Protective Effect of Cimetidine on Aspirin-Induced Gastric Mucosal Damage PETER A. MacKERCHER, M.D.; KEVIN J. IVEY, M.D., and WILLIAM J. KRAUSE,
Ph.D.; Columbia,
Aspirin alters the gastric mucosal barrier as measured by ionic flux and potential difference. The effect of cimetidine on aspirin-induced alterations in gastric mucosa was studied in five normal male volunteers. Aspirin effects were studied with and without previous treatment with cimetidine. Mean ( + SEM) basal potential difference was - 4 8 + 1 mV. After 6 0 0 mg of aspirin in 1 dl of isotonic saline, potential difference decreased in 10 min to - 3 9 ± 1 mV (P < 0.001) and returned to baseline within 60 min. Control biopsies showed 2 % damaged mucosal cells compared with 2 0 % damaged at the time of maximal drop in potential difference (P < 0.001) after aspirin. Recovery to 9 % damage occurred by 6 0 min. In subjects pretreated with 3 0 0 mg cimetidine, potential difference rose during 1 h to - 6 2 ± 1 mV (P < 0.001). After aspirin potential difference fell to - 4 8 ± 1 mV compared with - 3 9 + 1 mV with aspirin alone (P < 0.01) and returned to - 6 2 ± 1 mV at 60 min. The cimetidinetreated group showed 4 % mucosal damage at the peak potential difference fall after aspirin, significantly less (P < 0.02) than in the untreated subjects.
A S P I R I N has been shown to break the human gastric mucosal barrier, as evidenced by changes in ionic flux (1-4) and potential difference (5-7). Aspirin is well absorbed by gastric mucosal cells at p H values below its p K a of 3.5 (2, 8). Penetration of the mucosal cell by aspirin is considered the first step in mucosal damage, which is accentuated by back-diffusion of luminal hydrogen ions. Histamine has been implicated as a mediator of this injury, through increased back-diffusion of acid, and increased acid production (9). Cimetidine, an H-2 receptor antagonist, is known to inhibit histamine, vagal, and pentagastrin-stimulated acid secretion in man (10-13). Accurate evaluation of this mucosal damage is dependent on the sensitivity of the method used. Gastroscopic examination of subjects given 600 mg of aspirin orally showed lesions in only 17% of subjects (14). Studies of subjects given doses of 1 g of aspirin three times per day for 3 days have shown increased fecal blood loss in 5 0 % to 7 0 % of subjects (15). Microscopic examination of biopsy tissue has shown aspirin-induced damage in 100% of normal subjects tested after ingestion of only 600 mg of aspirin (5). • From the Departments of Medicine and Anatomy, Harry S. Truman Memorial Veterans Hospital; and the University of Missouri; Columbia, Missouri.
676
Downloaded from https://annals.org by Tulane University user on 01/13/2019
F.A.C.P.; WILLIAM IN. BASKIN,
M.D.;
Missouri
To evaluate whether cimetidine might protect against aspirin damage, we tested the effect of cimetidine pretreatment on aspirin-induced alterations in gastric mucosal structure and potential difference in healthy volunteers. Methods
Five healthy volunteers, aged 21 to 26 years, were studied four times at least 1 week apart, after informed consent was obtained. The studies were [1] continuous potential difference measurement before, during, and after aspirin administration; [2] gastric biopsy before, during, and after aspirin administration; and [3] repeat of the two previous studies after cimetidine administration. The studies were done in a random order, all at least 7 days apart and all after a 12-h fast. POTENTIAL DIFFERENCE
Potential difference was measured by methods previously described (5, 16). Polyethylene tubes filled with saturated KC1 in 3% agar were used as electrodes. A reference electrode, gas sterilized, was placed in a peripheral vein. The intraluminal electrode was attached to a nasogastric tube and placed fluoroscopically in the fundus near the greater curvature of the stomach. The electrodes were connected via KC1 beakers to a Keithley 610B electrometer (Keithley Instruments, Inc.; Cleveland, Ohio) and a strip chart recorder. The stomach was washed with 2 dl of isotonic saline to remove residual contents. One decilitre of isotonic saline was then instilled and potential difference recorded continuously for 30 min to determine a baseline value. The stomach was aspirated and 600 mg of powdered aspirin (Aldrich Chemical Co.; Milwaukee, Wisconsin) suspended in 1 dl of isotonic saline, pH 3.5, was instilled; dosage was chosen to be the rough equivalent of two aspirin tablets in a glass of water. After 30 min, gastric contents were again aspirated and 1 dl of isotonic saline instilled. At the end of another 30-min period the stomach was aspirated and the study terminated. All aspirates were tested for pH, using a Beckman Expandomatic meter (Beckman Instruments, Inc.; Fullerton, California). The study was then repeated on a separate day with cimetidine pretreatment. After the initial saline wash, 300 mg of cimetidine in 1 dl of isotonic saline was instilled and left for 1 h. Gastric contents were aspirated and 600 mg of aspirin suspended in 1 dl of isotonic saline administered. Potential difference was again recorded as outlined above. BIOPSY STUDY
Biopsy studies were carried out on a separate day using an established technique (5). A nasogastric tube was placed fluoroscopically in the fundus and the stomach washed with 2 dl of isotonic saline. One decilitre of saline was then introduced. After 30 min, the gastric contents were aspirated and a Quinton hydraulic multibiopsy tube (Quinton Instruments; Seattle, Washington) was passed and positioned fluoroscopically in the fundus. Duplicate biopsies were obtained with the tube being flushed with tyrode solution under C0 2 pressure. Then 600 mg of aspirin suspended in 1 dl of isotonic saline was introduced. Annals of Internal Medicine 87:676-679, 1977
After approximately 10 min a second set of duplicate biopsies was obtained. A third set of duplicate biopsies was taken 60 min after aspirin administration. The stomach was aspirated just before each biopsy and the pH of all aspirated solutions measured. On a separate day these studies were repeated, with 300 mg of cimetidine being administered 1 h before the first biopsy. The second biopsy was taken at 15 min after aspirin administration to correlate with maximum potential difference change. The third biopsy was done again at 60 min. TISSUE PREPARATION
Biopsies were fixed for 4 h in 3.5% gluteraldehyde buffered in 1.0 M phosphate to a pH of 7.4. They were washed in buffer and placed in 1.0% osmium tetroxide at 0 °C for 2 h (17). Tissue for light microscopy was processed routinely and embedded in Epon 812 (Fisher Scientific Co.; Fairlawn, New Jersey). Thick sections, 0.5 to 3 jutms, were cut and stained with toluidine blue. Duplicate specimens were similarly fixed and prepared for scanning electron microscopy. They were dehydrated in alcohol and then transferred to amyl acetate before critical point drying with liquid C0 2 substitution (18, 19). The dried tissues were placed on spinner stubs and coated with gold-palladium alloy to a depth of approximately 200 A in a vacuum evaporator. They were viewed with a Cambridge Stereoscan Mark II electron microscope (Cambridge Scientific Instruments Co.; Cambridge, England). QUANTITATIVE DAMAGE DETERMINATION
The light microscopic sections were coded and read in a blind manner by a histopathologist. The amount of damage was quantitated by counting 1000 consecutive surface epithelial cells per section and recording the percentage of damaged cells present (5). Cells were considered damaged if they showed apical membrane disruption associated with nuclear destruction (5). The entire blind biopsy sample (2 mm in diameter) was also examined by light microscopy for focal microscopic erosions (loss of surface epithelium). STATISTICS
The paired T-test was used to determine significance. Results GASTRIC PH
The mean p H values of gastric aspirates taken during the potential difference studies were: control aspirin study, basal p H was 2.3; when aspirin was administered, p H fell to 2.0 in 10 min and returned to 2.5 at 60 min after aspirin. In the cimetidine study the basal p H rose to
Figure 2. Graph of percentage of damaged cells seen with light microscopy per 1 0 0 0 cells counted. There was no significant difference in mucosal damage after cimetidine alone compared with saline control biopsy. After cimetidine pretreatment (hashed bar) followed by aspirin, significantly less damage occurred in 1 0 min than in untreated control subjects (P < 0 . 0 0 1 ) .
6.9. Aspirin administration resulted in a fall of p H to 3.8 at 10 min, which returned to 5.6 at 1 h. The values after cimetidine were significantly higher than after aspirin alone at each time period (P < 0.01). POTENTIAL DIFFERENCE
After saline instillation, basal potential difference (mean ± SEM) was — 48 ± 1 mV (Figure 1). The aspirin test solution caused a drop in potential difference to - 3 9 + 1 m V i n less than 10 min (P < 0.001), which recovered to normal in 30 to 60 min. After a similar saline baseline, cimetidine caused a significant rise in potential difference to —63 + 1.9 mV (P < 0.001) within 30 min, which persisted at least 30 min. After this, aspirin instillation caused a drop in potential difference to — 47 ± 3 mV. The resultant potential difference after aspirin was not significantly different from the basal potential difference of — 48 + 1 mV of either study. Moreover, it was significantly higher than the — 39 ± 1 mV found in the aspirin control studies (P < 0.01). The time required for maximal aspirin-induced drop in potential difference was significantly prolonged in the cimetidine pretreated subjects, from 10 ± 0.6 min to 19 ± 1.5 min ( P < 0.01). MICROSCOPIC D A M A G E
Figure 1 . Gastric potential difference (PD) before and after cimetidine, 3 0 0 mg in 1 dl of normal saline {dark line). Mucosal PD is negative to serosal. Note rise in PD after cimetidine (mean ± SEM, P < 0 . 0 0 1 ) . The PD in cimetidine-treated subjects (dark line) was significantly higher than in control subjects (dashed line) at ail time intervals after aspirin (P < 0 . 0 1 ) .
Baseline biopsies of normal gastric mucosa, taken before aspirin administration, showed only 2.4% abnormal surface epithelial cells (Figure 2). Biopsies taken 10 min after aspirin administration showed 19% cell damage (P < 0.001). One hour after aspirin the percentage of damaged cells had decreased to 9 % (P < 0.001). Cimetidine treatment for 1 h caused no histologic change in surface epithelial cells. Biopsies postcimetidine showed 2 . 3 % damage, identical to those obtained in the preaspirin control period (Figure 2). Aspirin administration after cimetidine increased cell damage to only 4.4% (P < 0.05). This is significantly less than the 19% damage seen without cimetidine pretreatment (P < 0.001). MacKercher
Downloaded from https://annals.org by Tulane University user on 01/13/2019
et al. • Cimetidine and Aspirin-Induced Mucosal Damage
677
Figure 3A. Scanning electron micrograph of normal gastric mucosa. Foveolae (F) are shown, with individual rounded cell apices. (Original magnification, X 5 0 0 . ) B. Scanning electron micrograph of mucosal surface 15 min after aspirin instillation. Note the honey-combed appearance (arrows) due to foci of apical cell damage. (Original magnification, X 5 0 0 . ) C. Scanning electron micrograph of mucosal surface 15 min after aspirin, but after cimetidine pretreatment. Note the qualitative decrease in apical cell damage seen in Figure 3B. (Original magnification, X 5 0 0 . )
At 1 h after aspirin, 6% damage was seen, which was still significantly less than at 10 min after aspirin alone (P < 0.05). Figure 3A shows scanning electron microscopy of a representative gastric biopsy obtained in the control period. On light microscopy the mucous granules are located apically and cell membranes and nuclei are intact (5). Scanning electron microscopy shows the regular cobblestone appearance of the mucosal cell apices. Figure 3B shows representative biopsies taken after aspirin administration when no cimetidine had been given. Light microscopy showed many cells that had undergone damage, demonstrated by disruption of apical membranes and loss of mucous granules. Nuclear degeneration was present in many of these cells, evidence for cell death (5). Scanning electron microscopy of duplicate biopsies (Figure 31?) shows a honeycomb pattern of holes created by loss of these cell apices. Figure 3C shows representative biopsies taken after aspirin administration when the subject had been pretreated with cimetidine. Though damaged cells can be seen on scanning electron microscopy, the degree of mucosal damage is qualitatively less, reflecting the quantitative protective effect shown in Figure 2. EROSIONS
Microscopic erosions (complete loss of the surface epithelial cells over an area) were also counted. With aspirin alone, four of five subjects were found to have erosions. When cimetidine had been administered, only one of five subjects was found to have an erosion. In this subject gastric pH was 3.5 at 10 min after aspirin during the biopsy study whereas in all other subjects gastric pH was greater than 3.5 at that time period. 678
December
1977
• Annals of Internal Medicine • Volume 87 • Number
Downloaded from https://annals.org by Tulane University user on 01/13/2019
Discussion
The gastric mucosal barrier is a physiologic concept of mucosal defense against disease or drug-induced damage. Aspirin is an established "barrier breaker" at a gastric pH less than its dissociation constant of 3.5, as shown by ion flux, potential difference, and histologic studies (1-7, 16, 20). Cimetidine, by inhibiting acid secretion, maintained intragastric pH above this level, at a mean of 3.8 during the potential difference studies and 4.5 during the biopsy studies. At these pH values less than 50% of the administered aspirin was in the lipid soluble undissociated state necessary for absorption by the epithelial cells. As cellular absorption of aspirin is required for damage (1), less injury would be anticipated. The amount of damage is proportional to the H + ion concentration present (1) and by reducing this from mean 12.8 meq/litre (pH, 2.0) in controls, to 0.3 meq/litre (pH, 3.8) after cimetidine, damage should be further reduced. Because at pH 3.8 some aspirin is in the undissociated state, it could be anticipated that aspirin might still cause some damage. As shown in Figure 2, our results strongly support these concepts; cimetidine dramatically protected against aspirin damage. The percentage of damaged cells dropped from 19% with aspirin alone to only 4% after treatment with cimetidine. Apart from their acid inhibitory action it is tempting to speculate that H-2 blockers may protect by means of their antihistamine effect. Histamine is considered a mediator of aspirin-induced injury, promoting increased acid output and hydrogen ion back-diffusion (increased ionic permeability) (9, 21). Some animal studies support the hypothesis that H-2 blockers may have an effect on ionic permeability (22) and protect against mucosal ulceration independent of their inhibitory effect on acid se6
cretion (23, 24). On the other hand, O'Brien and Carter (25) were unable to show any reduction in ionic permeability to H + and Na- ions after H-2 blockers in dogs, a finding we have confirmed in unpublished studies in man. The elevation of potential difference by this H-2 blocker suggests either increased active ion transport (of chloride ion into the gastric lumen or sodium ion out of the lumen) or decreased permeability of ions across the mucosal surface (26). T h e ability of aspirin (which increases ionic permeability) to cause the same absolute fall in potential difference in the presence of cimetidine argues against cimetidine's decreasing ionic permeability. This supports the concept that cimetidine protection is based predominantly on reducing acid secretion, limiting the number of £T ions available to back-diffuse through the damaged mucosa. The establishment of increased potential difference across the mucosa (lumen more electronegative) does have an inhibiting effect on cation (hydrogen ion) movement out of the gastric lumen into the gastric mucosal cells (27). In this respect, the increased potential difference after the H-2 blocker would tend to reduce hydrogen ion back-diffusion and so reduce aspirin-induced cell damage. Although potential difference fell after aspirin in the presence of cimetidine, the level did not fall below normal. Therefore, the electronegative effect of potential difference restricting hydrogen ion back-diffusion was still present. In conclusion, the H-2 blocker, cimetidine, given orally, protects gastric mucosa from aspirin damage. Both gastric p H and potential difference are increased. The elevation in potential difference after cimetidine appears due to changes in active transport of Na + or Cl~ ions rather than to decreased ionic permeability. Cimetidine protection against aspirin damage seems mainly related to decreased acid secretion.
4. S M I T H BM, SKILLMAN JJ, E D W A R D S BG, SILEN W: Permeability of the
human gastric mucosa. Alteration by acetylsalicylic acid and ethanol. TV EnglJMed 285:716-721, 1971 5. BASKIN W N , I V E Y KJ, K R A U S E WJ, G E M M E L L RT: Aspirin-induced
ultrastructural changes in human gastric mucosa. Correlation with potential difference. Ann Intern Med 85:299-303, 1976 6. G E A L L M G , P H I L L I P S SE, SUMMERSKILL WH: The profile of gastric
potential difference in man: effect of aspirin, alcohol, bile and exogenous acid. Gastroenterology 58:437-443, 1970 7. M U R R A Y HS, STROTTMEN MP, COOKE AR: Effect of several drugs on gastric potential difference in man. Br Med J 1:19-2!, 1974 8. H O G B E N CAM, SCHANKER LS, T o c c o DJ, B R O D I E BB: Absorption of
drugs from the stomach. II. The human. / Pharmacol Exp Ther 120:540-545, 1957 9. D A V E N P O R T HW: Fluid produced by the gastric mucosa during damage by acetic and salicylic acids. Gastroenterology 50:487-499, 1966 10.
14. T H O R S E N WB, W E S T E R N D, T A N A K A Y, MORRISSEY J F : Aspirin inju-
ry to the gastric mucosa. Arch Intern Med 121:499-506, 1968 15. GROSSMAN MI, MATSUMOTO KK, LICHTER RJ: Fecal blood loss produced by oral and intravenous administration of various salicylates. Gastroenterology 40:383-388, 1961 16. IVEY KJ, PARSONS C, W E A T H E R B Y R: Effect of prednisolone and salicylic acid on ionic flux studies across the human stomach. Aust NZ J Med 5:408-412, 1975 17. MILLONIG G: Advantages of a phosphate buffer for Os0 4 solutions in fixation. J Appl Physics 32:1637, 1961 18. M E L L E R SM, C O P P E MR, I T O S, W A T E R M A N RE: Transmission elec-
tron microscopy of critical point dried tissue after observation in the scanning electron microscope. Anat Rec 176:245-252, 1973 19. K R A U S E WJ, LEESON CR: The postnatal development of the respiratory system of the opossum. Light and scanning electron microscopy. Am J Anat 137:337-356, 1973 20. IVEY KJ: Gastric mucosal barrier. Gastroenterology 61:247-257, 1971 21. JOHNSON LR, O V E R H O L T BF: Release of histamine into gastric venous blood following injury by acetic or salicylic acid. Gastroenterology 52:505-509, 1967 22.
ACKNOWLEDGMENTS: Received 21 March 1977; revision accepted 8 August 1977.
23.
• Requests for reprints should be addressed to Kevin J. Ivey, M.D., F.A.C.P.; Division of Gastroenterology, Department of Medicine, University of Missouri; Columbia, MO 65201.
24.
25.
References 1. D A V E N P O R T HW: Gastric mucosal injury by fatty and acetylsalicylic acids. Gastroenterology 46:245-253, 1964 2. IVEY KJ, MORRISON S, G R A Y C: Effect of salicylates on the gastric mucosal barrier in man. J Appl Physiol 33:81-85, 1972 3. O V E R H O L T BF, POLLARD HM: Acetylsalicylic acid and ionic fluxes across the gastric mucosa of man. Gastroenterology 54:538-542, 1968
H E N N BM, ISENBERG JI, M A X W E L L V, S T U R D E V A N T R A L : Inhibition
of gastric acid secretion by cimetidine in patients with duodenal ulcer. TV EnglJMed 293:371-375, 1975 11. IVEY KJ: Control of gastric acid secretion by histamine H-2 receptor antagonists; a pharmacological dream come true? Aust NZ J Med 4:521524, 1974 12. LONGSTRETH GF, G o VLW, MALAGELADA J-R: Cimetidine suppresion of nocturnal gastric secretion in active duodenal ulcer. N Engl J M e d 294:801-804, 1976 13. IVEY KJ, BASKIN W, JEFFREY G: Effect of cimetidine on gastric potential difference in man. Lancet 2:1072-1073, 1975
26.
27.
R E E S W D , R H O D E S J, W H E E L E R M H , M E E K EM, N E W C O M B E RG:
The role of histamine receptors in the pathophysiology of gastric mucosal damage. Gastroenterology 1'2:61-11, 1977 M A N N NS, SACHDEV A: Prevention of aspirin, ketoprofen and ibuprofen induced acute erosive gastritis by metiamide and cimetidine. Gastroenterology 60:914, 1976 BUGAJSKI J, H A N O J, D A N E K L: Effect of metiamide, a histamine H 2 -receptor antagonist, on the development of gastric stress ulcers and acid secretion. Eur J Pharmacol 36:237-240, 1976 O ' B R I E N PE, CARTER DC: Effect of gastric secretory inhibitors on the gastric mucosal barrier. Gut 16:437-442, 1975 K I T H A R A S, F o x KR, H O G B E N CA: Acid secretion, Na~ absorption, and the origin of the potential difference across isolated mammalian stomachs. Am J Dig Dis 14:221-238, 1969 D U R B I N RP: Electrical potential difference of the gastric mucosa, in Handbook of Physiology, Alimentary Canal, Vol. II Secretion, edited by C O D E C. Washington, D.C., American Physiological Society, 1967, pp. 879-888
MacKercher et a/. • Cimetidine and Aspirin-Induced Mucosal Damage
Downloaded from https://annals.org by Tulane University user on 01/13/2019
679
Ph.D.; Columbia,
Aspirin alters the gastric mucosal barrier as measured by ionic flux and potential difference. The effect of cimetidine on aspirin-induced alterations in gastric mucosa was studied in five normal male volunteers. Aspirin effects were studied with and without previous treatment with cimetidine. Mean ( + SEM) basal potential difference was - 4 8 + 1 mV. After 6 0 0 mg of aspirin in 1 dl of isotonic saline, potential difference decreased in 10 min to - 3 9 ± 1 mV (P < 0.001) and returned to baseline within 60 min. Control biopsies showed 2 % damaged mucosal cells compared with 2 0 % damaged at the time of maximal drop in potential difference (P < 0.001) after aspirin. Recovery to 9 % damage occurred by 6 0 min. In subjects pretreated with 3 0 0 mg cimetidine, potential difference rose during 1 h to - 6 2 ± 1 mV (P < 0.001). After aspirin potential difference fell to - 4 8 ± 1 mV compared with - 3 9 + 1 mV with aspirin alone (P < 0.01) and returned to - 6 2 ± 1 mV at 60 min. The cimetidinetreated group showed 4 % mucosal damage at the peak potential difference fall after aspirin, significantly less (P < 0.02) than in the untreated subjects.
A S P I R I N has been shown to break the human gastric mucosal barrier, as evidenced by changes in ionic flux (1-4) and potential difference (5-7). Aspirin is well absorbed by gastric mucosal cells at p H values below its p K a of 3.5 (2, 8). Penetration of the mucosal cell by aspirin is considered the first step in mucosal damage, which is accentuated by back-diffusion of luminal hydrogen ions. Histamine has been implicated as a mediator of this injury, through increased back-diffusion of acid, and increased acid production (9). Cimetidine, an H-2 receptor antagonist, is known to inhibit histamine, vagal, and pentagastrin-stimulated acid secretion in man (10-13). Accurate evaluation of this mucosal damage is dependent on the sensitivity of the method used. Gastroscopic examination of subjects given 600 mg of aspirin orally showed lesions in only 17% of subjects (14). Studies of subjects given doses of 1 g of aspirin three times per day for 3 days have shown increased fecal blood loss in 5 0 % to 7 0 % of subjects (15). Microscopic examination of biopsy tissue has shown aspirin-induced damage in 100% of normal subjects tested after ingestion of only 600 mg of aspirin (5). • From the Departments of Medicine and Anatomy, Harry S. Truman Memorial Veterans Hospital; and the University of Missouri; Columbia, Missouri.
676
Downloaded from https://annals.org by Tulane University user on 01/13/2019
F.A.C.P.; WILLIAM IN. BASKIN,
M.D.;
Missouri
To evaluate whether cimetidine might protect against aspirin damage, we tested the effect of cimetidine pretreatment on aspirin-induced alterations in gastric mucosal structure and potential difference in healthy volunteers. Methods
Five healthy volunteers, aged 21 to 26 years, were studied four times at least 1 week apart, after informed consent was obtained. The studies were [1] continuous potential difference measurement before, during, and after aspirin administration; [2] gastric biopsy before, during, and after aspirin administration; and [3] repeat of the two previous studies after cimetidine administration. The studies were done in a random order, all at least 7 days apart and all after a 12-h fast. POTENTIAL DIFFERENCE
Potential difference was measured by methods previously described (5, 16). Polyethylene tubes filled with saturated KC1 in 3% agar were used as electrodes. A reference electrode, gas sterilized, was placed in a peripheral vein. The intraluminal electrode was attached to a nasogastric tube and placed fluoroscopically in the fundus near the greater curvature of the stomach. The electrodes were connected via KC1 beakers to a Keithley 610B electrometer (Keithley Instruments, Inc.; Cleveland, Ohio) and a strip chart recorder. The stomach was washed with 2 dl of isotonic saline to remove residual contents. One decilitre of isotonic saline was then instilled and potential difference recorded continuously for 30 min to determine a baseline value. The stomach was aspirated and 600 mg of powdered aspirin (Aldrich Chemical Co.; Milwaukee, Wisconsin) suspended in 1 dl of isotonic saline, pH 3.5, was instilled; dosage was chosen to be the rough equivalent of two aspirin tablets in a glass of water. After 30 min, gastric contents were again aspirated and 1 dl of isotonic saline instilled. At the end of another 30-min period the stomach was aspirated and the study terminated. All aspirates were tested for pH, using a Beckman Expandomatic meter (Beckman Instruments, Inc.; Fullerton, California). The study was then repeated on a separate day with cimetidine pretreatment. After the initial saline wash, 300 mg of cimetidine in 1 dl of isotonic saline was instilled and left for 1 h. Gastric contents were aspirated and 600 mg of aspirin suspended in 1 dl of isotonic saline administered. Potential difference was again recorded as outlined above. BIOPSY STUDY
Biopsy studies were carried out on a separate day using an established technique (5). A nasogastric tube was placed fluoroscopically in the fundus and the stomach washed with 2 dl of isotonic saline. One decilitre of saline was then introduced. After 30 min, the gastric contents were aspirated and a Quinton hydraulic multibiopsy tube (Quinton Instruments; Seattle, Washington) was passed and positioned fluoroscopically in the fundus. Duplicate biopsies were obtained with the tube being flushed with tyrode solution under C0 2 pressure. Then 600 mg of aspirin suspended in 1 dl of isotonic saline was introduced. Annals of Internal Medicine 87:676-679, 1977
After approximately 10 min a second set of duplicate biopsies was obtained. A third set of duplicate biopsies was taken 60 min after aspirin administration. The stomach was aspirated just before each biopsy and the pH of all aspirated solutions measured. On a separate day these studies were repeated, with 300 mg of cimetidine being administered 1 h before the first biopsy. The second biopsy was taken at 15 min after aspirin administration to correlate with maximum potential difference change. The third biopsy was done again at 60 min. TISSUE PREPARATION
Biopsies were fixed for 4 h in 3.5% gluteraldehyde buffered in 1.0 M phosphate to a pH of 7.4. They were washed in buffer and placed in 1.0% osmium tetroxide at 0 °C for 2 h (17). Tissue for light microscopy was processed routinely and embedded in Epon 812 (Fisher Scientific Co.; Fairlawn, New Jersey). Thick sections, 0.5 to 3 jutms, were cut and stained with toluidine blue. Duplicate specimens were similarly fixed and prepared for scanning electron microscopy. They were dehydrated in alcohol and then transferred to amyl acetate before critical point drying with liquid C0 2 substitution (18, 19). The dried tissues were placed on spinner stubs and coated with gold-palladium alloy to a depth of approximately 200 A in a vacuum evaporator. They were viewed with a Cambridge Stereoscan Mark II electron microscope (Cambridge Scientific Instruments Co.; Cambridge, England). QUANTITATIVE DAMAGE DETERMINATION
The light microscopic sections were coded and read in a blind manner by a histopathologist. The amount of damage was quantitated by counting 1000 consecutive surface epithelial cells per section and recording the percentage of damaged cells present (5). Cells were considered damaged if they showed apical membrane disruption associated with nuclear destruction (5). The entire blind biopsy sample (2 mm in diameter) was also examined by light microscopy for focal microscopic erosions (loss of surface epithelium). STATISTICS
The paired T-test was used to determine significance. Results GASTRIC PH
The mean p H values of gastric aspirates taken during the potential difference studies were: control aspirin study, basal p H was 2.3; when aspirin was administered, p H fell to 2.0 in 10 min and returned to 2.5 at 60 min after aspirin. In the cimetidine study the basal p H rose to
Figure 2. Graph of percentage of damaged cells seen with light microscopy per 1 0 0 0 cells counted. There was no significant difference in mucosal damage after cimetidine alone compared with saline control biopsy. After cimetidine pretreatment (hashed bar) followed by aspirin, significantly less damage occurred in 1 0 min than in untreated control subjects (P < 0 . 0 0 1 ) .
6.9. Aspirin administration resulted in a fall of p H to 3.8 at 10 min, which returned to 5.6 at 1 h. The values after cimetidine were significantly higher than after aspirin alone at each time period (P < 0.01). POTENTIAL DIFFERENCE
After saline instillation, basal potential difference (mean ± SEM) was — 48 ± 1 mV (Figure 1). The aspirin test solution caused a drop in potential difference to - 3 9 + 1 m V i n less than 10 min (P < 0.001), which recovered to normal in 30 to 60 min. After a similar saline baseline, cimetidine caused a significant rise in potential difference to —63 + 1.9 mV (P < 0.001) within 30 min, which persisted at least 30 min. After this, aspirin instillation caused a drop in potential difference to — 47 ± 3 mV. The resultant potential difference after aspirin was not significantly different from the basal potential difference of — 48 + 1 mV of either study. Moreover, it was significantly higher than the — 39 ± 1 mV found in the aspirin control studies (P < 0.01). The time required for maximal aspirin-induced drop in potential difference was significantly prolonged in the cimetidine pretreated subjects, from 10 ± 0.6 min to 19 ± 1.5 min ( P < 0.01). MICROSCOPIC D A M A G E
Figure 1 . Gastric potential difference (PD) before and after cimetidine, 3 0 0 mg in 1 dl of normal saline {dark line). Mucosal PD is negative to serosal. Note rise in PD after cimetidine (mean ± SEM, P < 0 . 0 0 1 ) . The PD in cimetidine-treated subjects (dark line) was significantly higher than in control subjects (dashed line) at ail time intervals after aspirin (P < 0 . 0 1 ) .
Baseline biopsies of normal gastric mucosa, taken before aspirin administration, showed only 2.4% abnormal surface epithelial cells (Figure 2). Biopsies taken 10 min after aspirin administration showed 19% cell damage (P < 0.001). One hour after aspirin the percentage of damaged cells had decreased to 9 % (P < 0.001). Cimetidine treatment for 1 h caused no histologic change in surface epithelial cells. Biopsies postcimetidine showed 2 . 3 % damage, identical to those obtained in the preaspirin control period (Figure 2). Aspirin administration after cimetidine increased cell damage to only 4.4% (P < 0.05). This is significantly less than the 19% damage seen without cimetidine pretreatment (P < 0.001). MacKercher
Downloaded from https://annals.org by Tulane University user on 01/13/2019
et al. • Cimetidine and Aspirin-Induced Mucosal Damage
677
Figure 3A. Scanning electron micrograph of normal gastric mucosa. Foveolae (F) are shown, with individual rounded cell apices. (Original magnification, X 5 0 0 . ) B. Scanning electron micrograph of mucosal surface 15 min after aspirin instillation. Note the honey-combed appearance (arrows) due to foci of apical cell damage. (Original magnification, X 5 0 0 . ) C. Scanning electron micrograph of mucosal surface 15 min after aspirin, but after cimetidine pretreatment. Note the qualitative decrease in apical cell damage seen in Figure 3B. (Original magnification, X 5 0 0 . )
At 1 h after aspirin, 6% damage was seen, which was still significantly less than at 10 min after aspirin alone (P < 0.05). Figure 3A shows scanning electron microscopy of a representative gastric biopsy obtained in the control period. On light microscopy the mucous granules are located apically and cell membranes and nuclei are intact (5). Scanning electron microscopy shows the regular cobblestone appearance of the mucosal cell apices. Figure 3B shows representative biopsies taken after aspirin administration when no cimetidine had been given. Light microscopy showed many cells that had undergone damage, demonstrated by disruption of apical membranes and loss of mucous granules. Nuclear degeneration was present in many of these cells, evidence for cell death (5). Scanning electron microscopy of duplicate biopsies (Figure 31?) shows a honeycomb pattern of holes created by loss of these cell apices. Figure 3C shows representative biopsies taken after aspirin administration when the subject had been pretreated with cimetidine. Though damaged cells can be seen on scanning electron microscopy, the degree of mucosal damage is qualitatively less, reflecting the quantitative protective effect shown in Figure 2. EROSIONS
Microscopic erosions (complete loss of the surface epithelial cells over an area) were also counted. With aspirin alone, four of five subjects were found to have erosions. When cimetidine had been administered, only one of five subjects was found to have an erosion. In this subject gastric pH was 3.5 at 10 min after aspirin during the biopsy study whereas in all other subjects gastric pH was greater than 3.5 at that time period. 678
December
1977
• Annals of Internal Medicine • Volume 87 • Number
Downloaded from https://annals.org by Tulane University user on 01/13/2019
Discussion
The gastric mucosal barrier is a physiologic concept of mucosal defense against disease or drug-induced damage. Aspirin is an established "barrier breaker" at a gastric pH less than its dissociation constant of 3.5, as shown by ion flux, potential difference, and histologic studies (1-7, 16, 20). Cimetidine, by inhibiting acid secretion, maintained intragastric pH above this level, at a mean of 3.8 during the potential difference studies and 4.5 during the biopsy studies. At these pH values less than 50% of the administered aspirin was in the lipid soluble undissociated state necessary for absorption by the epithelial cells. As cellular absorption of aspirin is required for damage (1), less injury would be anticipated. The amount of damage is proportional to the H + ion concentration present (1) and by reducing this from mean 12.8 meq/litre (pH, 2.0) in controls, to 0.3 meq/litre (pH, 3.8) after cimetidine, damage should be further reduced. Because at pH 3.8 some aspirin is in the undissociated state, it could be anticipated that aspirin might still cause some damage. As shown in Figure 2, our results strongly support these concepts; cimetidine dramatically protected against aspirin damage. The percentage of damaged cells dropped from 19% with aspirin alone to only 4% after treatment with cimetidine. Apart from their acid inhibitory action it is tempting to speculate that H-2 blockers may protect by means of their antihistamine effect. Histamine is considered a mediator of aspirin-induced injury, promoting increased acid output and hydrogen ion back-diffusion (increased ionic permeability) (9, 21). Some animal studies support the hypothesis that H-2 blockers may have an effect on ionic permeability (22) and protect against mucosal ulceration independent of their inhibitory effect on acid se6
cretion (23, 24). On the other hand, O'Brien and Carter (25) were unable to show any reduction in ionic permeability to H + and Na- ions after H-2 blockers in dogs, a finding we have confirmed in unpublished studies in man. The elevation of potential difference by this H-2 blocker suggests either increased active ion transport (of chloride ion into the gastric lumen or sodium ion out of the lumen) or decreased permeability of ions across the mucosal surface (26). T h e ability of aspirin (which increases ionic permeability) to cause the same absolute fall in potential difference in the presence of cimetidine argues against cimetidine's decreasing ionic permeability. This supports the concept that cimetidine protection is based predominantly on reducing acid secretion, limiting the number of £T ions available to back-diffuse through the damaged mucosa. The establishment of increased potential difference across the mucosa (lumen more electronegative) does have an inhibiting effect on cation (hydrogen ion) movement out of the gastric lumen into the gastric mucosal cells (27). In this respect, the increased potential difference after the H-2 blocker would tend to reduce hydrogen ion back-diffusion and so reduce aspirin-induced cell damage. Although potential difference fell after aspirin in the presence of cimetidine, the level did not fall below normal. Therefore, the electronegative effect of potential difference restricting hydrogen ion back-diffusion was still present. In conclusion, the H-2 blocker, cimetidine, given orally, protects gastric mucosa from aspirin damage. Both gastric p H and potential difference are increased. The elevation in potential difference after cimetidine appears due to changes in active transport of Na + or Cl~ ions rather than to decreased ionic permeability. Cimetidine protection against aspirin damage seems mainly related to decreased acid secretion.
4. S M I T H BM, SKILLMAN JJ, E D W A R D S BG, SILEN W: Permeability of the
human gastric mucosa. Alteration by acetylsalicylic acid and ethanol. TV EnglJMed 285:716-721, 1971 5. BASKIN W N , I V E Y KJ, K R A U S E WJ, G E M M E L L RT: Aspirin-induced
ultrastructural changes in human gastric mucosa. Correlation with potential difference. Ann Intern Med 85:299-303, 1976 6. G E A L L M G , P H I L L I P S SE, SUMMERSKILL WH: The profile of gastric
potential difference in man: effect of aspirin, alcohol, bile and exogenous acid. Gastroenterology 58:437-443, 1970 7. M U R R A Y HS, STROTTMEN MP, COOKE AR: Effect of several drugs on gastric potential difference in man. Br Med J 1:19-2!, 1974 8. H O G B E N CAM, SCHANKER LS, T o c c o DJ, B R O D I E BB: Absorption of
drugs from the stomach. II. The human. / Pharmacol Exp Ther 120:540-545, 1957 9. D A V E N P O R T HW: Fluid produced by the gastric mucosa during damage by acetic and salicylic acids. Gastroenterology 50:487-499, 1966 10.
14. T H O R S E N WB, W E S T E R N D, T A N A K A Y, MORRISSEY J F : Aspirin inju-
ry to the gastric mucosa. Arch Intern Med 121:499-506, 1968 15. GROSSMAN MI, MATSUMOTO KK, LICHTER RJ: Fecal blood loss produced by oral and intravenous administration of various salicylates. Gastroenterology 40:383-388, 1961 16. IVEY KJ, PARSONS C, W E A T H E R B Y R: Effect of prednisolone and salicylic acid on ionic flux studies across the human stomach. Aust NZ J Med 5:408-412, 1975 17. MILLONIG G: Advantages of a phosphate buffer for Os0 4 solutions in fixation. J Appl Physics 32:1637, 1961 18. M E L L E R SM, C O P P E MR, I T O S, W A T E R M A N RE: Transmission elec-
tron microscopy of critical point dried tissue after observation in the scanning electron microscope. Anat Rec 176:245-252, 1973 19. K R A U S E WJ, LEESON CR: The postnatal development of the respiratory system of the opossum. Light and scanning electron microscopy. Am J Anat 137:337-356, 1973 20. IVEY KJ: Gastric mucosal barrier. Gastroenterology 61:247-257, 1971 21. JOHNSON LR, O V E R H O L T BF: Release of histamine into gastric venous blood following injury by acetic or salicylic acid. Gastroenterology 52:505-509, 1967 22.
ACKNOWLEDGMENTS: Received 21 March 1977; revision accepted 8 August 1977.
23.
• Requests for reprints should be addressed to Kevin J. Ivey, M.D., F.A.C.P.; Division of Gastroenterology, Department of Medicine, University of Missouri; Columbia, MO 65201.
24.
25.
References 1. D A V E N P O R T HW: Gastric mucosal injury by fatty and acetylsalicylic acids. Gastroenterology 46:245-253, 1964 2. IVEY KJ, MORRISON S, G R A Y C: Effect of salicylates on the gastric mucosal barrier in man. J Appl Physiol 33:81-85, 1972 3. O V E R H O L T BF, POLLARD HM: Acetylsalicylic acid and ionic fluxes across the gastric mucosa of man. Gastroenterology 54:538-542, 1968
H E N N BM, ISENBERG JI, M A X W E L L V, S T U R D E V A N T R A L : Inhibition
of gastric acid secretion by cimetidine in patients with duodenal ulcer. TV EnglJMed 293:371-375, 1975 11. IVEY KJ: Control of gastric acid secretion by histamine H-2 receptor antagonists; a pharmacological dream come true? Aust NZ J Med 4:521524, 1974 12. LONGSTRETH GF, G o VLW, MALAGELADA J-R: Cimetidine suppresion of nocturnal gastric secretion in active duodenal ulcer. N Engl J M e d 294:801-804, 1976 13. IVEY KJ, BASKIN W, JEFFREY G: Effect of cimetidine on gastric potential difference in man. Lancet 2:1072-1073, 1975
26.
27.
R E E S W D , R H O D E S J, W H E E L E R M H , M E E K EM, N E W C O M B E RG:
The role of histamine receptors in the pathophysiology of gastric mucosal damage. Gastroenterology 1'2:61-11, 1977 M A N N NS, SACHDEV A: Prevention of aspirin, ketoprofen and ibuprofen induced acute erosive gastritis by metiamide and cimetidine. Gastroenterology 60:914, 1976 BUGAJSKI J, H A N O J, D A N E K L: Effect of metiamide, a histamine H 2 -receptor antagonist, on the development of gastric stress ulcers and acid secretion. Eur J Pharmacol 36:237-240, 1976 O ' B R I E N PE, CARTER DC: Effect of gastric secretory inhibitors on the gastric mucosal barrier. Gut 16:437-442, 1975 K I T H A R A S, F o x KR, H O G B E N CA: Acid secretion, Na~ absorption, and the origin of the potential difference across isolated mammalian stomachs. Am J Dig Dis 14:221-238, 1969 D U R B I N RP: Electrical potential difference of the gastric mucosa, in Handbook of Physiology, Alimentary Canal, Vol. II Secretion, edited by C O D E C. Washington, D.C., American Physiological Society, 1967, pp. 879-888
MacKercher et a/. • Cimetidine and Aspirin-Induced Mucosal Damage
Downloaded from https://annals.org by Tulane University user on 01/13/2019
679
