Gastric Secretion and Emptying after Ordinary Meals in Duodenal Ulcer

GASTROENTEROLOGY

73:989-994, 1977

Vol. 73 , No.5

Copyright © 1977 by the American Gastroenterological Association

Printed in U.S A.

GASTRIC SECRETION AND EMPTYING AFTER ORDINARY MEALS IN DUODENAL ULCER JUAN-R. MALAGELADA, M .D., GEORGE M . D.,

W. H. J .

F.

LONGSTRETH , M .D., TIMOTHY

SuMMERSKILL, M.D., AND VAY LIANG

W.

B.

DEERING,

Go, M.D.

Gastroenterology Unit, Mayo Clinic, R ochester, Minnesota

We have studied the gastric response to an ordinary solid-liquid meal in 12 patients with active duodenal ulcer and 8 healthy volunteers. Our method employs gastric and duodenal markers to quantify acid, pepsin, and volume outputs in response to the meal, without manipulating intragastric pH. Intragastric volume, rate of gastric emptying, delivery of acid into the duodenum, and serum gastrin response were also measured simultaneously. On a separate day, peak acid output in response to betazole (1.5 mg per kg subcutaneously) was determined. Our results indicate an inappropriately prolonged gastric secretory response to meals in duodenal ulcer disease, without a concomitant increase in peak postprandial secretory rates or an increase in serum immunoreactive gastrin levels. Further, the stomach in duodenal ulcer disease did not "retain" the additional acid secreted in the later postprandial period, and abnormally high rates of acid delivery into the duodenum occurred. Our data are consistent with a dual defect in the duodenal mechanisms regulating both acid secretion and acid delivery into the duodenum. Although it is generally accepted that acid plays an important pathogenetic role in duodenal ulcer, we have much to learn about the character and significance of the gastric secretory and motor abnormalities in this disease. 1 Particularly, current knowledge of the physiological gastric response to meals is quite limited because: (1) acid output after solid meals has been measured reliably only by intragastric titration,2 which artificially maintains the pH of the stomach at 5.5 and therefore alters the physiologic response to food; 3 (2) previous assessment of postprandial gastric emptying has had conflicting results and has often ignored the dilutional effect produced by variable rates of endogenous secretion on intragastric meal and marker concentrations;
Our present observations on postprandial gastric secretion and gastric emptying in patients with duodenal ulcer obviate some of these deficiencies. Gastric function after a mixed solid and liquid meal was quantified by a previously developed and validated method7 under circumstances resembling physiological conditions. Our approach has yielded substantial new information and increased our understanding of pathophysiological interactions among gastric acid secretion, intragastric acidity, and duodenal acid load as well as the serum gastrin response to meals in patients with duodenal ulcer. Methods Patients. Twelve patients (1 female), ages 21 to 57 years, mean 43.9, were studied. Their height was 171 .7 ± 1.9 em (mean ± SE) and their weight, 77.3 ± 2.9 kg. All had current typical symptoms and endoscopic or radiological evidence of duodenal ulcer. None was being treated with H 2-histamine receptor antagonist or anticholinergic drugs, had other significant disease, or had undergone prior gastrointestinal or biliary tract surgery. Eight healthy volunteers (1 female) served as controls, comparable to the patients with duodenal ulcer. Their ages ranged from 22 to 54 years, mean 41.4; their height was 174.7 ± 2.5 centimeters and their weight, 81.7 ± 3.9 kg. Quantification of postprandial gastric function. We used our method7 for quantification of the postprandial acid, pepsin, and total secretory outputs, the volume of the gastric contents and its rate of emptying into the duodenum, and the duodenal acid load. In summary, patients took no antacid or other medication during the 12 hr preceding the studies. Mter an overnight fast, they swallowed a triple lumen duodenal tube and a double lumen gastric sump tube. We used one lumen of the duodenal tube for continuous perfusion of normal saline (2 ml per min) containing a nonabsorbable marker, 10 i-J.Ci per liter of [' 4 C]polyethylene glycol ([14 C]PEG), at the level of the

Vol. 73 , No.5

MALAGELADA ET AL .

990

papilla of Vater. We used a second lumen for continuous sampling of duodenal contents 20 em distal to the perfusion site, at the level of the angle ofTreitz. The third lumen was a duodenal air vent to facilitate suction. For 1 hr (basal period), we continuously aspirated gastric and duodenal samples with a suction pump ( -40 mm Hg). Previous studies by us8 had excluded any effect of transpyloric intubation on the gastric motor or secretory response to a similar meal. At the end of the 1-hr base line period, we stopped gastric aspiration, and the patients ate a meal consisting of 90 g (uncooked weight) of ground tenderloin steak seasoned with 0.1 g of salt, 25 g of white bread with 8 g of butter, and 60 g of vanilla ice cream with 35 g of chocolate syrup. They also drank one glass of water (240 ml) containing 15 g of polyethylene glycol (PEG). Thereafter, we sampled gastric and duodenal contents every 10 min as previously described 7 until the gastric samples no longer contained food particles. We obtained blood for serum gastrin determination 30 min and immediately before the meal was served; further samples were obtained, beginning 15 min after the meal, at 30-min intervals until the end of the study. The pH of the samples was measured with a Fisher Accument model 520 (Fisher Scientific Company, Pittsburgh, Pa.) and the previously described methods were used to determine concentrations of acid, pepsin, [14 C]PEG, and PEG.' Serum gastrin concentrations were determined by radioimmunoassay. 9 For calculation of volume and secretory output, we used the concentration of markers, acid, and pepsin as previously described. 7 Peak postprandial gastric acid output (milliequivalents per hour) was calculated by doubling the sum of the three consecutive highest 10-min outputs. Quantification of gastric secretory response to betazole. On a separate day, each control volunteer and 10 of the 12 patients who participated in the meal studies underwent standard gastric analysis with betazole stimulation. They were intubated under fluoroscopic control with a gastric sump tube that had an accessory lumen for the infusion of a nonabsorbable marker, 51 CrCl 3 , in isotonic saline. Mter a basal hour, betazole, 1.5 mg per kg of body weight, was injected subcutaneously and gastric juice was continuously aspirated and collected during seven consecutive 15-min periods. Acid output was determined by titrating the samples with NaOH and correcting the volume aspirated by the marker recovery. 10 Peak acid output (milliequivalents per hour) was calculated by doubling the two consecutive highest 15-min outputs. Statistical analysis . To compare data obtained in patients with duodenal ulcer with the respective control studies in healthy volunteers, we applied three-way analysis of variance11 to disease, time, and individual patient variables. When interaction between disease and time was significant (P < 0.05) or when there was a significant difference over-all (P < 0.05) with no interaction, we applied the paired t-test 11 to the data of each postprandial interval. Peak acid output in response to meals or to betazole was compared by nonparametric methodsY

Results

Response to Meal Postprandial gastric secretory output. During the first postprandial hour, gastric acid output rose similarly in the volunteers and patients with duodenal ulcer, and peaked at approximately the same level (fig. 1). Thereafter, the acid output declined rapidly but continued at significantly higher rates in the ulcer patients. The secretory volume (fig. 2) and pepsin (fig. 3) output paralleled the acid output. The over-all profile of the

42

Meal

~

10 8

Acid

output,

6

meq/40min

2

Hours

FIG. 1. Gastric acid output after meals in pa tients with duodenal ulcer (D .U .) and in healthy volunteers. (*), significant differences (P < 0.05) at each 10-min interval. Hourly outputs for second, third, and fourth postcibal hours were significantly greater (P < 0.05) in duodenal ulcer and over-all profiles were also significantly different by analysis of variance (P < 0.02) . Meal

t

120 100 Secretory volume output, mi/10min

80 60

J

Mean :!:.SE

40 20 0 -1

/:

Y"r·i 0

Hearth

tY1~-1f'£1 . tY..

**** * 2

y

3

£f./;, 4

Hours

FIG. 2. Gastric secretory volume after meals in patients with duodenal ulcer (D .U .) and in hea lthy volunteers. (*), significant differences (P < 0.05) at each 10-min interval. Hourly outputs for second , third, and fourth postcibal hours were significantly greater (P < 0.05) in duodenal ulcer and over-all profiles were also significantly different by analysis of variance (P < 0.02) .

acid, volume, and pepsin response to the meal was significantly different between duodenal ulcer disease and health (analysis of variance, P < 0.02) . Hourly output (second, third, and fourth postcibal hours) was sig.-.:ificantly greater (P < 0.05) in duodenal ulcer disease than in health. In both, the intragastric pH increased to about pH 5.0 after ingestion of the meal and then gradually declined and stabilized at pH 1.5 to 2.0 by the end of the second hour (fig. 4). Interactions between gastric acid secretion and intragastric acidity in health and duodenal ulcer disease were examined by converting the postprandial pH measurements to hydrogen ion activity and plotting them against the simultaneously measured acid output (fig. 5). The rate of acid secretion was much higher in duodenal ulcer disease than in health when the intragastric hydrogen ion activity ranged approximately between 10 and 30 mEq per liter. Above and below this hydrogen ion activity range, the acid output was similar in both groups.

November 1977

991

GASTRIC PATHOPHYSIOLOGY IN DUODENAL ULCER

Postprandial volume of gastric contents and rate of gastric emptying. Mter the meal, the volume of the gastric contents rose to approximately 400 ml (the volume of the meal) and during the next 4 hr it decreased gradually toward the basal volume (fig. 6). No significant differences were detected between duodenal ulcer disease and health. The rate of emptying of gastric contents peaked both in health and duodenal ulcer disease during the first hour and declined thereafter. Beyond 90 min the mean rate of delivery of the gastric contents into the duodenum was higher in duodenal ulcer patients than in normal volunteers, but the differences did not reach statistical significance. Postprandial acid content of the stomach and duodenal acid load. The acid content of the stomach and the duodenal acid load were quantified as titratable acid (comprising both undissociated "protein-buffered" and dissociated hydrogen ion). The duodenal acid load was also measured using the hydrogen ion activity as a close

8 7 6

5 Total acid output, 4 meq/40min 3

2

40

20

30

40

50

H+ activity, !Jeq/ml FIG. 5. Relationship between postprandial total acid output and intragastric hydrogen ion activity measured simultaneously in patients with duodenal ulcer and in healthy volunteers.

80

60 Pepsin output, mg/10min

40

D.U.

/

20

rr··i 0 -1

0

*

2

Hours

3

4

*P
FIG. 3. Gastric pepsin output after meals in patients with duodenal ulcer (D.U .) and in healthy volunteers. (*) , significant differences (P < 0.05) at each 10-min interval. Hourly outputs for second, third, and fourth postcibal hours were significantly greater (P < 0.05) in duodenal ulcer and over-all profiles were also significantly different by analysis of variance (P < 0.02).

6 5

4 Gastric pH

Meal(pH s.ol

'

3

2

FrG. 6. Postprandial volume of gastric contents and rate of gastric emptying in patients with duodenal ulcer (D .U.) and healthy volunteers.

TMeon :tSE

0

-1

0

2

3

4

Hours

FIG. 4. Postprandial gastric pH in patients with duodenal ulcer W.U.) and in healthy controls.

approximation of the unbuffered hydrogen ion concentration.12 The acid content of the stomach was similar in duodenal ulcer disease and health throughout the postpran-

Vol . 73 , No.5

MALAGELADA ET AL.

992

dial period (data not shown). The duodenal acid load was the same in both groups during the first postprandial hour, but during the subsequent 3 hr it became significantly greater in the duodenal ulcer group (fig. 7). The hydrogen ion load, both in health and in duodenal ulcer disease, was minimal during the first postprandial hour, when most hydrogen ion was buffered by meal protein. However, during the second postprandial hour the hydrogen ion delivery into the duodenum in duodenal ulcer disease increased rapidly and became much greater than in health for the remainder of the observation petiod (fig. 8) . Serum gastrin. The serum immunoreactive gastrin level increased after the meal, peaked during the first hour, and then gradually decreased toward basal (fig. 9) . No significant difference between duodenal ulcer disease and health was noted at any interval.

Response to Betazole The peak acid output response to standard betazole stimulation (1.5 mg per kg subcutaneously) was significantly greater in duodenal ulcer patients than in healthy subjects (P < 0.05). In contrast, the postprandial peak acid output was similar for both groups (fig. 10). There was little correlation between the peak acid output after meals and that after betazole injection both in duodenal ulcer patients (r = 0.34) and in healthy subjects (r = 0.27). The ratio between the peak response to betazole and to meals was slightly greater in the duodenal ulcer group, but the difference was not statistically significant.

2.2

Meal

t

~.8

H+ activity 1.4 entering duodenum, ~ .o meq/10min 0.6

I

0.2

Mean

±SE

..

o~--~

-1

~~~~~~~~~-L­

0

1

2

4

3

*P<0.05 Hours FIG. 8. Duodenal hydrogen ion load after meals in patients with duodenal ulcer (D.U.) and healthy volunteers. (*), significant differ· ences (P < 0.05) at each 10-min interval.

200

Serum 1!10 gastrin, pg/ml

100

I±

Mean SE

~1.~--~o----~--~z----L3----4L--Discussion A key abnormality that we have found in patients Hours with duodenal ulcer was their high rate of acid delivery 13 FIG. 9. Serum immunoreactive gastrin response to meals in duointo the duodenum after meals. Rhodes et al., using a denal ulcer (D.U.) disease and in health. glass electrode, had previously recorded a longer reduction in postprandial bulbar pH in patients with duo-

Meal- stimulated

Betazole- stimulated

60 6

Meal

t

5 Total acid entering duodenum, meq/10min

N.S.

50 Peak acid output, meq/30'

4 3 2

40

0

30

0 0

20

15'"'" cP 0

~0

0

Hours Fra. 7. Total duodenal acid load in patients with duodenal ulcer (D .U.) and healthy volunteers. (*),significant differences (P < 0.05) at each 10-min interval. Hourly outputs for second, third, and fourth postcibal hours were significantly greater (P < 0.05) in duodenal ulcer and over-all profiles were also significantly different by analysis of variance (P < 0. 02).

P
• •

..,.

..

0 0

~ f9 0

Health

DU

Health

~ ••

•

DU

FIG. 10. Comparison between peak acid output after meals and after betazole (1.5 mg per kg subcutaneously) in patients with duodenal ulcer (DU) and healthy volunteers.

denal ulcer than in healthy controls. Our current data suggest that the greater postprandial delivery of acid is at least partially responsible for the lower pH in the duodenal bulb in duodenal ulcer disease. The two important determinants of duodenal acid load are: (1) gas-

November 1977

GASTRIC PATHOPHYSIOLOGY IN DUODENAL ULCER

tric acid output, a major determinant of intragastric acidity; and (2) gastric emptying, which determines the rate of delivery of acidified gastric contents into the duodenum. We found that the postprandial gastric acid output in duodenal ulcer was characterized by normal peak acid secretion during the first postprandial hour and abnormally prolonged acid secretion throughout the second to fourth postprandial hours. On the other hand, peak gastric acid output in response to the conventionaP 4 "maximal" betazole stimulus was higher in duodenal ulcer than in health, in accordance with established concepts.' Further, despite the relatively small number of individuals involved, our series are probably quite representative of the respective populations at large, at least in terms of gastric secretory capacity, inasmuch as mean peak acid outputs after betazole in our duodenal ulcer patients and healthy controls closely matched values previously reported for a much larger number of individuals similarly studied. 15 • 16 Because we found poor correlation between the peak acid output after the meal and after betazole, we must conclude that the peak postprandial acid output is not strictly related to parietal cell mass, which correlates well with the response to betazoleY Instead, the response to the meal should be regarded as the net effect of multiple stimulatory and inhibitory regulatory mechanisms that produce output at or below the true maximal secretory capacity of the stomach. Pathophysiologically this implies that, in duodenal ulcer disease, an inappropriately prolonged acid secretory response to a meal can occur in the absence of an absolute increase in acid secretory rate. This suggests that the altered response to food in this disease is the expression of an abnormal regulatory process rather than a simple manifestation of an increased secretory capacity of the stomach, as earlier studies with chemical or hormonal stimuli had led one to suspect. Our results are partially at variance with those of Fordtran and Walsh2 who found that patients with duodenal ulcer had both a higher peak and a more prolonged acid secretory response to a meal. Further, they reported that the peak histamine and the maximum meal-stimulated acid output correlated well in duodenal ulcer disease and in health, and that the response ratio of meal to histamine was greater in duodenal ulcer disease. 2 These authors measured the postprandial acid output by an intragastric titration method, artificially maintaining the intragastric pH at 5.5, and thus they abolished the feedback regulatory mechanisms that are normally triggered by acid. 3 It is possible that they measured the maximal acid secretory response (predictably higher in patients with duodenal ulcer who have a larger mean parietal cell mass)' rather than the spontaneous response to food as quantified by our method, but other differences in methodology, meal used, and/or patient population studied might have also contributed to the discrepancy. The data of Fordtran and Walsh 2 and our own do concur, however, in that in duodenal ulcer the secretory response to a meal is abnormally prolonged into the late postprandial period, whereas it is quickly inhibited in healthy persons after the initial

993

peak. Whether this represents a failure of normal inhibition or a persistent stimulation by an, as yet, indeterminate factor cannot be ascertained. 3• 18 Our study indicates that immunoreactive gastrin release was normal in duodenal ulcer, as opposed to larger series in which a slightly higher postprandial response has been often noted. 1 ~21 Nevertheless, it is remarkable that in our patients an abnormal gastric secretory response to the meal coincided with a normal serum gastrin response. Although this observation tends to minimize the potential pathogenetic role of gastrin, we cannot exclude the possibility that an increased sensitivity to circulating gastrin 18 may account for the greater acid output measured during the second through fourth postcibal hours. The greater gastric secretory rate in the late postcibal hours cannot entirely account for the greater duodenal acid load as the stomach, acting as a reservoir, should be able to retain any excess of acid and empty it at normal rates over a longer time. This did not happen in duodenal ulcer as shown by the failure of the total acid content of the stomach to rise above normal at any time during the postprandial period. The acid, secreted at above normal rates, rapidly emptied into the duodenum and abnormally increased the duodenal acid load. It seems unlikely that a primary motor disorder of the stomach produced this effect because the total emptying of the gastric contents was barely above normal rates. Further, both the titratable acid and the dissociated hydrogen ion delivered into the duodenum were increased in duodenal ulcer disease. Thus, the more rapid emptying of meal buffer, as postulated by Fordtran and Walsh, 2 cannot entirely account for the increased delivery of hydrogen ion, although it may have magnified the effect. We believe that our data are most consistent with a dual postprandial regulatory defect that causes: (1) an inappropriately prolonged gastric secretory response to food that is not inhibited by normal acidification of the gastric contents, and (2) an inappropriately high rate of delivery of acid by the stomach into the duodenum. Failure of the duodenal regulatory mechanisms that simultaneously control acid secretion and acid emptying may explain these abnormal postprandial responses in patients with duodenal ulcer disease. REFERENCES 1. Fordtran JS: Acid secretion in peptic ulcer. In Gastrointestinal Disease. Edited by MH Sleisenger, JS Fordtran. Philadelphia, WB Saunders Co, 1973, p 174-188 2. Fordtran JS, Walsh JH: Gastric acid secretion rate and buffer content of the stomach after eating. J Clin Invest 52:645-657 , 1973 3. Walsh JH, Richardson CT, Fordtran JS: pH dependence of acid secretion and gastrin release in normal and ulcer subjects. J Clin Invest 55:462-468, 1975 4. Griffith GH, Owen GM, Campbell H , et al: Gastric emptying in health and in gastroduodenal disease. Gastroenterology 54:1-7, 1968 5. Cobb JS, Bank S, Marks IN: Gastric emptying after vagotomy and pyloroplasty. Am J Dig Dis 16:207-215, 1971

994

MALAGELADA ET AL.

6. Barber DC, Duthie HL, Howlett PJ, eta!: Principal components: a new approach to the analysis of gastric emptying. In Proceeding of Symposium on Dynamic Studies with Radioisotopes in Clinical Medicine and Research. International Atomic Energy Agency, Knoxville, TN, 1974, p 44-60 7. Malagelada JR, Longstreth GF, Summerskill WHJ, eta!: Measurement of gastric functions during digestion of ordinary solid meals in man. Gastroenterology 70:203-210, 1976 8. Longstreth GF, Malagelada JR, Go VLW: The gastric response to a transpyloric duodenal tube. Gut 16:777-780, 1975 9. Sizemore GW, Go VLW, Kaplan EL, eta!: Relations of calcitonin and gastrin in the Zollinger-Ellison syndrome and medullary carcinoma of the thyroid. N Eng! J Med 288:641-644, 1973 10. Go VLW, Hofmann AF, Summerskill WHJ: Simultaneous measurement of total pancreatic, biliary and gastric outputs in man using a perfusion technique. Gastroenterology 58:321-328, 1970 11. Dixon WJ, Massey FJ Jr: Introduction to Statistical Analysis. Third Edition. New York, McGraw-Hill, 1969 12. Moore EW, Scarlata RW: The determination of gastric acidity by the glass electrode. Gastroenterology 49:178-188, 1965 13. Rhodes J, Apsimon HT, Lawrie JH: pH of the contents of the

Vol. 73,No.5

duodenal bulb in relation to duodenal ulcer. Gut 7:507-508, 1966 14. Makhlouf GM: Measures of gastric acid secretion in man. Gastroenterology 55:423-429, 1968 15. Wormsley KG, Grossman MI: Maximal histalog test in control subjects and patients with peptic ulcer. Gut 6:427-435, 1965 16. Hood JM , Spencer EFA, MacRae KD, eta!: Predictive value of perioperative gastric acid tests. Gut 17:998-1000, 1976 17. Card WI, Marks IN: The relationship between acid output of the stomach following "maximal" histamine stimulation and the parietal cell mass. Clin Sci 19:147-163 , 1960 18. Isenberg JI, Grossman MI, Maxwell V, eta!: Increased sensitivity to stimulation of acid secretion by pentagastrin in duodenal ulcer. J Clin Invest 55:330-337, 1975 19. McGuigan JE , Trudeau WL: Differences in rates of gastrin release in normal persons and patients with duodenal ulcer disease. N Eng! J Med 288:64-66, 1973 20. Korman MG, Soveny C, Hansky J : Serum gastrin in duodenal ulcer. Gut 12:899-902, 1971 21. Stern DH, Walsh JH: Gastrin release in postoperative ulcer patients: evidence for release of duodenal gastrin. Gastroenterol· ogy 64:363-369, 1973