Role of Pepsin and Pepsinogen: Linking Laryngopharyngeal Reflux With Otitis Media With Effusion in Children

Transcription

1 The Laryngoscope VC 2013 The American Laryngological, Rhinological and Otological Society, Inc. Role of Pepsin and Pepsinogen: Linking Laryngopharyngeal Reflux With Otitis Media With Effusion in Children Hua-Nan Luo, MD; Qi-Mei Yang, MS; Ying Sheng, BS; Zheng-Hui Wang, MD; Qing Zhang, MD; Jing Yan, MS; Jin Hou, MS; Kang Zhu, MS; Ying Cheng, MS; Bo-Tao Wang, MS; Ying-Long Xu, MS; Xiang-Hong Zhang, MS; Xiao-Yong Ren, MD; Min Xu, MD Objectives/Hypothesis: To analyze the relationship between laryngopharyngeal reflux (LPR) represented by pepsin and pepsinogen, and pathogenesis of otitis media with effusion (OME). Study Design: Prospective case-control study. Methods: Children with OME who required adenoidectomy and tympanostomy/tympanostomy tubes placement were enrolled in OME group, whereas children with adenoid hypertrophy (AH) who required adenoidectomy and individuals who required cochlear implantation (CI) were enrolled in AH and CI groups, respectively. Pepsinogen mrna and protein levels were assessed by real-time fluorescence-based quantitative polymerase chain reaction and immunohistochemistry in adenoid specimens from the OME and AH groups. Pepsin and pepsinogen concentrations were evaluated by enzyme-linked immunosorbent assay in middle ear fluid and plasma from the OME and CI groups. Results: The levels of pepsinogen protein expressed in cytoplasm of epithelial cells and clearance under epithelial cells in adenoid specimens from the OME group were significantly higher than those in the AH group. Furthermore, the concentrations of pepsin and pepsinogen in the OME group were ng/ml and ng/ml, respectively, which were significantly higher than those in the CI group (P <.001). In addition, the concentrations of pepsin in dry ears were significantly lower than those in serous and mucus ears in the OME group (F , P <.001).Finally, the concentration of pepsinogen in middle ear effusion was positively correlated with the expression intensity of pepsinogen protein in cytoplasm of epithelial cells(r , P <.05) in the OME group. Conclusions: Pepsin and pepsinogen in middle ear effusion are probably caused by LPR and may be involved in the pathogenesis of OME. Key Words: Otitis media with effusion, laryngopharyngeal reflux, pepsin, pepsinogen. Level of Evidence: 3b Laryngoscope, 124:E294 E300, 2014 INTRODUCTION Otitis media with effusion (OME), a kind of inflammatory disease of the middle ear that is characterized by the retention of middle ear effusion (MEE) and hearing loss, occurs frequently in 2 to 8 year olds. 1 It is one of the important causes of hearing loss, which can lead From the Department of Otolaryngology Head and Neck Surgery (H.N.L., Y.S., Z.-H.W., Q.Z., J.Y., J.H., K.Z., Y.C., B.-T.W., Y.-L.X., X.-H.Z., X.-Y.R., M.X.), The Second Hospital, Xi an Jiao Tong University, Xi an; and the Department of Otolaryngology Head and Neck Surgery (Q.-M.Y.), Shan Xi Provincial People s Hospital, Xi an, China Editor s Note: This Manuscript was accepted for publication November 25, This work was supported by scientific research funds from the Second Hospital, Xi an JiaoTong University (YJ [QN] to Hua-Nan Luo). The authors have no other funding, financial relationships, or conflicts of interest to disclose. Hua-Nan Luo, MD, and Qi-Mei Yang, MS, contributed equally to this work. Send correspondence to Xiao-Yong Ren, MD, Department of Otolaryngology Head and Neck Surgery, The Second Hospital, Xi an Jiao Tong University, Xi an , China. renxiaoyong@vip.sina.com; Min-Xu, Department of Otolaryngology Head and Neck Surgery, The Second Hospital, Xi an Jiao Tong University, Xi an , China. ent551205@163.com DOI: /lary E294 to profound effects on language and cognitive development of children. 2 There are many factors involved in the development of OME in children, including bacterial infection, formation of bacterial biofilms, anatomical characteristics of the eustachian tube (wide, short, and straight), an immature immune system, and changes in anatomy and physiology with regard to respiratory and digestive tracts, resulting in children who are prone to vomit because of a weak closure function of the sphincter in the stomach and prone to be infected because of a confined pediatric airway and poor ciliary movement. 3 5 Laryngopharyngeal reflux (LPR) results from the reflux of stomach contents to the pharynx and leads to stimulation and injury of laryngopharyngeal mucosa, causing uncomfortable symptoms. In fact, reflux of stomach contents is common in infants and young children, and occurs in about two-thirds of 4-month-old infants, but it decreases in frequency during the first year of life. 6 LPR is different from classic gastroesophageal reflux (GER). The primary defect in LPR is thought to be dysfunction of the upper esophageal sphincter, rather than dysfunction of the lower esophagus as in GER. 7 Although 24-hour double probe (pharyngeal and esophageal) ph monitoring is preferentially chosen for diagnosis of LPR because of its highly sensitive and specific

2 characteristic (the gold standard for diagnosis of LPR), it should be performed by a specialist without anesthesia, and younger patients often need to be hospitalized for overnight observation to avoid accidental retraction of the probe. 8,9 In view of the discomfort of patients and the inconvenience to children, many researchers have used related scales, such as the reflux symptom index, reflux finding score, and other indicators, such as pepsinogen levels, to evaluate LPR Pepsin often exists in the form of pepsinogen, which is secreted mainly by stomach mucosa cells and completely deactivated in alkaline environments such as the middle ear cavity. 13,14 The process of pepsinogen activation requires cleavage of 44 amino acids from the primary structure of pepsinogen to transform it into active pepsin form in acidic conditions. Therefore, pepsinogen may exert biological activity when activated if the ph is decreased by conversion to pepsin. Pepsin is not detected in the middle ear under normal conditions, but pepsinogen (inactive) may be activated as pepsin (active) under stimulation of hydrochloric acid in gastric juice during reflux. 15 LPR was recently implicated in the etiology of OME, especially in children. 16,17 Aydin et al. showed the presence of higher LPR rates in OME patients with adenoid hypertrophy (AH), compared with control subjects who had AH (with or without tonsillar hypertrophy) without any ear problems, indicating that LPR may be involved in the pathogenesis of OME in children. 18 Recently, Al-Saab et al. provided the first line of evidence linking LPR to OME, by detecting pepsinogen in adenoid tissue and middle ear fluid. 19 Meanwhile, O Reilly et al. concluded that recovery of pepsin in the middle ear space of pediatric patients with otitis media (OM) is an independent risk factor, and gastroesophageal reflux and extraesophageal reflux had been postulated to be involved in the pathogenesis of OM Their results, which indicated that adenoid tissues from the OME group had significantly higher pepsinogen immunoreactivity compared with those from the control group without OME, are suggestive of the involvement of LPR in OME. However, the study did not address whether there is physiologic reflux of pepsinogen that reaches the middle ear through the nasopharynx and eustachian tube in normal children, or whether there is physiologic accumulation of pepsinogen in the sera of normal children. Further evaluation of the involvement of LPR in OME is warranted, and pepsinogen levels in OME and control groups, the latter of which should include normal children (a true control group), need to be compared to fully address this issue and delineate the precise role of pepsinogen in OME. Based on the above data, we hypothesized that the appearance of pepsinogen in the middle ear cavity and in adenoid specimens may be caused, in part, by LPR and involved in the pathogenesis of OME. To test this hypothesis, we analyzed the expression of pepsinogen mrna and/or protein in adenoid specimens from OME and AH groups, and measured the pepsin/pepsinogen concentration in MEE and serum samples from OME and cochlear implantation (CI) groups. We also determined whether the detection rates and concentrations of pepsinogen vary with degree of inflammation in OME to explore the precise role of pepsinogen in the pathogenesis of OME in children, with the aim of providing a theoretical basis for the origin and course of OME and the possibility of clinical application of antireflux therapeutic strategies for OME in children. MATERIALS AND METHODS Study Subjects Children diagnosed as AH with or without OME and as having congenital/prelingual profound deafness without OME indicated for CI were consecutively enrolled in this prospective casecontrol study from May of 2011 to December of 2012, undertaken at the Department of Otolaryngology Head and Neck Surgery, The Second Hospital, Xi an Jiao Tong University. The subjects were divided into the following groups: OME, AH, and CI groups, where the latter two groups were established as control groups. Children who were between 2 and 8 years old and included in the study (OME) group required tympanotomy/tympanostomy tube insertion and adenoidectomy. Exclusion criteria for the OME group were children with craniofacial deformities or acute and chronic OM. Inclusion criteria for the AH group were children between 2 and 8 years old, clinically diagnosed as AH without any ear problems, and suitable for adenoidectomy. Inclusion criteria for the CI group were children between 2 and 8 years old, clinically diagnosed as congenital/prelingual profound deafness without OME, and requiried CI. Exclusion criteria for the control groups were craniocerebral injury disease, tympanic membrane perforation, chronic suppurative otitis media, and other middle ear disease. Finally, the patients were excluded who were on medication for reducing stomach acidity, such as proton pump inhibitors, in the past 1 month preceding the start date of the study in the OME group and control groups. Informed consent was obtained from parents of each participating patient and the research protocols were approved by the ethics committee of the Second Hospital, Xi an Jiao Tong University (registration number ). Sampling Methods Collection of adenoid specimens. Adenoid tissue specimens were obtained from the OME and AH groups during their scheduled operation at the Second Hospital, Xi an Jiao Tong University. After completion of adenoidectomy, adenoid samples were carefully cleaned with sterile brine and the surfaces of the samples, including the epithelial layer, were removed by excision. All specimens were divided into two portions: one portion was kept at 280 C for performing polymerase chain reaction (PCR), whereas the other portion was fixed in 10% formalin solution, embedded in paraffin, and microdissected into 4- to 6-lmthick sections for performing immunohistochemical analyses. Collection of middle ear fluid from the tympanotomy/ tympanostomy tube in the OME group. The aspirated secretions were collected during tympanotomy or tympanostomy tube insertion and stored at 280 C. If fluid was present, we collected it in the cannula, after which 0.5 ml of sterile brine was flushed into the middle ear cleft and aspirated into the same cannula. If there was no effusion in the middle ear cavity, to recover pepsin adherent to the middle ear mucosa, 0.5 ml of sterile brine was lavaged into the middle ear cavity, collected, and stored at 280 C. Effusions were divided into the dry ear, serous ear, and mucoid ear subgroups according to appearance of the aspirated material. Collection of middle ear lavage fluid in the CI group. CI was carried out via the mastoid path. Lavage was stopped when the tympanic antrum inlet was exposed, guaranteeing that the lavage fluid did not flow outside the middle ear cavity. E295

3 A small amount of sterile brine (0.5 ml) was infiltrated into the middle ear through the tympanic antrum inlet with a vascular probe, and the fluid was collected and stored at 280 C. Collection of plasma. Blood samples were collected from all subjects enrolled in this study. The supernatant of plasma was acquired from fasting venous blood at in the after allowing the samples to stand for 1 hour at 4 C and then centrifuged for 10 minutes at 3,000 rpm, followed by storage at 280 C until use. Quantitative Real-Time PCR and Immunohistochemistry Adenoid samples were fully ground and then total RNA was isolated with Trizol (Invitrogen, Carlsbad, CA). The RNA was reverse-transcribed into cdna with a random primer from the First Strand cdna Synthesis Kit (Invitrogen). The mrna levels of pepsinogen and glyceraldehyde 3-phosphate dehydrogenase, the internal control, were evaluated by quantitative PCR (Q-PCR), and the relative expression levels were calculated as described before using the SYBR Green PCR master mix(applied Biosystems Inc., Foster City, CA) with an ABI 7500HT system (Applied Biosystems). 23 The streptavidin-perosidase detection method was used for immunohistochemical analysis of pepsinogen protein. 24 Amouse polyclonal anti-pepsinogen antibody (R&D Systems, Minneapolis, MN) was diluted at 1:100. A monoclonal rabbit anti-mouse immunoglobulin G antibody was used as the second antibody (OriGene Technologies, Inc., Rockville, MD). Tissue sections of gastric mucosa from children who had undergone stomach surgeries served as the positive control, and phosphate-buffered saline was used instead of the primary antibody as the negative control. The results were scored by intensity (1 4) and percentage of positive cells (1 4). Relative expression levels were obtained by multiplying intensity by percentage of positively stained cells. Pepsin and Pepsinogen Detection by Enzyme- Linked Immunosorbent Assay in MEE and Middle Ear Lavage Fluid Commercially-available enzyme-linked immunosorbent assay (ELISA) kits were used for detection of pepsin and pepsinogen (Fujian Blueprint Technology Company, FuZhou, China). The specific experimental procedures were carried out according to the manufacturer s instructions. Absorbance values of the samples at 450 nm were determined using a microplate reader (BioTek Instruments, Inc., Winooski, VT). Results were classified as positive if pepsin or pepsinogen was successfully detected in MEE from either ear. Statistical Analysis The SPSS 17.0 statistical software package (IBM, Armonk, NY) was used for data processing. Measurement data are expressed as mean 6 standard deviation, and the Student t test and one-way analysis of variance were used for comparing means of two or multiple independent and normal continuous variables, respectively. The v 2 test was used to compare the rates of different groups. Pearson correlation analysis was performed to determine the relationship between pepsinogen expression intensity and its concentration. All of the differences were considered statistically significant if P values were <.05. RESULTS Clinical Features of Patients in OME, AH, and CI Groups Forty-eight children (30 male and 18 female; mean age, years), 50 children (31 male and 19 female; E296 mean age, years), and 30 children (18 male and 12 female; mean age, years) were included in the OME, AH, and CI groups, respectively. A total of 98 adenoids samples (48 and 50 in the OME and AH groups, respectively) were obtained. Differences in the ages and sex distribution of patients among the three groups were not significant (P >.05). Pepsinogen mrna and Protein Detection in Adenoid Specimens Pepsinogen mrna was not detected in adenoids samples from the OME and AH groups using Q-PCR. Pepsinogen protein was strongly expressed in the positive control (Fig. 1A) and expressed either in the nuclei (Fig. 1B) or cytoplasm of epithelial cells (Fig. 1C) and clearance under the epithelial cells (Fig. 1D) of adenoid samples in the OME group. The pepsinogen protein levels were significantly higher in the OME group than in the AH group (Fig. 1E,F) (Z , P <.001 and Z , P <.001, respectively [Tables I and II]), suggesting that pepsinogen expression is associated with the pathogenesis of OME. Pepsin and Pepsinogen Detection in Middle Ear Fluid in OME and CI Groups To determine the concentration and distribution of pepsin and pepsinogen in the middle ear cavity, we detected them in the OME and CI groups with ELISA. The positivity rates for pepsin in the OME and CI groups were 71% (34/48 cases) and 16.67% (8/30 cases), respectively. The pepsin concentration in the OME group was ng/ml, which was significantly higher than in the CI group ( ng/ml; t 5 8.4, P <.001). On the other hand, the positivity rates for pepsinogen in the OME and CI groups were 81.25% (39/48 cases) and 20% (6/30 cases), respectively. The pepsinogen concentration in the OME group was ng/ml, which was significantly higher than in the CI group ( ng/ml; t , P <.001). Furthermore, not only was pepsinogen concentration in the middle ear cavity significantly higher (about times) than in the plasma ( ng/ml) from the OME group (t , P <.001), but the concentration of pepsinogen in the middle ear lavage was significantly lower than in the plasma ( ng/ml;t53.5,p &LT;.05) from the CI group, suggesting that pepsinogen likely originated from processes such as LPR, rather than from an effusion from serum. However, there was no statistically significant difference in plasma pepsinogen concentrations between the OME and CI groups (t , P >.05), further verifying that pepsinogen in serum, without conversion to pepsin, is not significantly related with OME (Table III and Fig. 2A,B). Correlation Analysis of Pepsinogen in MEE and Expression of Pepsinogen in Adenoid Specimens in the OME Group Next, we performed a Pearson correlation analysis and found that the concentration of pepsinogen in MEE was positively correlated with the expression intensity of

4 Fig. 1. Expression of pepsinogen protein in adenoid samples in the otitis media with effusion (OME) group is much higher than that in the adenoid hypertrophy (AH) group. (A) Expression of pepsinogen protein in stomach mucosa was used as the positive control. Pepsinogen is mainly expressed in the cytoplasm and nucleus of cells in stomach mucosa (red arrow). Magnification, (B D) Representative images of immunohistochemical staining for pepsinogen protein in adenoid samples from the OME group. Magnification, Pepsinogen is expressed in the nucleus (B, black arrow), cytoplasm of epithelial cells (C, green arrow), and in clearance under epithelial cells (D, yellow arrow). (E, F) Representative images of immunohistochemical staining for pepsinogen protein in adenoid samples in the AH group. Pepsinogen is barely detectable in epithelial cells and in clearance under epithelial cells. Magnification, [Color figure can be viewed in the online issue, which is available at pepsinogen protein in the cytoplasm of epithelial cells in adenoid specimens in the OME group (r , P <.05) (Fig. 2C). However, there was no significant relationship between the concentration of pepsinogen in MEE and pepsinogen expression in clearance under epithelial cells in adenoid specimens in the OME group (P >.05, data not shown). Comparison of Pepsin and Pepsinogen Concentrations Among the Dry, Serous, and Mucoid Ear Subgroups of the OME Group Sixty-nine cases of MEE were collected from the OME group, including 12 cases with dry ear (17.39%), 38 cases with serous ear (55.07%), and 19 cases with mucoid ear (27.54%). The differences in pepsin concentrations among the three subgroups were statistically significant (F , P <.001). In particular, the concentration of pepsin in the dry ear subgroup ( ng/ml) was significantly lower than that in serous ( ng/ ml; t , P <.01) and mucoid ( ng/ml; t , P <.01) ear subgroups (Table IV and Fig. 2). However, there was no significant difference between the serous and mucoid ear subgroups (t , P >.10) (Table IV and Fig. 2). In addition, the differences in pepsinogen concentrations among the three subgroups were not statistically significant (F 5 1.2, P 5.31), illustrating that active pepsin, rather than inactive pepsinogen, plays a more important role in the pathogenesis of OME. DISCUSSION In this study, we found that the expression levels of pepsinogen protein in adenoid samples in the OME TABLE I. Expression of Pepsinogen in Cytoplasm of Epithelial Cells in Adenoid Specimens From the OME and AH Groups. Pepsinogen Group Z P OME, N (31.25%) 12 (25%) 18 (37.5%) 3 (6.25%) AH, N (72%) 11 (22%) 3 (6%) <.001 Z and P values were calculated using the v 2 test. AH 5 adenoid hypertrophy; OME 5 otitis media with effusion. E297

5 TABLE II. Expression of Pepsinogen in Clearance Under Epithelial Cells in Adenoid Samples From the OME and AH Groups. Pepsinogen Group Z P OME, N (31.25%) 21 (43.75%) 12 (25%) AH, N (76%) 7 (14%) 5 (10%) <.001 Z and P values were calculated using the v 2 test. AH 5 adenoid hypertrophy; OME 5 otitis media with effusion. TABLE III. Comparison of Pepsin and Pepsinogen Concentrations Between the OME and CI Groups. Pepsin (ng/ml) Pepsinogen (ng/ml) Specimen OME CI t P OME CI t P Middle ear fluid < * <.01 Plasma >.05 Data shown are mean 6 standard deviation. *Group compared with group, t , P <.01; group compared with group, t 5 3.5, P > not detected; CI 5 cochlear implantation; OME 5 otitis media with effusion. group were significantly higher than those in the AH group; however, pepsinogen mrna could not be detected in either group. These data suggest that the detected pepsinogen protein was not originally produced in adenoid samples, but likely originated from other processes, such as LPR. Furthermore, the positivity rates and concentrations of accumulated pepsin and pepsinogen that appeared in MEE in the OME group were significantly higher than those in the CI group and in corresponding serum samples, indicating that pepsin and pepsinogen are involved in the pathogenesis of OME. In theory, pepsin and pepsinogen are not detectable in the middle ear cavity except during physiologic reflux, accounting for the detection of pepsin/pepsinogen in the normal control group (CI group), in whom reflux of stomach contents is common especially in infants and young children. Meanwhile, the concentrations of pepsin and pepsinogen in the CI group were much lower than those in the OME group, further substantiating the hypothesis that pepsinogen/pepsin could not cause OME if they are not accumulated beyond a certain threshold concentration. However, the specific value of this threshold of pepsinogen/pepsin from LPR is still unclear and needs further evaluation. In fact, the role of LPR in the presence and detectability of pepsinogen in the middle ear cavity and surface of adenoids samples is controversial. The reverse hypothesis considered that pepsinogen spreads from serum, becomes concentrated in the middle ear cavity, and flows toward adenoid tissues through the nasopharynx and eustachian tube. 25 However, this presumption is difficult to address using empirical methods. As mentioned before, the concentration of pepsin in MEE in the OME group was significantly higher than that in the corresponding serum, which indicates that the spread of Fig. 2. Pepsin and pepsinogen concentrations in the middle ear fluid. (A) Distribution of pepsin concentration in middle ear fluid. (B) Distribution of pepsinogen concentrations in middle ear fluid. (C) Correlation analysis of pepsinogen concentration in middle ear effusions (MEEs) and expression of pepsinogen in adenoid specimens. The concentration of pepsinogen in MEE is positively correlated with the expression intensity of pepsinogen protein in the cytoplasm of epithelial cells, rather than clearance under epithelial cells of adenoid specimens (r , P <.05) in the OME group. CI 5 cochlear implantation; OME 5 otitis media with effusion. [Color figure can be viewed in the online issue, which is available at E298

6 TABLE IV. Comparison of Pepsin and Pepsinogen Concentrations Among Dry, Serous, and Mucoid Ear Subgroups Within the Otitis Media With Effusion Group. Group Dry [12 Ears] Serous [38 Ears] Mucoid [19 Ears] F P Pepsin-positive [36 ears] 6 (16.67%) 19 (52.78%) 11 (30.55%) Pepsin, ng/ml * Pepsinogen-positive [48 ears] 6 (12.5%) 27 (56.25%) 15 (31.25%) Pepsinogen, ng/ml k Data shown are mean 6 standard deviation. *Group compared with the group, P <.01; *group compared with group, P <.01; group compared with group, P >.10; group and k group compared with the group, P >.10. pepsinogen from a low concentration in serum to a higher concentration in the middle ear cavity is highly unlikely. We also considered whether pepsin/pepsinogen was endogenously produced by middle ear mucosa, or the pepsinogen isozyme was secreted by adenoid samples or nasopharyngeal tissues. Lieu et al. found that pepsinogen I/A genes were encoded in gastric mucosa, but not in the mastoid mucous of middle ear cavity, using RT- PCR methods. 26 Furthermore, although it is known that the pepsinogen isozyme exists in other organs outside the stomach, it is limited only to the intestines, lung, pancreas, and malignant tumor In this study, we found that pepsinogen mrna was not detected in adenoid specimens, indicating that adenoid tissues or nasopharyngeal tissues do not produce the pepsinogen isozyme. In addition, we found that the concentration of pepsinogen in MEE is positively correlated with expression of pepsinogen protein in adenoid specimens in the OME group, showing that the pepsinogen in the middle ear cavity and surface of adenoid samples was exogenously produced. These data further demonstrate that pepsinogen, as an indicator of LPR, is involved in the pathogenesis of OME from the nasopharynx and eustachian tube to the middle ear. The severity of the lesion in the middle ear cavity varies, as the consistency of MEE changes in patients with OME. 31 In this study, we found that the concentration of pepsin gradually increased when the viscosity of the middle ear cavity fluid increased. These data suggest that pepsin aggravates OME, and that MEE appearing in the middle cavity is likely caused by pepsin, because the pepsin level in the dry ear subgroup was significantly lower than that in the serous and ear mucoid subgroups. We also found that there were no differences in the pepsinogen concentrations in the different OME subgroups, further indicating that pepsin, as an active form, rather than pepsinogen, is involved in the pathogenesis of OME. Interestingly, there were no differences in the pepsin concentrations between the serous and mucoid subgroups. We surmise that the small sample size or the dilution of pepsin during the collection process used in our study may account for this lack of difference. CONCLUSION In this study we found that pepsin and pepsinogen that appeared in the MEE of children with OME may be caused by LPR, by comprehensively ruling out other possibilities. In addition, our data showed that pepsin and pepsinogen could lead to aggravation of OME, especially when their concentrations are elevated. However, we note that further studies in a larger cohort and in animal models of OME need to be carried out to address more thoroughly the cause effect relationship between pepsin/pepsinogen in the middle ear and OME. Pepsin and pepsinogen in MEE are probably caused by LPR and may be involved in the pathogenesis of OME. BIBLIOGRAPHY 1. Zhao HX, Wang JB, Kong WJ. Practical Otolaryngology Head and Neck Surgery. 2nd ed. Beijing: People s Medical Publishing House; Browning GG, Rovers MM, Williamson I, et al. Grommets (ventilation tubes) for hearing loss associated with otitis media with effusion in children. Cochrane Database Syst Rev 2010;(10):CD Hall-Stoodley L, Stoodley P. Evolving concepts in biofilm infections. Cell Microbiol 2009,11: Saafan ME, Ibrahim WS, Tomoum MO. Role of adenoid biofilm in chronic otitis media with effusion in children. Eur Arch Otorhinolaryngol 2013; 270: Erdivanli OC, Coskun ZO, Kazikdas KC, et al. Prevalence of otitis media with effusion among primary school children in Eastern Black Sea, in Turkey and the effect of smoking in the development of otitis media with effusion. Indian J Otolaryngol Head Neck Surg 2012;64: Czinn SJ, Blanchard S. Gastroesophageal reflux disease in neonates and infants: when and how to treat. Paediatr Drugs 2013;15: Koufman JA. The otolaryngologic manifestations of GERD: laryngopharyngeal reflux is different from classic gastroesophageal reflux disease. Ear Nose Throat J 2002;81: Postma GN, Belafsky PC, Aviv JE, et al. Laryngopharyngeal reflux testing. Ear Nose Throat J 2002;81: Little JP, Matthews BL, Glock MS. Extraesophageal pediatric reflux: 24- hour double-probe ph monitoring of 222 children. Ann Otol Rhinol Laryngol Suppl 1997;169: Bortoli N, Nacci A, Savarino E, et al. How many cases of laryngopharyngeal reflux suspected by laryngoscopy are gastroesophageal reflux disease-related? World J Gastroenterol 2012;18: Wang L, Liu X, Liu YL, et al. Correlation of pepsin-measured laryngopharyngeal reflux disease with symptoms and signs. Otolaryngol Head Neck Surg 2010;143: Randhawa PS, Mansuri S, Rubin JS. Is dysphonia due to allergic laryngitis being misdiagnosed as laryngopharyngeal reflux? Logoped Phoniatr Vocol 2010;35: Yufera M, Moyano FJ, Astola A, et al. Acidic digestion in a teleost: postprandial and circadian pattern of gastric ph, pepsin activity,and pepsinogen and proton pump mrnas expression. PLoS One 2012;7:e Wezyk MT, Makowski A. PH of fluid collected from middle ear in the course of otitis media in children. Otolaryngol Polska 2000;54: Magliulo G, Plateroti R, Plateroti AM, et al. Gastroesophageal reflux disease and the presence of pepsin in the tears. Med Hypotheses 2013;80: Around A, Dettmar PW, Panetti M, et al. The reflux of gastric juice and glue ear in children. Lancet 2002;359: El Fattah, Maksoud G, Ramadan A, et al. Pepsin assay: a marker for reflux in pediatric glue ear. J Otolaryngol Head Neck Surg 2007;136: Aydin E, Tastan E, Aydogan F, et al. Role of nasopharyngeal reflux in the etiology of otitis media with effusion. J Otolaryngol Head Neck Surg 2011;40: Al-Saab F, Manoukian JJ, Al-Sabah B, et al. Linking laryngopharyngeal reflux to otitis media with effusion: pepsinogen study of adenoid tissue and middle ear fluid. J Otolaryngol Head Neck Surg 2008;37: E299

7 20. He Z, O Reilly RC, Bolling L, et al. Detection of gastric pepsin in middle ear fluid of children with otitis media. Otolaryngol Head Neck Surg 2007;137: O Reilly RC, He Z, Bloedon E, et al. The role of extraesophageal reflux in otitis media in infants and children. Laryngoscope 2008;118: He Z, O Reilly RC, Mehta D. Gastric pepsin in middle ear fluid of children with otitis media: clinical implications. Curr Allergy Asthma Rep 2008;8: Arinton IG. Serum gastrin level and pepsinogen I/II ratio as biomarker of Helicobacter pylori chronic gastritis. Acta Med Indones 2010;42: Martinho O, Pinto F, Granja S. RKIP inhibition in cervical cancer is associated with higher tumor aggressive behavior and resistance to cisplatin therapy. PLoS One 2013;8:e Heo KW, Hur DY, Park SK, et al. Expression of chitinases in hypertrophied adenoids of children. Otolaryngol Head Neck Surg 2011;145: Lieu JE, Muthappan PG, Uppaluri R. Association of reflux with otitis media in children. Otolaryngol Head Neck Surg 2005;133: Wells M, Brown B, Hall J. Pepsinogen C expression in intestinal IEC-6 cells. Cell Physiol Biochem 2003;13: Elabiad MT, Zhang J. Detection of pepsinogen in the neonatal lung and stomach by immunohistochemistry. J Pediatr Gastroenterol Nutr 2011; 53: Galaviz MA, Garcia-Ortega A, Gisbert E, et al. Expression and activity of trypsin and pepsin during larval development of the spotted rose snapper Lutjanus guttatus. Comp Biochem Physiol B Biochem Mol Biol 2012;161: Antunes AA, Leite KR, Sousa-Canavez JM, et al. The role of prostate specific membrane antigen and pepsinogen C tissue expression as an adjunctive method to prostate cancer diagnosis. J Urol 2009;181: Seth R, Discolo CM, Palczewska GM, et al. Ultrasound characterization of middle ear effusion. Am J Otolaryngol 2013;34: E300