Duodenal mucosal resistance to intraluminal acid in the rat: Role of adaptive cytoprotection

GASTROENTEROLOGY

1992;102:1129-1135

Duodenal Mucosal Resistance to Intraluminal Acid in the Rat: Role of Adaptive Cytoprotection AURELIA LUGEA, ANTONIO SALAS, FRANCISCO FERNANDO AZPIROZ, and JUAN-R. MALAGELADA

GUARNER,

Digestive System Research Unit and Department of Pathology, Hospital General Vall d’Hebron, Autonomous University of Barcelona, Barcelona, Spain

The duodenal mucosa is normally challenged by intermittent exposure to acid because of periodic gastric emptying. We studied the mechanisms of duodenal adaptation to acid in anesthetized rats. A polyvinyl chloride tube passing through a ligated pylorus was used for duodenal pulse instillations of 1 mL of saline or acid (100 or 400 pmol HCl) at 30-minute intervals. Duodenal lesions were blindly assessed using a combined macroscopic and histological score. Mucosal damage after exposure to saline or 100 ylmol HCl was negligible in intact, vaand indomethacin-pretreated rats, gotomized, whereas 400 pm01 induced noticeable macroscopic and microscopic lesions. Interestingly, in intact and vagotomized rats, previous exposure to a 100~pmol HCl bolus significantly prevented mucosal damage by a subsequent 400~pmol bolus. This effect was not observed in indomethacin-pretreated rats. In these rafs, however, intraduodenal instillation of exogenous 16,lBdimethyl prostaglandin E, (16,16-dmPGE,) prevented the damage induced by 400 pmol of HCI. A second protocol investigated the luminal release of bicarbonate and PGE, in response to intraduodenal perfusion with 100 pm01 of HCl. Duodenal bicarbonate release was stimulated by acid in all groups, whereas the release of PGE, increased in intact and vagotomized rats but not in the indomethacin-pretreated group. In summary, these data suggest that adaptive cytoprotection plays a significant role in protecting the duodenal mucosa from acid. Vagal innervation and bicarbonate release do not appear to be as critical as cycle-oxygenase activity for this mechanism.

T

he resistance of the gastric mucosa to injury markedly increases after a brief exposure to a mild irritant. This phenomenon, originally described and named “adaptive cytoprotection” by Robert et a1.,lW3has been subsequently confirmed by several groups in different models of gastric damage.4-g Some evidence suggests that the ability of the gastric mu-

cosa to synthesize prostaglandins plays an important role in mediating the protection.‘-5 Other investigators claim that endogenous prostaglandins may not be essential because gastric cytoprotection persists after pretreatment with prostaglandin-synthesis inhibitors.m In any case, the occurrence of adaptive cytoprotection in experimental models of gastric injury seems well established, although its precise physiological role remains to be determined.” Most experimental studies of gastric cytoprotection have used necrotizing agents such as ethanol at high concentrations (7O%-100%) or even boiling water. These strong agents induce severe necrotic hemorrhagic lesions when applied to the gastric mucosa and cannot be regarded as physiological. In contrast, the duodenal mucosa of most mammalians including humans is normally challenged by intermittent exposure to acid because of the periodic emptying of the gastric contents into the duodenal lumen.” Under physiological conditions, the intraduodenal pH during fasting is subjected to wide fluctuations that may range from 1 to 7,” largely because of the phasic gastroduodenal motility13 and the intermittent pancreatic secretion of neutralizing bicarbonate. Hence, we reasoned that in the setting of intermittent duodenal exposure to acid, adaptive cytoprotection might play a significant role in maintaining a physiological equilibrium between duodenal mucosal resistance and luminal acidity. To investigate the mechanisms of duodenal adaptation to acid and their regulation, we developed a rat model of duodenal damage induced by periodic instillation of acid directly into the duodenal lumen. The model was used to test the effect of a brief exposure to acid on the prevention of duodenal damage by subsequent instillation of stronger doses of acid. The role of vagal innervation, duodenal bicarbonate secretion, and mucosal prostaglandin synthesis on 0 1992 by

the American Gastroenterological 0016-5065/92/$3.00

Association

1130

LUGEA ET AL.

the adaptive cytoprotection was also explored.

GASTROENTEROLOGY Vol. 102, No. 4

of the duodenal

mucosa

Materials and Methods Animals

and Protocols

Experiments were performed in male SpragueDawley rats weighing 200-300 g (Interfauna, Barcelona, Spain). The animals were housed under controlled conditions of temperature, humidity, and illumination (from 7 AM to 7 PM) and were maintained on standard laboratory chow and tap water ad libitum. The study included two protocols. The first protocol investigated whether intermittent instillation of acid into the duodenal lumen would change mucosal resistance to injury and explored putative mechanisms. The second protocol quantified changes in intraluminal release of bicarbonate and prostaglandin E, (PGE,) in response to duodenal perfusion with acid. Protocol 1: Duodenal resistance to acid. Seventyeight rats were included in this study. After an overnight fast in appropriate cages to avoid coprophagia, rats were anesthetized [urethane, 1.25 g/kg intraperitoneally (IP); Fluka Chemie, Buchs, Switzerland] and a tracheostomy was performed. The abdomen of each rat was incised and the stomach and duodenum exposed. A double-lumen tube, prepared with polyvinyl chloride tubing (ID, 1.40 mm; OD, 1.90 mm; Abbot, Sligo, Ireland), was inserted through the forestomach into the gastric lumen and passed through the pylorus. The tube was held in place by ligatures around the entry point (forestomach) and the pylorus. One of the lumens was used to drain the gastric contents by siphonage. The second lumen opened just beyond the pylorus and was used for pulse instillations of normal saline (0.15 mol/L) or acid into the duodenal lumen. Finally, the abdominal wall was sutured. Previous work by Yonei and Guth14 has shown that laparotomy stimulates an endogenous mechanism of protection of the gastric mucosa that persists for approximately 1-2 hours and is blocked by indomethacin. In our protocol, the rats received a 1-mL saline (pH 6.15) bolus through the duodenal line every 30 minutes for 2 hours to achieve full equilibration and to avoid such postlaparotomy effects. Thereafter, two additional l-mL pulses were instilled at 30-minute intervals as follows, each group consisting of five rats: sham group, saline and saline; mild group, 100 umol HCl (pH 1.03) and saline; strong group, saline and 400 pmol HCl (pH 0.68); and mild-strong group, 100 and 400 umol HCl. The complete series of experiments was performed in three sets of 20 rats each: (a) intact, (b) vagotomized, and (c) indomethacin-pretreated rats. Truncal vagotomy was achieved by esophageal transection performed just after the incision of the abdominal wall, hence, 2 hours before the instillation of acid through the duodenal line. In every case, a marked decrease in gastric acidity was confirmed by checking the pH of the effluents from the gastric line before and after vagotomy. In the indomethacin-pretreated group, indomethacin (Sigma, St. Louis, MO) was given subcutaneously (SC) at 5 mg/kg 60 minutes before the last duodenal instillation.

In an additional set of experiments, we studied the effects of exogenous administration of the prostaglandin analogue 16,16-dm-PGE, (Sigma). Three mild-strong groups of six rats each were included, and all of them received 100 and 400 umol HCl as above. A control group received no previous treatment, an indomethacin-pretreated group received indomethacin, 5 mg/kg SC, 30 minutes before the 100ymol HCl pulse and 1 mL of intraduodenal saline vehicle 15 minutes before the 400 pmol HCl pulse, and an indomethacin plus 16,16-dm-PGE,-pretreated group received indomethacin as above plus 30 pg/kg of 16,16-dmPGE, in 1 mL of saline 15 minutes before the 4OOymol HCl pulse. All rats were killed by cervical dislocation under anesthesia 1 hour after the last pulse. The stomach and duodenum were removed and immediately fixed by intraluminal perfusion of a neutral buffered 10% formalin solution for 30 minutes via the double-lumen tube used in the experiments. Thereafter, the specimens were carefully opened along the greater curvature, laid out, and pinned on a flat surface. The appearance of the duodenal mucosa was recorded and the length of the damaged area measured. Longitudinal sections along the duodenum were stained with H&E for histological examination. Macroscopic and microscopic scoring of the lesions was performed by one of the authors (AS.), a pathologist, who was unaware of the treatment administered to each rat. A global score of duodenal damage was obtained by summation of macroscopic (mucosal appearance, length of damaged area) and histological (depth of mucosal necrosis, vascular thrombosis) findings (Table 1). Protocol 2: Duodenal release of bicarbonate and PGE,. Fifteen rats were included in this study. After an overnight fast, rats were anesthetized (urethane, 1.25 mg/ kg IP) and a tracheostomy was performed. The abdomen of each rat was incised and a perfused duodenal loop was prepared using a double-lumen tube as previously described.15 Briefly, the tube was inserted through the forestomach into the gastric lumen, passed through the pylorus, and held in place by ligatures around the forestomach and pylorus. The first lumen was used for drainage of gastric contents, and the second, which opened just beyond the pylorus, was used for duodenal perfusion. The intestinal wall was transected 3 cm distal to the pylorus and ligated to a glass cannula that collected the perfusate. Pancreatic and bile ducts were ligated, and the abdominal wall was then sutured. Isotonic saline adjusted to pH 7.0 and heated to 37’C was continuously perfused at a rate of 8 mL/h through the duodenal loop; the perfusate was recovered by siphonage through the outlet tube and collected in 15-minute fractions after a 30-minute equilibration, Following a 60-minute baseline period, we instilled a 1OOymol HCl pulse (volume, 1 mL) through the duodenal line, and the perfusion was stopped. After a 2-minute exposure to acid, the perfusion was resumed for a further go-minute period. Samples collected during the first 10 minutes after the acid pulse were discarded. The 15 rats included in this protocol were divided into three groups of 5 rats each: control, vagotomized, and indomethacin-pretreated rats. Truncal vagotomy was per-

DUODENAL RESISTANCE TO ACID

April 1992

Table 1. Criteria for Assessment of Duodenal

Results

Damage

Macroscopic

score

Mucosal appearance Normal Petechiae Necrotic Length of damaged area (cm) 0 along the duodenum <3 3-7 >7 Maximum score Histological score Depth of mucosal necrosis None <33% of villi 33%-66% >66% Thrombotic phenomena” Absent Rare Frequent Maximum score NOTE. The macroscopic score was obtained by the product of mucosal appearance times length of damaged area scores. Likewise, the histological score was the product of depth of necrosis times thrombotic phenomena scores. A global score was obtained by summation of macroscopic and histological scores (O-15). “Presence of vessels occluded by fibrin thrombi.

formed as described above, and indomethacin pretreatment was given SC 30 minutes before the duodenal instillation of acid. Bicarbonate concentration in the collected perfusates was measured using the back-titration method (Autoburette, Radiometer, Copenhagen, Denmark), and duodenal bicarbonate output was calculated in 15-minute periods. PGE, concentration in the perfusates was measured by specific radioimmunoassay.” Tritiated standard was purchased from Amersham International (Buckinghamshire, England). Antiserum was kindly donated by Dr. J. A. Salmon (Wellcome Research Laboratories, Beckenham, England). Perfusate samples were assayed without prior extraction and high-performance liquid chromatography purification. Statistical

1131

Methods

Results of duodenal damage scores are expressed as the median and interquartile range for each experimental group. Statistical differences between groups were determined using the Kruskal-Wallis test for multiple comparisons and the Mann-Whitney U test for single comparisons. Bicarbonate and prostaglandin outputs are expressed as mean + SEM. Differences were tested by two-way analysis of variance. A paired t test was used to compare values within the same group, such as changes before and after acid exposure.

Duodenal

Resistance

to Acid

Instillation of saline pulses into the duodenal lumen (sham group) did not induce damage to the mucosa in intact, vagotomized, or indomethacinpretreated rats (Table 2). Likewise, pulses of 100 pmol of HCI (mild group) did not induce duodenal mucosal lesions, and a score of zero was recorded in all experiments including vagotomized and indomethacin-pretreated rats. In contrast, instillation of 400 prnol of HCl (strong group) invariably induced severe duodenal lesions in all animals (intact, vagotomized, and indomethatin pretreated). The mucosa showed large necrotic areas descending from the pylorus down to the beginning of the jejunum. Under light microscopy, the surface epithelial cells covering the villi showed signs of coagulation necrosis with lightly stained cytoplasm and indistinct nuclei. Cell exfoliation was present. In most animals the lesions involved the whole height of the villi, sparing the crypts. The lamina propria showed edema, congestion, some vessels occluded by fibrin thrombi, and hemorrhagic foci. There were no significant differences in strong group scores between intact, vagotomized, and indomethatin-pretreated rats (Figure 1). However, instillation of 400 pmol of HCl 30 minutes after a 100~pmol HCl pulse (mild-strong groups) failed to induce severe necrotic lesions in either intact or vagotomized rats. Hence, previous exposure of the duodenal mucosa to acid significantly reduced the macroscopic and histological lesions induced by instillation of strong acid (Table 2). The mucosa appeared normal or with small petechiae, and remarkably, the mucous layer covering the mucosa was prominent. The microscopic findings revealed cellular damage of the superficial epithelium close to the lumen, but most of the villus showed normal staining. Mucosal edema or vascular thrombosis were almost absent. Vagotomy had no apparent effect on the adaptive cytoprotection of the rat duodenal mucosa. In contrast, indomethacin-pretreated rats showed the full macroscopic and microscopic picture of necrohemorrhagic lesions induced by strong acid as described above. Indomethacin abolished the protective effect of 100 pmol of HCl, and thus duodenal damage scores were similar in the strong and mildstrong groups (Figure 1). Effect of 16,16-dm-PGE, In this series tection by exposure dent in control but rats. As shown in

of experiments, significant proto 100 pmol of HCl was also evinot in indomethacin-pretreated damage inFigure 2, duodenal

1132

LUGEA ET AL.

Table 2. Duodenal

GASTROENTEROLOGY Vol. 102, No. 4

Damage Scores Macroscopic Sham

Intact Vagotomy Indomethacin

0 (O-O) 0 (O-O) 0 (O-O)

Mild 0 (O-O) 0 (O-O) 0 (O-O)

scores

Strong

Histological scores Mild-strong

6 (4-6) 6 (6-6) 6 (6-6)

Sham

2 (1-2)' 1 (l-1y 6 (6-6)

Mild

0 (O-O) 0 (O-O) 0 (O-O)

Strong

0 (O-O) 0 (O-O) 0 (O-O)

9 (9-9) 9 (4-9) 9 (2-9)

Mild-strong 3 (2-4)" 2 (2-4)' 6 (4-6)

NOTE. Duodenal damage scores in intact, vagotomized, or indomethacin-pretreated rats subjected to intraduodenal instillation of a l-mL bolus of saline (sham), 100~pmol of HCl (mild), 400 pmol of HCI (strong), or 100 Fmol of HCl plus 400 Fmol30 minutes later (mild-strong). Lesions were blindly assessed using the criteria shown in Table 1. Results are expressed as the median and the interquartile range of five rats per group. “P < 0.05 vs. strong.

duced by acid instillation was more severe in the indomethacin-pretreated group than in controls, suggesting that indomethacin pretreatment suppressed the adaptive cytoprotection induced by 100 pmol of HCl. Intraduodenal administration of 16,16dm-PGE, to indomethacin-pretreated rats significantly reduced the mucosal damage induced by strong acid compared with damage in the group receiving indomethacin only (Figure 2). Therefore, the prostaglandin analogue mimicked the adaptive cytoprotection induced by 100 pmol of HCI and reversed the impairment of mucosal resistance produced by indomethacin pretreatment. Duodenal Release of Bicarbonate and PGE,

in response to a pulse instillation of 100 ymol of HCl compared with baseline saline perfusion (Figure 3). The response was significant in intact and indomethacin-pretreated rats (P < O.O5),although the indomethacin-pretreated group showed a slightly lower increase. Vagotomized rats showed a high basal bicarbonate output, and the response to HCl was not significant. The time course of the release of PGE, into a perfused duodenal loop is shown in Figure 4. An intraduodenal pulse of 100 pmol of HCl significantly increased the luminal release of PGE, in intact and vagotomized rats compared with baseline output. However, the prostaglandin response was completely blocked by indomethacin pretreatment. 15

In intact, vagotomized and indomethacin-pretreated rats, duodenal bicarbonate release increased

1

t

Damage

15 -

score

10 . Damage score

Control 5

.

0 . Intact

Vagotomy

lndomethacin

Figure 1. Duodenal damage in intact, vagotomized, or indomethacin-pretreated rats subjected to intraduodenal instillation of 400 pmol of HCl (strong; ?? ) or 100 pmol of HCl plus 400 pmol 30 minutes later (mild-strong; ?).? Duodenal lesions were blindly assessed using the combined macroscopic and histological scores shown in Table 1. Differences between strong and mildstrong groups are significant among intact and vagotomized rats (*P < 0.01) but not among indomethacin-pretreated rats (median and interquartile range of five rats per group).

lndomethacin

lndomethacin + 16.16 dm PGE 2

Figure 2. Duodenal damage in rats subjected to intraduodenal instillation of 100 pmol of HCl plus 400 pmol 30 minutes later (mild-strong; 0). The control group received no previous treatment, the indomethacin group received 5 mg/kg SC indomethatin 30 minutes before the lOO-pmol HCl pulse and 1 mL of intraduodenal saline vehicle 15 minutes before the 400~pmol HCl pulse, and the indomethacin plus 16,16-dm-PGE,-pretreated group received indomethacin as above plus 30 pg/kg 16,16-dmPGE, in 1 mL of saline by intraduodenal instillation 15 minutes before the 460-pmol HCl pulse. Combined scores were obtained as described in Table 1. Duodenal damage was significantly higher in the indomethacin-pretreated group than in the control or indomethacin plus 16,16-dm-PGE,-pretreated groups (*P < 0.01; median and interquartile range of six rats per group).

DUODENAL

April 1992

15

??

10 kico; kmol/h 5

0

??

Intact

Vagotomy

lndomethacin

Figure 3. Luminal release of bicarbonate into a perfused duodenal loop. Stimulation with intraluminal HCl(lO0 pmol; 0) elicited a significant bicarbonate response compared with perfusion with saline (m) in intact and indomethacin-pretreated rats (*P < 0.05; mean + SEM of five rats per group). Vagotomized rats showed a high basal bicarbonate output, and the response to HCI was not significant.

Discussion This report presents evidence that adaptive cytoprotection in response to intermittent acid exposure takes place in the normal rat duodenum. Previous studies by different investigators had described a property of the gastric mucosa, generally termed adaptive cytoprotection, that referred to the enhanced resistance to injury developed in response to the challenge with mild intraluminal irritants.‘-’ The current study was conducted to explore the relevance of such a protective mechanism in the rat duodenum, because the duodenal mucosa is constantly challenged by intermittent exposure to gastric acid. Our results show that the exposure of the proximal duodenum to acid prevents the mucosal damage induced by strong acid. This effect can be suppressed by indomethacin, but it is not influenced by truncal vagotomy. In addition, we have shown that the exposure to acid induces the duodenal release of PGE, and bicarbonate. Indomethacin pretreatment does not change bicarbonate release but inhibits the prostaglandin response. Neither the bicarbonate nor the prostaglandin response is influenced by vagotomy. These data suggest that adaptive cytoprotection plays an important role in the defense of the duodenal mucosa from gastric acid. Vagal innervation and bicarbonate release do not appear to be as critical as mucosal prostaglandin synthesis in the mediation of this mechanism. Emptying of gastric contents into the duodenal lumen is a complex process determined by the coordinated motor activity of the stomach and the proxi-

RESISTANCE

TO ACID

1133

During fasting, acid is the major mal intestine.” component of gastric contents, and thus gastric emptying delivers a certain amount of acid directly onto the duodenal mucosal surface. In mammahans, the gut shows a pattern of cyclic variations, alternating periods of quiescence with periods of activity, that is constantly present during normal fasting conditions. The interdigestive activity has a dual motor and secretory component so that during periods of quiescence and early periods of activity, the stomach accumulates a certain amount of gastric secretion that is cleared during the peak of motor activity. Hence, it has been shown in humans that the intraduodenal pH decreases to 1 or 2 during the phase of motor activity but remains neutral during the periods of quiescence.13 In the rat, the interdigestive motor cycle may have a duration of 17-30 minutes.17 According to previous data from our own laboratory on neural-stimulated acid secretion in the rat,‘* up to 50-150 umol of acid may accumulate in the stomach during that time, to be delivered as an acid pulse into the duodenal lumen in association with the periods of motor activity. In the conscious unstimulated rat, the average concentration of acid in the gastric juice ranges from 85 to 125 pmol/mL,lg similar to the HCl concentration used as a mild irritant in our study (100umol/mL). Thus, we have reproduced the intermittent exposure to acid that occurs physiologically and observed that pulses of acid are able to trigger a response of adaptive cytoprotection by the duodenal mucosa and to stimulate the release of cytoprotective prostaglandins into the lumen. Remarkably, such a protective response renders the duodenal mucosa able to withstand strong necrotizing

200

PGE, pg/min

100 umol HCI

100

60

120

min

Figure 4. Luminal release of PGE, into a perfused duodenal loop. An intraduodenal HCl pulse induced a significant increase of the luminal release of PGE, in intact (-) and vagotomized (-- ---) rats compared with baseline output (P i 0.01; mean + SEM of five rats per group). The response was not observed in indomethacin-pretreated rats (- - -).

1134 LUGEA ET AL.

concentrations of acid far beyond the physiological range. Previous work by RobertzO showed a protective effect of prostaglandins on the duodenal mucosa. Duodenal ulcers produced in rats by gastric secretagogues such as histamine, carbachol, pentagastrin, or cysteamine were inhibited by treatment with prostaglandins, albeit given at antisecretory doses. It was also shown that nonsteroidal anti-inflammatory agents could induce multiple perforating ulcers of the small intestine that were preventable by concomitant treatment with exogenous prostaglandins.20 These experiments suggested a cytoprotective effect of prostaglandins on the rat small intestinal mucosa. Data from Leung et a1.21 showed protection of the duodenal mucosa against acid by exogenous 16,16dm-PGE,. Our study supports these observations by showing a beneficial effect of exogenous 16,16-dmPGE, on acid-induced duodenal damage in indomethacin-treated rats. In addition, our study suggests that endogenous prostaglandin synthesis plays an important role in the adaptive cytoprotection of the duodenal mucosa to acid. An original conclusion of our work is that adaptive cytoprotection mediated by prostaglandin synthesis appears to be a key mechanism for the defence of the duodenal mucosa against gastric acid. Our findings are consistent with the clinical evidence that drugs that inhibit prostaglandin synthesis, such as nonsteroidal anti-inflammatory drugs, may cause ulcerative lesions of the duodenal mucosa. The prevalence of duodenal ulcers in subjects taking long-term nonsteroidal anti-inflammatory drugs was estimated at ll%, whereas this figure is 1.4% in the general population.22 Two clinical studies 23,24involving a considerable number of patients (297 and 144) showed that 8% of patients with normal results of upper endoscopy before treatment developed duodenal ulcers after 4-6 weeks of taking nonsteroidal anti-inflammatory drugs as assessed by regular endoscopic controls. The incidence of gastric lesions ranged from 6% to 12% in these studies. Interestingly, prophylactic inhibition of gastric acid secretion with the H, antagonist ranitidine prevented the formation of duodenal ulcers, but its effect was not significant with respect to gastric mucosal damage. These data would suggest that, in the absence of gastric acid, nonsteroidal anti-inflammatory drugs may not be directly injurious to the duodenal mucosa. Only in the presence of an acid challenge would the failure of the duodenal mucosa to normally synthesise prostaglandins result in the development of duodenal ulcerations. Such an explanation may not be valid for the gastric mucosa, where significant lesions may develop in the absence of acid, perhaps by a direct damaging effect of the

GASTROENTEROLOGY Vol. 102, No. 4

drugs. The concept that a normal prostaglandin response of the duodenal mucosa is mandatory for effective protection against acid was also suggested by previous studies showing impaired mucosal prostaglandin synthesis in response to acid exposure in patients with duodenal ulcer disease.25g26 However, other studies that have measured luminal release of prostaglandin E, did not find evidence of impaired acid responseU2’ The concept of impaired prostaglandin-mediated defense against injury as a pathogenic mechanism remains, therefore, a tantalizing but unproven hypothesis. In conclusion, the present work shows that adaptive cytoprotection plays a significant role among the mechanisms of defense of the duodenal mucosa from the action of gastric acid. In the rat, this mechanism implies the mediation of a preserved ability to synthesize prostaglandins, whereas the local release of bicarbonate and the role of the vagus appear to be less relevant for the prevalence of a normal adaptation to acid. Impairment of this mechanism by cyclooxygenase inhibitors may seriously compromise the capacity of the duodenal mucosa to defend itself and lead to ulceration. References 1. Robert A, Lancaster

C, Hanchar AJ, Nezamis JE. Mild irritants prevent gastric necrosis through prostaglandin formation: histological study (abstr). Gastroenterology 1978;74:1086. 2. Chaudhury TK, Robert A. Prevention by mild irritants of gastric necrosis produced in rats by sodium taurocholate. Dig Dis Sci 1980;25:830-836. 3. Robert A, Nezamis JE, Lancaster C, Davis JP, Field SO, Hanchar AJ. Mild irritants prevent gastric necrosis through “adaptive cytoprotection” mediated by prostaglandins. Am J Physiol 1983;245:G113-G121. 4. Henagan JM, Smith GS, Seidel ER, Miller TA. Influence of vagotomy on mucosal protection against alcohol induced gastric damage in the rat. Gastroenterology 1984;87:903-908. 5. Tarnawski A, Hollander D, Krause WJ, Zipser RD, Stachura J, Gergely H. Does sucralfate affect the normal gastric mucosa? Histological, ultrastructural and functional assessment in the rat. Gastroenterology 1986;90:893-905. 6. Wallace JL. Increased resistance of the rat gastric mucosa to hemorrhagic damage after exposure to an irritant. Role of the “mucoid cap” and prostaglandin synthesis. Gastroenterology 1988;94:22-32. 7. Hawkey CJ, Kemp RT, Walt RP, Bhaskar NK, Davies J, Filipowicz B. Evidence that adaptive cytoprotection in rats is not mediated by prostaglandins. Gastroenterology 1988;94:948954. 8. MacNaughton WK, Wallace JL. A role for dopamine as an endogenous protective factor in the rat stomach. Gastroenterology 1989;96:972-980, 9. Tepperman BL, Soper BD, Morris GP. Effect of sialoadenectomy on adaptive cytoprotection in the rat. Gastroenterology 1989;97:123-129. 10. Silen W. What is cytoprotection of the gastric mucosa? Gastroenterology 1988;94:232-235. 11. Malagelada JR, Azpiroz F. Determinants of gastric emptying and transit in the small intestine. In: Schultz SG, Wood JD,

April 1992

12.

13.

14.

15.

16.

17.

18.

19.

20.

Rauner BB, eds. Handbook of physiology. Section 6: The gastrointestinal system. New York: Oxford University, 1989:909937. Ovesen L, Bendtsen F, Tage-Jensen U, Pedersen NT, Gram BR, Rune SJ. Intraluminal pH in the stomach, duodenum, and proximal jejunum in normal subjects and patients with exocrine pancreatic insufficiency. Gastroenterology 1986;90:958962. Nishioka T, Sekiguchi T, Kogure M, Kusano M, Arai H, Katoh R, Okamura S, Matsuzaki T, Fukazawa T, Akiyama T, Ohwada H, Kobayashi S. A study on continuous simultaneous recordings of gastrointestinal motility and intraluminal pH in human healthy subjects. Jpn J Gastroenterol 1981;78:20652073. Yonei Y, Guth PH. Laparotomy stimulates an endogenous gastric mucosal protective mechanism in the rat. Gastroenterology 1988;95:600-611. Guarner F, Vilaseca J, Salas A, Rodriguez R, Malagelada JR. Reduction of cysteamine-induced duodenal ulcers by dietary fish oil. Eur J Gastroenterol Hepatol 1991;3:239-243. Vilaseca J, Salas A, Guarner F, Rodriguez R, Malagelada JR. Participation of thromboxane and other eicosanoid synthesis in the course of experimental colitis. Gastroenterology 1990;98:269-277. Ruckebusch M, Fioramonti J. Electrical spiking activity and propulsion in small intestine in fed and fasted rats. Gastroenterology 1975;68:1500-1508. Cucala M, Salas A, Guarner F, Malagelada JR. Chronic, centrally-mediated stimulation of gastric secretion worsens duodenal ulcer in the rat (abstr). Gastroenterology 1989;96:A103. Barocelli E, Impicciatore M, Seaton J, Conter R, Kauffman G. Localization of central prostaglandin E, antisecretory effects. Gastroenterology 1991;100:320-327. Robert A. Prostaglandins and the gastrointestinal tract. In: Johnson LR, ed. Physiology of the gastrointestinal tract. Volume 2. New York: Raven, 1981:1407-1434.

DUODENAL RESISTANCE TO ACID

1135

21. Leung FW, Miller JC, Reedy TJ, Guth PH. Exogenous prostaglandin protects against acid-induced deep mucosal injury by stimulating alkaline secretion in rat duodenum. Dig Dis Sci 1989;34:1686-1691. 22 McCarthy D. Nonsteroidal antiinflammatory drug-induced ulcers: management by traditional therapies. Gastroenterology 1989;96:662-674. 23. Ehsanullah RSB, Page MC, Tildesley G, Wood JR. Nonsteroida1 anti-inflammatory drug-induced gastroduodenal damage: prophylactic effect of inhibition of gastric acid secretion. Br Med J 1988;297:1017-1021. 24. Robinson MG, Griffin JW, Bowers J, Kogan FJ, Kogut DG, Lanza FL, Warner CW. Effect of ranitidine on gastroduodenal damage induced by nonsteroidal antiinflammatory drugs. Dig Dis Sci 1989;34:424-428. 25. Ahlquist DA, Dozois RR, Zinsmeister AR, Malagelada JR. Duodenal prostaglandin synthesis and acid load in health and in duodenal ulcer disease. Gastroenterology 1983;85:522-528. 26. Malagelada JR, Ahlquist DA, Moore SC. Defects in prostaglandin synthesis and metabolism in ulcer disease. Dig Dis Sci 1986;31:203-273. 27. Bukhave K, Rask-Madsen J, Hogan DL, Koss MA, Isenberg JI. Proximal duodenal prostaglandin E, release and mucosal bicarbonate secretion are altered in patients with duodenal ulcer. Gastroenterology 1990;99:951-955.

Received March 15,1991. Accepted September 10,1991. Address requests for reprints to: F. Guarner, M.D., Digestive System Research Unit, Hospital General Vall d’Hebron, Barcelona 08035, Spain. Dr. Lugea is the recipient of a Fellowship from the Fondo de Investigaciones Sanitarias de la Seguridad Social (Madrid, Spain). Part of this study was presented at the Annual Meeting of the American Gastroenterological Association, May 1990, in San Antonio, Texas.