SciELO - Scientific Electronic Library Online

vol.54 issue2Indicators of inflammation and cellular damage in chronic asymptomatic or oligosymptomatic alcoholics: correlation with alteration of bilirubin and hepatic and pancreatic enzymesClinical and radiological aspects in Melnick-Needles syndrome author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Revista do Hospital das Clínicas

On-line version ISSN 1678-9903

Rev. Hosp. Clin. vol.54 n.2 São Paulo Mar./Apr. 1999 




Rowilson Flora Filho and Bruno Zilberstein



FLORA FILHO, R. et al. - Reflux esophagitis and the gastroesophageal reflux disease: across-sectional study of gastroesophageal reflux disease patients by age group. Rev. Hosp. Clín. Fac. Med. S. Paulo 54 (2): 61 - 67, 1999.


SUMMARY: The purpose of this study was to explore the relationship between the intensity of acid reflux and severity of esophageal tissue damage in a cross-sectional study of patients with gastroesophageal reflux disease (GERD). Seventy-eight patients with were selected in accordance with the strict 24-hour ambulatory esophageal pHmetry (24h-pHM) criteria and distributed into three age groups: Group A: 14 - 24 years of age. Group B: 25 - 54; and Group C: 55 - 64. The 24h-pHM was carried out in accordance with DeMeester standardization, and the Savary-Miller classification for the diagnosis of reflux esophagitis was used. The groups were similar in 24h-pHM parameters (p > 0.05), having above normal values. For the study group as a whole, there was no correlation between age group and intensity of acid reflux, and there was no correlation between intensity of acid reflux and severity of esophageal tissue damage. However, when the same patients were sub-grouped in accordance with the depth of their epithelial injury and then distributed into age groups, there was a significant difference in esophagitis without epithelial discontinuity. Younger patients had less epithelial damage than older patients. Additionally, although there was a significant progression from the least severe to the moderate stages of epithelial damage among the age groups, there was no apparent difference among the age groups in the distribution between the moderate stages and most severe stages. The findings support the conclusion that the protective response of individuals to acid reflux varies widely. Continued aggression by acid reflux appears to lead to the exhaustion of individual mechanisms of epithelial protection in some patients, but not others, regardless of age or duration of the disease. Therefore, the diagnosis and follow-up of GERD should include both measurements of the quantity of refluxed acid and an assessment of the damage to the esophageal epithelium.


DESCRIPTORS: Esophagitis, Peptic. Esophagus. Gastroesophageal Reflux. Age. Monitoring. Diagnosis. Natural history. Assessments. Upper Digestive Endoscopy. Hydrogen-Ion Concentration. 24-pHmetry.



The criteria for diagnosis of gastroesophageal reflux disease (GERD) have changed over the years. Currently the gold standard10 for diagnosis involves the demonstration of presence or absence of macroscopic distal esophageal mucosa damage as determined through upper digestive endoscopy (UDE), with histological confirmation, and presence of changes in esophageal 24h-pHmetry (24h-pHM).

These gold standard criteria determine at esophageal epithelium level the result of the sum of two factors: the intensity of the tissue injury (individual response factor to aggression) and the characterization and quantification of the refluxed acid (aggressor factor).

Since GERD was first described in the literature by Chevalier Jackson (1929)16 and by Asher Winkelstein (1935)25 as peptic esophagitis, the acid secretion from the stomach was considered as a primary and disease perpetrating fact. Though the biliary reflux has more recently been implicated as the possible pathogen in GERD patients with Barrett's Esophagus, it is thought that the refluxed bile would lead to esophageal epithelium damage only if there is a synergetic interaction with the refluxed acid3,17.

Once it became possible to determine the aggressor factor's intensity and variation with a diagnostic method as highly sensitive and highly specific as the 24h-pHM2, other methods for this purpose have been rejected by clinicians7,10.

The UDE method has also greatly progressed. The use of high definition cameras coupled to devices with a comfortable size and good tolerance has resulted in recorded images that decrease the subjective difference in staging macroscopic injuries. With the use of chromoscopy, it is possible to document minimal esophageal mucosa damage11.

Though the association of more severe esophagitis with gastroesophageal reflux patterns (combined < supine > orthostatic)12 is well defined, there is no published correlation between the refluxed acid amount (measured by the varying 24h-pHM parameters) and the tissue injury intensity (esophagitis severity).

In this cross-sectional study we describe both the aggressor factor in GERD in objectively measurable amounts and the respective tissue responses. Since GERD is a chronic disease with little possibility of spontaneous remission9,18, the evolution of the disease is of interest. In order to study the evolution of the disease, we grouped the patients in the study into age groups (young, middle aged, pre-senescent).



a) Patients

The patients were retrospectively selected between September 1996 and July 1998 and were initially grouped according to complaints compatible with GERD. All the patients underwent UDE followed by 24h-pHM. The criteria for inclusion in the study was based on 24h-pHM results: % reflux in orthostatic position > 8.2 % and/or reflux in supine position > 3.0 %. Seventy-eight patients fulfilled the inclusion criteria (NT = 78). The patients' minimum age was 14 and the maximum age was 64 (average = 40.6 ± 13.7). There were 53 male (68.0%) and 25 female (32.0%) participants. The patients were divided into three arbitrarily defined age groups:

Group A: 14 to 24 years (NA = 10)
Group B: 25 to 54 years (NB = 54)
Group C: 55 to 64 years (NC = 14)

b) Methods

(1) Upper Digestive Endoscopy

All patients underwent upper digestive video-endoscopy with a Video-Pentax® EGD-29901 device. Oropha-ryngeal topic anesthesia was given with 10% Xylocain spray followed by sedation through endovenous application of 10 mg Diazepam. Endoscopic esophagitis was classified from grade 0 thru 4, in accordance with the Savary-Miller classification22,23: Grade 0 - presence of changes in the distal third of the mucosa such as erythema, edema, friability or paleness, without mucosa erosion; grade 1 - flat erosions (either single or multiple); grade 2 - several flat non-confluent erosions, including those present in the medial esophageal third or confluent flat erosions in the distal third; grade 3 - confluent erosions encompassing all the distal esophagus circumference; grade 4 - presence of reflux esophagitis complications such as ulcerations, esophageal size cicatricial reductions and/or presence of Barrett's Esophagus.

Patients were grouped by severity of epithelial damage: grade 0 - esophagitis without epithelial discontinuity (REwD); grades 1, 2 and 3 - esophagitis with epithelial discontinuity only (RED); and grade 4 and patients with Barrett's Esophagus - esophagitis with extra-epithelial discontinuity (REeD).

(2) Esophageal 24h-pHmetry

A Digitrapper Mark III ® (Synetics Medical Inc) device was used for monitoring acid esophageal pH. The standardization method was the same as used by DeMeester et al (1976, 1980)7,8 and has already been presented in other publications12,13.

c) Statistical analyses

For determination of the significance of statistical differences of the DeMeester's Score values and the total time % between the groups distributed in accordance with esophagitis grades or age groups, the Kruskal-Wallis6 non-parametric test was used. Fisher1 exact test was used to determine the significance of statistical differences between the esophagitis grades in relation to age group. Values indicated in parentheses preceded by the "±" sign refer to the standard deviation. p was considered significant when found to be smaller than 0.05 (p > 0.05).



The gender distribution among the age groups was as follows: Group A: 7/10 (70.0%) male and 3/10 (30.0%) female. Group B: 39/54 (72.2%) male and 15/54 (27.8%) female, and Group C: 7/14 (50.0%) male and 7/14 (50.0%) female.

The group age average was: Group A = 19.9 years; Group B = 38.9 and Group C = 61.6. In Group A the mean and standard deviation of the total < 4.0 reflux episodes was 215.5 ± 122.4 episodes, while in Group B it was 202.2 ± 96.0 and in Group C 225.9 ± 73.6 episodes (NS p > 0.05) (Figure 1). DeMeester Score averaged 49.5 ± 24.0 in Group A, 40.6 ± 16.2 in Group B and 49.5 ± 24.0 in Group C (NS p > 0.05) (Figure 2). The time percentage in which pH remained below 4.0 in relation to the total examination time was 8.8 ± 4.4 in Group A, 9.9 ± 3.5 in Group B and 12.8 ± 8.2 in Group C (NS p > 0.05) (Figure 3) (Table 1).









When these 24h-pHM parameters are grouped according to the severity of the reflux esophagitis, they yield the results on table 2, and again no significant differences are noticed among the various esophagitis grades (NS p < 0.05) (Figures 4, 5 and 6).









When patients are grouped according to the endoscopic injury depth into the esophageal mucosa epithelial layers, the results are: REwD - 12/78 patients, RED - 50/78 patients; and REeD - 16/78 patients. When these patients are sub-grouped and divided into age groups, the following distribution results: Group A/REwD - 4/10 (40.00%), Group A/RED - 5/10 (50.0%); Group A/REeD - 1/10 (10.00%); Group B/REwD - 8/54 (14.81%); Group B/RED - 34/54 (62.96%); Group B/REeD - 12/54 (22.22%); Group C/REwD - 0/14 (0.00%): Group C/RED - 11/14 (78.57%); Group C/REeD -3/14 (21.43%), summarized in table 3.



The difference between the age group {14 to 24 years} and {25 to 54 years} as to REwD and RED grades was significant (p = 0.0005). The difference between groups {25 to 54 years} and {55 to 64 years} as to REwD and RED grades also was significant (p = 0.0001). The statistical differences between age groups and other esophagitis grades were not significant (p < 0.05) (Figure 7).




The limitations of clinical research regarding the possibility of controlling some or all of the variables are well known. When we deal with clinical observations, the indications run the risk of being falsified by the enormous quantity of incident variables. In inductive reasoning, the adaptation of measurable conditions to an experiment is acceptable and, as a consequence, they are also acceptable in a clinical observation21. Thus, the included patients were retrospectively and cross-sectionally selected from a similarity of an important variable to the disease under study. As the inclusion criteria were an unquestionable change in the 24h-pHM parameters, the group in this study had 24h-pHM quantitative standards that may be considered as having almost 100% specificity and sensitivity to GERD. In other words, all the included patients had acid reflux levels outside the normal range and were unlikely to be "false-positives".

The determination of the exact beginning of the GERD is practically impossible. The symptomatology criteria (either typical or atypical) are failure prone2,7,9. Roughly, it is considered that over 50.0% of patients with GERD are asymptomatic5,19, a percentage which increases depending on the study group24. GERD prevalence studies are estimative, and it is even considered that about 20.0% of the population may present GERD to a certain degree5,19.

There is a convergent thought that GERD is a chronic disease with little possibility for spontaneous remission4,7,13,24. Some authors that retrospectively studied a significant number of patients from sedimented databases have concluded that GERD hardly changes over the years9.

As hypotheses and implications related to GERD also suggest that the disease begins at an early age and continues over the years, it is valid to accept such hypothesis even as ad hoc15.

Therefore, the division of patients in age groups in this study tends to reflect GERD's temporal evolution. As demonstrated in previous studies13, this division follows a logical approach: 14 to 24 years - young; 25 to 54 years - middle aged; 55 to 64 years - pre-senescent.

These age groups showed no significant differences regarding important parameters of 24h-pHM reflecting the refluxed acid quantity (number of reflux episodes with pH < 4.0, DeMeester Score and % time with pH <4.0 in relation to total examination time). Therefore, for discussion purposes, it is considered that a variable taken as significant for GERD (quantity of refluxed acid = aggressor factor) was very similar for the groups studied.

When these same 24h-pHM parameters were applied to the esophagitis grades, it was again observed that the value differences were not significant, i.e., different reflux grades occurred with similar quantities of the aggressor factor. The supposition that there are other factors determining GERG’s evolution is hereby reinforced. And the corollary conclusion is: under gold standard criteria we can not dispense with the tissue injury for the clinical staging of GERD.

The sub-classification of esophagitis grades according to epithelial injury (REwD = Reflux esophagitis without discontinuity; RED = Reflux esophagitis with epithelial discontinuity only; REeD = Reflux esophagitis with extra-epithelial discontinuity) helps in the following inductions, mainly when confronted with the age groups as representative of GERD’s chronological evolution. Therefore, REwD is present in 40.00 % of patients in Group A, and in 14.81% in Group B patients, and is absent (0.00%) in Group C patients, these differences being significant. That is, with older age or the increase of the esophageal epithelium’s exposition time to the aggressor stimulus, in some patients a higher probability of exhaustion of individual epithelial protection mechanisms can be noted, given an aggressor factor of the same intensity.

Patients with RED are distributed without significant differences in the age groups (Group A = 50.00%; Group B = 62.96%; Group C = 78.57%). Though not proven by this study, clinical experience shows that patients with erosive esophagitis without effective treatment oscillate evolutionarily between Savary-Miller22 classification grades 1, 2, and 3 during endoscopic follow-up. In this study, the majority of the patients were in this RED group during endoscopic follow-up. It is induced that the maintenance of the aggressor factor leads to the checking the progress of the disease by individual's protective adaptation, unless there is some new factor yielding unbalance.

For patients with Barrett's Esophagus, there is a consensus in several studies that the epithelial metaplasm appears early in some GERD patients and remains unchanged over the years4,14,20. In this study, a number of patients classified as REeD (including those with Barrett's Esophagus) were present in all three age groups. This finding reinforces the hypotheses that once a complication occurs, its disappearance is very unlikely to occur spontaneously without therapeutic interference.



The findings of this study are consistent with the hypothesis that GERD is a chronic disease with very little chance of spontaneous remission. No relationship between the 24h-pHM standards and grades of endoscopic esophagitis was found. With the similar 24h-pHM parameters (control of an important variable to GERD) all the reflux esophagitis grades were observed. Although the gold standard for GERD diagnosis take into consideration both the agressor factor and local individual factors, the individual factors, are even more important than the refluxed acid quantity in determining GERD's evolution.

Most of the patients were in the group with erosive esophagitis in several grades, thus confirming the clinical experience. The number of patients without epithelial injury (Grade 0 esophagitis) undergoing similar acid aggression significantly decreased in older age groups. We conclude that continued aggression leads, in some patients, to the exhaustion of individual mechanisms of epithelial protection. Patients with complicated forms of GERD were represented in a similar number in the three age groups. This finding reinforces the hypothesis that in some forms of GERD, the tissue injury occurs early in life and remains unchanged over the years.




FLORA FILHO, R. e col. - A esofagite de refluxo e a doença do refluxo gastroesofageano. Estudo transversal em portadores da doença do refluxo gastroesofageano estratificados por grupos etários. Rev. Hosp. Clín. Fac. Med. S. Paulo 54 (2): 61-67, 1999.


Foram selecionados transversalmente 78 pacientes portadores de doença do refluxo gastroesofageano (DRGE) segundo critérios rigorosos de pHmetria esofageana ambulatorial de 24h (pHM-24h) e distribuídos em três grupos etários: Grupo A = 14 e = 24 anos, Grupo B = 25 e = 54 anos e Grupo C = 55 e = 64 anos. A pHM-24h foi realizada segundo a padronização de DeMeester e para diagnóstico da esofagite de refluxo utilizou-se classificação de Savary-Miller. Os grupos foram semelhantes em parâmetros da pHM-24h (p>0,05) cujas valores médios estavam acima da normalidade. A comparação entre grupos etários diferentes com parâmetros semelhantes de refluxo ácido não possibilitou qualquer inferência entre intensidade de refluxo ácido e grau de esofagite. Quando os mesmos pacientes foram subagrupados conforme profundidade da lesão epitelial e confrontados nos grupos etários, houve diferença significativa nas esofagites sem solução de continuidade epitelial. Os achados reforçam que a DRGE é uma doença crônica e sem resolução espontânea. As bases de diagnóstico e seguimento da DRGE não podem prescindir da defesa individual em detrimento da quantidade de ácido refluído, sendo ambos importantes.


DESCRITORES: Esofagite. Refluxo gastro-esofageano. pHM-24h. Classificação Savary-Miller. Endoscopia Digestiva Superior.



1. ANDERSON, S; AUQUER, A; HAUCK, W.W. et al. - Choise of procedure. In: ANDERSON, S. et al. - Statistical Methods for Comparative Studies. New York, Wiley, 1980.         [ Links ]

2. BREMMER, R.M. & DEMEESTER, T.R. - Pre- and postoperative assessments in gastroesophageal reflux disease. In: SCARPIGNATO, C. & GALMICHE, J.P. - Functional Evaluation in Esophageal Diasease. Baser, Karger, 1994. v.22, p.260-287.         [ Links ]

3. BREMMER, C.G. & DEMEESTER, T.R. - Proceedings from an international conference on ablation therapy for Barrett's mucosa. Dis Esophagus 1998; 11: 1-27.         [ Links ]

4. CAMERON, A .J. & LOMBOY, C.T. - Barrett's esophagus - an acquired but nonprogressive disorder. Gastroenterology 1991; 100: A38.         [ Links ]

5. CAMERON, A.J. - Epidemiology of columnar-lined esophagus and adenocarcinoma. Gastroenterol Clin N Am 1997; 26: 487-494.         [ Links ]

6. CONOVER, W.J. - Contingency tables. Methods based on ranks. In: CONOVER, W.J. Practical Nonparametric Statistics. New York, Wiley, 1980.         [ Links ]

7. DEMEESTER, T.R; JOHNSON, L.F; JOSEPH, G.J. et al. - Patterns of Gastroesophageal Reflux in Health and Disease. Ann Surg 1976; 184: 459-470.         [ Links ]

8. DEMEESTER, T.R; WANG, C.I; WERNLY, A.J. - Technique, indications, and clinical use of 24 hour esophageal pH monitoring. J Thorac Cardivasc Surg 1980; 79: 656-670.         [ Links ]

9. EL-SERAG, H.B; SONNENBERG, A. - Associations between different forms of gastro-oesophageal reflux disease. Gut 1997; 41: 594-599.         [ Links ]

10. ERGUN, G.A & KAHRILAS, J. - Clinical Applications of esophageal manometry and pH monitoring. Am J Gastroenterology 1996; 91: 1077-1089.         [ Links ]

11. FLORA FILHO, R. & ZILBERSTEIN, B. - Aplicação do ácido acético a 0,5% no evidenciamento da esofagite de refluxo. GED 1998; 17: 93-95.         [ Links ]

12. FLORA FILHO, R. & ZILBERSTEIN, B. - Relação entre pHmetria esofageana ambulatorial de 24h e endoscopia digestiva alta em pacientes portadores de esofagite de refluxo. Arq Bras Gastroenterologia 1998; 38 (5).         [ Links ]

13. FLORA FILHO, R. & ZILBERSTEIN, B. - Importância do fator etário no determinismo da hérnia hiatal por deslizamento e a doença do refluxo gastroesofageano. Estudo transversal. Arq Bras Gastroenterol (in press).         [ Links ]

14. GILLEN, P.J; KEELING, P & BYRNE, P.J. - Implication of duodenogastric reflux in the pathogenosis of Barrett's esophagus. B J Surg 1988; 75: 540-543.         [ Links ]

15. HEMPEL, C.G. - Hipóteses "Ad Hoc". In: HEMPEL, C.G. - Filosofia da Ciência Natural. Rio de Janeiro, Zahar, 1974. p. 43-45.         [ Links ]

16. JACKSON, C. - Peptic ulcer of the esophagus. JAMA 1929; 92: 369.         [ Links ]

17. JOHNSON, L.F. & HARMON, J.W. - Experimental esophagitis in a rabbit model. Clinical relevance. J Clin Gastroenterol 1986; 1 (Suppl 8): 26-44.         [ Links ]

18. MCDOUGALL, N.I; JOHNSTON, B.T. & KEE, F. - Natural history of reflux esophagitis: a 10 year follow up of its effects on patient symptomatology and quality of life. Gut, 1996; 38: 481-486.         [ Links ]

19. NEBEL, O .T; FORNES, M.F. & CASTELL, D.O. - Symptomatic gastroesophageal reflux: Incidence and precipitating factors. Dig Dis Sci 1976; 21: 953-958.         [ Links ]

20. OLIVEIRA, C.A; LIMA, J.R.G.F. & ANDRAD, E. FILHO, J.S. - Contribuição da endoscopia ao estudo da doença do refluxo gastroesofágico. In: SOCIEDADE BRASILEIRA DE ENDOSCOPIA DIGESTIVA. Endoscopia Digestiva. Rio de Janeiro, Medsi, 1994. p. 19-41.         [ Links ]

21. POPPER, K. - Falseabilidade. In: KARL, R. - A lógica da pesquisa científica, São Paulo, Cultrix, 1989. p. 82-98.         [ Links ]

22. SILVERSTEIN, F.E; TYTGAT, G.N.J. & ISSELBACHER, K.J. - Endoscopic features of reflux esophagitis. In: TYTGAT, G.N.J. - Atlas of Gastrointestinal Endoscopy. Edinburgh, United Kingdom, 1990. p. 2.10-2.12.         [ Links ]

23. TRIADAFILOPOULOS, G. & SHARMA, B.A. - Features of symptomatic gastroesophageal reflux disease in elderly patients. Am J Gastroenterol 1997; 92: 2007-2011.         [ Links ]

24. WINKELSTEIN, A . - Peptic esophagitis, a new clinical entity. JAMA 1935; 104: 906-909.         [ Links ]



Received for publication on the 25/09/99


Study Undertaken at the Gastroenterology Department of the Hospital das Clínicas - São Paulo University Medical School.

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License