INTRODUCTION

Gastroesophageal reflux disease (GERD) is caused by the combination between contact of irritant refluxed factors (especially chloride acid and pepsin) with the mucosa and the deficiency of mechanisms to protect this mucosa. It has been known for over 50 years in its digestive classic form, being considered one of the most common gastrointestinal diseases currently diagnosed 11. The first report of acid reflux reaching the larynx was made by CHERRY & MARGUILLES, in 1968 2; however, only in the past two decades, especially after the dissemination of the clinical use of optical fibers to examine the larynx and the pharynx, that GERD started to be considered as responsible for a wide range of benign and malignant diseases 4, 6, 14. This form of GERD was named laryngopharyngeal reflux (LPR) in 1991 by KOUFMAN 14.

Despite the advances in understanding GERD, there are many questions still to be answered, especially concerning LPR. It is not known whether patients with the same level of proximal acid reflux (larynx and pharynx) present manifestations and clinical findings of distinct manifestations. One of the explanations would be absence or deficiency of some local protective factor of the larynx and pharynx, making some people more sensitive to reflux.

We know that the esophagus has mechanical and chemical factors to protect it, but we know little about the mechanisms of protection of the laryngeal mucosa. The larynx is deprived from most of the defense mechanisms that the esophagus has. In recent years, many studies about the contribution of saliva in the protection of the digestive tube has been developed, especially about the biologically active proteins produced by salivary glands, which are capable of accelerating healing by stimulating replication of DNA 21, 24, 25. These factors of laryngeal organic protection have not been comprehensively studied so far. The existing studies relate neoplasm diseases; there are very few descriptions of these types of polypeptides in normal mucosa, especially because there are ethical issues concerning the conduction of biopsy in normal larynges.

In a careful review of the literature for the past 50 years, we did not find descriptions of the function or behavior of these organic elements in the larynx of patients with LPR. There are, however, many studies describing the reduction of the organic elements such as Epidermal Growth Factor (EGF), Prostaglandins (PG), Tumor growth Factor (especially its sub-fraction TGF2 alpha), among others, in patients with reflux esophagitis and dyspeptic disease 7, 12, 15, 21, 22, 23. Other studies showed early recovery of the gastric and esophageal mucosa after intravenous infusion of EGF, suggesting an important activity of the polypeptide in healing and epithelial regeneration 9, 18.

The Epidermal Growth Factor (EGF) is a single chain peptide, comprising 53 amino acids, which was discovered accidentally by COHEN in 1961 3. This factor has a wide biological action, being implied in the induction of epithelial growth, angiogenesis, inhibition of gastric acid saliva and acceleration of healing. In the digestive mucosa and in the saliva, it is known that EGF is reduced in cases of dyspeptic disease and esophagitis, suggesting a possible protective action of the protein. In the larynx, there are not sufficient studies about the expression of these protective factors of the epithelium in non-neoplastic diseases, and there are no reports about the behavior of gastroesophageal reflux disease with laryngeal or pharyngeal repercussions (LPR).

The purpose of the present study was to assess the concentration of the Epidermal Growth Factor (EGF) in the saliva of subjects with chronic laryngitis caused by LPR comparing it to normal subjects' saliva.

MATERIAL AND METHOD

We analyzed the saliva of subjects with LPR diagnosed with suggestive clinical history and compatible videolaryngoscopy and compared it to the saliva of normal controls (without LPR). Based on the hypothesis that subjects with reflux laryngitis associated with reflux esophagitis would have more GERD than subjects only with LPR, the subjects in the group were divided into two subgroups, one of patients only with LPR and the other with patients with LPR associated with reflux esophagitis, with macroscopic diagnosis made by upper digestive endoscopy, but still considering them as one single group for comparison purposes with the control group (without disease).

Between January 2001 and June 2002, we studied 59 subjects, being 39 adults with LPR and 20 normal controls matched by age and gender. After saliva collection, 39 subjects were submitted to upper digestive endoscopy to better characterize the form of clinical presentation and intensity of GERD, creating three groups:

 Group A - Eighteen patients (6 men and 12 women) with videolaryngoscopic symptoms and signs of laryngopharyngeal reflux (LPR) with endoscopic macroscopic esophagitis.
 Group B - Twenty-one patients (4 men and 17 women) with videolaryngoscopic signs and symptoms of laryngopharyngeal reflux (LPR) with no endoscopic macroscopic esophagitis.
 Group C - Twenty adults (6 men and 14 women) volunteers and controls without videolaryngoscopic signs and symptoms of laryngopharyngeal reflux (LPR).

The mean age in Group A was 50.056 years (minimum age of 36 and maximum of 62), Group B was 51.238 years (minimum of 38 and maximum of 72), and Group C was 50.150 years (minimum of 26 and maximum of 85). There were no statistically significant differences between the studied groups and the control group concerning gender and age (p=0.696 and p=0.675, respectively).

The exclusion criteria included smoking, alcohol abuse, exposure to inhalation abrasive chemicals, and treatment in the past 7 days with gastric secretion blockers, pro-kinetic, antihistaminic and beta-blockers. We also excluded patients with white lesions, papillomas or larynx and pharynx neoplasms (present or previously treated). They were examined following the protocol of the study, with rigid laryngoscope and/or flexible nasofibrolaryngoscope. We included in the study only the patients that responded to the questionnaire and presented laryngeal exam compatible with LPR, after having signed the informed consent term including objectives, procedures used and risks involved. The healthy volunteers were submitted to the process of previous informed consent, answering the questionnaire and also undergoing diagnostic videolaryngoscopy. The project was approved by the Research Ethics Committee of the institution.

All participants in the study responded to an extensive questionnaire about the supraesophageal and gastroesophageal symptoms related to GERD, as well as about other diseases capable of causing posterior laryngitis (allergic rhinitis and chronic sinusitis).

As to the main laryngopharyngeal complaints, the most frequent ones were throat clearing, globus pharyngeus and dysphonia. Throat clearing and globus pharyngeus were related to all patients in group B (100%) and in 89% and 78% of the patients in group A, respectively. Dysphonia was present in 12 of the 18 patients (66%) in group A and in 16 of 21 patients in group B (76%). Only two patients (10%) in group C presented dysphonia, being that one of them had it since childhood and it was associated with vocal fold sulcus; in the other one, dysphonia was intermittent and associated with vocal abuse (Figure 1).

As to gastric complaints, the most common ones were pyrosis, epigastralgia, poor digestion and eructation. All patients in Group A had gastric complaints. Nineteen of the 21 patients (90%) in Group B also had some gastric complaints, being that the two patients without complaints of gastric symptoms had had the laryngopharyngeal complaints for less than 6 months (Figure 2).

ENT examination
The subjects were submitted to general ENT examination and we did not detect any anatomical or pathological abnormality that could interfere in the study.
The videolaryngoscopic exam (rigid and flexible) was conducted after the general ENT examination. We considered signs of laryngopharyngitis by gastroesophageal reflux based on ECKLEY & COSTA (2002):

 edema, hyperemia or increase of lymphoid tissue on the posterior pharyngeal wall, conveying a granulous aspect, called posterior pharyngitis (FP) (Figure 3). These signs were scored as mild, moderate and severe.
 Larynx edema and hyperemia, diffusely and specifically, in the posterior third, in which we observed the presence and intensity of hyperemia and edema of the arytenoids, retrocricoid and interarytenoid areas, being that the latter was named pachyderma (PQ) (Figure 3). These signs were also scored as mild, moderate and severe.
 Salivary stasis in the retrocricoidarytenoid area and pyriform sinuses (Figure 4).

The videonasofibrolaryngoscopic examination showed many abnormalities of the laryngopharyngeal mucosa, which ranged in intensity and location in groups A and B (Figure 5). The patients in the control group (C) presented few findings in the nasofibrolaryngoscopic exam.

A very frequent finding in the affected groups was increase in nausea reflex. It was present in 9 patients in Group A (50%), 9 patients in group B (43%) and 2 patients in group C (10%), with statistically significant difference between the studied groups and the control group (p = 0.068).

Complementary tests
Patients with PLR were submitted to upper digestive endoscopy, 24-hour ph-meter with two channels and esophageal manometry.

Upper digestive endoscopic exams were conducted preferably on the same day of salivary collection and only after it, so that there were no interferences of medication or mechanical stimulation to the esophagus in salivary production. In group A, all patients presented esophagitis (mandatory condition to be included in the group), being that 13 were mild (72%) and in 5 (28%) it was moderate. In this group, esophagitis was associated with gastritis in 10 patients, all of them of mild level. Eight of the 18 patients (58%) presented hiatus hernia, being small in all of them (below 3cm). H. pylori investigation was positive in only 4 cases (24%), being that 3 of them had mild esophagitis and one had moderate esophagitis, but all four had mild associated gastritis. In Group B, only 4 patients (19%) presented normal examination. All the 17 patients with endoscopic abnormalities presented gastritis, being mild in 14 patients (93%) and moderate in 3 patients (7%). Three patients in group B presented duodenitis associated with gastritis, being mild in 2 patients and moderate in one (the patient presented mild gastritis). H. pylori was present in 7 patients (33%) in group B, all of them with mild gastritis.

We found abnormalities in esophageal motility, characterized by hypocontractility of the esophagus in only 3 of the 18 patients (22%) in Group A and in 6 of the 21 patients (28%) in group B. We noticed hypotonia of lower esophageal sphincter in 13 of the 18 patients (72%) in Group A and in 6 of 21 (28%) patients in group B. There were no changes in tone of the upper esophageal sphincter in any of the 39 studied subjects.

The exam of 24-hour ph-meter was also performed in all patients with LPR, using two channels of antimony in a single catheter: one situated 5 cm above the lower sphincter and the other at the level of the upper sphincter, or right above it. The position of the reading channels of pH was determined previously by esophageal manometry. Considering the pH meter exams with pathological reflux, both distal and proximal, as well as the pathological reflux associated with both channels, we had 15 patients in group A (83%) and 11 in group B (52%) with positive exams.

Collection and processing of saliva for the study of EGF
Both the subjects with disease and the normal controls had saliva collected in the morning after a period of 8-hour fast, being instructed not to use tooth paste in the morning of the collection. Two distinct salivary samples were collected on the same day:

 Total spontaneous saliva (SET): after washing the mouth with running water to eliminate epithelial desquamation and bacterial debris, the subjects were instructed to remain seated and not to swallow the saliva produced spontaneously during 15 minutes, which should overflow from the mouth (avoiding sputum not to add material from the rhinopharynx) to a glass funnel, coupled to a test tube in a foam box with ice.
 Retrolaryngeal region saliva (SRL): During the examination with the nasofibrolaryngoscope we collected the secretion of the retrocricoarytenoid region, between both arytenoids, through endoscope scrub using a recovered cytology brush (Olympus Optical Co. Ltd., Japan) of 0.3 X 0.1 cm introduced in a polyethylene catheter to prevent contamination during the introduction and extrusion of the device.

Processing of the saliva
SET was centrifuged for 10 minutes at a speed of 5000-8000 rpm, for sedimentation of bacteria, epithelial cells, nuclear debris and other debris. The supernatant was collected by a pipette, and transferred to another test tube with a threaded tap and stored at minus 70 degrees Celsius up to analysis 86.

Each sample was identified by an alphanumeric code.

Analysis of saliva
Determination of the salivary concentration of EGF
Salivary concentration of EGF was determined by a commercial kit of ELISA (Quantikine, R & D Systems, Minneapolis, MN, USA), in which a buffered protein base was used as diluting agent. EGF concentration was expressed by the reading of the optical density of the sample compared to the standard curve, as well as the function of total saliva protein concentration. Thus, the changes in concentration of salivary EGF were related to total concentration of saliva proteins, regardless of total salivary volume.

Restratification of the sample for the effect of the analysis of salivary results
Since the subjects with LPR included in group A and B presented clinical and laryngoscopic characteristics that were similar as a whole, we tried to group them again according to intensity of laryngeal findings, to study the possible salivary differences. We, therefore, created three groups:
 Group with mild laryngitis: comprising patients with laryngeal signs of at least 3 analyzed parameters (retrocricoid edema, arytenoid hyperemia, interarytenoid edema and pachyderma), classified as mild;
 Group with moderate laryngitis: comprising patients with at least 3 laryngeal parameters classified as moderate;
 Group with severe laryngitis: comprising patients with at least two of the laryngeal parameters classified as severe and the others with moderate level.
Statistical analysis

The data were statistically analyzed by non-parametric and parametric tests, such as chi-square. ANOVA and Kruskal-Wallis tests, if applicable.






RESULTS

The collection of saliva of the 59 participants in the study was made according to the methodology described above; there were no unexpected events. There were no reports of discomfort or other significant symptoms after salivary collection. The minimum fast time was 9 hours and the maximum was 12 hours.

The mean volume of total saliva was 15 minutes for 4.7ml in Group A, 4.2 ml in group B and 3.5ml in group C. It was not possible to define a correlation between age and salivary volume owing to the great variation in volume in the three groups.

The dosage of concentration of EGF in total spontaneous saliva and retrocricolaryngeal saliva proved to be significantly greater in Group C when compared to groups A and B together (p = 0.85 and p = 0.002, respectively) and there were no statistically significant differences between groups A and B (Table 1).

Endoscopic findings such esophagitis, gastritis and hiatus hernia and infection with H. pylori were not correlated with salivary concentration of EGF. There was no statistically significant correlation between findings of ph meter (number of episodes of distal and proximal canal, duration of episodes, associated symptoms) and salivary concentration of EGF, both in total spontaneous saliva and retrocricolaryngeal saliva. The presence of esophageal dismotility did not show correlation with concentration of EGF, both for total and retrocricolaryngeal saliva.

The analysis of salivary concentration in EGF after regrouping of the subjects with LPR (n = 39) according to severity of chronic laryngitis (mild, moderate and serve) did not result in statistically significant differences between the groups, both for total spontaneous saliva (p = 0.544) and retrocricolaryngeal saliva (p = 0.544), even though mean levels of EGF (total and retrocricolaryngeal) showed a tendency to reduction as laryngitis worsened.

In our study we did not include smokers. However, six of the patients in group A , two in group B and two in group C had been smokers, and all of the had quit for over 12 months. The concentration of EGF in total spontaneous and retrocricolaryngeal saliva of ex-smokers was statistically greater than in those that had never smoked (p = 0.043 and p=0.057, respectively). There was no statistically significant difference between former smokers in groups A and B, both in SET (p=0.689) and in SRL (p=0.821). When the smokers were excluded from the study and the control groups they maintained the salivary concentration of EGF seen in the total groups, both for total spontaneous and retrocricolaryngeal saliva, that is, subjects with the disease presented significantly less EGF than normal subjects (p=0.004 and p=0.009, respectively).






DISCUSSION AND CONCLUSIONS

Gastroesophageal reflux disease is considered the most prevalent gastrointestinal disease of modern times. Symptoms associated with reflux are weekly reported in 3 to 6% of the subjects in the general population 1, 17, 20. Multicenter studies with large populations have shown how this disease can manifest in a silent form from the digestive perspective 20. However, its symptomatology causes significant comprise of quality of life. It is believed that this great prevalence of the disease is owed to a confluence of factors, from poor eating habits to obesity and genetic factors 20.
In the past 12 years, many researchers and clinicians have proposed an association between gastroesophageal reflux disease (GERD) and chronic laryngitis 2, 4, 6, 11, 14, 20.

It is still not known how many episodes of reflux are necessary to produce inflammatory abnormalities and damage to the larynx. However, studies such as the ones by DELAHUNTY & CHERRY in 1968 and KOUFMAN in 1991 demonstrated that the application of chloride acid and pepsin to laryngeal mucosa of dogs cause contact granulomas in the first study, and laryngeal stenosis, in the second study, after some weeks 14. These and other studies indicate that the laryngeal mucosa is very sensitive to chemical substances and suggest that LPR, even if intermittent, can cause severe inflammatory lesions in the larynx 14. The role of non-acid substances such as bile and pancreatic juice, in the formation of laryngeal lesions, is still very controversial since there are no available techniques to measure them in clinical practice 2.

It is interesting to note that a large number of patients with laryngopharyngeal reflux do not have esophagitis or other signs of GERD in the digestive tube 6. Certainly, the protection mechanisms of gastric and esophageal mucosa play a decisive role in the ability these organs have to support mechanical and chemical aggressions to which they are submitted daily, being that various of these mechanisms can be measured by saliva 8, 15, 22. Saliva contains many organic and inorganic substances that contribute to protecting against physical and chemical assaults and to maintain the integrity of the mucosa, not only the oral cavity, but also the digestive tube 8, 15, 22, 24. Mechanical and chemical stimuli to the esophagus are capable of changing the volume, viscosity and pH of the saliva. This is the so-called salivary esophagus reflex, very important for the maintenance of the integrity of esophageal and gastric mucosa in patients with GERD.

Among the biologically active organic factors produced by salivary glands, the epidermal growth factor (EGF) owed to the power of stimulating the DNA cell synthesis, showed the one that is more directly associated with quick regeneration of oral and digestive epithelia, after aggressions 12, 15, 18, 19, 22. Quantitative and qualitative studies of the biological action of salivary polypeptides such as prostaglandins (PGE), tumor growth transformation factor (TGF), alpha and beta sub-fractions, and epidermal growth factor (EGF) showed that despite the fact that they all act directly or indirectly to protect the digestive epithelium, the most active of the polypeptide, both in salivary concentration and biological action, was EGF 15.

To conduct the present study we were based on the following premises:

 Considering that the digestive tube has apparent deficiency of EGF, both salivary and esophageal, in dyspeptic disease and reflux esophagitis 7, 9, 12, 15, 18, 19, 22, 24, and other alterations of the oropharynx and larynx mucosa 5, 13, 16 also have deficiency in salivary concentration of this polypeptide, it is assumed that there is a similar lack in the larynx.

 There are evidences of the action of EGF in the integrity of the mucosa of the digestive tract in studies that administered EGF, be them topical 5, 7 or systemic 9, 18, 19, for the treatment of gastric and duodenal ulcers and in experimental animal and human models, confirming the acceleration in healing of digestive tract lesions. In the study by OKITA et al., in 1991 18, it was also confirmed that pre-treatment of animals with intravenous EGF prevented the onset of gastric ulcers after the continuous irrigation of chloride acid.

 Recent population studies showed that over 30% of the general population presented occasional dyspeptic and laryngopharyngeal symptoms associated with gastroesophageal reflux and that in up to 40% of these subjects the endoscopic exam of the digestive mucosa was normal 20. Such findings attract attention to the fact that there is a silent behavior in GERD, but also it makes us infer that there are possibly more factors involved in the pathophysiology of GERD than simply the reflux of gastroduodenal content, be it acid or alkali.

· The focus of current medicine practice is on the operation of the body at the biomolecular level. We know that in addition to aggressors and pathogens, the organism is defended at the cell and molecular levels through biologically active substances; health and disease will depend on the dynamic balance between the aggression and the defense mechanisms.

 The clinical behavior of GERD and specifically of LPR presents different grades of laryngotracheitis, with similar amounts of reflux, which makes us wonder whether it would not be only the chemical reflux to cause the mucosa affections found, but there would rather be a deficiency of some factors of local protection that would act as co-factor of the disease.

We designed this study in order to assess whether there were differences in salivary concentration of EGF between subjects with LPR and normal controls. Since the literature emphasizes the existence of two clinical manifestations of GERD, one with laryngitis and/or other ENT symptoms (LPR) and other predominantly digestive (GER), we decided to stratify the patients with LPR in two subgroups (with and without esophagitis) to study whether there were clinical differences and differences of salivary concentration of EGF between these subjects. However, throughout the study, we did not find significant differences between these two groups, concerning the disease, symptoms and laryngeal signs, and concentration of EGF in the saliva. Probably it was owed to the fact that these patients had been screened from ENT units, being that the complaints that motivated the medical search were located in the head and neck. It is important to emphasize that despite the fact that the participants of the groups reported gastric and dyspeptic complaints, they were not provided upon induction, but most of them were actually mild or moderate and occurred only occasionally.

It was evident that the patients with LPR presented, as a whole, less EGF in the saliva than the normal controls. This finding suggests that the smaller concentration of EGF in the saliva of subjects with LPR can take to larger mucosa affections, since the control group did not present laryngitis or other significant abnormalities of the pharyngolaryngeal epithelium.

Once defined the difference in salivary concentration of EGF between sick and normal patients, we decided to assess whether the intensity of laryngitis, that is, of the laryngeal mucosa, would be influenced by the salivary concentration of EGF. With that objective in mind, we divided the patients with LPR in three new groups, according to the severity of laryngitis: mild, moderate and severe. There was no statistically significant difference in salivary concentration (both total spontaneous saliva and retrolaryngeal saliva) of EGF between these three groups.

This study suggests that reduced salivary concentration of EGF can act as a co-factor in the genesis of LPR. If future studies are able to associate the salivary concentration of EGF with intensity of LPR, we will be faced by a test that has diagnostic and prognosis power, possibly sensitive and cheap, in addition to being less invasive than the currently available ones.

REFERENCES

1. Johnason JF. Epidemiology of esophageal and supraesophageal injuries. Am J Med 2000; 108(4A):99S-103S.
2. Cherry J, Margulies S.I. Contact ulcer of the larynx. Laryngoscope 1968; 78:1937-40.
3. Costa HO, Eckley CA, Fernandes AMF, Destailleur D, Villela PH. Refluxo gastroesofágico: comparação entre achados laríngeos e digestivos. Rev Port ORL 1997; 35(1):21-6.
4. Eckley CA & Costa HO. Manifestações Otorrinolaringológicas da Doença do Refluxo Gastroesofágico. In: Martinez JC. Affecções Cirúrgicas do Estômago e Intestino Delgado. São Paulo: Atheneu; 2002. No prelo.
5. Koufman JA. The otolaryngologic manifestations of gastroesophageal reflux disease (GERD): a clinical investigation of 225 patients using ambulatory 24-hour pH monitoring and an experimental investigation of the role of acid and pepsin in the development of laryngeal injury. Laryngoscope 1991; 101(Suppl):1-78.
6. Rourk RM, Namiot Z, Saroziek J, Yu Z, Mccallum RW. Impairment of Salivary Epidermal Growth Factor Secretory Response to Esophageal Mechanical and Chemical Stimulation in Patients with Reflux Esophagitis. Am J Gastroenterol 1994; 89(2):237-44.
7. Sonnenberg A, Steinkamp U, Weise A, Berges W, Weinbeck M, Rohner HG, Peter P. Salivary Secretion in Reflux Esophagitis. Gastroenterol 1982; 83:889-95.
8. Starkey RH. & Orth DN. Radioimmunoassay of human epidermal growth factor (urogastrone). J Clin Endocrinol Metab 1977; 45(6):1144-53.
9. Gray MR, Donneley RJ, Kingsnorth AN. Role of salivary epidermal growth factor in the pathogenisis of Barrett's columnar lined oesophagus. Br J Surg 1991; 78:1461-6.
10. Konturek JW, Bielanski W, Konturek SJ, Konturek JW, Oleksy J, Yamazaki J. Distribution and release of epidermal growth factor in man. Gut 1989; 30:1194-200.
11. Marcinkiewicz M, Han K, Zbroach T, Poplawski C, Gramley W, Goldin G, Sarosiek J. The Potential Role of the Esophageal Pre-Epithelial Barrier Components in the Maintenance of Integrity of the Esophageal Mucosa in Patients with Endoscopically Negative Gastroesophageal Reflux Disease. Am J Gastroenterol 2000; 95(7):1652-60.
12. Sarosiek J & Mccallum RW. Do salivary Organic Components Play a Protective Role in Health and Disease of the Esophageal Mucosa? Digestion 1995; 56 (Suppl. 1):32-7.
13. Sarosiek J, Scheurich CJ, Marcinkiewicz M, Mccallum R.W. Enhancement of Salivary Esophagoprotection: Rationale for a Physiological Approach to Gastroesophageal Reflux Disease. Am J Gastroenterol 1996; 110:675-81.
14. Itoh M, Joh T, Imai S, Miyamoto T, Matsusako K, Iwai A, Katsumi K, Endo K, Goto K, Takeuchi T. Experimental and clinical studies on epidermal growth factor for gastric mucosal protection and healing of gastric ulcers. J Clin Gastroenterol 1988; 10 (Suppl.1):S7-12.
15. Okita K, Karita M, Nakanishi N. Role of Epidermal Growth Factor in Protection and Repair of Gastric Mucosal Injury. J Clin Gastroenterol 1991; 13(Suppl.1):S103-8.
16. Cohen S. Isolation of a mouse submaxillary gland protein accelerating incisor eruption and eyelid opening in the new-born animal. J Biol Chem 1962; 237(5):1555-62.
17. Anelli W. In: Aspectos perceptivos-auditivos e acústicos da Doença do Refluxo Gastroesofágico. São Paulo; 2002. p. 97. (Tese Mestrado - UNIFESP).
18. Moraes-Filho JPP, Cecconello I, Gama-Rodrigues J, Castro LP. Brazilian consensus on gastroesophageal reflux disease: proposals for assessment classification and management. Am J Gastroenterol 2002; 97(2):241-8.
19. Ronkainen JA, Aro P, Storskrubb T. Prevalence of esophagitis and endoscopy-negative reflux disease in a population. A report from the Kalixandra Study. Abstracts of the Digestive Disease Week 2002; S1357:A-269.
20. Helm JF, Dodds WJ, Hogan WJ. Salivary response to esophageal acid in normal subjects and patients with reflux esophagitis. Gastroenterol 1987; 93:1393-7.
21. Olsen PS, Poulsen SS, Kirkegaard P, Nexo E. Role of Submandibular Saliva and Epidermal Growth Factor in Gastric Cytoprotection. Gastroenterol 1984; 87:103-8.
22. Dumbrigue HB, Sandow PL, Nguyen KT, et al. Salivary epidermal growth factor levels decrease in patients receiving radiation therapy to the head and neck. Oral Surg Oral Med Oral Pathol 2000; 89(6):710-6.
23. Korsten MA, Rosman AS, Fishbein S, Shlein RD, Goldberg HE, Biener A. Chronic xerostomia increases esophageal acid exposure and is associated with esophageal injury. Am J Med 1991; 90:701-6.
24. Moazzez R, Anggiansah A, Bartlett D, Owen W. Tooth wear saliva and symptoms of GERD: is there a relationship? Abstracts of the Digestive Disease Week 2002 W1164:A-816.
25. Joh T, Itoh M, Katsumi K, Yokoyama Y, Takeuchi T, Kato T, Wada Y, Tanaka R. Physiolocical concentration of human epidermal growth factor in biological fluids: use of a sensitive enzyme immunoassay. Clin Chem Acta 1986; 158:81-90.

---

1Master in Medicine, Medical School, Santa Casa de São Paulo, Assistant Professor, Department of Otorhinolaryngology, FCMSCSP
2Ph.D. in Medicine, University of São Paulo, Joint Professor, Department of Otorhinolaryngology, FCMSCSP
Study conducted at Medical School, Santa Casa de São Paulo.
Address correspondence to: Dra. Claudia A. Eckley - Rua Sabará 566, cj. 23 - Higienópolis CEP: 01239-010 - Sao Paulo - SP
Tel/Fax: (55 11) 3257-2686 - e-mail: ceckley@unisys.com.br
Study classified among the 4 top finalists for Prêmio Byk- Excelência em Saúde Versão 2002.
Presented as Free Communication at the Annual Meeting of the American Academy of Otolaryngology, Head and Neck Surgery & OTO EXPO 2002 in San Diego, CA, USA.
This study was financed by Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP).
Article submitted on March 17, 2002. Article accepted on July 24, 2003.