risk of obstructive sleep apnea, but nonallergic
asthma did not have a causal effect on the risk of obstructive sleep apnea. Although different subtypes of asthma exist clinically, the main focus of exploration and research in the past has been on allergic asthma, which is the most common form of asthma. Correlations between nonallergic asthma and obstructive sleep apnea have rarely been reported, and our finding reveal that, at least at the genetic level, nonallergic asthma and obstructive sleep apnea may have different underlying causative processes, and more studies are needed in the future to further validate this result.
Despite adopting an MR design that can effectively avoid confounding bias and reverse causality effects, there are still some limitations of our study. First, the results of this study are based on data from individuals of European ancestry, and whether this result is consistent in other populations remains to be investigated. Second, we used a more relaxed P value threshold (P<5E-06) as our instrument to identify enough SNP instruments for each exposure group to perform a Mendelian randomization analysis, which may have led to an increased rate of false positives. However, we included all strong IVs, and this process is common in previous Mendelian randomization analyses[17, 18].
Table 4
Summary genetic instruments between nonallergic asthma and obstructive sleep apnea.
SNP
|
Chr
|
EA
|
OA
|
Nonallergic asthma
|
Obstructive sleep apnea
|
R2
|
F-statistic
|
eaf
|
beta
|
se
|
pval
|
eaf
|
beta
|
se
|
pval
|
rs10515145
|
17
|
T
|
C
|
0.002882
|
1.1625
|
0.2546
|
4.98E-06
|
0.002864
|
-0.1668
|
0.1155
|
0.15
|
0.007767056
|
1089.293083
|
rs11593338
|
10
|
T
|
G
|
0.1182
|
0.1713
|
0.0369
|
3.39E-06
|
0.1188
|
0.0052
|
0.0192
|
0.79
|
0.006116913
|
856.4439167
|
rs117456590
|
9
|
T
|
C
|
0.01548
|
0.4722
|
0.1015
|
3.30E-06
|
0.0156
|
0.0449
|
0.0501
|
0.37
|
0.006796377
|
952.2283384
|
rs117781262
|
11
|
A
|
C
|
0.01443
|
0.6297
|
0.109
|
7.59E-09
|
0.01472
|
0.0594
|
0.0528
|
0.26
|
0.011278496
|
1587.373572
|
rs12655575
|
5
|
C
|
T
|
0.1091
|
0.181
|
0.0388
|
3.03E-06
|
0.1088
|
0.0036
|
0.02
|
0.86
|
0.006368554
|
891.9026704
|
rs13405741
|
2
|
C
|
T
|
0.09154
|
0.2148
|
0.0431
|
6.19E-07
|
0.09253
|
0.0426
|
0.0218
|
0.05
|
0.007673885
|
1076.125155
|
rs147967955
|
8
|
C
|
T
|
0.004711
|
0.8937
|
0.1958
|
4.99E-06
|
0.004746
|
0.0704
|
0.0921
|
0.44
|
0.007489897
|
1050.129407
|
rs1888648
|
13
|
A
|
G
|
0.001682
|
1.8429
|
0.402
|
4.56E-06
|
0.001746
|
0.3898
|
0.1579
|
0.01
|
0.01140587
|
1605.507498
|
rs2075382
|
7
|
A
|
G
|
0.3146
|
0.1194
|
0.0262
|
4.99E-06
|
0.3148
|
-0.0151
|
0.0135
|
0.26
|
0.006148108
|
860.8386054
|
rs2139194
|
3
|
C
|
A
|
0.3903
|
-0.114
|
0.0247
|
4.00E-06
|
0.3902
|
0.0149
|
0.0128
|
0.24
|
0.00618521
|
866.0658726
|
rs34816745
|
7
|
T
|
C
|
0.1977
|
-0.1431
|
0.0301
|
2.04E-06
|
0.1985
|
0.0013
|
0.0156
|
0.93
|
0.0064961
|
909.8820243
|
rs35783488
|
3
|
G
|
A
|
0.01418
|
0.4955
|
0.1078
|
4.32E-06
|
0.01421
|
0.0146
|
0.0534
|
0.78
|
0.00686422
|
961.7993443
|
rs61823434
|
1
|
T
|
C
|
0.1967
|
-0.1394
|
0.0303
|
4.22E-06
|
0.1962
|
-0.0182
|
0.0157
|
0.25
|
0.00614098
|
859.8344133
|
rs61933750
|
12
|
T
|
C
|
0.01022
|
0.6572
|
0.133
|
7.82E-07
|
0.01038
|
0.014
|
0.0597
|
0.81
|
0.008738053
|
1226.671223
|
rs73641258
|
9
|
T
|
G
|
0.1221
|
0.2137
|
0.0368
|
6.24E-09
|
0.1222
|
-0.0311
|
0.019
|
0.10
|
0.009790385
|
1375.860968
|
rs749112
|
11
|
C
|
T
|
0.7777
|
0.139
|
0.0291
|
1.78E-06
|
0.7773
|
0.0189
|
0.015
|
0.21
|
0.006680534
|
935.8885799
|
rs76428105
|
19
|
T
|
C
|
0.01408
|
0.5059
|
0.1052
|
1.53E-06
|
0.01422
|
-0.0461
|
0.0527
|
0.38
|
0.007105648
|
995.8698453
|
rs7814083
|
8
|
G
|
A
|
0.7196
|
0.1235
|
0.0268
|
3.93E-06
|
0.72
|
0.0164
|
0.0139
|
0.24
|
0.006155071
|
861.8196367
|
rs78917329
|
15
|
T
|
C
|
0.04677
|
-0.2776
|
0.06
|
3.80E-06
|
0.04656
|
-0.0249
|
0.0306
|
0.42
|
0.006871222
|
962.7873181
|
Figure 2 Scatter plots of genetic associations between OSA and asthma or subtypes. a. OSA on Allergic asthma; b. OSA on Nonallergic asthma; c. Allergic asthma on OSA; d. Nonallergic asthma on OSA. The slope of the straight line indicates the magnitude of the causal association.