clinical significance of minimal change lesions of the esophagus in a healthy korean population: a...

GASTROENTEROLOGY

Clinical significance of minimal change lesions of theesophagus in a healthy Korean population: A nationwidemulti-center prospective studyJun Haeng Lee,* Nayoung Kim,† Il Kwun Chung,‡ Yun-Ju Jo,§ Geom Seog Seo,¶ Sang Wook Kim,**Eui Hyeog Im,†† Hye Rang Kim,‡‡ Soo Hyun Park,‡‡ So-Young Lee,‡‡ Hyun-Min Cha,‡‡ Kyoung Soo Lee,§§

Dong Hyo Hyun,¶¶ Hyun Young Kim,† Sun-Mi Kim,† Jeong Eun Shin,*** Soo-Heon Park,‡‡

Hyun Chae Chung† and In-Sik Chung‡‡ and The H. pylori and GERD Study Group of Korean College ofHelicobacter and Upper Gastrointestinal Research

*Department of Medicine, Sungkyunkwan University School of Medicine, Samsung Medical Center, †Department of Internal Medicine, SeoulNational University College of Medicine, §Department of Internal Medicine, Eulji University, ‡‡Department of Internal Medicine, The CatholicUniversity of Korea, College of Medicine, Seoul, ‡Department of Internal Medicine, Soonchunhyang University, ***Department of InternalMedicine, Dankook University, Cheonan, ¶Department of Internal Medicine, Wonkwang University, Iksan, **Department of Internal Medicine,Chonbuk National University, Chungju, ††Department of Internal Medicine, Kunyang University, Daejeon, §§Department of Internal Medicine, HallaGeneral Hospital and ¶¶Department of Internal Medicine, Hanmaeum Hospital, Jeju, Korea

AbstractBackground and Aim: At least half of the patients with typical reflux symptoms havenon-erosive reflux disease (NERD). Minimal change lesions are commonly seen in thescreening endoscopic examinations for individuals without clinically significant symp-toms. We evaluated the correlation between minimal changes and symptoms in individualsvisiting the hospital for routine health check-up by a nationwide survey in 2006.Methods: Upper gastrointestinal endoscopic examinations as a health check-up wereperformed for 25 536 patients. Among them, symptom questionnaires were given in 23 350patients without mucosal break or Barrett’s esophagus. Endoscopic findings of the loweresophagus were divided into normal or minimal changes. Minimal changes in the presentstudy included white turbid discoloration and Z-line blurring.Results: Among a total of 25 536 subjects, reflux esophagitis was found in 2019 subjects(7.91%) and 3043 patients (11.9%) were classified as having minimal changes. History ofgastroesophageal reflux disease (GERD) was more commonly found in individuals withminimal changes. Among the reflux-related symptoms, heartburn, acid regurgitation, globussensation, and epigastric soreness were related to the minimal changes of the esophagus.Especially, individuals with globus sensation or epigastric soreness were more likely to haveminimal changes compared to individuals without respective symptoms. Male gender,current smoker, history of H. pylori eradication, frequent stooping at work, hiatal hernia, andatrophic/metaplastic gastritis were found to be risk factors for minimal changes.Conclusion: The minimal changes were closely related with upper gastrointestinal symp-toms and had similar risk factors for GERD, suggesting that minimal changes could beconsidered as early endoscopic findings of GERD.

Key words

gastrointestinal reflux disease, healthcheck-up, minimal change lesion.

Accepted for publication 26 June 2007.

Correspondence

Dr Soo-Heon Park, Department of InternalMedicine, The Catholic University of Korea,College of Medicine, Seoul, Korea. Email:[email protected]

IntroductionAt least half of the patients with typical reflux symptoms do nothave endoscopic evidence of gastroesophageal reflux disease(GERD).1–6 For this group of patients, the diagnosis of non-erosive reflux disease (NERD) is commonly used. However,various endoscopic findings have been described in the literature(Table 1). For example, Armstrong et al.4 listed seven findings,such as localized area(s) of erythema in one or more segments atthe mucosal junction, indistinctness or blurring of all parts of themucosal junction, friability at the mucosal junction, diffuseerythema in the distal esophagus, patchy erythema in the distalesophagus, increased vascularity in the distal esophagus, and

edema/accentuation of mucosal folds. The interobserver agree-ment for these descriptions is so low that almost no clinical sig-nificance was given to them.9

In the Eastern literature,7,13,15 endoscopic findings for NERDpatients have been divided into two groups: normal and minimalchange lesion (MCL). Although the interobserver agreement wasreported to be very low,12 some authors report a high rate ofminimal change lesions in GERD patients.10,11,14 One of the prob-lems with MCL is that there is no standard definition (Table 1).Most definitions are too complex to use in clinical practice.Recently, a simplified definition of MCL was suggested byHongo.13 In the present study, findings of white turbid discolora-tion and/or Z-line blurring are considered as MCL.

doi:10.1111/j.1440-1746.2008.05299.x

1153Journal of Gastroenterology and Hepatology 23 (2008) 1153–1157 © 2008 The Authors

Journal compilation © 2008 Journal of Gastroenterology and Hepatology Foundation and Blackwell Publishing Asia Pty Ltd

mailto:[email protected]

Esophagogastroduodenoscopy (EGD) is frequently performedfor the early detection of gastric cancer in Korea. Minimalchange lesions are frequently seen in screening endoscopicexaminations for individuals who have GERD symptoms. Weevaluated whether or not there is any correlation betweenminimal change lesions and GERD symptoms in individuals vis-iting the hospital for routine health check-ups by a nationwidesurvey in 2006.

MethodsThis is a secondary analysis of the data from a study performed bythe Korean College of Helicobacter and Upper Gastrointestinal

Research. The design of the overall study was prepared by theScientific Committee of the same group. Participating gastroenter-ologists’ workshop was held in June 2005 to approve the design ofthe study and to unify the description of endoscopic findings. In thisworkshop, the simplified working definition of MCL (e.g. whiteturbid discoloration and Z-line blurring) was proposed to decreasethe interobserver variation in the endoscopic diagnosis of MCL.After feedback from the participating endoscopists and experts inthis field, the study started in January 2006 and finished in July2006.

During the health check-up, EGD were performed for 25 536subjects in 40 Healthcare Centers, located in South Korea. Amongthem, symptom questionnaires were carried out in 23 350 subjectswithout mucosal break or Barrett’s esophagus. Patients with malig-nancy or peptic ulcer (active and healing stages) were also excluded.

The gastroesophageal reflux questionnaire was designed toevaluate symptom categories such as heartburn, acid regurgitation,chest pain, hoarseness, globus sensation, cough, and epigastricsoreness. In addition, the frequency of the most bothersomesymptom and its severity was asked. Several questions aboutlifestyle factors that can affect GERD were included, such asalcohol use, smoking, income, job title, eradication history ofHelicobacter pylori, history of seven major categories of drug, andpresent status of seven categories of disease. Biochemical testresults such as glucose, cholesterol, triglyceride, H. pylori (any oneof the three following tests: anti-H. pylori IgG test, rapid urease test,and histology), and bone densitometry were also carried out.

From the endoscopic findings of the lower esophagus, the indi-viduals were divided into three groups: (i) normal; (ii) MCL; and(iii) reflux esophagitis. For this grouping, the presence of mucosalbreak(s) was initially determined during the endoscopic examina-tion of the lower esophagus. Individuals with mucosal break(s)were not included in the present study. For individuals withoutmucosal break(s), the endoscopic findings of the lower esophaguswere further divided into two groups (normal vs MCL). Minimalchange lesions in this study include white turbid discoloration andZ-line blurring (Fig. 1).

Statistical analysis

The data were analyzed by c2-tests and logistic regression modelsusing the SAS statistical package.

ResultsA total of 25 536 subjects (male 15 180, 59.5%; female 10 358,40.5%) were included. The mean age was 46.7 years and theparticipating subjects came from all over South Korea, roughlyproportional to the population of Seoul and six other provinces(Gyeonggi, Kangwon, Chungcheong, Kyungsang, Cholla, andJeju). Reflux esophagitis was found in 2019 subjects (7.91%) and3043 subjects (11.9%) were classified as having minimal changes.

The analysis was localized into normal and minimal changes(n = 23 350). Symptom analysis between normal and minimalchanges was possible in 22 923 subjects (98.2%) based on theresults of the questionnaire (Table 2). History of GERD was morecommonly found in individuals with minimal changes than inthe normal group. Among the seven reflux-related symptoms,

Table 1 Minimal changes in the distal esophagus in the selectedliterature

Reference Description

Armstrong et al. (1996)4 Localized area(s) of erythema in one ormore segments at the mucosal junction

Indistinctness or blurring of all parts of themucosal junction

Friability at the mucosal junctionDiffuse erythema in the distal esophagusPatchy erythema in the distal esophagusIncreased vascularity in the distal

esophagusEdema/accentuation of mucosal folds

Hoshihara (1996)7 Erythema and/or whitish turbidityKouzu et al. (1997)8 Erythema and/or whitish turbidity,

including those who have lesionsunstained by iodine spray

Lundell et al. (1999)9 Excessive reddening of the cardiaErythema of the squamocolumnar junctionFriability of the squamocolumnar junctionBlurring of the squamocolumnar junctionDiffuse erythema of the distal esophagusPatchy erythema of the distal esophagusIncreased vascularity of the distal

esophagusEdema or accentuation of the mucosal

foldsKiesslich et al. (2004)10 Presence of vascular injection or vascular

spots above the Z-line, villous mucosalsurface and islands of squamous cellepithelium below the Z-line†

Nakamura et al. (2005)11 Whitish or reddish colors, edematouschange and erosion that is not regardedas mucosal break

Amano et al. (2006)12 Undemarcated discoloration of the loweresophageal mucosa (‘grade M’)

Hongo (2006)13 Erythematous changes (red) andacanthotic changes (white)

Shimizu et al. (2006)14 Presence of prominent erythema withoutclear demarcation or whitish cloudinessof the lower esophageal mucosaobscuring the longitudinal blood vessels

Present study Turbid discoloration and Z-line blurring

†Using magnifying endoscopes.

Minimal changes in healthy population JH Lee et al.

1154 Journal of Gastroenterology and Hepatology 23 (2008) 1153–1157 © 2008 The Authors


heartburn, acid regurgitation, globus, and epigastric soreness wererelated to minimal changes of the esophagus.

Odds ratios for having minimal changes in the esophagus wereestimated based on individual symptoms (Table 3). Individualswith globus symptoms or epigastric soreness were more likely tohave minimal changes compared to individuals without respectivesymptoms (OR = 1.320 and 1.162, respectively).

Odds ratios for having minimal changes in the esophagus wereestimated based on individual characteristics (Table 4). Individ-uals with male gender, current smoker, history of H. pylori eradi-cation, frequent stooping at work, hiatal hernia of �1 cm, and

atrophic/metaplastic gastritis were more likely to have minimalchanges compared to the normal group without minimal changes.

DiscussionThe definition of GERD has been changing. By the definition in theGenval workshop report,1 the term ‘GERD’ should be used toinclude all individuals who are exposed to the risk of physicalcomplications from gastroesophageal reflux, or who experienceclinically significant impairment of health-related well-being(quality of life) due to reflux-related symptoms, after adequatereassurance of the benign nature of their symptoms. Because of the

(a)

(b)

Figure 1 Sample pictures of minimal changes of the lower esophagus.(a) White turbid discoloration, (b) Z-line blurring.

Table 2 Comparison of symptoms of 22 923 individuals with orwithout minimal changes

Normal(n = 19 896)

Minimal changes(n = 3027)

P-value

History of GERD 1355 (6.8%) 321 (10.6%) < 0.01Any seven† symptoms 9950 (50.0%) 1662 (54.9%) < 0.001Heartburn 4891 (24.6%) 812 (26.8%) 0.02Acid regurgitation 7034 (35.4%) 1164 (38.5%) < 0.01Chest pain 2870 (14.4%) 458 (15.1%) 0.25Hoarseness 1914 (9.6%) 313 (10.3%) 0.18Globus sensation 2363 (11.9%) 450 (14.9%) < 0.01Cough 1434 (7.2%) 203 (6.7%) 0.39Epigastric soreness 4680 (23.5%) 788 (26.0%) < 0.01

†Seven symptoms: heartburn, acid regurgitation, chest pain, hoarse-ness, globus sensation, cough and epigastric soreness.GERD, gastroesophageal reflux disease.

Table 3 Odds ratio for the presence of minimal changes in personswith or without individual symptom(s)

Symptom Odds ratio 95% CI P-value

Heartburn 1.123 0.982–1.285 0.906Acid regurgitation 1.040 0.901–1.201 0.591Chest pain 0.981 0.817–1.178 0.838Hoarseness 0.974 0.765–1.242 0.834Globus sensation 1.320 1.158–1.505 < 0.001Cough 0.881 0.683–1.136 0.329Epigastric soreness 1.162 1.034–1.305 0.012

Table 4 Risk factors for minimal changes (n = 23 341)

Risk factor Odds ratio 95% CI P-value

Male gender 1.339 1.237–1.449 < 0.0001Smoking 1.269 1.165–1.383 < 0.0001Alcohol 0.937 0.824–1.065 0.3180Diabetes mellitus 0.970 0.822–1.145 0.7175History of H. pylori

eradication1.222 1.075–1.395 0.0030

Stooping posture duringwork

1.235 1.122–1.358 < 0.0001

Hiatal hernia 4.444 3.759–5.263 < 0.0001Atrophic or metaplastic

gastritis1.679 1.446–1.991 < 0.0001

CI, confidence interval.

JH Lee et al. Minimal changes in healthy population



ambiguity of the terminology in this definition, the diagnosis ofGERD is commonly made when a patient has typical symptom(s) ofgastroesophageal reflux, such as heartburn or acid regurgitation.However, some patients have complications of gastroesophagealreflux with no typical symptoms. In 2005, the American Collegeof Gastroenterology (ACG) published a practice guideline onGERD.5 According to this guideline, the diagnosis of GERD can bemade when there are symptoms or mucosal damage produced by theabnormal reflux of gastric contents into the esophagus. This defi-nition made it clear that typical symptoms of gastroesophagealreflux are not mandatory for the diagnosis of GERD.5 EGD iscommonly performed in Korea for the early detection of gastriccancer. During the screening endoscopy, findings of erosive esoph-agitis are frequently found.16 The diagnosis of GERD is possible forthese patients by the ACG practice guideline.5 However, as yet, themeaning of MCL found during screening endoscopy is unclear.

When investigation is needed for a patient with reflux symp-toms, endoscopy is the first choice, as it is the only test that cangive sensitive recognition and grading of esophagitis and reliablediagnosis of esophageal columnar metaplasia.5 Endoscopy alsoallows for the effective identification of significant peptic stric-tures, other types of esophagitis, and other upper gastrointestinaldisorders such as peptic ulcer disease and esophageal and gastriccarcinoma. However, most patients with reflux disease do not haveendoscopically visible mucosal damage, so a negative endoscopydoes not exclude the diagnosis of reflux disease. In other words,endoscopy appears to be an insensitive test for GERD. Tradition-ally, patients with GERD without endoscopic evidence of thedisease were given the diagnosis of NERD. Patients with NERDcommonly demonstrate a negative pH test.2 Acid reflux character-istics and symptom patterns suggest a heterogeneous group ofpatients. However, this terminology has caused some confusionamong endoscopists. For a description of the endoscopic findingsin patients with GERD, the LA classification is most commonlyused.4,9,17 In this classification, the endoscopic evidence of GERDis the mucosal break, which is defined as an area of slough orerythema clearly demarcated from the adjacent normal-appearingmucosa. When a discrete non-confluent area of erythema withoutdefinite erosion or ulcer is found during endoscopy for patientswith reflux symptoms, this finding (‘non-erosive mucosal break’ or‘red streak’18) can be called LA group A or B depending on thelength of the area.4 Strictly speaking, the term NERD means refluxdisease without esophageal erosion(s), so a non-erosive mucosalbreak is one of the reasons for the confusion in the diagnosis ofNERD. In order to avoid this confusion, we grouped the patientsby the existence of esophageal erosion. For patients withoutesophageal erosion, endoscopic findings of the lower esophaguswere divided into normal or MCL.

Recently, it was reported that the interobserver agreement onclassifying endoscopic diagnoses of non-erosive esophagitis wastoo low to have clinical significance.12 However, the images usedin that study seem to be suboptimal. They used 30 endoscopicimages projected onto a screen and they used still images only.12 Inour opinion, a larger study using high-quality motion pictures arenecessary to exclude the clinical significance of MCL. Recently,lugol chromoendoscopy was reported to be useful as a diagnostictool in NERD.19 In this study, several unstained streaks wereobserved in the distal esophagus in approximately half the patientswith NERD. Another tool for objective evaluation of NERD is

histology. By a recent review,20 at least two-thirds of NERDpatients have microscopic esophageal mucosal damage, such asdilated intercellular space (DIS). This microscopic esophagealinjury responded to proton pump inhibitor (PPI) therapy.

In spite of these findings, the characteristics of patients withMCL are still unclear. For instance, Nakamura et al. reported thatmost NERD patients can be classified as MCL.11 In individualswith MCL, symptomatic patients were more likely to have hiatalhernia than those who were asymptomatic.11 In the present study,we found that various upper gastrointestinal symptoms, whichsuggested GE reflux, were related to minimal changes of theesophagus (Table 2). These included heartburn, acid regurgitation,globus sensation, and epigastric soreness. In addition, individualswith globus symptoms or epigastric soreness were more likely tohave minimal changes compared to individuals without respectivesymptoms. Individuals with male gender, current smoker, historyof H. pylori eradication, frequent stooping at work and hiatalhernia, which have been reported as risk factors for GERD,21–24

were more likely to have minimal changes compared to the normalgroup without MCL findings. In our opinion, these findingssuggest that minimal changes could be considered as early endo-scopic findings of GERD.

As far as we understand, this is the first report evaluating theclinical significance of MCL found in screening endoscopy. Theprevalence of various GI symptoms is quite high in the generalpopulation. The prevalence of functional upper gastrointestinalsymptoms is especially high in patients with GERD. In a recentstudy,25 72% of patients with GERD had associated functional GIdisorders, with a mean of 4.1 � 1.9 functional digestive symptomsper patient. Among patients with functional GI symptoms, 27%had symptoms suggestive of irritable bowel syndrome (IBS), 16%were suggestive of dyspepsia while 57% had both upper and lowerfunctional digestive symptoms. However, individuals with mildupper GI symptoms usually do not search for medical evaluations.As a result, the prevalence of upper GI symptoms is quite highamong populations seeking screening endoscopy mainly forgastric cancer. Because the prevalence of various upper GI symp-toms was higher in patients with MCL, we think that the presenceof MCL in patients with screening endoscopy may indicate thenecessity for further evaluations. In this way, patients with unde-tected GERD may be selected.

Our study has some limitations. First, we did not perform sys-tematic biopsies of the gastric mucosa to evaluate the gastricatrophy and/or metaplasia. So, the unexpectedly high risk of MCLin individuals with atrophic and/or metaplastic gastritis is difficultto interpret. Because, the interobserver variation of the endoscopicdiagnosis of atrophic and/or metaplastic gastritis may be high,further studies of the systematic biopsies are necessary to draw afirm conclusion about the relationship between atrophic/metaplastic gastritis and MCL. Second, we did not evaluate therelationship between current H. pylori infection and MCL. Becausethe H. pylori infection status may influence gastric acid secretion, itshould be included in future studies. Third, we did not evaluate thefindings suggesting MCL for patients with mucosal break or Bar-rett’s esophagus. Finally, in the present study, there are no follow-updata for patients with MCL. The natural history of GERD is stillunclear. Some investigators proposed that GERD might be dividedinto three categories: non-erosive GERD, erosive GERD andBarrett’s esophagus.26,27 However, other investigators believe that

Minimal changes in healthy population JH Lee et al.

1156 Journal of Gastroenterology and Hepatology 23 (2008) 1153–1157 © 2008 The Authors


GERD is a progressive disease.28,29 In order to study the naturalhistory of MCL, long-term follow-up studies are necessary.

In summary, the prevalence of MCL was 11.9% in Koreanpopulations seeking screening endoscopy. Individuals with globussensation or epigastric soreness were more likely to have minimalchanges compared to individuals without respective symptoms.MCL in these individuals may help diagnose hidden cases withGERD.

AcknowledgmentsWe would like to thank the participating gastroenterologists.Sang Woo Lee and Jong-Jae Park, Korea University; Jae Woo Kim,Wonju University; Hyun Jin Kim, Kyungsang University; Moon GiChung, Gachon Medical School, Seon Mee Park, ChungbukNational University; Gwang Ho Baik and Joon Yong Park, HallymUniversity; Byung Kyu Nah, University of Ulsan; Su Youn Nam,Korean National Cancer Center, Kang Seok Seo, Kwangju Chris-tian Hospital; Byung Sung Ko, Eulji University; Jae-Young Jang,Kyung Hee University; Byeong Gwan Kim and Ji Won Kim, SeoulNational University; Su Jin Hong, Soon Chun Hyang University;Jae Kyu Sung, Chungnam National University; Kyung Sik Park,Keimyung University; Seong-Eun Kim, Ewha Womens University,Hyun-Shin Park, Inha University; Young Sun Kim, Seon Hee Lim,Chung Hyeon Kim, Min Jung Park, Jeong Yoon Yim, Kyung RanCho and Dong Hee Kim, Seoul National University, Health CareSystem Gangnam Center; Seun Ja Park, Kosin University; SamRyong Ji and Jeong Seop Moon, Inje University; Chan-Guk Park,Chosun University; Chang-HunYang, Dong Guk University; HyunSoo Kim, Chonnam National University; Sungkook Kim, Kyung-pook University; Geun Am Song, Busan National University.

References1 Dent J, Brun J, Fendrick AM et al. An evidence-based appraisal of

reflux disease management - the Genval Workshop Report. Gut1999; 44 (Suppl. 2): S1–16.

2 Martinez SD, Malagon IB, Garewal HS, Cui H, Fass R. Non-erosivereflux disease (NERD) – acid reflux and symptom patterns. Aliment.Pharmacol. Ther. 2003; 17: 537–45.

3 Quigley EM, DiBaise JK. Non-erosive reflux disease: the realproblem in gastro-oesophageal reflux disease. Dig. Liver Dis. 2001;33: 523–7.

4 Armstrong D, Bennett JR, Blum AL et al. The endoscopicassessment of esophagitis: a progress report on observer agreement.Gastroenterology 1996; 111: 85–92.

5 DeVault KR, Castell DO. Updated guidelines for the diagnosis andtreatment of gastroesophageal reflux disease. Am. J. Gastroenterol.2005; 100: 190–200.

6 Fujiwara Y, Higuchi K, Shiba M et al. Differences in clinicalcharacteristics between patients with endoscopy-negative refluxdisease and erosive esophagitis in Japan. Am. J. Gastroenterol. 2005;100: 754–8.

7 Hoshihara Y. Diagnosis of GERD (in Japanese). Rinsho ShokakiNaika 1996; 11: 1563–8.

8 Kouzu T, Miyazaki S, Yoshimura K, Hishikawa E, Arima M,Harada N. Barrett esophagus and reflux esophagitis. EndoscopiaDigestiva 1997; 9: 991–7 (in Japanese).

9 Lundell LR, Dent J, Bennett JR et al. Endoscopic assessment ofoesophagitis: clinical and functional correlates and further validationof the Los Angeles classification. Gut 1999; 45: 172–80.

10 Kiesslich R, Kanzler S, Vieth M et al. Minimal change esophagitis:prospective comparison of endoscopic and histological markersbetween patients with non-erosive reflux disease and normal controlsusing magnifying endoscopy. Dig. Dis. 2004; 22: 221–7.

11 Nakamura T, Shirakawa K, Masuyama H, Sugaya H, Hiraishi H,Terano A. Minimal change oesophagitis: a disease with characteristicdifferences to erosive oesophagitis. Aliment. Pharmacol. Ther. 2005;21 (Suppl. 2): 19–26.

12 Amano Y, Ishimura N, Furuta K et al. Interobserver agreement onclassifying endoscopic diagnoses of nonerosive esophagitis.Endoscopy 2006; 38: 1032–5.

13 Hongo M. Minimal changes in reflux esophagitis: red ones and whiteones. J. Gastroenterol. 2006; 41: 95–9.

14 Shimizu Y, Dobashi K, Kobayashi S et al. High prevalence ofgastroesophageal reflux disease with minimal mucosal change inasthmatic patients. Tohoku J. Exp. Med. 2006; 209: 329–36.

15 Lee JH, Lee JS, Rhee PL et al. Interobserver variation in theendoscopic diagnosis of gastroesophageal reflux disease. Korean J.Gastrointest. Endosc. 2006; 33: 197–203.

16 Sung KC, Shim SC, Choi YS et al. Esophagogastroduodenoscopicfindings in 9,137 healthy subjects examined for the secondaryprevention. Korean J. Gastrointest. Endosc. 1998; 18: 161–8.

17 Kusano M, Ino K, Yamada T et al. Interobserver and intraobservervariation in endoscopic assessment of GERD using the ‘LosAngeles’ classification. Gastrointest. Endosc. 1999; 49: 700–4.

18 Vieth M, Haringsma J, Delarive J et al. Red streaks in theoesophagus in patients with reflux disease: is there ahistomorphological correlate? Scand. J. Gastroenterol. 2001; 36:1123–7.

19 Yoshikawa I, Yamasaki M, Yamasaki T, Kume K, Otsuki M.Lugol chromoendoscopy as a diagnostic tool in so-calledendoscopy-negative GERD. Gastrointest. Endosc. 2005; 62:698–703; quiz 752, 754.

20 Dent J. Microscopic esophageal mucosal injury in nonerosive refluxdisease. Clin. Gastroenterol. Hepatol. 2007; 5: 4–16.

21 Labenz J, Jaspersen D, Kulig M et al. Risk factors for erosiveesophagitis: a multivariate analysis based on the ProGERD studyinitiative. Am. J. Gastroenterol. 2004; 99: 1652–6.

22 Rosaida MS, Goh KL. Gastro-oesophageal reflux disease, refluxoesophagitis and non-erosive reflux disease in a multiracial Asianpopulation: a prospective, endoscopy based study. Eur. J.Gastroenterol. Hepatol. 2004; 16: 495–501.

23 Avidan B, Sonnenberg A, Schnell TG, Sontag SJ. Risk factors forerosive reflux esophagitis: a case-control study. Am. J. Gastroenterol.2001; 96: 41–6.

24 Fujiwara Y, Higuchi K, Shiba M et al. Association betweengastroesophageal flap valve, reflux esophagitis, Barrett’s epithelium,and atrophic gastritis assessed by endoscopy in Japanese patients.J. Gastroenterol. 2003; 38: 533–9.

25 Guillemot F, Ducrotte P, Bueno L. Prevalence of functionalgastrointestinal disorders in a population of subjects consulting forgastroesophageal reflux disease in general practice. Gastroenterol.Clin. Biol. 2005; 29: 243–6.

26 Fass R, Ofman JJ. Gastroesophageal reflux disease - should weadopt a new conceptual framework? Am. J. Gastroenterol. 2002; 97:1901–9.

27 Tack J, Fass R. Review article: approaches to endoscopic-negativereflux disease: part of the GERD spectrum or a unique acid-relateddisorder? Aliment. Pharmacol. Ther. 2004; 19 (Suppl. 1): 28–34.

28 Pace F, Bianchi Porro G. Gastroesophageal reflux disease: a typicalspectrum disease (a new conceptual framework is not needed). Am.J. Gastroenterol. 2004; 99: 946–9.

29 Pace F, Pallotta S, Vakil N. Gastroesophageal reflux disease is aprogressive disease. Dig. Liver Dis. 2007; 39: 409–14.

JH Lee et al. Minimal changes in healthy population



clinical significance of minimal change lesions of the esophagus in a healthy korean population: a...

Documents