- Email: [email protected]
24-hour gastric pH and extent of duodenal gastric metaplasia in Helicobacter pylori-positive patients
GASTROENTEROLOGY 1997;113:741–745
ALIMENTARY TRACT 24-Hour Gastric pH and Extent of Duodenal Gastric Metaplasia in Helicobacter pylori–Positive Patients VINCENZO SAVARINO,* GIUSEPPE SANDRO MELA,* PATRIZIA ZENTILIN,* GABRIELLA LAPERTOSA,‡ PAOLA CEPPA,‡ SERGIO VIGNERI,§ MARIA RAFFAELLA MELE,* CARLO MANSI,* DANIELA TRACCI,* GIULIANA BISSO,* and GUIDO CELLE* *Dipartimento di Medicina Interna, Cattedra di Gastroenterologia, and ‡Istituto di Anatomia Patologica, Universita` di Genova, Genova; and § Istituto di Medicina Interna e Geriatria, Universita` di Palermo, Palermo, Italy
Background & Aims: Gastric metaplasia (GM) is essential to explain duodenal colonization by Helicobacter pylori. It seems to be acid induced but also occurs in H. pylori–positive patients with nonulcer dyspepsia (NUD), who are not acid hypersecretors. The aim of this study was to assess the circadian gastric acidity of 47 patients with duodenal ulcers (DUs) and 32 patients with NUD, both H. pylori positive, and its correlation to duodenal GM extent. Methods: H. pylori was detected by histology and CLOtest, and GM was diagnosed and graded on four bulb biopsy specimens. Each patient underwent 24-hour gastric pH-metry, and the relation between gastric pH and GM extent was assessed by factorial analysis. Results: Gastric pH was greater in patients with NUD than in patients with DU during 24 hours, night and daytime (P õ 0.03–0.005). Gastric pH differed significantly (P õ 0.0002) in relation to GM extent between the two populations, whereas no difference was found among the pH values of GM degrees. A significant increase in 24-hour gastric pH was associated with greater GM in patients with DU, whereas the opposite occurred in patients with NUD (P õ 0.007). Conclusions: The lower gastric acidity in patients with NUD than in patients with DU and the lack of correlation between gastric pH and the various GM degrees in the two H. pylori–positive populations suggest that gastric hyperacidity is not associated with duodenal GM.
I
t has been suggested that the presence of gastric metaplasia (GM) in the duodenum is necessary to allow Helicobacter pylori to colonize this region and to favor ulceration.1 The development of duodenal gastric-type epithelium is thought to be acquired mainly in response to acid hypersecretion.2 However, we have already shown that there is no difference in the pattern of circadian gastric acidity between patients with duodenal ulcer (DU) disease with and without GM in the duodenum.3 / 5e20$$0003
08-08-97 13:37:18
gasa
This finding seems to rule out a major role for gastric acid in the development of this histological change. Moreover, duodenal GM may occur frequently in H. pylori–positive patients with nonulcer dyspepsia (NUD), which is rarely associated with acid hypersecretion.4 The different patterns of gastric acidity in patients with DU disease and NUD compared with that of healthy subjects5,6 seem to reduce further the relevance of acid in determining the genesis of this histological alteration. Probably the best way to define the role of acid in this field is the direct correlation of gastric pH with the extent of gastric-type epithelium in the duodenum. Therefore, we performed this study to assess the circadian pattern of gastric acidity in H. pylori–positive patients with DU disease and NUD and to correlate this variable with the extent of GM in the duodenum of these two populations. The hypothesis tested was that gastric acidity should increase in parallel with the extent of duodenal GM if acid is the most important factor in inducing this histological change.
Patients and Methods Patients Forty-seven patients with DU disease (35 men and 12 women; mean age, 52 { 12 years) and thirty-two patients with NUD (23 men and 9 women; mean age, 49 { 14 years) were enrolled into the study. Thirty-three patients in the first group and 21 patients in the second group were smokers. The ulcer crater was ascertained endoscopically, and the diagnosis of NUD was based on the presence of persistent or recurrent dyspeptic symptoms without any abnormality of the upper gastrointestinal tract at endoscopy or systemic and metabolic diseases. Dyspeptic patients complained of unexplained epigastric pain and/or abdominal discomfort (postprandial fullness, Abbreviations used in this paper: DU, duodenal ulcer; GM, gastric metaplasia; NUD, nonulcer dyspepsia. q 1997 by the American Gastroenterological Association 0016-5085/97/$3.00
WBS-Gastro
742 SAVARINO ET AL.
GASTROENTEROLOGY Vol. 113, No. 3
early satiety, nausea, and vomiting) centered in the upper abdomen7 of at least 2 months’ duration. Patients who were administered nonsteroidal anti-inflammatory drugs or who were undergoing long-term treatment with antisecretory compounds were excluded from the investigation. None of the patients was treated with H2-receptor antagonists or proton pump inhibitors during the 2 weeks preceding the inclusion into the study. No patient had undergone previous gastric surgery. All patients were H. pylori positive and had duodenal GM defined as the occurrence of foci of gastric epithelial cells containing apical periodic acid–Schiff–positive neutral mucine together with the absence of a brush border. All patients gave their informed consent to the study, which was also approved by our local ethical committee.
Methods The diagnosis of H. pylori infection was based on histology and rapid urease test. These procedures were performed on antral and angular biopsy specimens8 using the well-established Wartin–Starry and modified Giemsa methods for the former and CLO test (Delta West Ltd., Bentley, Western Australia, Australia) for the latter. GM was diagnosed from four duodenal biopsy specimens that were taken from the following locations: two from the ulcer edge and two from the opposite wall in patients with an ulcer crater, and from the roof, the floor, and the anterior and posterior wall of the bulb in patients with NUD. The presence of H. pylori was also sought for in these bulb biopsy specimens. The extent of GM was estimated independently by two histopathologists as õ5%, 5%–20%, 20%–50%, or ú50% of the total epithelial surface present by averaging the scores of the four specimens obtained from the same patient. The interobserver agreement of the two investigators over the grading of the extent of GM was high (k Å 0.83). Each patient underwent 24-hour continuous intragastric pH-metry; in the DU disease group, this was performed within 2 days of the endoscopic diagnosis of the crater. The examination was performed with two closely adjacent pH electrodes (Ingold 440-M 3 and M 1.5; Medical Instruments Corp., Solothurn, Switzerland) connected to a two-channel solid-state data logger (Digitrapper ED; Synectics AB, Stockholm, Sweden) and positioned fluoroscopically approximately 10 cm below the cardia in the gastric corpus. The simultaneous recording with two paired pH electrodes has been shown to improve the accuracy of the test by removing noise, interference, and artifacts.9 The electrodes were calibrated using commercial buffer solutions at pH 7.0 and pH 1.0 in accordance with an already validated procedure,10 and gastric pH was measured with a 6-second sampling rate. Patients were fully ambulant, and smoking was permitted because it does not affect the gastric pH pattern.11 Standardized meals were given at the preset times of 6 PM, 8 AM, and noon.12 Extra food or water and alcoholic or carbonated beverages were forbidden during the test day.
Data Processing and Statistics Experimental data obtained from continuous gastric pH-metry were downloaded into a calculator (Proliant; Com-
/ 5e20$$0003
08-08-97 13:37:18
gasa
Figure 1. Twenty-four–hour mean pH profiles of H. pylori –positive patients with DU disease ( ) and NUD ( ).
paq, Bo¨bingen, Germany) using service routines. The pH readings of the two channel tracings of each patient were averaged every minute so that 1440 data points per 24 hours were obtained.9 Acidity data were expressed as mean pH13 for three different time intervals: 24 hours (5 PM to 4:59 PM), night (8 PM to 7:59 AM), and daytime (8 AM to 7:59 PM). These findings were compared by means of two-way analysis of variance, and P values were corrected for multiple testing. The relationship between 24-hour mean pH and the extent of duodenal GM was assessed by means of two-factor nonparametric factorial analysis for unequal cell size. x Test was used to compare the percentage of GM between ulcer edges and other bulb areas as well as the percentage of H. pylori detection in the bulb of patients with DU disease and NUD. The significance level was set at P value of õ0.05.
Results In 2 patients, gastric pH-metry was repeated because of prolonged discrepancy (more than 1 hour longer than the circadian cycle) between the pH readings of the two closely adjacent minielectrodes. H. pylori was detected in the areas of duodenal GM of 12 of 47 patients with DU disease (25%) and 7 of 32 patients with NUD (21%) (P Å NS). GM was observed at the edge of the ulcer in all of the 47 patients with DU disease (100%) and in the areas distant from the ulcer crater in 39 of them (82%). The difference was significant (P õ 0.05). The 24-hour gastric mean pH profiles of the two populations we examined are reported in Figure 1. It is clearly visible that the pH curve pertaining to patients with NUD runs above that of patients with DU disease during almost the whole circadian period. Although the difference can be easily observed after each postprandial period, it is particularly evident during the night. The mean pH values of patients with NUD are significantly greater than those of patients with DU for each WBS-Gastro
September 1997
GASTRIC PH AND BULB GASTRIC METAPLASIA 743
similar to that of patients with DU disease. This result, however, was obtained with a stimulus comparable to food-stimulated acid secretion and is unable to represent the actual influence of H. pylori infection on the natural secretion of acid by the stomach during the whole circadian cycle.16 In NUD, on the other hand, it has been shown that there is no difference in terms of basal and maximal acid output between H. pylori–positive and –negative patients, as well as between these two subgroups and healthy subjects.17 This further confirms that H. pylori infection does not affect the output of acid in this disorder. GM is believed to be an essential prerequisite for the colonization of duodenal mucosa by H. pylori and, then, for explaining how a germ that is located electively within the stomach influences the pathogenesis of DU more than gastric ulcer.1 This gastric-type epithelium is found more commonly at the edge of the ulcer crater than in the other areas of the duodenal cap,18,19 as we confirmed in our study. H. pylori, however, was detected in GM areas of only 25% and 21% of our patients with DU disease and NUD, respectively. Other investigators20,21 have observed this disappointing low percentage of positivity compared with the success of isolating the germ in areas of gastritis. Probably the presence of bile in the duodenum makes it a hostile environment for H. pylori survival.22 Duodenal GM seems to be a protective mechanism against an injury mainly secondary to increased secretion of gastric acid.2 However, GM has been shown both in healthy subjects and in patients with NUD,23,24 who are not acid hypersecretors. Our study is the first to correlate the pattern of 24hour gastric acidity to the extent of GM in the duodenum of H. pylori–positive patients. The reliability of our technique in differentiating the level of gastric acidity in various clinical conditions associated with altered gastric acid secretion has been shown in several physiological studies.5,25,26 It is well known that intraluminal pH monitoring ignores the volume of acid secretion, but gastric acidity may reflect more than acid output the pathophysiological conditions present at the site of peptic lesions27 and then the factors inducing the histological alterations
Table 1. Mean pH Values and SD During Three Different Time Periods in the Two Populations of Patients With DU Disease and NUD With H. pylori Positivity Time period
DU disease
NUD
5 PM to 4:59 PM 8 PM to 7:59 AM 8 AM to 7:59 PM
a
2.1 { 1.0 (1.8–2.4) 2.1 { 1.2 (1.7–2.5) 2.2 { 0.8 (1.9–2.5)
1.6 { 0.4 (1.5–1.7) 1.5 { 0.6b (1.3–1.7) 1.7 { 0.3a (1.6–1.8)
NOTE. Values are expressed as means { SD. Values in parentheses represent 95% confidence intervals. a P õ 0.005. b P õ 0.03.
time interval we analyzed (24 hours, P õ 0.005; night, P õ 0.03; and daytime, P õ 0.005), as shown in Table 1. The relationship between the 24-hour mean gastric pH and the extent of duodenal GM is reported in Table 2. There is a remarkable difference in gastric pH values related to the extent of gastric-type epithelium in the duodenum between the two populations (P õ 0.0002), whereas pH does not differ among the various degrees of GM within the same populations. There is a highly significant interaction between the two populations and the percentages of bulb GM (P õ 0.007). This means that a significant increase in 24-hour mean gastric pH is associated with greater GM in patients with DU disease, whereas the opposite occurs in patients with NUD.
Discussion Our study shows that circadian gastric acidity is significantly lower in H. pylori–positive patients with NUD than in H. pylori–positive patients with DU disease. Despite the claim that H. pylori infection significantly affects gastric physiology in terms of increase in fasting and especially postprandial gastrinemia and then in parietal cell production of acid,14 our findings show that the gastric acidity patterns of patients with DU disease and NUD, both infected with H. pylori, remain greatly different. Recently, the adoption of gastrin-releasing peptide as a stimulant has led El-Omar et al.15 to conclude that a substantial proportion of H. pylori– positive patients with NUD has an increased acid output
Table 2. Relationship Between 24-Hour Mean pH and the Extent of Duodenal GM in the Two Populations With H. pylori Infection GM (% of total epithelia) Populations
õ5%
5%–20%
20%–50%
ú50%
DU disease NUD
1.4 { 0.4 (n Å 19) 2.8 { 0.8 (n Å 10)
1.6 { 0.4 (n Å 14) 2.5 { 0.7 (n Å 9)
1.7 { 0.3 (n Å 8) 1.6 { 0.2 (n Å 5)
2.0 { 0.5 (n Å 6) 1.9 { 0.7 (n Å 8)
NOTE. Values are expressed as means { SD.
/ 5e20$$0003
08-08-97 13:37:18
gasa
WBS-Gastro
744 SAVARINO ET AL.
GASTROENTEROLOGY Vol. 113, No. 3
implicated in the formation of the ulcer crater. The main result of our investigation is that 24-hour gastric pH does not differ significantly in relation to the various degrees of GM within the two populations of patients with DU disease and NUD. Even more, pH values progressively increase as we move from the lowest to the highest degrees of histological alteration in DU disease, thus contradicting our initial hypothesis that gastric acidity should increase in parallel with the degree of gastric metaplasia if acid is so relevant to GM development. In patients with NUD, on the contrary, pH values tend to decrease with the highest extent of GM, but the 24-hour levels of acidity found at the greatest degrees of histological change are still comparable to those of normal subjects.3,25,26 This rules out gastric acid hypersecretion as a central factor in determining the extent of duodenal GM also in NUD. Our results are at variance with those recently published by Harris et al.,28 who showed that there is a correlation between basal and stimulated acid output and the degree of GM. However, they found this correlation in patients with DU disease and not in normal subjects, both H. pylori positive and H. pylori negative. If the role of acid is crucial in the pathogenesis of duodenal GM, it is very surprising that this factor correlates with GM only in one population and not in the other. Harris et al. do not try to explain this evident discrepancy, but the low prevalence of H. pylori infection29 and the high variability of GM23 in healthy subjects probably make this population unsuitable for assessing a correlation between acid output and GM. For this reason, we preferred to evaluate the relationship between these two variables in two populations with a high prevalence of both H. pylori infection and duodenal GM.23 An additional weakness of the study by Harris et al. is that GM does not change 6 months after the significant reduction of acid output secondary to H. pylori eradication. Khulusi et al.30 showed that GM decreases by ú40% after 6 months of acid suppression obtained with a continuous antisecretory treatment of 40 mg/die omeprazole. Moreover, Harris et al. exclude that H. pylori eradication produces a decrease in the extent of GM; once again, this is in contrast with the findings by Khulusi et al. who showed a 42% GM reduction as the exclusive result of the disappearance of the germ. In our opinion, these conflicting data do not contribute to the clarification of the complicated matter of the relationship between acid and GM in H. pylori– positive patients. Our findings related to hydrogen ion activity instead of acid output suggest a negative correlation between these two variables. Other factors may be involved in the production of GM, such as the germ-induced inflammation30 or the / 5e20$$0003
08-08-97 13:37:18
gasa
impaired neutralization of normal acid load to the duodenum. In fact, the secretion of duodenal bicarbonates needed to neutralize the acid entering into the duodenum has been shown to be impaired in H. pylori–infected patients and, more importantly, is restored after bacterial eradication.31 Thus, the development of duodenal GM seems to be a nonspecific response to mucosal injury caused by multiple factors and not exclusively to acid hypersecretion. Hogan et al.31 did not find any difference in the prevalence of duodenal GM between healthy subjects and patients with DU disease (79% vs. 75%), as further confirmation that this histological alteration can be observed in groups with completely divergent gastric acidity patterns.5 Also, the occurrence of duodenal gastric-type epithelium in several different populations, including healthy subjects and patients with NUD and DU disease, casts many doubts on its importance in DU pathogenesis. In summary, our study shows that the circadian gastric acidity of patients with NUD is significantly lower than that of patients with DU disease, although both groups were infected with H. pylori. There is no correlation between 24-hour mean pH values and the various degrees of duodenal GM within the two populations. Furthermore, gastric acidity does not increase in parallel with the degree of histological change in DU disease, whereas it tends to increase in patients with NUD. These findings suggest that gastric hyperacidity is not a key factor in patients with duodenal GM.
References 1. Walker MM, Dixon MF. Gastric metaplasia: its role in duodenal ulceration. Aliment Pharmacol Ther 1996;10(Suppl 1):119– 128. 2. Carrick J, Lee A, Hazell S, Ralston M, Daskalopoulos G. Campylobacter pylori, duodenal ulcer, and gastric metaplasia: possible role of functional heterotopic tissue in ulcerogenesis. Gut 1989; 30:790–797. 3. Savarino V, Mela GS, Zentilin P, Mele MR, Lapertosa G, Patetta R, Dallorto E, Vassallo A, Mansi C, Vigneri S, Celle G. Circadian gastric acidity in Helicobacter pylori positive ulcer patients with and without gastric metaplasia in the duodenum. Gut 1996;39: 508–512. 4. Collen MJ. Gastric analysis (basal acid output) in non-ulcer dyspepsia. Dig Dis Sci 1990;35:540–541. 5. Savarino V, Mela GS, Scalabrini P, Sumberaz A, Fera G, Celle G. 24-hour study of intragastric acidity in duodenal ulcer patients and normal subjects using continuous intraluminal pH-metry. Dig Dis Sci 1988;33:1077–1080. 6. Savarino V, Mela GS, Celle G, Vigneri S. Acid and gastric metaplasia in the duodenum. Gut 1994;35:1151–1152. 7. Talley NJ, Colin-Jones D, Koch KL, Koch M, Nyren O, Stanghellini V. Functional dyspepsia: a classification with guidelines for diagnosis and management. Gastroenterol Int 1991;4:145–160. 8. Genta RM, Graham DY. Comparison of biopsy sites for the histopathologic diagnosis of Helicobacter pylori: a topographic study of H. pylori density and distribution. Gastrointest Endosc 1994; 40:342–345.
WBS-Gastro
September 1997
GASTRIC PH AND BULB GASTRIC METAPLASIA 745
9. Mela GS, Savarino V, Malesci A, Di Mario F, Sossai P, Vigneri S, Zambotti A. New method for improving accuracy of 24-hour continuous intragastric pH-metry. Reflections on physiological and pharmacological studies. Dig Dis Sci 1994;39:1416–1424. 10. Savarino V, Mela GS, Zentilin P, Magnolia MR, Scalabrini P, Valle F, Moretti M, Bonifacino G, Celle G. Gastric aspiration versus antimony and glass pH electrodes. A simultaneous comparative in vivo study. Scand J Gastroenterol 1989;24:434–439. 11. Bauerfeind P, Cilluffo T, Fimmel CJ, Emde C, Von Ritter C, Kohler W, Gugler R, Gasser T, Blum AL. Does smoking interfere with the effect of histamine H2-receptor antagonists on intragastric acidity in man? Gut 1987;28:549–556. 12. Savarino V, Mela GS, Scalabrini P, Magnolia MR, Di Timoteo E, Percario G, Celle G. Continuous 24-hour intragastric pH monitoring: focus on reproducibility in duodenal ulcer patients. Gastroenterol Clin Biol 1986;10:826–830. 13. Mela GS, Savarino V, Sumberaz A, Bonifacino G, Zentilin P, Villa G, Caputo E. Continuous acidity monitoring in the study of gastric antisecretory drugs: pH or antilog transformation of pH, mean or median? Am J Gastroenterol 1990;85:1105–1108. 14. Peterson WL. Gastrin and acid in relation to Helicobacter pylori. Aliment Pharmacol Ther 1996;10(Suppl 1):97–102. 15. El-Omar E, Penman I, Ardill JES, McColl KEL. A substantial proportion of non-ulcer dyspepsia patients have the same abnormality of acid secretion as duodenal ulcer patients. Gut 1995;36:534– 538. 16. Pounder RE. Helicobacter pylori and gastroduodenal secretory function. Gastroenterology 1996;110:947–950. 17. Tucci A, Corinaldesi R, Stanghellini V, Tosetti C, Di Febo G, Paparo GF, Varoli O, Paganelli GM, Morselli Labate AM, Masci C, Zoccoli G, Monetti N, Barbara L. Helicobacter pylori infection and gastric function in patients with chronic idiopathic dyspepsia. Gastroenterology 1992;103:768–774. 18. Malfertheiner P, Bode G, Stanescu A, Ditschuneit H. Gastric metaplasia and Campylobacter pylori in duodenal ulcer disease: an ultrastructural analysis. Gastroenterol Clin Biol 1989;13: 71B–74B. 19. Wyatt JI, Rathbone BJ, Sobala GM, Shallcross T, Heatley RV, Axon ATR, Dixon MF. Gastric epithelium in the duodenum: its association with Helicobacter pylori and inflammation. J Clin Pathol 1990;43:981–986. 20. Madsen JE, Vetvik K, Aase S. Helicobacter pylori and chronic inflammation of the duodenum and stomach in duodenal ulcer patients treated with ranitidine, misoprostol or an acid neutralizing agent. Scand J Gastroenterol 1991;26:465–470.
/ 5e20$$0003
08-08-97 13:37:18
gasa
21. Satoh K, Kimura K, Yoshida Y, Kasano T, Kihara K, Taniguchi Y. Relationship between Helicobacter pylori colonization and acute inflammation of duodenal mucosa. Am J Gastroenterol 1993;88: 360–363. 22. O’Connor HJ, Newbold KN, Alexander-Williams J, Thompson H, Drumm J, Donovan IA. Effect of Roux-en-Y biliary diversion on Campylobacter pylori. Gastroenterology 1989;97:958–963. 23. Noach LA, Rolf TM, Bosma NB, Schwartz MP, Oosting J, Rauws EAJ, Tytgat GNJ. Gastric metaplasia and Helicobacter pylori infection. Gut 1993;34:1510–1514. 24. Wyatt JI, Rathbone BJ, Dixon MF, Heatley RV. Campylobacter pyloridis and acid induced gastric metaplasia in the pathogenesis of duodenitis. J Clin Pathol 1987;40:841–848. 25. Savarino V, Mela GS, Zentilin P, Mansi C, Vigneri S, Di Mario F, Malesci A, Calabro` A, Sossai P, Celle G. Circadian acidity pattern in gastric ulcers at different sites. Am J Gastroenteol 1995;90: 254–258. 26. Savarino V, Mela GS, Zentilin P, Mansi C, Mele MR, Vigneri S, Cutela P, Vassallo A, Dallorto E, Celle G. Evaluation of 24-hour gastric acidity in patients with hepatic cirrhosis. J Hepatol 1996; 25:152–157. 27. Hunt RH, Cederberg C, Dent J, Halter F, Howden C, Solly Marks IN, Rune S, Walt RP. Optimizing acid suppression for treatment of acid-related diseases. Dig Dis Sci 1995;40(Suppl):24S–49S. 28. Harris AW, Gummett PA, Walker MM, Misiewicz JJ, Baron JH. Relation between gastric acid output, Helicobacter pylori, and gastric metaplasia in the duodenal bulb. Gut 1996;39:513– 520. 29. Ormand JE, Talley NJ. Helicobacter pylori: controversies and an approach to management. Mayo Clin Proc 1990;65:414–426. 30. Khulusi S, Badve S, Patel P, Lloyd R, Marrero JM, Finlayson C, Mendall MA, Northfield TC. Pathogenesis of gastric metaplasia of the human duodenum: role of Helicobacter pylori, gastric acid and ulceration. Gastroenterology 1996;110:452–458. 31. Hogan DL, Rapier RC, Dreilinger A, Koss MA, Basuk PM, Weinstein WM, Nyberg LM, Isenberg JI. Duodenal bicarbonate secretion: eradication of Helicobacter pylori and duodenal structure and function in humans. Gastroenterology 1996;110:705– 716. Received September 6, 1996. Accepted April 3, 1997. Address requests for reprints to: Vincenzo Savarino, M.D., Dipartimento di Medicina Interna, Cattedra di Gastroenterologia, Universita` di Genova, Viale Benedetto XV, n. 6, 16132 Genova, Italy. Fax: (39) 10-3538956.
WBS-Gastro
ALIMENTARY TRACT 24-Hour Gastric pH and Extent of Duodenal Gastric Metaplasia in Helicobacter pylori–Positive Patients VINCENZO SAVARINO,* GIUSEPPE SANDRO MELA,* PATRIZIA ZENTILIN,* GABRIELLA LAPERTOSA,‡ PAOLA CEPPA,‡ SERGIO VIGNERI,§ MARIA RAFFAELLA MELE,* CARLO MANSI,* DANIELA TRACCI,* GIULIANA BISSO,* and GUIDO CELLE* *Dipartimento di Medicina Interna, Cattedra di Gastroenterologia, and ‡Istituto di Anatomia Patologica, Universita` di Genova, Genova; and § Istituto di Medicina Interna e Geriatria, Universita` di Palermo, Palermo, Italy
Background & Aims: Gastric metaplasia (GM) is essential to explain duodenal colonization by Helicobacter pylori. It seems to be acid induced but also occurs in H. pylori–positive patients with nonulcer dyspepsia (NUD), who are not acid hypersecretors. The aim of this study was to assess the circadian gastric acidity of 47 patients with duodenal ulcers (DUs) and 32 patients with NUD, both H. pylori positive, and its correlation to duodenal GM extent. Methods: H. pylori was detected by histology and CLOtest, and GM was diagnosed and graded on four bulb biopsy specimens. Each patient underwent 24-hour gastric pH-metry, and the relation between gastric pH and GM extent was assessed by factorial analysis. Results: Gastric pH was greater in patients with NUD than in patients with DU during 24 hours, night and daytime (P õ 0.03–0.005). Gastric pH differed significantly (P õ 0.0002) in relation to GM extent between the two populations, whereas no difference was found among the pH values of GM degrees. A significant increase in 24-hour gastric pH was associated with greater GM in patients with DU, whereas the opposite occurred in patients with NUD (P õ 0.007). Conclusions: The lower gastric acidity in patients with NUD than in patients with DU and the lack of correlation between gastric pH and the various GM degrees in the two H. pylori–positive populations suggest that gastric hyperacidity is not associated with duodenal GM.
I
t has been suggested that the presence of gastric metaplasia (GM) in the duodenum is necessary to allow Helicobacter pylori to colonize this region and to favor ulceration.1 The development of duodenal gastric-type epithelium is thought to be acquired mainly in response to acid hypersecretion.2 However, we have already shown that there is no difference in the pattern of circadian gastric acidity between patients with duodenal ulcer (DU) disease with and without GM in the duodenum.3 / 5e20$$0003
08-08-97 13:37:18
gasa
This finding seems to rule out a major role for gastric acid in the development of this histological change. Moreover, duodenal GM may occur frequently in H. pylori–positive patients with nonulcer dyspepsia (NUD), which is rarely associated with acid hypersecretion.4 The different patterns of gastric acidity in patients with DU disease and NUD compared with that of healthy subjects5,6 seem to reduce further the relevance of acid in determining the genesis of this histological alteration. Probably the best way to define the role of acid in this field is the direct correlation of gastric pH with the extent of gastric-type epithelium in the duodenum. Therefore, we performed this study to assess the circadian pattern of gastric acidity in H. pylori–positive patients with DU disease and NUD and to correlate this variable with the extent of GM in the duodenum of these two populations. The hypothesis tested was that gastric acidity should increase in parallel with the extent of duodenal GM if acid is the most important factor in inducing this histological change.
Patients and Methods Patients Forty-seven patients with DU disease (35 men and 12 women; mean age, 52 { 12 years) and thirty-two patients with NUD (23 men and 9 women; mean age, 49 { 14 years) were enrolled into the study. Thirty-three patients in the first group and 21 patients in the second group were smokers. The ulcer crater was ascertained endoscopically, and the diagnosis of NUD was based on the presence of persistent or recurrent dyspeptic symptoms without any abnormality of the upper gastrointestinal tract at endoscopy or systemic and metabolic diseases. Dyspeptic patients complained of unexplained epigastric pain and/or abdominal discomfort (postprandial fullness, Abbreviations used in this paper: DU, duodenal ulcer; GM, gastric metaplasia; NUD, nonulcer dyspepsia. q 1997 by the American Gastroenterological Association 0016-5085/97/$3.00
WBS-Gastro
742 SAVARINO ET AL.
GASTROENTEROLOGY Vol. 113, No. 3
early satiety, nausea, and vomiting) centered in the upper abdomen7 of at least 2 months’ duration. Patients who were administered nonsteroidal anti-inflammatory drugs or who were undergoing long-term treatment with antisecretory compounds were excluded from the investigation. None of the patients was treated with H2-receptor antagonists or proton pump inhibitors during the 2 weeks preceding the inclusion into the study. No patient had undergone previous gastric surgery. All patients were H. pylori positive and had duodenal GM defined as the occurrence of foci of gastric epithelial cells containing apical periodic acid–Schiff–positive neutral mucine together with the absence of a brush border. All patients gave their informed consent to the study, which was also approved by our local ethical committee.
Methods The diagnosis of H. pylori infection was based on histology and rapid urease test. These procedures were performed on antral and angular biopsy specimens8 using the well-established Wartin–Starry and modified Giemsa methods for the former and CLO test (Delta West Ltd., Bentley, Western Australia, Australia) for the latter. GM was diagnosed from four duodenal biopsy specimens that were taken from the following locations: two from the ulcer edge and two from the opposite wall in patients with an ulcer crater, and from the roof, the floor, and the anterior and posterior wall of the bulb in patients with NUD. The presence of H. pylori was also sought for in these bulb biopsy specimens. The extent of GM was estimated independently by two histopathologists as õ5%, 5%–20%, 20%–50%, or ú50% of the total epithelial surface present by averaging the scores of the four specimens obtained from the same patient. The interobserver agreement of the two investigators over the grading of the extent of GM was high (k Å 0.83). Each patient underwent 24-hour continuous intragastric pH-metry; in the DU disease group, this was performed within 2 days of the endoscopic diagnosis of the crater. The examination was performed with two closely adjacent pH electrodes (Ingold 440-M 3 and M 1.5; Medical Instruments Corp., Solothurn, Switzerland) connected to a two-channel solid-state data logger (Digitrapper ED; Synectics AB, Stockholm, Sweden) and positioned fluoroscopically approximately 10 cm below the cardia in the gastric corpus. The simultaneous recording with two paired pH electrodes has been shown to improve the accuracy of the test by removing noise, interference, and artifacts.9 The electrodes were calibrated using commercial buffer solutions at pH 7.0 and pH 1.0 in accordance with an already validated procedure,10 and gastric pH was measured with a 6-second sampling rate. Patients were fully ambulant, and smoking was permitted because it does not affect the gastric pH pattern.11 Standardized meals were given at the preset times of 6 PM, 8 AM, and noon.12 Extra food or water and alcoholic or carbonated beverages were forbidden during the test day.
Data Processing and Statistics Experimental data obtained from continuous gastric pH-metry were downloaded into a calculator (Proliant; Com-
/ 5e20$$0003
08-08-97 13:37:18
gasa
Figure 1. Twenty-four–hour mean pH profiles of H. pylori –positive patients with DU disease ( ) and NUD ( ).
paq, Bo¨bingen, Germany) using service routines. The pH readings of the two channel tracings of each patient were averaged every minute so that 1440 data points per 24 hours were obtained.9 Acidity data were expressed as mean pH13 for three different time intervals: 24 hours (5 PM to 4:59 PM), night (8 PM to 7:59 AM), and daytime (8 AM to 7:59 PM). These findings were compared by means of two-way analysis of variance, and P values were corrected for multiple testing. The relationship between 24-hour mean pH and the extent of duodenal GM was assessed by means of two-factor nonparametric factorial analysis for unequal cell size. x Test was used to compare the percentage of GM between ulcer edges and other bulb areas as well as the percentage of H. pylori detection in the bulb of patients with DU disease and NUD. The significance level was set at P value of õ0.05.
Results In 2 patients, gastric pH-metry was repeated because of prolonged discrepancy (more than 1 hour longer than the circadian cycle) between the pH readings of the two closely adjacent minielectrodes. H. pylori was detected in the areas of duodenal GM of 12 of 47 patients with DU disease (25%) and 7 of 32 patients with NUD (21%) (P Å NS). GM was observed at the edge of the ulcer in all of the 47 patients with DU disease (100%) and in the areas distant from the ulcer crater in 39 of them (82%). The difference was significant (P õ 0.05). The 24-hour gastric mean pH profiles of the two populations we examined are reported in Figure 1. It is clearly visible that the pH curve pertaining to patients with NUD runs above that of patients with DU disease during almost the whole circadian period. Although the difference can be easily observed after each postprandial period, it is particularly evident during the night. The mean pH values of patients with NUD are significantly greater than those of patients with DU for each WBS-Gastro
September 1997
GASTRIC PH AND BULB GASTRIC METAPLASIA 743
similar to that of patients with DU disease. This result, however, was obtained with a stimulus comparable to food-stimulated acid secretion and is unable to represent the actual influence of H. pylori infection on the natural secretion of acid by the stomach during the whole circadian cycle.16 In NUD, on the other hand, it has been shown that there is no difference in terms of basal and maximal acid output between H. pylori–positive and –negative patients, as well as between these two subgroups and healthy subjects.17 This further confirms that H. pylori infection does not affect the output of acid in this disorder. GM is believed to be an essential prerequisite for the colonization of duodenal mucosa by H. pylori and, then, for explaining how a germ that is located electively within the stomach influences the pathogenesis of DU more than gastric ulcer.1 This gastric-type epithelium is found more commonly at the edge of the ulcer crater than in the other areas of the duodenal cap,18,19 as we confirmed in our study. H. pylori, however, was detected in GM areas of only 25% and 21% of our patients with DU disease and NUD, respectively. Other investigators20,21 have observed this disappointing low percentage of positivity compared with the success of isolating the germ in areas of gastritis. Probably the presence of bile in the duodenum makes it a hostile environment for H. pylori survival.22 Duodenal GM seems to be a protective mechanism against an injury mainly secondary to increased secretion of gastric acid.2 However, GM has been shown both in healthy subjects and in patients with NUD,23,24 who are not acid hypersecretors. Our study is the first to correlate the pattern of 24hour gastric acidity to the extent of GM in the duodenum of H. pylori–positive patients. The reliability of our technique in differentiating the level of gastric acidity in various clinical conditions associated with altered gastric acid secretion has been shown in several physiological studies.5,25,26 It is well known that intraluminal pH monitoring ignores the volume of acid secretion, but gastric acidity may reflect more than acid output the pathophysiological conditions present at the site of peptic lesions27 and then the factors inducing the histological alterations
Table 1. Mean pH Values and SD During Three Different Time Periods in the Two Populations of Patients With DU Disease and NUD With H. pylori Positivity Time period
DU disease
NUD
5 PM to 4:59 PM 8 PM to 7:59 AM 8 AM to 7:59 PM
a
2.1 { 1.0 (1.8–2.4) 2.1 { 1.2 (1.7–2.5) 2.2 { 0.8 (1.9–2.5)
1.6 { 0.4 (1.5–1.7) 1.5 { 0.6b (1.3–1.7) 1.7 { 0.3a (1.6–1.8)
NOTE. Values are expressed as means { SD. Values in parentheses represent 95% confidence intervals. a P õ 0.005. b P õ 0.03.
time interval we analyzed (24 hours, P õ 0.005; night, P õ 0.03; and daytime, P õ 0.005), as shown in Table 1. The relationship between the 24-hour mean gastric pH and the extent of duodenal GM is reported in Table 2. There is a remarkable difference in gastric pH values related to the extent of gastric-type epithelium in the duodenum between the two populations (P õ 0.0002), whereas pH does not differ among the various degrees of GM within the same populations. There is a highly significant interaction between the two populations and the percentages of bulb GM (P õ 0.007). This means that a significant increase in 24-hour mean gastric pH is associated with greater GM in patients with DU disease, whereas the opposite occurs in patients with NUD.
Discussion Our study shows that circadian gastric acidity is significantly lower in H. pylori–positive patients with NUD than in H. pylori–positive patients with DU disease. Despite the claim that H. pylori infection significantly affects gastric physiology in terms of increase in fasting and especially postprandial gastrinemia and then in parietal cell production of acid,14 our findings show that the gastric acidity patterns of patients with DU disease and NUD, both infected with H. pylori, remain greatly different. Recently, the adoption of gastrin-releasing peptide as a stimulant has led El-Omar et al.15 to conclude that a substantial proportion of H. pylori– positive patients with NUD has an increased acid output
Table 2. Relationship Between 24-Hour Mean pH and the Extent of Duodenal GM in the Two Populations With H. pylori Infection GM (% of total epithelia) Populations
õ5%
5%–20%
20%–50%
ú50%
DU disease NUD
1.4 { 0.4 (n Å 19) 2.8 { 0.8 (n Å 10)
1.6 { 0.4 (n Å 14) 2.5 { 0.7 (n Å 9)
1.7 { 0.3 (n Å 8) 1.6 { 0.2 (n Å 5)
2.0 { 0.5 (n Å 6) 1.9 { 0.7 (n Å 8)
NOTE. Values are expressed as means { SD.
/ 5e20$$0003
08-08-97 13:37:18
gasa
WBS-Gastro
744 SAVARINO ET AL.
GASTROENTEROLOGY Vol. 113, No. 3
implicated in the formation of the ulcer crater. The main result of our investigation is that 24-hour gastric pH does not differ significantly in relation to the various degrees of GM within the two populations of patients with DU disease and NUD. Even more, pH values progressively increase as we move from the lowest to the highest degrees of histological alteration in DU disease, thus contradicting our initial hypothesis that gastric acidity should increase in parallel with the degree of gastric metaplasia if acid is so relevant to GM development. In patients with NUD, on the contrary, pH values tend to decrease with the highest extent of GM, but the 24-hour levels of acidity found at the greatest degrees of histological change are still comparable to those of normal subjects.3,25,26 This rules out gastric acid hypersecretion as a central factor in determining the extent of duodenal GM also in NUD. Our results are at variance with those recently published by Harris et al.,28 who showed that there is a correlation between basal and stimulated acid output and the degree of GM. However, they found this correlation in patients with DU disease and not in normal subjects, both H. pylori positive and H. pylori negative. If the role of acid is crucial in the pathogenesis of duodenal GM, it is very surprising that this factor correlates with GM only in one population and not in the other. Harris et al. do not try to explain this evident discrepancy, but the low prevalence of H. pylori infection29 and the high variability of GM23 in healthy subjects probably make this population unsuitable for assessing a correlation between acid output and GM. For this reason, we preferred to evaluate the relationship between these two variables in two populations with a high prevalence of both H. pylori infection and duodenal GM.23 An additional weakness of the study by Harris et al. is that GM does not change 6 months after the significant reduction of acid output secondary to H. pylori eradication. Khulusi et al.30 showed that GM decreases by ú40% after 6 months of acid suppression obtained with a continuous antisecretory treatment of 40 mg/die omeprazole. Moreover, Harris et al. exclude that H. pylori eradication produces a decrease in the extent of GM; once again, this is in contrast with the findings by Khulusi et al. who showed a 42% GM reduction as the exclusive result of the disappearance of the germ. In our opinion, these conflicting data do not contribute to the clarification of the complicated matter of the relationship between acid and GM in H. pylori– positive patients. Our findings related to hydrogen ion activity instead of acid output suggest a negative correlation between these two variables. Other factors may be involved in the production of GM, such as the germ-induced inflammation30 or the / 5e20$$0003
08-08-97 13:37:18
gasa
impaired neutralization of normal acid load to the duodenum. In fact, the secretion of duodenal bicarbonates needed to neutralize the acid entering into the duodenum has been shown to be impaired in H. pylori–infected patients and, more importantly, is restored after bacterial eradication.31 Thus, the development of duodenal GM seems to be a nonspecific response to mucosal injury caused by multiple factors and not exclusively to acid hypersecretion. Hogan et al.31 did not find any difference in the prevalence of duodenal GM between healthy subjects and patients with DU disease (79% vs. 75%), as further confirmation that this histological alteration can be observed in groups with completely divergent gastric acidity patterns.5 Also, the occurrence of duodenal gastric-type epithelium in several different populations, including healthy subjects and patients with NUD and DU disease, casts many doubts on its importance in DU pathogenesis. In summary, our study shows that the circadian gastric acidity of patients with NUD is significantly lower than that of patients with DU disease, although both groups were infected with H. pylori. There is no correlation between 24-hour mean pH values and the various degrees of duodenal GM within the two populations. Furthermore, gastric acidity does not increase in parallel with the degree of histological change in DU disease, whereas it tends to increase in patients with NUD. These findings suggest that gastric hyperacidity is not a key factor in patients with duodenal GM.
References 1. Walker MM, Dixon MF. Gastric metaplasia: its role in duodenal ulceration. Aliment Pharmacol Ther 1996;10(Suppl 1):119– 128. 2. Carrick J, Lee A, Hazell S, Ralston M, Daskalopoulos G. Campylobacter pylori, duodenal ulcer, and gastric metaplasia: possible role of functional heterotopic tissue in ulcerogenesis. Gut 1989; 30:790–797. 3. Savarino V, Mela GS, Zentilin P, Mele MR, Lapertosa G, Patetta R, Dallorto E, Vassallo A, Mansi C, Vigneri S, Celle G. Circadian gastric acidity in Helicobacter pylori positive ulcer patients with and without gastric metaplasia in the duodenum. Gut 1996;39: 508–512. 4. Collen MJ. Gastric analysis (basal acid output) in non-ulcer dyspepsia. Dig Dis Sci 1990;35:540–541. 5. Savarino V, Mela GS, Scalabrini P, Sumberaz A, Fera G, Celle G. 24-hour study of intragastric acidity in duodenal ulcer patients and normal subjects using continuous intraluminal pH-metry. Dig Dis Sci 1988;33:1077–1080. 6. Savarino V, Mela GS, Celle G, Vigneri S. Acid and gastric metaplasia in the duodenum. Gut 1994;35:1151–1152. 7. Talley NJ, Colin-Jones D, Koch KL, Koch M, Nyren O, Stanghellini V. Functional dyspepsia: a classification with guidelines for diagnosis and management. Gastroenterol Int 1991;4:145–160. 8. Genta RM, Graham DY. Comparison of biopsy sites for the histopathologic diagnosis of Helicobacter pylori: a topographic study of H. pylori density and distribution. Gastrointest Endosc 1994; 40:342–345.
WBS-Gastro
September 1997
GASTRIC PH AND BULB GASTRIC METAPLASIA 745
9. Mela GS, Savarino V, Malesci A, Di Mario F, Sossai P, Vigneri S, Zambotti A. New method for improving accuracy of 24-hour continuous intragastric pH-metry. Reflections on physiological and pharmacological studies. Dig Dis Sci 1994;39:1416–1424. 10. Savarino V, Mela GS, Zentilin P, Magnolia MR, Scalabrini P, Valle F, Moretti M, Bonifacino G, Celle G. Gastric aspiration versus antimony and glass pH electrodes. A simultaneous comparative in vivo study. Scand J Gastroenterol 1989;24:434–439. 11. Bauerfeind P, Cilluffo T, Fimmel CJ, Emde C, Von Ritter C, Kohler W, Gugler R, Gasser T, Blum AL. Does smoking interfere with the effect of histamine H2-receptor antagonists on intragastric acidity in man? Gut 1987;28:549–556. 12. Savarino V, Mela GS, Scalabrini P, Magnolia MR, Di Timoteo E, Percario G, Celle G. Continuous 24-hour intragastric pH monitoring: focus on reproducibility in duodenal ulcer patients. Gastroenterol Clin Biol 1986;10:826–830. 13. Mela GS, Savarino V, Sumberaz A, Bonifacino G, Zentilin P, Villa G, Caputo E. Continuous acidity monitoring in the study of gastric antisecretory drugs: pH or antilog transformation of pH, mean or median? Am J Gastroenterol 1990;85:1105–1108. 14. Peterson WL. Gastrin and acid in relation to Helicobacter pylori. Aliment Pharmacol Ther 1996;10(Suppl 1):97–102. 15. El-Omar E, Penman I, Ardill JES, McColl KEL. A substantial proportion of non-ulcer dyspepsia patients have the same abnormality of acid secretion as duodenal ulcer patients. Gut 1995;36:534– 538. 16. Pounder RE. Helicobacter pylori and gastroduodenal secretory function. Gastroenterology 1996;110:947–950. 17. Tucci A, Corinaldesi R, Stanghellini V, Tosetti C, Di Febo G, Paparo GF, Varoli O, Paganelli GM, Morselli Labate AM, Masci C, Zoccoli G, Monetti N, Barbara L. Helicobacter pylori infection and gastric function in patients with chronic idiopathic dyspepsia. Gastroenterology 1992;103:768–774. 18. Malfertheiner P, Bode G, Stanescu A, Ditschuneit H. Gastric metaplasia and Campylobacter pylori in duodenal ulcer disease: an ultrastructural analysis. Gastroenterol Clin Biol 1989;13: 71B–74B. 19. Wyatt JI, Rathbone BJ, Sobala GM, Shallcross T, Heatley RV, Axon ATR, Dixon MF. Gastric epithelium in the duodenum: its association with Helicobacter pylori and inflammation. J Clin Pathol 1990;43:981–986. 20. Madsen JE, Vetvik K, Aase S. Helicobacter pylori and chronic inflammation of the duodenum and stomach in duodenal ulcer patients treated with ranitidine, misoprostol or an acid neutralizing agent. Scand J Gastroenterol 1991;26:465–470.
/ 5e20$$0003
08-08-97 13:37:18
gasa
21. Satoh K, Kimura K, Yoshida Y, Kasano T, Kihara K, Taniguchi Y. Relationship between Helicobacter pylori colonization and acute inflammation of duodenal mucosa. Am J Gastroenterol 1993;88: 360–363. 22. O’Connor HJ, Newbold KN, Alexander-Williams J, Thompson H, Drumm J, Donovan IA. Effect of Roux-en-Y biliary diversion on Campylobacter pylori. Gastroenterology 1989;97:958–963. 23. Noach LA, Rolf TM, Bosma NB, Schwartz MP, Oosting J, Rauws EAJ, Tytgat GNJ. Gastric metaplasia and Helicobacter pylori infection. Gut 1993;34:1510–1514. 24. Wyatt JI, Rathbone BJ, Dixon MF, Heatley RV. Campylobacter pyloridis and acid induced gastric metaplasia in the pathogenesis of duodenitis. J Clin Pathol 1987;40:841–848. 25. Savarino V, Mela GS, Zentilin P, Mansi C, Vigneri S, Di Mario F, Malesci A, Calabro` A, Sossai P, Celle G. Circadian acidity pattern in gastric ulcers at different sites. Am J Gastroenteol 1995;90: 254–258. 26. Savarino V, Mela GS, Zentilin P, Mansi C, Mele MR, Vigneri S, Cutela P, Vassallo A, Dallorto E, Celle G. Evaluation of 24-hour gastric acidity in patients with hepatic cirrhosis. J Hepatol 1996; 25:152–157. 27. Hunt RH, Cederberg C, Dent J, Halter F, Howden C, Solly Marks IN, Rune S, Walt RP. Optimizing acid suppression for treatment of acid-related diseases. Dig Dis Sci 1995;40(Suppl):24S–49S. 28. Harris AW, Gummett PA, Walker MM, Misiewicz JJ, Baron JH. Relation between gastric acid output, Helicobacter pylori, and gastric metaplasia in the duodenal bulb. Gut 1996;39:513– 520. 29. Ormand JE, Talley NJ. Helicobacter pylori: controversies and an approach to management. Mayo Clin Proc 1990;65:414–426. 30. Khulusi S, Badve S, Patel P, Lloyd R, Marrero JM, Finlayson C, Mendall MA, Northfield TC. Pathogenesis of gastric metaplasia of the human duodenum: role of Helicobacter pylori, gastric acid and ulceration. Gastroenterology 1996;110:452–458. 31. Hogan DL, Rapier RC, Dreilinger A, Koss MA, Basuk PM, Weinstein WM, Nyberg LM, Isenberg JI. Duodenal bicarbonate secretion: eradication of Helicobacter pylori and duodenal structure and function in humans. Gastroenterology 1996;110:705– 716. Received September 6, 1996. Accepted April 3, 1997. Address requests for reprints to: Vincenzo Savarino, M.D., Dipartimento di Medicina Interna, Cattedra di Gastroenterologia, Universita` di Genova, Viale Benedetto XV, n. 6, 16132 Genova, Italy. Fax: (39) 10-3538956.
WBS-Gastro