The Effect of Ammonia on Omeprazole-Induced Reduction of Gastric Acidity in Subjects With Helicobacter pylori Infection

Размещено на http://www.allbest.ru/

Размещено на http://www.allbest.ru/

The Effect of Ammonia on Omeprazole-Induced Reduction of Gastric Acidity in Subjects With Helicobacter pylori Infection

OBJECTIVE: Omeprazole produces a higher intragastric pH during Helicobacter pylori (H. pylori) infection than after cure. We tested the hypothesis that this difference is due to the production of ammonia by H. pylori.

METHODS: Gastric acidity and acid output (AO) were measured overnight in 12 subjects, with and without omeprazole, before and 1 and 6 months after cure of H. pylori infection. Gastric ammonia ([NH3]), total bile acid ([TBA]) and protein concentrations and plasma omeprazole levels were measured.

RESULTS: During omeprazole, median AO were 0.0 mmol/h before, 0.86 mmol/h (p = 0.003 vs before cure) at 1 month, and 0.34 mmol/h (p = 0.02) at 6 months after cure; median NH3 output was 0.17 mmol/h before, 0.03 mmol/h (p = 0.002) at 1 month, and 0.02 mmol/h (p = 0.005) at 6 months after cure. AO and NH3 output were similar 1 and 6 months after cure. When corrected for [NH3], AO and gastric pH curves were similar before and after cure. Omeprazole plasma levels increased after cure and gastric [TBA] were unchanged.

CONCLUSIONS: The higher pH observed before cure of H. pylori during omeprazole administration is attributable, in large part, to ammonia production. Other acid-neutralizing substances and changes in acid secretion may also be important, but duodenogastric reflux and omeprazole pharma-cokinetics are not involved. (Am J Gastroenterol 2000;95: 947-955. © 2000 by Am. Coll. of Gastroenterology)

INTRODUCTION

We have reported previously that Helicobacter pylori (H. pylori) infection augments the rise in intragastric pH produced by omeprazole (1-3); the mechanism whereby H. pylori increases the apparent efficacy of omeprazole is, however, unknown. H. pylori is thought to survive in the acidic gastric milieu because it produces ammonia, which can neutralize surrounding acid and increase the local pH. The effect of ammonia in increasing gastric pH should be most evident during potent antisecretory therapy when the concentration of hydrogen ions is lower (2), but it is not known whether H. pylori produces sufficient ammonia to cause the augmented omeprazole-induced increase in gastric pH. Other than an effect mediated by ammonia, H. pylori infection might also act by increasing the concentration of other acid-neutralizing substances, such as bile acids or proteins, by modifying antroduodenal motility to promote duodenogastric reflux or by altering the bioavailability of omeprazole.

To examine the mechanisms responsible for the interaction between H. pylori and omeprazole, we studied the changes in nocturnal gastric acidity and acid output produced by omeprazole in 12 subjects with H. pylori infection, but without peptic ulcer disease, both before and after cure of their infection. The primary aim of the study was to test the hypothesis that ammonia produced by H. pylori is sufficient to account for the greater increase in gastric pH produced by omeprazole (20 mg daily) before eradication, compared with the omeprazole-induced increase observed after eradication. Acid output was measured by titrating gastric aspirates to pH 7, and ammonia output was also measured in gastric aspirates; the total acid output was then calculated by correcting measured acid output on the basis of ammonia output. Intragastric acidity was recorded continuously throughout the study periods, using an intragastric pH electrode, and the measured gastric acidity was also corrected on the basis of ammonia output.

To assess the overall acid-neutralizing capacity of gastric juice, both before and after cure of the infection, gastric aspirates were titrated to endpoints of pH 7 and pH 11; the latter endpoint was chosen to quantify the acid-neutralization effects of all bases or buffers active in the pH range from 7 to 11. The increased amount of sodium hydroxide needed to titrate the samples from pH 7 to pH 11 was considered to be a measure of the acid (Д[H+]) buffered by substances with a pKa between 7 and 11 and, hence, an indirect quantification of the buffers or acid neutralizers themselves. Concentrations of ammonia, total bile acids, and proteins in gastric juice aspirates were also assayed, as were plasma omeprazole levels.

MATERIALS AND METHODS

Subjects

Twelve H. pylori -positive subjects (seven men, five women, aged 22-45 yr) without peptic ulceration or gastroesophageal reflux disease participated in the study. Upper gastrointestinal lesions, including reflux esophagitis and peptic ulcer disease, were excluded endoscopically. Esophageal appearances relevant to reflux disease were graded using the “MUSE” classification (4); in brief, lesions of Metaplasia, Ulceration, Stricture, and Erosions were graded independently from 0 (absent) to 2 (severe). Subjects were healthy and had no history of gastrointestinal disease. At enrollment, no subject was taking any medication except oral contraceptives or paracetamol (acetaminophen). All subjects gave written, informed consent, and the study was conducted according to the Declaration of Helsinki. The protocol was approved by the Local Ethical Review Committee.

Study Protocol

H. pylori status was determined using a 13C-urea breath test (5-7) and confirmed by a specific enzyme-linked immu-nosorbant assay (EIA, Roche, Reinach, Switzerland). Upper gastrointestinal endoscopy (Olympus Q20, Olympus, Volketswil, Switzerland) was performed only if both tests were positive. CagA antibody status was determined by ELISA (8). On study days, subjects drank a standardized liquid meal (Ensure, Abbot, Cham, Switzerland) at 18:00 after which, they fasted. All subjects were studied on 12 separate occasions; an overnight, 8-h gastric pH-metry and an overnight 8-h acid output study were conducted, in random order on consecutive nights, without omeprazole administration and were repeated, again in random order, on days 7 and 8 of omeprazole administration. The pH and acid output studies were repeated before and after omeprazole administration at 1 month and at 6 months after cure of H. pylori infection. An upper gastrointestinal endoscopy was performed at the end of the study.

In addition to the nocturnal recordings, subjects came to the laboratory at 09:00 on days 1 and 5 of omeprazole administration, before and 1 month and 6 months after cure of H. pylori. Six consecutive blood samples were obtained hourly to determine omeprazole plasma levels.

Omeprazole Administration

Omeprazole 20 mg was taken, every morning for 1 wk, as granules dispensed in hard gelatin capsules, on weeks 1, 5, and 25 of the study (Astra Hassle AB, Molndal, Sweden).

Antimicrobial Therapy for H. pylori Infection

Subjects received amoxycillin 1 g b.i.d., clarithromycin 500 mg b.i.d., and omeprazole 40 mg b.i.d. for 14 days.

Gastric pH-Metry Protocol

Gastric pH was measured with an assembly that consisted of two glass pH-electrodes (MIC, Ingold, Urdorf, Switzerland) attached 7 cm apart (1). The assembly was inserted trans-nasally such that the proximal electrode was located in the corpus, 5 cm below the cardia. Recordings started at 22:00 and finished at 06:00 the following morning. The pH-electrodes were connected to a data-logger (Gastroscan, MIC, Solothurn, Switzerland) and, at the end of each recording, data were transferred to a personal computer for analysis (MIC, Solothurn, Switzerland).

Acid Output Protocol

A combined nasogastric tube (12 French) for aspiration and gastric perfusion was positioned in the most dependent part of the stomach (antrum) and its position checked by fluo-roscopy. Gastric contents were aspirated continuously from 22:00 to 6:00 using an aspiration pump (Medela AG, Baar, Switzerland), and aliquots were pooled every 15 min in glass tubes. To correct for pyloric losses, the stomach was perfused continuously at a rate of 150 ml/h with a solution containing phenol red (9 mg/L) (9).

Measurements in Gastric Aspirates

Two samples from each aliquot were assayed immediately for ammonia concentrations and titrated to a pH endpoint of 7 with 0.01 mol/L sodium hydroxide using an automatic titration system (Radiometer 85, Radiometer, Copenhagen, Denmark). Remaining samples were stored at --70°C for a maximum of 2 months, during which time they were assayed for TBA and total proteins, and titrated to a pH endpoint of 11. Additional samples, stored at --70°C for 2 months and titrated to pH 7, showed a good correlation with titration of samples performed immediately after experiments (r2 = 0.98), indicating that there had been no deterioration of samples during storage.

Phenol red concentrations were determined spectropho-tometrically at 560 nm in gastric aspirates after filtration and alkalinization with sodium hydroxide (0.4 ml, 2.5 mol/L). Corrected total acid output was calculated using the phenol red recovery (9), which was usually >90%.

Neutralizing Substances

Ammonia concentrations were determined using an ammonia-sensitive electrode (Orion, Boston, MA) connected to a pH-meter (Orion 920, Boston, MA) which was calibrated before each experiment using standard NH4Cl solutions at concentrations of 10, 1.0, 0.1 and 0.01 mmol/L (Orion, Boston, USA) (10). The gastric samples were alkalinized to pH 11 using 0.3 ml of sodium hydroxide (5 mol/L), and the ammonia concentration in the sample was recorded immediately afterward.

To quantify the effect on gastric pH of hydrogen ions that had been neutralized by ammonia, an “adjusted” pH value was calculated according to the following formula: Adjusted pH = -log10((0.8*[NH3]) + 10Tobserved ph). This calculation assumed that all ammonia in the samples was in the form of NH4 and that the hydrogen activity coefficient of the gastric juice was 0.8 (11). The adjusted pH value represents the predicted gastric pH value in the absence of any ammonia.

Table 1. Median (95% CI) Nocturnal pH Values, in the Corpus and in the Antrum, Spontaneously and During Omeprazole Administration

* p < 0.001 vs before cure † p < 0.05 vs before cure

Acid output corrected for the presence of ammonia was calculated as measured acid output plus ammonia output, based on measured ammonia concentrations and volumes of gastric aspirates.

Total bile acid (TBA) concentrations in gastric juice were measured by an enzymatic method (12); in this context, TBA was used as a marker of duodenogastric reflux. The limit of detection for TBA was 2 mmol/L. Total protein concentrations were measured in gastric aspirates during omeprazole administration, before cure and 1 month after cure of H. pylori using the QuanTtest (Quantimetrix, Hawthorne, CA) and a centrifugal analyzer (Cobas Bio, Roche, Basle, Switzerland). This assay, based on the interaction of proteins with Coomassie Blue dye, does not show any interference with ammonia; the limit of detection of the method is 10 mg/L.

Endoscopic Biopsies and Histology

Three biopsies were obtained from the corpus and three from the antrum of all subjects. One biopsy from each site was used to perform a rapid urease test (Jatrox®-Hp-Test, Rohn Pharm, Weiterstadt, Germany). All remaining biopsies were fixed in 10% buffered formalin, stained (hema-toxylin and eosin; Warthin-Starry technique), and examined by the same pathologist (M.S.). Gastritis severity was classified in both the antrum and corpus according to a modified Sydney system (13-14) with reference to H. pylori density, lymphocyte and plasma cell infiltration, neutrophil infiltration, replacement of foveolar epithelium by regenerative epithelium, and mucus depletion. All parameters were graded on a 5-point scale, and their sum yielded the gastritis score for the antrum and for the corpus. In addition, atrophy and intestinal metaplasia were sought in all biopsies.

Data Analysis and Statistical Evaluation

The change in hydrogen ion concentration (Д[H+]) was calculated as: ([H+] titrated to pH 11) -- ([H+] titrated to pH 7). To evaluate the strength of association between A[H+] and ammonia, proteins and bile acids, Spearman's rank correlation test (rs) was performed. Median pH values for the entire nocturnal recording were calculated, as were median hourly acid output data for each subject. All data are presented as group median values with 95% confidence intervals (95% CI), and all statistical testing was conducted using the Wilcoxon rank test for paired samples. Linear correlation-regression (r2) was used to correlate TBA concentrations with the pH of the gastric aspirates.

In the event that subjects dropped out during follow-up, corresponding data were excluded, and statistical testing was performed on paired data only.

RESULTS

Subjects

All 12 subjects completed the study up to 1 month after cure of the H. pylori infection; 10 subjects were available for review at the 6-month follow-up visit. One subject developed endoscopic esophagitis 6 months after cure of the infection (M0U0S0E2) (4). No serious adverse events were recorded during the study.

Six subjects tested positive for CagA antibodies. Subjects with a positive CagA status did not differ with respect to gastric pH values, acid output values, or any of the other parameters measured in the study from subjects with negative CagA status (data not shown).

Nocturnal Intragastric pH

Cure of the H. pylori infection was associated with lower spontaneous pH values (p < 0.05) in the antrum and with lower omeprazole-induced pH values (p < 0.01) in the antrum and corpus at the time of the 1-month and 6-month follow-up visits (Table 1).

Acid Output: Titration to Endpoint of pH 7

The median spontaneous acid outputs were 3.35 mmol/h (95% CI 2.02-4.68) before cure, 2.86 mmol/h (1.6-4.92) 1 month after cure, and 2.44 mmol/h (0.99-3.80) 6 months after cure of the H. pylori infection (p > 0.05 vs before cure, Fig. 1). The hourly acid outputs between 22:00 and 06:00 were also similar before cure and 1 month and 6 months after cure of the infection (Fig. 2).

During omeprazole administration, the median acid outputs were 0.0 mmol/h (0.0-0.08) before cure, 0.86 mmol/h (0.05-2.72) 1 month after cure (p = 0.009 vs before cure), and 0.34 mmol/h (0.07-2.84) 6 months after cure of H. pylori infection (p = 0.02 vs before cure, Fig. 1). In addition, the hourly acid outputs from 22:00 to 06:00 were higher 1 month and 6 months after cure than they had been before cure (p < 0.05 vs before cure, Fig. 2).

Figure 1. Nocturnal acid output (left panel), ammonia output (middle panel), and corrected acid output (right panel) for individual subjects in the absence of omeprazole (spontaneous acid secretion: upper panel) and in the presence of omeprazole (lower panel); horizontal lines indicate group median values under each condition.

Spontaneous acid outputs were similar before and after cure of H. pylori infection (upper left graph) but, during omeprazole administration, acid output was increased 1 month and 6 months after cure of H. pylori (lower left graph). Ammonia outputs fell significantly after cure ofH. pylori, although baseline outputs before cure were higher under spontaneous conditions (upper middle graph) than in the presence of omeprazole (lower middle graph). Corrected acid outputs, adjusted for ammonia production, showed no difference before cure and 1 month and 6 months after cure of H. pylori infection under spontaneous conditions (upper right graph) or in the presence of omeprazole (lower right graph). Open squares indicate the two subjects for whom 6-month data were not available. *p < 0.01 versus before cure.

Acid Output: Titration to Endpoint ofpH 11

The median spontaneous nocturnal acid outputs were 5.81 mmol/h (3.38-8.23) before cure, 4.20 mmol/h (2.09-6.66) 1 month after cure, and 3.87 mmol (2.04-9.09), 6 months after cure of H. pylori infection (Fig. 1). Thus, the median volumes of sodium hydroxide needed to titrate the samples from pH 7 to pH 11 (reflecting the amount of the acid (A[H+]) buffered by substances with a pKa between 7 and 11) under spontaneous conditions were 2.46 mmol before cure, 1.34 mmol 1 month after cure, and 1.43 mmol 6 months after cure.

During omeprazole administration, the median nocturnal acid outputs were 1.57 mmol/h (1.20-2.16) before cure, 1.83 mmol/h (0.90-3.94) 1 month after cure, and 1.09 mmol/h (0.86-5.11) 6 months after cure of the H. pylori infection. In the presence of omeprazole, the volumes of sodium hydroxide needed to titrate the samples from pH 7 to pH 11 were 1.57 mmol before cure, 0.97 mmol 1 month after cure, and 0.75 mmol 6 months after cure.

Ammonia Concentrations

The spontaneous ammonia concentrations in gastric samples were 1.53 mmol/L (1.12-1.92) before cure, 0.14 mmol/L (0.08-0.20) 1 month after cure (p = 0.002 vs before cure), and 0.09 (0.04-0.15) 6 months after cure of H. pylori infection (p = 0.002 vs before cure). During omeprazole administration, the ammonia concentrations were 0.88 mmol (0.61-1.17) before cure, 0.16 (0.06-0.31) 1 month after cure (p = 0.002 vs before cure), and 0.09 (0.04-0.17) 6 months after cure of the infection (p = 0.002 vs before cure). Before cure, spontaneous ammonia concentrations were higher than during omeprazole administration (p = 0.002).

Figure 2. Hourly acid output in the absence (spontaneous acid secretion: upper panel) and presence of omeprazole administration (lower panel). Spontaneous hourly acid output was not affected by cure of the H. pylori infection when samples were titrated to pH 7 (left upper panel), whereas, during omeprazole, the hourly acid output increased 1 month and 6 months after cure ofH. pylori (left lower panel). These differences were not observed when samples were titrated to a pH end point of 11 (right upper and lower panels). *p < 0.05 versus 1 month after cure; **p < 0.05 versus 1 and 6 months after cure; +p < 0.05 versus 1 month after cure.

Spontaneous ammonia outputs were 0.31 mmol/h (0.23-0.35) before cure, 0.03 mmol/h (0.03-0.04) 1 month after cure (p = 0.002 vs before cure), and 0.02 mmol/h (0.02-0.05) 6 months after cure of H. pylori infection (p = 0.005 vs before cure). During omeprazole administration, the ammonia outputs were 0.17 mmol/h (0.12-0.22) before cure, 0.03 mmol/h (0.01-0.04) 1 month after cure (p = 0.002 vs before cure), and 0.02 mmol/h (0.01-0.05), 6 months after cure of the infection (p = 0.005 vs before cure).

Corrected Acid Output

The median spontaneous acid outputs, corrected for the neutralizing effect of ammonia, were 3.81 mmol/h (2.26-5.9) before cure, 2.91 mmol/h (1.63-4.94) 1 month after cure, and 2.47 mmol/h (1.02-4.02) 6 months after cure of the H. pylori infection (p > 0.05 vs before cure, Fig. 1).

During omeprazole administration, median corrected acid outputs were 0.22 mmol/h (0.14-0.27 before cure, 0.90 mmol/h (0.06-2.74) 1 month after cure (p > 0.05 vs before cure), and 0.38 mmol/h (0.08-2.86) 6 months after cure of H. pylori infection (p > 0.05 vs before cure, Fig. 1).

Total Bile Acid and Total Protein Concentrations

There was no difference in spontaneous TBA concentrations before and after cure of H. pylori. During omeprazole administration, TBA concentrations were lower 1 month after cure than they were before cure of H. pylori, but they increased 6 months after cure of the infection (Table 2). During omeprazole administration, TBA concentrations did not correlate with the pH of gastric aspirates either before (r2 = 0.003) or 1 month after cure of H. pylori infection (r2 = 0.03).

During omeprazole administration, total protein concentrations were 238 mg/L (93-507) before cure and 96 mg/L (49-170) one month after cure of H. pylori (p = 0.02 vs before cure).

Neutralizing Capacity of Ammonia, Proteins, and Bile Acids

Before cure, there was a positive correlation between A[H+] and ammonia concentrations (rs = 0.73, p < 0.001); this was not the case 1 month after cure (rs = 0.31). There was no correlation between A[H+] and total protein concentration or [TBA], either before or after cure of the H. pylori. Recalculation of the [H+] observed before cure, during omeprazole administration, to account for the concentration of ammonia in the same samples produced an “adjusted” pH curve, which was significantly lower than the observed precure curve but similar to the observed postcure curve (Fig. 3). Recalculation of the [H+] observed, in the absence of omeprazole, showed that the observed ammonia concentrations did not modify spontaneous pH values.

Table 2. Median (95% CI) TBA Concentrations in 15-Min Gastric Aspirates and Median (95% CI) Volumes of 15-Min Gastric Aspirates in the Absence and Presence of Omeprazole Administration, Before and After Cure of H. pylori

* p < 0.05 vs before cure.

†p < 0.05 vs 1 month after cure.

Volumes of Gastric Aspirates

Gastric volumes were not affected either by administration of omeprazole or by cure of the H. pylori infection (Table 2).

Omeprazole Plasma Levels

On day 1 of omeprazole administration, omeprazole plasma levels were similar before, 1 month after, and 6 months after cure of H. pylori. On day 5 of omeprazole administration, plasma levels 1 h after administration of the drug were lower before cure than 6 months after cure of H. pylori (Fig. 4).

Figure 3. Overnight pH-metry recordings showing mean pH curves before cure (closed diamonds) and after cure (closed circles) of H. pylori during the administration of omeprazole. The adjusted pH curve (open diamonds) shows the pH corrected for hydrogen ions bound to ammonia before cure of H. pylori infection and during omeprazole administration. The adjusted pH curve is lower than the mean pH curve recorded before cure and is similar to the curve observed after cure, albeit somewhat lower in the early morning.

H. pylori-Associated Gastritis

The gastritis scores observed before cure in the antrum (17 [15-18]) and corpus (14 [9-15]) decreased markedly to 2 (1-3) and 1 (1-1), respectively, 6 months after cure (p < 0.001).

DISCUSSION

The data reported from the present study are consistent with previous reports that the increase in gastric pH produced by omeprazole inH. pylori-infected subjects is diminished after cure of the infection (1-3). This effect was evident in the titration studies of gastric aspirates and was confirmed in the gastric pH monitoring studies. Cure of H. pylori was associated with a decrease in the production of ammonia, and the documented changes in ammonia concentrations were sufficient to account for the apparent reduction in omeprazole efficacy after H. pylori cure.

Figure 4. Omeprazole plasma levels before cure and 1 month and 6 months after cure ofH. pylori infection. On day 5 of omeprazole administration, plasma levels 1 h after intake of the drug were lower before cure than 6 months after cure ofH pylori. *p < 0.05 versus before cure.

The present study was designed to determine why omeprazole appears to have a reduced antisecretory effect after cure of H. pylori infection. Nocturnal acid secretion was studied, in part, because it was during this period that H. pylori cure seemed to have the greatest effect on gastric pH and, in part, because of the need to minimize the effect of food on acid secretion and neutralization.

A number of possible explanations for the observation noted above were studied. One possibility is that H. pylori infection increases the absorption and bioavailability of omeprazole such that this drug has a greater antisecretory effect. However, cure of H. pylori did not decrease plasma omeprazole levels. On day 5 of administration, 2 days before the relevant acid secretion studies, the 1-h plasma omeprazole levels were actually higher when measured 6 months after cure than before cure. Thus, the higher pH produced by omeprazole during H. pylori infection is not due to an increased bioavailability of omeprazole.

The observation that the H. pylori-related change in omeprazole effect was most marked at night (1,2) prompted the hypothesis that the difference in omeprazole-induced gastric pH was due to the presence of H. pylori-related buffers or acid neutralizes and not directly to a previously reported nocturnal decrease in acid inhibition (15). This hypothesis is based on the premise that the increase in pH produced by a fixed quantity of acid neutralizer would be most evident when the ambient pH is high.

The effect of H. pylori on gastric acid secretion and acidity is controversial: depending on the circumstances, it has been reported that H. pylori increases, decreases, or has no effect on gastric acid secretion and acidity (16-23). Furthermore, the effect of H. pylori on the production of buffers and acid-neutralizing substances is not known. Under normal conditions, bicarbonate, proteins, and possibly bile acids are the major neutralizers of gastric acid. In the H. pylori-infected stomach, there is also significant production of ammonia. Normally, gastric acid output is measured by titrating acidity to pH 7.0; however, titration to pH 7.0 will not detect acid buffered or neutralized by ammonia, proteins, or other substances that have a pKa >7.0. The NH3/ NH4 buffer system, for, example, has a pKa of 9.4 (effective range ± 1 pH unit). For this reason, gastric samples were also titrated to pH 11, to determine whether comparable amounts of acid were secreted before and after cure of the H. pylori infection. The pH endpoint of 11 was chosen based on the effective range of NH3/NH4 buffer system and on in vitro experiments.

As in the previous studies (1,2), cure of H. pylori had no effect on spontaneous corpus pH (in the absence of omeprazole), whereas it was associated with a decrease in antral pH, 1 month and 6 months after cure. Gastric pH during omeprazole administration was also lower in both corpus and antrum after cure, although the pH differences of 3-4 units were significantly greater than those observed in the absence of omeprazole.

Irrespective of the titration endpoint used, titration studies showed that spontaneous nocturnal acid output was similar before and after cure of the H. pylori infection. This suggests that H. pylori infection does not influence spontaneous acid secretionper se in subjects without peptic ulcer disease or atrophic gastritis.

During omeprazole administration, acid output was apparently higher after cure of the infection when gastric samples were titrated to a pH endpoint of 7. Because a higher apparent acid output might also have been due to decreased duodenogastric reflux, gastric total bile acid concentrations (TBA) were measured before and after cure (24). Although TBA decreased 1 month after cure during omeprazole administration, the change was not sufficient to account for the observed change in acid output. Furthermore, TBA increased again to values higher than baseline 6 months after cure, but this was not accompanied by any significant change in gastric acid output and acidity compared with the values observed 1 month after cure. Thus, increased reflux of alkaline duodenal contents into the stomach is unlikely to be a cause of the higher pH observed during omeprazole administration and H. pylori infection.

Other possible explanations for the higher apparent acid output after cure of H. pylori include: a net increase in acid secretion, a decrease in H+ back diffusion after improvement of corpus gastritis, a decrease in the production of acid inhibitory substances, and a decrease in the production of acid-neutralizing substances. These mechanisms may be related to a chronic inflammatory process, or they may be due directly to H. pylori.

Despite the differences observed when samples were titrated to pH 7, acid output was similar, before and after cure, when gastric samples were titrated to a pH endpoint of 11. If the secretory capacity of the stomach had been reduced during H. pylori infection, one would have expected a comparable difference in acid output both during titration to pH 7 and to pH 11. This suggests that improvement of corpus gastritis after cure of H. pylori or decreased H+ back diffusion is a less probable explanation for the apparent increase in acid output observed during titration to pH 7. Titration to pH 11 does, indeed, lead to the dissociation of H+ ions from neutralizing substances with a pKa between 7 and 11. Furthermore, the amount of acid detected before cure by titration to pH 11 suggests that H. pylori infection is associated with high concentrations of neutralizing substances. Cure of H. pylori infection has been reported to lead, in a number of subjects with atrophic gastritis, to an increase or normalization of acid secretion. A recent study has shown that in 16 of 250 patients, both basal and stim-ulated acid secretion increased greatly after cure of the infection (23). However, in the present study, none of the subjects had hypochlorhydria or histological evidence of atrophic gastritis on examination of the endoscopic biopsies. Furthermore, spontaneous acid secretion was similar before, and 1 and 6 months after cure of the infection, which suggests that neither the secretory capacity of the stomach nor H+ back diffusion was influenced by the infection itself. Several studies have shown that potent antisecretory therapy induces migration of H. pylori to the gastric corpus and that this is associated with an increased severity of body gastritis (25, 26). This, in turn, may decrease acid secretion after prolonged acid suppression (27). However, in a recent study, we have shown that a 1-wk course of omeprazole 20 mg daily does not influence spontaneous and GRP-stimu-lated acid secretion in subjects withH. pylori infection (28). Thus, reported changes in corpus gastritis and acid secretion may also have been related both to the dosage and the duration of treatment with antisecretory agents. These observations notwithstanding (25-27), increased corpus gastritis and impaired acid secretion do not explain the increased antisecretory effect of omeprazole in patients with duodenal ulcer disease and H. pylori infection who have a predominant antral inflammation and a higher acid secretion (3). Similarly, they do not explain independent observations of a higher pH, during the first day of administration of novel reversible proton pump inhibitor, in subjects with H. pylori infection (29).

It is well established that proteins present in gastric juice can act as buffers (30). They are active over a wide pH range (2-10) because of their different terminal groups. In the present study, intragastric protein concentrations decreased after cure, possibly as a consequence of the decrease in severity of gastritis (lower immunoglobulin secretion and albumin leakage). Proteins have a demonstrable effect in buffering gastric acid, but the lack of correlation between A[H+] values and protein concentrations, both before and after cure in the present study, suggests strongly that proteins do not play a major role in modifying the effect of omeprazole on pH. In contrast, the strong positive correlation observed between A[H+] and ammonia concentrations before cure supports the hypothesis that ammonia is important in determining the effect of omeprazole on gastric pH in H. pylori-infected subjects.

Before cure of H. pylori, NH3 concentrations were lower during omeprazole administration than in the spontaneously secreting stomach. This may be attributable to a transient reduction in bacterial load during omeprazole administration (1); alternatively, it may be related to observations in vitro that H. pylori urease is inhibited competitively by omeprazole (31) and that its activity is decreased at high pH (32). One study, using the 13C urea breath test, has suggested that there is inhibition of urease activity by omeprazole in vivo, although this was reported at doses higher than those used in the present study (33). After cure, ammonia concentrations decreased very considerably; the residual ammonia in the gastric aspirates probably originated from the host, as cure of the H. pylori infection was confirmed in all subjects. The decrease in ammonia concentration after cure of H. pylori and the positive correlation between A[H+] and ammonia suggest that ammonia is the most probable cause of the observed pH change. In addition to the neutralizing effect of ammonia, it has been reported that ammonia can directly affect parietal cells, decreasing acid secretion (34), and that it can disrupt tight junctions, facilitating back diffusion of H+ ions (35). Thus, it is also possible that cure of H. pylori may be associated with an apparent increase in acid output due to lower back diffusion of H+ ions, or to a net increase in acid secretion as a consequence of the disappearence of acid inhibitory factors acting at the H+K+-ATPase level (36).

To test our hypothesis that ammonia produced by H. pylori was responsible for the apparent change in the omeprazole effect after cure, we calculated “adjusted” pH curves as an estimate of the gastric pH that would have been expected if the ammonia measured before cure had not been present. The adjusted pH curve was significantly lower than the pH curve observed before cure and was comparable to the pH curve observed after cure. Indeed, the adjusted precure pH curve was somewhat lower than the observed postcure pH curve, suggesting that the ammonia concentrations observed during H. pylori infection are more than sufficient to account for the differing effects of omeprazole seen before and after cure. Finally, the acid output, corrected for the ammonia production during omeprazole, was similar before and after cure of the infection, providing further support for the hypothesis that ammonia is a major factor involved in the observed pH difference.

In conclusion, the higher gastric pH observed during short-term administration of omeprazole 20 mg daily in subjects with H. pylori infection seems to be related predominantly to the production of ammonia by H. pylori. Ammonia concentrations are sufficiently high to produce a marked increase in gastric pH when acid secretion is reduced in the early morning or suppressed by an H+-K+ATPase inhibitor, but they have no discernible influence on spontaneous gastric acidity when hydrogen ion concentrations are several-fold higher than those of ammonia. It is, however, possible that ammonia plays a lesser role if higher doses of antisecretory medication are used for longer periods of time. Further studies are therefore needed to examine the roles of corpus gastritis, hydrogen ion back diffusion, acid-inhibitory factors and ammonia during H. pylori infection, and long-term antisecretory therapy.

ACKNOWLEDGMENTS

This work was supported by Swiss National Fund 32-33626.92 and by a grant from Astra Hassle AB, Molndal, Sweden.

Reprint requests and correspondence: PrЎemysl Bercґэk, M.D., Intestinal Diseases Research Programme, McMaster University, 1200 Main Street West, Hamilton, Ontario L8N 3Z5, Canada. Received July 14, 1999; accepted Dec. 1, 1999.

REFERENCES

omeprazole helicobacter pylori stomach

1. Verduґ EF, Armstrong D, Fraser R, et al. Effect of Helicobacter pylori status on intragastric pH during treatment with omeprazole. Gut 1995;36:539-43.

2. Verduґ EF, Armstrong D, IdstroЁm J-P, et al. Effect of curing Helicobacter pylori infection on intragastric pH during omeprazole treatment. Gut 1995;37:743- 48.

3. Labenz J, Tillenburg B, Peitz U, et al. Helicobacter pylori augments the pH rising effect of omeprazole in duodenal ulcer patients. Gastroenterology 1996;110:725-32.

4. Armstrong D, Monnier P, Nicolet M, et al. Endoscopic assessment of oesophagitis. Gullet 1991;1:63-7.

5. Klein PD. Clinical applications of 13CO2 measurements. Fed Proc 1982;41:2698-701.

6. Klein PD, Graham DY. Minimum analysis requirements for the detection of Helicobacter pylori infection by the 13C-urea breath test. Am J Gastroenterol 1993;88:1865-9.

7. Logan R, Polson R, Misiewicz J, et al. Simplified single sample 13carbon urea breath test for Helicobacter pylori: Comparison with histology, culture, and ELISA serology. Gut 1991;32:1461-4.

8. Xiang Z, Bugnoli M, Ponzetto A, et al. Detection in an enzyme immunoassay of an immune response to a recombinant fragment of the 128 kilodalton protein (CagA) of Helicobacter pylori. Eur J Microbiol Infect Dis 1993;12:739-45.

9. Lind T, Cederberg C, Ekenved G, et al. Inhibition of basal and betazol-sham-feeding-induced acid secretion by omeprazole in man. Scand J Gastroenterol 1986;21:1004-10.

10. Verduґ EF, Armstrong D, Sabovcikova L, et al. High concentrations of ammonia, but not volatile amines, in gastric juice of subjects with Helicobacter pylori infection. Helicobacter 1998;3:97-102.

11. Moore EW, Scarlata T. The determination of gastric acidity by the glass electrode. Gastroenterology 1965;49:178-88.

12. Hanson NQ, Freier EF. Enzymic measurement of total bile acids adapted to the Cobas Fara centrifugal analyzer. Clin Chem 1989;35:1538-9.

13. Price AB. The Sydney System: Histological division. J Gas-troenterol Hepatol 1991;6:209-22.

14. Stolte M, Stadelmann O, Bethke B, et al. Relationships between the degree of Helicobacter pylori colonization and the degree of activity of gastritis, surface epithelial degeneration and mucus secretion. Z Gastroenterol 1995;33:89-93.

15. Moore JG, Englert E Jr. Circadian rhythm of gastric acid secretion in man. Nature 1970;226:1261-2.

16. El-Omar E, Penman I, Dorrian CA, et al. Eradicating Helico-bacter pylori infection lowers gastrin mediated acid secretion by two thirds in patients with duodenal ulcer. Gut 1993;34: 1060-5.

17. Hamlet A, Olbe L. The influence of Helicobacter pylori infection on postprandial duodenal acid load and duodenal bulb pH in humans. Gastroenterology 1996;111:391-400.

18. Tarnasky PR, Kovacs TOG, Sytnik B, et al. Asymptomatic H. pylori infection impairs pH inhibition of gastrin and acid secretion during second hour of peptone meal stimulation. Dig Dis Sci 1993;38:1681-7.

19. Goldschmiedt M, Barnett CC, Schwarz BE, et al. Effect of age on gastric acid secretion and serum gastrin concentrations in healthy men and women. Gastroenterology 1991;101:1-9.

20. Peterson WL, Barnett CC, Evans DJ, et al. Acid secretion and serum gastrin in normal subjects and patients with duodenal ulcer: The role of Helicobacter pylori. Am J Gastroenterol 1993;88:2038-43.

21. McColl KEL, Fullarton GM, Chittajalu R, et al. Plasma gastrin, daytime intragastric pH, and nocturnal acid output before and at 1 and 7 months after eradication of Helicobacter pylori in duodenal ulcer subjects. Scand J Gastroenterol 1991;26:339-46.

22. Katelaris PH, Seow F, Lin BPC, et al. Effect of age, Helicobacter pylori infection, and gastritis with atrophy on serum gastrin and gastric acid secretion in healthy men. Gut 1993; 34:1032-7.

23. El-Omar EM, Oien K, El-Nujumi A, et al. Helicobacter pylori infection and chronic gastric acid hyposecretion. Gastroenterology 1997;113:15-24.

24. Verdu EF, Fraser R, Blum AL, et al. The origin of nocturnal pH rises in healthy volunteers. Scand J Gastroenterol 1995; 30:935-43.

25. Logan RPH, Walker MM, Misiewicz JJ, et al. Changes in intragastric distribution of Helicobacter pylori during treatment with omeprazole. Gut 1995;36:12-6.

26. Solcia E, Fiocca R, Villani L, et al. Effects of permanent eradication or transient clearance of Helicobacter pylori on histology of gastric mucosa using omeprazole with or without antibiotics. Scand J Gastroenterol 1996;31:105-10.

27. Gillen D, Wirz A, Neithercut WD, et al. Helicobacter pylori infection potentiates the inhibition of gastric acid secretion by omeprazole. Gut 1999;44:468-75.

28. Bercґэk P, Verdu EF, Armstrong D, et al. Apparent increase in acid output during omeprazole after cure of Helicobacter pylori infection. Gastroenterology 1997;112A70 (abstract).

29. Riedel T. The reversible H+K+-ATPase antagonist BY 841 is more effective in elevating gastric pH in Helicobacter pylori positive than H. pylori negative volunteers. Gastroenterology 1998;114A268.

30. Makhlouf GM, Blum AL, Moore EW. Undissociated acidity of human gastric juice. Gastroenterology 1970;58:345-51.

31. Bugnoli M, Bayeli P, Rappuoli R, et al. Inhibition of Helicobacter pylori urease by omeprazole. Eur J Gastroenterol Hepa-tol 1993;5:683-5.

32. Mederer SE, GruЁbel P. Profound increase of Helicobacter pylori urease activity in gastric antral mucosa at low pH. Dig Dis Sci 1996;5:994-9.

33. Stoschus B, Dominguez-Munoz JE, Kalhori N, et al. Effect of omeprazole on Helicobacter pylori urease activity in vivo. Eur J Gastroenterol Hepatol 1996;8:811-3.

34. Lorentzon P, Jackson R, Wallmark B, et al. Inhibition of proton potassium ATPase by omeprazole in isolated gastric vesicles requires proton transport. Biochim Biophys Acta 1987;897:41-51.

35. Hazell SL, Lee A. Campylobacter pylori, urease, hydrogen ion back diffusion and gastric ulcers. Lancet 1986;2:15-7.

36. Winfield B, Wagner S, Obst B, et al. Helicobacter pylori augments the inhibitory effect of omeprazole in gastric H+K+-ATPase membrane vesicles. Gastroenterology 1999; 116A1550.

Размещено на Allbest.ru