BACKGROUND
Previous studies have linked sleep quality to temporomandibular joint disorders (TMD), suggesting a role for snoring in this association. However, the directionality of this relationship remains a subject of debate. This investigation aimed to elucidate the connections between snoring and TMD. Methods: This research employed a two-sample Mendelian randomization (MR) approach, leveraging publicly available large-scale genome-wide association study (GWAS) data on snoring and TMD. We utilized a suite of analytical methods, including the inverse variance weighted (IVW) method, maximum likelihood estimation, adjusted profile score, weighted median, MR‒Egger regression, and a series of sensitivity analyses, to rigorously assess the existence of relationships. Results: Our findings indicate that a greater genetic predisposition to snoring is significantly associated with a reduced risk of TMD (IVW method; odds ratio [OR] = 0.156, 95% confidence interval [CI] = 0.028 to 0.843, p = 0.0309). Conversely, the analysis did not support a potential influence of TMD on snoring susceptibility (IVW method; 95% CI = 0.990 to 1.002, p = 0.1926). Additionally, our sensitivity analyses did not reveal any significant pleiotropy that could bias these findings. Conclusion: This MR study provides preliminary but novel genetic evidence supporting a potential causal link between snoring and a decreased risk of developing TMD. On the other hand, it does not substantiate a effect of TMD on the likelihood of snoring.
OBJECTIVE
The objective of this study was to employ a MR analytical framework to investigate the impact of snoring on TMD. This approach aims to provide complementary information to that obtained from RCTs, contributing to a more comprehensive understanding of the relationship between snoring and TMD.
METHODS
Mendelian randomization methods.Genetically predicted.Genetic associations with temporomandibular joint disorders.
RESULTS
First, a total of 38 SNPs predicting snoring at genome-wide significance (p < 5 × 10−8) were obtained in the published GWAS, but 35 SNPs were used to assess the association with TMD after merging with the outcome dataset. Then, 7 SNPs that were palindromic with intermediate allele frequencies were removed via the "harmonize_data" tool. Next, we excluded 5 SNPs with a single F-statistic < 10 or linked to phenotypes related to potential confounding factors. The PhenoScanner database was queried to examine phenotypes associated with each genetic variant. Variants linked to phenotypes related to potential confounding factors, such as OSA, pain, smoking, or psychosocial conditions, were excluded based on PhenoScanner[64]. Finally, 23 SNPs (F ranging from 15.730 to 1070.37) were selected for subsequent MR analysis. The information gathered for the SNPs for snoring used in the current study is summarized in Supporting Information Table S1, mainly including effect alleles, other alleles and summary statistics.
Fig. 2 summarizes the results in the flow chart.
The relationships between snoring and TMD were investigated independently using IVW, MR‒Egger, Weighted median, weighted mode, and simple mode approaches. Table 1 and Fig. 3 show that statistical significance was observed in the IVW analysis [odds ratio (OR) = 0.156, 95% confidence interval (CI) = 0.028 to 0.843, p = 0.0309]. A forest plot showing the effects of snoring on TMD is shown in Fig. 4. No significant effects were detected in the MR‒Egger (OR = 0.210, 95% CI = 0.038 to 1.649, p = 0.7266), WM (OR = 0.091, 95% CI = 0.0074 to 1.129, p = 0.0506), weighted mode (OR = 0.017, 95% CI = 0.065 to 4.696, p = 0.1703) or simple mode (OR = 0.028, 95% CI = 0.0138 to 6.000, p = 0.2063) analyses. What’s more, a high statistical power (90%) was identified in our study analyzed in mRnd (https://shiny.cnsgenomics.com/mRnd/)[60].
CONCLUSIONS
The study utilized MR to explore the relationship between snoring and TMD, finding that a genetic predisposition to snoring may reduce the risk of TMD. No link was found in the opposite direction. It hypothesized that snoring could have a protective effect on TMD through several possible mechanisms like intermittent hypoxia and vibratory disruption of pain signals, offering a new perspective on their association.
