Causal relationships between rhinosiusitis and gastroesophageal reflux disease： a bidirectional two-sample Mendelian randomization study

doi:10.21203/rs.3.rs-3782277/v1

Download PDF

Research Article

Causal relationships between rhinosiusitis and gastroesophageal reflux disease： a bidirectional two-sample Mendelian randomization study

https://doi.org/10.21203/rs.3.rs-3782277/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Gastroesophageal reflux disease (GERD) is associated with acute and chronic rhinosinusitis (ARS/CRS), although the causal relationship between them is not clear.

Methods

We conducted a bidirectional two-sample Mendelian randomized study to investigate the causal effects between GERD and rhinosinusitis (RS). The data for GERD and RS were obtained from large sample size genome-wide association studies (GWAS). Single-nucleotide polymorphisms (SNPs) associated with each exposure were considered instrumental variables in this study. We conducted a causal effect analysis using the inverse-variance weighted (IVW)，Weighted median，MR Egger，Simple mode and Weighted mode methods. The inverse-variance weighted and MR - Egger methods were used for the heterogeneity analysis. The MR - Egger model method was adopted for Pleiotropic effects.

Results

MR analysis showed that GERD had a statistically significant causal effect on ARS (IVW, OR = 1.324, 95% CI 1.177-1.490, p <0.001) and CRS (IVW, OR = 1.365, 95% CI 1.185-1.572, p <0.001).MR analysis showed that CRS had a statistically significant causal effect on GERD (IVW, OR = 1.033, 95% CI 1.009-1.056, p =0.006). However, the MR-Egger model revealed the evidence of horizontal pleiotropy for the causal effect of CRS on GERD (Egger-intercept=0.007, p=0.032). In addition，we did not observe a causal effect of ARS on GERD (IVW, OR = 1.019, 95% CI 0.986-1.053, p=0.269).

Conclusions

We confirmed the causal effect of GRED on RS, while no causal effect of RS on GERD was found. These studies are expected to provide high-quality causal evidence for the pathogenesis of RS and GERD, and additional ideas for their primary prevention and treatment.

Gastroesophageal reflux disease

rhinosinusitis

Mendelian randomization

Causal effect

Rhinosiusitis is one of the most common mucosal inflammatory diseases of the upper respiratory tract, including acute rhinosiusitis and chronic rhinosiusitis. Epidemiological studies have found that the incidence of CRS in our country is as high as 8%[1], and about 10% in western countries[2]. So far, the main medical treatment for RS has been drug therapy and surgery[2]. RS imposes a heavy economic burden on society in terms of health care and loss of productivity[3, 4]. According to most guidelines, topical corticosteroids, oral antibiotics, and intranasal saline irrigation are recommended for patients with RS;Surgery is recommended for patients with severe RS or who have poor response with medical treatment[5]. However, some patients with RS continue to relapse after regular medication and surgery. Therefore, it is necessary to explore the potential modifiable risk factors of RS from the etiological perspective, so as to provide scientific basis for the daily prevention and clinical treatment of RS.

Similarly, Gastroesophageal reflux disease is a common disease in both adults and children, and the global incidence of GERD is increasing year by year[6]. In GERD, risk factors include lifestyle factors (e.g eating before sleep, lifting heavy objects, overeating), types of food (e.g high-fat foods, carbonated beverages, caffeine, chocolate and alcohol), and typical medications (e.g Anticholinergics, SSRI antidepressant and contraceptives)[7]. These may trigger acid reflux symptoms. In addition, some diseases such as sinusitis, asthma and pharyngitis have also been suggested to be related to GERD[8–10]. Both gastroesophageal reflux and sinusitis are common in the clinic, it is difficult to establish a direct relationship between them. The causal relationships between these diseases and GERD remain elusive and thus need verification.

The Mendelian randomized (MR) Study is an emerging approach to obtain the causal relationship between exposure (potential risk factors) and outcome(disease). It seeks to explore the causal relationship between exposure and outcome by obtaining expose-related single-nucleotide polymorphisms (SNPs) as instrumental variables. Based on literature research, no studies have used the MR studies to explore a causal relationship between GERD and RS. In this study, we used the MR study to analyze the causal relationship between GERD and RS.

Study design

In clinical work, we often see patients with both sinusitis and acid reflux, which aroused our interest in investigating whether there is a relationship between them. Therefore, we used a bidirectional MR study to explore the causal effect between GERD and RS. The MR Study is based on three assumptions: First, instrumental variables are correlated with exposure factors; Second, instrumental variables can affect outcome only through exposure factors; Third, instrumental variables are independent of other confounding factors[11].

Data sources

The data for this study were derived from GWAS studies on people of European ancestry. Data on GERD included 602,604 samples, including 473,524 controls and 129,080 cases. Data on CRS included 176,373 samples, including 167,849 controls and 8524 cases. Data on ARS included 193,861 samples, including 182,945 controls and 10,916 cases. All original studies received ethical approval and informed consent from all participants was collected by the original investigator.

Instrumental variables (genetic variants) selection

When GERD was used as exposure and RS as outcome, the P-value of genome-wide significance threshold we adopted was less than 5×10^− 8. When RS was used as the exposure and GERD as outcome, the P-value of genome-wide significance threshold we used was less than 5×10^− 5, since few SNPs could achieve P-value less than 5×10^− 8. We removed SNPs with palindromic structure and linkage disequilibrium (r2 > 0.001).

Statistics analysis

Five methods were used to assess the causal effect, of which IVW was the main analytical method. IVW, MR-Egger methods and funnel plot were used for heterogeneity analysis. In addition, horizontal pleiotropy was verified by MR - Egger model method. The overall causal effect was evaluated again by removing different SNP each time using the leave-one-out method and forest plot.

The causal effect of exposure on outcomes can be described using odds ratio (OR) value and a 95% confidence interval (95% CI). To further confirm the reliability of our results, the Bonferroni adjusted p-value of significance was less than 0.0125 (0.05/4 = 0.0125). In this study, the causal effect of exposure on outcome was considered to have strong evidence when p-value was less than 0.0125. When p-value is greater than 0.0125 but less than 0.05, the causal effect of exposure on the outcome is considered to have suggestive evidence.

The R software (version 4.2.3) is used for data extraction, analysis, and visualization. The R package used is TwoSampleMR (version 0.5.6).

Causal relationship between GERD and RS in forward direction

MR analysis showed that GERD had a statistically significant causal effect on ARS and CRS.

As presented in Table 1, the risk of ARS increased with the increase in GERD (IVW, OR = 1.324, 95% CI 1.177–1.490, p < 0.001). Figure 1A depicted the relationship between GERD and risk of ARS. The MR-Egger model revealed no evidence of horizontal pleiotropy for the causal effect of GERD on ARS (Egger-intercept=-0.015, p = 0.184). The funnel plot showed little heterogeneity between them (Fig.S1). Forest plot revealed that GRED increases the risk of ARS with IVW method (Fig.S2). Leave-one-out sensitivity analysis revealed that no individual SNP significantly affected the causal effect of GERD on ARS (Fig.S3).

Table 1

Causal effects of gastroesophageal reflux disease on rhinosinusitis
Exposure	Outcome	No. of SNPs	Method	β	SE	p	OR	Lower 95% CI for OR	Upper 95% CI for OR	Intercept p value
GERD	ARS	75	MR Egger	0.747	0.352	0.037	2.111	1.058	4.213	0.184
			WM¹	0.282	0.082	< 0.001	1.326	1.129	1.557	NA
			IVW	0.281	0.060	< 0.001	1.324	1.177	1.490	NA
			SM	0.211	0.209	0.317	1.235	0.819	1.862	NA
			WM²	0.278	0.182	0.131	1.320	0.924	1.885	NA
GERD	CRS	75	MR Egger	0.493	0.426	0.252	1.636	0.710	3.773	0.667
			WM¹	0.371	0.098	< 0.001	1.450	1.196	1.757	NA
			IVW	0.311	0.072	< 0.001	1.365	1.185	1.572	NA
			SM	0.654	0.249	0.01	1.924	1.182	3.133	NA
			WM²	0.645	0.233	0.007	1.907	1.207	3.012	NA
GERD gastroesophageal reflux disease, ARS acute rhinosinusitis, CRS chronic rhinosinusitis, MR Mendelian randomization, WM¹ weighted median, IVW inverse-variance weighted, SM Simple mode, WM² Weighted mode, CI confidence interval, OR odds ratio, SE standard error, SNP single-nucleotide polymorphism, NA not applicable

The risk of CRS increased with GERD (IVW, OR = 1.365, 95% CI 1.185–1.572, p < 0.001). Figure 1B depicted the relationship between GERD and risk of CRS. The MR-Egger model revealed no evidence of horizontal pleiotropy for the causal effect of GERD on CRS (Egger-intercept=-0.006, p = 0.667). The funnel plot showed little heterogeneity between them (Fig.S4). Forest plot revealed that GRED increases the risk of CRS with IVW method (Fig.S5). Leave-one-out sensitivity analysis revealed that no individual SNP significantly affected the causal effect of GERD on CRS (Fig.S6).

Causal relationship between GERD and RS in backward direction

MR analysis showed that CRS had a statistically significant causal effect on GERD. As presented in Table 2, the risk of GERD increased with CRS (IVW, OR = 1.033, 95% CI 1.009–1.056, p = 0.006). Figure 1C depicted the relationship between CRS and risk of GERD. However, the MR-Egger model revealed the evidence of horizontal pleiotropy for the causal effect of CRS on GERD (Egger-intercept = 0.007, p = 0.032). Therefore, the causal effect of CRS on GERD was not credible in the present study.

Table 2

Causal effects of rhinosinusitis on gastroesophageal reflux disease
Exposure	Outcome	No. of SNPs	Method	β	SE	p	OR	Lower 95% CI for OR	Upper 95% CI for OR	Intercept p value
ARS	GERD	40	MR Egger	-0.049	0.042	0.257	0.952	0.877	1.035	0.092
			WM¹	0.021	0.019	0.272	1.021	0.984	1.059	NA
			IVW	0.019	0.017	0.269	1.019	0.986	1.053	NA
			SM	0.084	0.045	0.070	1.088	0.996	1.189	NA
			WM²	0.056	0.037	0.134	1.058	0.984	1.137	NA
CRS	GERD	55	MR Egger	-0.034	0.032	0.290	0.966	0.907	1.029	0.032
			WM¹	0.018	0.014	0.178	1.019	0.992	1.046	NA
			IVW	0.032	0.012	0.006	1.033	1.009	1.056	NA
			SM	0.015	0.033	0.653	1.015	0.951	1.082	NA
			WM²	0.006	0.025	0.795	1.006	0.959	1.056	NA
GERD gastroesophageal reflux disease, ARS acute rhinosinusitis, CRS chronic rhinosinusitis, MR Mendelian randomization, WM¹ weighted median, IVW inverse-variance weighted, SM Simple mode, WM² Weighted mode, CI confidence interval, OR odds ratio, SE standard error, SNP single-nucleotide polymorphism, NA not applicable

In addition, we did not observe a causal effect of ARS on GERD (IVW, OR = 1.019, 95% CI 0.986–1.053, p = 0.269).

This study showed that GERD had a causal effect on both ARS and CRS. In the current literature search, this is the first time that MR study has been used to explore the causal effect between GERD and RS. RS remains a major health problem in the world. The latest European Position Paper on Rhinosinusitis and Nasal Polyps (EPOS 2020) stated that the four symptoms in the diagnosis of RS were nasal blockage/obstruction/congestion, nasal discharge (anterior / posterior nasal drip), facial pain/pressure and reduction or loss of smell[12]. The current broad consensus on recommended treatment for sinusitis is the appropriate use of corticosteroids and functional endoscopic nasal surgery (FESS). However, there are a number of factors that are thought to contribute to refractory sinusitis or recurrent sinusitis, including allergic rhinitis, asthma, smoking, and acid reflux.

GERD may be a negative factor for the treatment effect and recurrence of CRS; the treatment of GERD can contribute to the treatment of RS and reduce the recurrence[13]. Our findings suggested that GERD had a causal effect on ARS and CRS. At present, there are different theories on the pathogenesis of RS caused by GERD, mainly as follows: First, the influence of acid reflux on nasal cavity and sinus mucosa. Acid reflux directly irritate the nasal and sinus mucosa, causing damage to the movement of the mucosa and cilia, and eventually leading to sinus opening obstruction and infection[14, 15]. Currently, reflux contents that may cause mucosal damage may include gastric acid and pepsin. After acid exposed to nasal epithelial cells, mature E-cadherin decreased and cleaved E-cadherin increased; It has been suggested that acid exposure to alter the expression of E-cadherin in nasal mucosal epithelium may be a possible mechanism of nasal tissue damage associated with chronic nasal inflammation and acid reflux[16]. Ren JJ et al. measured the levels of pepsin A in nasal mucosa, nasal secretions, and plasma, and found that patients with sinusitis had higher levels of pepsin A compared to controls[17]. Acid reflux may play a role in sinusitis through pepsin, and the increased expression of heat shock protein 70 may be related to the pathological mechanism of mucosal injury in sinusitis[18]. Second, neurogenic mechanisms mediated by the vagus nerve. This mechanism involves disturbance of the self-meridian system, leading to edema of the naso-sinuses and subsequent obstruction of the sinuses opening. An enhanced vagal response was found in patients with asthma and GERD compared with controls, and the vagal disturbance occurred in the respiratory tract[19]. Wong et al. found that in gastro-nasal reflex, the infusion of hydrochloric saline in the lower esophagus increased nasal mucus production and nasal symptom score[20]. In recent years, there has been a lack of research on GERD causing CRS by disrupting autonomic nervous function, and no new studies have further confirmed the mechanism. Third, Helicobacter pylori (H. pylori) infection. H. pylori was associated with the severity of GERD[21, 22]. Siupsinskiene N et al. found that patients with CRS have higher prevalence of H. pylori infection in the nasal cavity than the normal population[23]. Polymerse chain reaction technology was used to find a significant correlation between CRS and the presence of H. pylori in the nasal mucosa; This relationship may reflect the possibility of HP as a risk factor for CRS[24]. Nasal mucociliary clearance (NMC) dysfunction plays an essential role in the pathogenesis of CRS). Sahin E et al. found that H. pylori-positive patients had a longer duration of NMC than H. pylori-negative patients[25]. Therefore, they speculated that eradication of H. pylori would have a positive effect on the prognosis of CRS.

There were some literatures suggesting that treatment of RS was helpful in improving the symptoms of acid reflux, including laryngopharyngeal reflux (LPR) and GERD. Nam-Kyung Yeo et al. conducted a retrospective analysis of 91 patients with simultaneous LPR and CRS. The results showed that subjective CRS symptoms were correlated with subjective LPR symptoms; Moreover, FESS can effectively reduce the symptoms and signs of LPR in patients with CRS[26]. Yang Xu et al. evaluated the impact of GERD on the quality of life of CRS patients before and after FESS, and the results showed that compared with CRS patients without GERD, GERD patients with CRS had worse quality of life, and after FESS, Quality of life improved in these GERD patients with CRS[27]. In this study, as shown in Table 2, we saw that the risk of GERD rised with CRS occurrence in the IVW method. Unfortunately, the MR-Egger model revealed the evidence of horizontal pleiotropy for the causal effect of CRS on GERD. Therefore, the causal effect of CRS on GERD has not been proven.

This paper revealed the causal effect between gastroesophageal reflux and sinusitis using MR study method for the first time. It should also be pointed out that there are some shortcomings in this study. First, the data we used in the GWAS database found that the number of patients with sinusitis was relatively low compared to the number of patients diagnosed with gastroesophageal reflux. In the study of causality, big data may get more convincing results. Second, to explore the causal effect of sinusitis on gastroesophageal reflux. Because we could not get a valid number of SNPs, we changed the P-value of genome-wide significance threshold from 5×10^− 8 to 5×10^− 6, which may result in a more desirable positive result. Third, the volume of this study is relatively thin, and more longitudinal studies are needed in the later stage to explore the causal relationship between GERD and RS. Fourth, the mechanism of interaction between GERD and RS has not been further explored in this paper. Fifth, the data used in this paper are all European population data, and do not include data of other populations and races. Other data can be collected in the later stage to make the results more authentic and reliable.

In conclusion, We confirmed the causal effect of GRED on RS, while no causal effect of RS on GERD was found. These studies are expected to provide high-quality causal evidence for the pathogenesis of RS and GERD, and additional ideas for their primary prevention and treatment.

Acknowledgements

We are grateful to the researchers who conducted GWAS as well as to all participants, whose contributions were invaluable to the study.

Authors’ contributions

FC, JZ and XF contributed to the study conception and design. The literature was searched by all authors. Data collection was performed by all authors. Data analysis was performed by FC, JZ and XF. The manuscript was drafted and revised by FC, JZ and XF. The final manuscript was read and approved by all authors.

Funding

None

Availability of data and materials

All data generated or analyzed and materials during this study are available from the corresponding author on reasonable request.

Competing interests

All remaining authors declare that they have no competing interests.

Ethics approval and consent to participate

The data used in this paper are publicly available, ethically approved, and the subjects have given their informed consent.

Consent for publication

Not applicable

Shi JB, Fu QL, Zhang H, Cheng L, Wang YJ, Zhu DD, Lv W, Liu SX, Li PZ, Ou CQ et al: Epidemiology of chronic rhinosinusitis: results from a cross-sectional survey in seven Chinese cities. Allergy 2015, 70(5):533-539.
Sedaghat AR, Kuan EC, Scadding GK: Epidemiology of Chronic Rhinosinusitis: Prevalence and Risk Factors. J Allergy Clin Immunol Pract 2022, 10(6):1395-1403.
Hirsch AG, Nordberg C, Bandeen-Roche K, Tan BK, Schleimer RP, Kern RC, Sundaresan A, Pinto JM, Kennedy TL, Greene JS et al: Radiologic sinus inflammation and symptoms of chronic rhinosinusitis in a population-based sample. Allergy 2020, 75(4):911-920.
Kato A, Schleimer RP, Bleier BS: Mechanisms and pathogenesis of chronic rhinosinusitis. J Allergy Clin Immunol 2022, 149(5):1491-1503.
Wallace DV: Treatment options for chronic rhinosinusitis with nasal polyps. Allergy Asthma Proc 2021, 42(6):450-460.
Fass R, Boeckxstaens GE, El-Serag H, Rosen R, Sifrim D, Vaezi MF: Gastro-oesophageal reflux disease. Nat Rev Dis Primers 2021, 7(1):55.
Chen J, Brady P: Gastroesophageal Reflux Disease: Pathophysiology, Diagnosis, and Treatment. Gastroenterol Nurs 2019, 42(1):20-28.
Finocchio E, Locatelli F, Sanna F, Vesentini R, Marchetti P, Spiteri G, Antonicelli L, Battaglia S, Bono R, Corsico AG et al: Gastritis and gastroesophageal reflux disease are strongly associated with non-allergic nasal disorders. BMC Pulm Med 2021, 21(1):53.
Robinson LB, Ruffner MA: Proton Pump Inhibitors in Allergy: Benefits and Risks. J Allergy Clin Immunol Pract 2022, 10(12):3117-3123.
Zhao X, Li ST, Chen LH, Liu K, Lian M, Wang HJ, Fang YJ: Identification of independent risk factors for intraoperative gastroesophageal reflux in adult patients undergoing general anesthesia. World J Clin Cases 2021, 9(35):10861-10870.
Davies NM, Holmes MV, Davey Smith G: Reading Mendelian randomisation studies: a guide, glossary, and checklist for clinicians. BMJ 2018, 362:k601.
Fokkens WJ, Lund VJ, Mullol J, Bachert C, Alobid I, Baroody F, Cohen N, Cervin A, Douglas R, Gevaert P et al: European Position Paper on Rhinosinusitis and Nasal Polyps 2012. Rhinol Suppl 2012, 23:3 p preceding table of contents, 1-298.
Lechien JA-O, Debie G, Mahillon V, Thill MP, Rodriguez A, Horoi M, Kampouridis S, Muls V, Saussez S: A 10-Year Follow-Up of a Randomized Prospective Study of 2 Treatments for Chronic Rhinosinusitis Without Nasal Polyps and Investigation of the Impact of Gastroeosophageal Reflux Disease in the Resistance to Treatment. Ear Nose Throat J 2021, 100(5_suppl)(1942-7522 (Electronic)):569S-577S.
DelGaudio JM: Direct nasopharyngeal reflux of gastric acid is a contributing factor in refractory chronic rhinosinusitis. Laryngoscope 2005, 115(6)(0023-852X (Print)):946-957.
Delehaye E, Dore Mp Fau - Bozzo C, Bozzo C Fau - Mameli L, Mameli L Fau - Delitala G, Delitala G Fau - Meloni F, Meloni F: Correlation between nasal mucociliary clearance time and gastroesophageal reflux disease: our experience on 50 patients. Auris Nasus Larynx 2009, 36(2)(1879-1476 (Electronic)):157-161.
Kim B, Lee HJ, Im NR, Lee DY, Kang CY, Park IH, Lee SH, Lee SH, Baek SK, Kim TH: Effect of matrix metalloproteinase inhibitor on disrupted E-cadherin after acid exposure in the human nasal epithelium. Laryngoscope 2018, 128(1):E1-E7.
Ren JJ, Zhao Y, Wang J, Ren X, Xu Y, Tang W, He Z: PepsinA as a Marker of Laryngopharyngeal Reflux Detected in Chronic Rhinosinusitis Patients. Otolaryngol Head Neck Surg 2017, 156(5)(1097-6817 (Electronic)):893-900.
Wang J, Yu Z, Ren J, Xu Y, Zhang Y, Lei L, Zheng Y, Huang L, He Z: Effects of pepsin A on heat shock protein 70 response in laryngopharyngeal reflux patients with chronic rhinosinusitis. Acta Otolaryngol 2017, 137(12):1253-1259.
Lodi U, Harding SM, Coghlan HC, Guzzo MR, Walker LH: Autonomic regulation in asthmatics with gastroesophageal reflux. Chest 1997, 111(1):65-70.
Wong IW, Rees G, Greiff L, Myers JC, Jamieson GG, Wormald PJ: Gastroesophageal reflux disease and chronic sinusitis: in search of an esophageal-nasal reflex. Am J Rhinol Allergy 2010, 24(4):255-259.
Scida S, Russo M, Miraglia C, Leandro G, Franzoni L, Meschi T, De' Angelis GL, Di Mario F: Relationship between Helicobacter pylori infection and GERD. Acta Biomed 2018, 89(8-S):40-43.
Yucel O: Interactions between Helicobacter pylori and gastroesophageal reflux disease. Esophagus 2019, 16（1）(1612-9067 (Electronic)):52-62.
Siupsinskiene N, Katutiene I, Jonikiene V, Janciauskas D, Vaitkus S: Intranasal Helicobacter pylori infection in patients with chronic rhinosinusitis with polyposis. J Laryngol Otol 2018, 132(9):816-821.
Nikakhlagh S, Samarbafzadeh AR, Jahani M, Poostchi H, Kayedani GA, Naghashpoor M, Saki N: Determining the Role of Helicobacter pylori in Chronic Sinus Infections Using the Polymerase Chain Reaction. Jundishapur J Microbiol 2015, 8(3):e20783.
Sahin E, Katar MK, Haberal Can I: Impact of gastric Helicobacter pylori infection on nasal mucociliary clearance. Eur Arch Otorhinolaryngol 2020, 277(10):2761-2765.
Yeo NK, Park SJ, An TH: Laryngopharyngeal reflux in chronic rhinosinusitis patients and the role of endoscopic sinus surgery. Auris Nasus Larynx 2022, 49(4):663-669.
Xu Y, Yang W, Song Y, Cheng D, Cao M, Huang J, Ren J, Lee JW, Hueniken K, Chen Y et al: Association of gastro-oesophageal reflux disease and quality of life in patients with chronic rhinosinusitis. Acta Otorhinolaryngol Ital 2021, 41(5):450-460.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Causal relationships between rhinosiusitis and gastroesophageal reflux disease： a bidirectional two-sample Mendelian randomization study

Status:

Version 1

Abstract

Figures

Introduction

Materials and methods

Study design

Data sources

Instrumental variables (genetic variants) selection

Statistics analysis

Results

Causal relationship between GERD and RS in forward direction

Causal relationship between GERD and RS in backward direction

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1