Protection Against Aspirin-Induced Antral and Duodenal Damage With Enprostil

Transcription

1 GASTROENTEROLOGY 1985;88:382-6 Protection Against Aspirin-Induced Antral and Duodenal Damage With Enprostil A Double-Blind Endoscopic Study MAX M. COHEN, DAVID R. McCREADY, LORENNA CLARK, and HILLI SEVELIUS Department of Surgery, University of Toronto, and Mount Sinai Hospital, Toronto, Ontario, Canada, and Syntex Research, Inc., Palo Alto, California Prostaglandins protect against aspirin-induced damage to the gastrointestinal tract. This study tested the ability of enprostil, a synthetic analog of prostaglandin E 2, given concurrently to prevent gastroduodenal injury. Twenty-four healthy subjects were randomly assigned to one of three groups. All received aspirin 650 mg q.i.d. for 5 days. One group received placebo and the other groups were given either 7 or 70 p.,g of en pros til b.i.d. for 5 days. Upper endoscopy was performed at entry and 2 h after the final dose of aspirin. Enprostil 70 p.,g b.i.d. afforded significant protection of both the antral and duodenal mucosa. The 7-p.,g dose protected only the antral mucosa. Side effects were not observed with the lower dose of enprostil. Serum salicylate levels did not differ significantly between the groups. Prostaglandins, found normally in the human gastric mucosa, have been shown to possess an array of potentially protective properties (1). In particular, they are extremely potent antiulcer agents even in the absence of acid inhibition (2). This phenomenon has been termed "cytoprotection," but its mechanism is not clearly understood (3). Synthetic analogs of prostaglandins in acid-inhibitory doses are now being subjected to trial as therapeutic agents for peptic ulcer. We have shown previ- Received March 16, Accepted May 14, Address requests for reprints to: Dr. M. M. Cohen, 600 University Avenue, Suite 440, Toronto, Ontario, M5G lx5, Canada. This work was supported by the Medical Research Council of Canada, grant No. MT-5316, and by grants from the Mount Sinai Institute and Syntex Research. The authors thank Ms. Cedma Clyne, R.N. for technical assistance and Ms. Helen Gigas for secretarial assistance by the American Gastroenterological Association /85/$3.30 ously that orally administered natural prostaglandin E2 (PGE 2 ) (which does not inhibit acid secretion in humans) can prevent aspirin damage to the human gastric mucosa as determined by fecal blood loss (4) and potential difference (5). In most of the studies of prostaglandin protection in animals and in humans, the prostaglandin was given as pretreatment before administration of the damaging agent. We have reported that simultaneous administration of either natural PGE 2 or a synthetic analog is equally effective (6,7). The only published endoscopic study of prostaglandin protection against aspirin injury used a 24-h treatment period followed by a single dose of aspirin (8). The purposes of this study were as follows: (a) to directly and separately assess aspirininduced mucosal damage of fundus, antrum, and duodenum endoscopically; (b) to determine if a synthetic prostaglandin analog given both at potent acid-inhibitory and weak acid-inhibitory doses could provide protection; and (c) to determine if the concurrent administration of this analog would be effective in reducing aspirin damage. Materials and Methods Twenty-four healthy (71.8 ± 2.3 kg) subjects (19 men and 5 women) aged yr (mean 27.4 ± 1.2 yr) participated in the study, which was approved by the University of Toronto Human Experimentation Committee on September 20, All subjects provided informed written consent. The subjects selected had no significant past medical history nor any symptoms related to the gastrointestinal tract. No functional abnormalities were detected after a thorough physical and laboratory examination that included complete blood count, SMA 12, and urinalysis. Female volunteers had a negative serum human chorionic gonadotrophin J3-subunit assay pregnancy test. Serum salicylate concentration was measured by high-

2 January 1985 PROSTAGLANDIN AND ASPIRIN INJURY 383 pressure liquid chromatography and was initially zero in every case. No medications were allowed for at least 1 wk before entry. During the study period, all were instructed to abstain from coffee, alcohol, and all drugs not included in the study regimen. Smoking history was noted but subjects were not advised to alter their smoking habits. All subjects were screened by gastroduodenoscopy using an Olympus GIF Type Q gastroscope (Olympus Corporation of America, New Hyde Park, N.Y.). Two endoscopists agreed that the stomach and duodenum were entirely normal. At least 5 days and not more than 2 wk after the baseline endoscopy, subjects were randomly assigned to one of three groups. This time was allowed to be quite certain that aspirin would not be given to any subject before possible minor trauma from endoscopy had resolved. Two groups were given oral enprostil [( ± )-11", 15,,-dihydroxy-9- keto-16-phenoxy-17, 18, 19,20-tetranorprosta-4,5,13( + )-trienoic acid, methyl ester] prepared in a liquid vehicle (propylene carbonate) either as 7 or 70 f-lg b.i.d. before lunch and at bedtime. The third group received the vehicle only. All test medications were packaged identically and coded by the manufacturer (Syntex Research, Inc., Palo Alto, CaliL) so that neither subject nor investigator knew whether drug or placebo was being taken. In addition, all subjects were given 650 mg of aspirin (Bayer aspirin, Glenbrook Laboratories, New York, N.Y.) four times daily to be taken before meals and at bedtime. The prostaglandin was taken with the aspirin starting with the first dose of aspirin. Thus, each alternate aspirin dosage was taken alone. The first dose of enprostil and aspirin was taken before lunch on day 1. The last dose of enprostil was taken at bedtime on day 5 and the last dose of aspirin between 6 and 7 AM on day 6. Each dose of enprostil or placebo was prepared freshly for administration that day by adding 9 ml of purified water (USP) to the 1 ml of liquid contained in an amber glass vial. Each dose of medication was taken with 30 ml of water. All subjects received 20 doses of aspirin and 10 doses of the oral test solution. Subjects reported each day to the laboratory to receive the next day's study drugs and record any untoward symptoms. Side effects were monitored daily by the same nurse (L. C.). Care was taken to elicit symptoms not spontaneously offered. The time of onset, severity, and duration of each complaint was recorded. To prevent possible bias, the endoscopists did not question subjects about side effects. At the end of the 5-day treatment period, subjects again underwent endoscopy. All endoscopies were performed in a similar manner between 8 and 9 AM, ~ h 2 after the last dose of aspirin. Premedication consisted of topical 20% benzocaine to the oropharynx and intravenous diazepam (0-5 mg). The appearance of the fundus, antrum, and duodenum were sequentially and separately noted and scored independently and concurrently by the two endoscopists, one of whom used the teaching arm. Representative photographs were taken. The scoring system used to assess damage was a modification of that of Lanza et al. (9) (Table 1). For each subject, the two endoscopists' scores were averaged to yield one score for each of the anatomic regions and the sum of the regional scores was termed the "aggregate score." All endoscopies were per- Table 1. Endoscopic Scoring System Score Characteristics o No evidence of injury 1 One submucosal hemorrhage 2 Two or more isolated submucosal hemorrhages 3 Numerous areas of submucosal hemorrhage 4 Large area of submucosal hemorrhage with bleeding or a definite erosion or ulcer formed by the same two endoscopists (M. M. C. and D. R. McC.). Approximately 2 h after the second endoscopy, the physical examination and the laboratory evaluation of blood and urine were repeated. Although none of the subjects were alcoholics or even moderately heavy drinkers, serum ethanol was also determined to check compliance with the instruction that no alcohol be taken. Grouped data were nonparametrically analyzed by Kruskal-Wallis rank score analysis. Other data were tested by analysis of variance. Exact p values were obtained. Results There were no dropouts from the study. The three groups were comparable. No significant differences existed in age, sex, weight, or history of smoking. All subjects appeared to comply well with the study regimen. All subjects had serum salicylate concentrations on day 6 in the expected range. The mean salicylate level was 7.20 ± 0.58 mg/dl. No significant differences were found between the groups (p = 0.23). None of the subjects had detectable serum ethanol. Enprostil protected the gastroduodenal mucosa from aspirin damage. The aggregate scores of both treated groups (7 J.Lg, 4.5 ± 1.3; 70 J.Lg, 3.9 ± 1.0) were significantly lower (p < 0.05) than for placebo (8.0 ± 1.0). Although the group taking the 70-J.Lg dose of enprostil had the lowest score, there was no significant difference between the scores for the two doses of prostaglandin. Analysis of the regional scores is illustrated in Figure 1. The fundal damage was slightly reduced by increasing the dosage of enprostil but this was not significant (p = 0.3). Significant protection was afforded to the antral mucosa (p = 0.007). The doses of prostaglandin were equally effective. In the duodenum a dose-related effect was observed. A reduction in duodenal mucosal damage occurred with the low dose of enprostil but this was of marginal significance (p = 0.08). The protection afforded by the 70-J.Lg enprostil dose when compared with placebo was more significant (p = 0.03). Several of the subjects receiving 70 J.Lg of enprostil bj.d. experienced mild diarrhea that was transient despite continued administration of the drug. In the group receiving 7 J.Lg of enprostil, side effects were

3 384 COHEN ET AL. GASTROENTEROLOGY Vol. 88, No.1, Part ENDOSCOPIC 2 SCORE o PLACEBO ~ 7VQ ENPROSTll 810 FUNDUS ANTRUM DUODENUM Figure 1. Mean endoscopic scores for each anatomic region for each treatment group. The bars represent the standard errors of the means. There were 8 subjects in each group. *p < 0.01; +p < 0.05; compared with the corresponding control group. not reported. There were no significant changes in laboratory results in any subject. Discussion For more than a decade, it has been known that prostaglandins reduce gastric acid secretion in animals (10) and in humans (11). Prostaglandin protection against aspirin-induced damage to the canine gastric mucosal barrier was shown to be independent of gastric acid inhibition (12). Robert (2) showed that protection from gross hemorrhagic lesions caused by a wide variety of damaging agents such as acetylsalicylic acid and ethanol was afforded by orally administered prostaglandins in the absence of gastric acid inhibition, and termed this action "cytoprotection." This powerful protective property of the prostaglandins has been demonstrated using a variety of experimental models and has recently been reviewed (13). The mechanism of prostaglandin protection has been intensively investigated but remains poorly understood (3). Prostaglandin protection in humans has been less well documented. Natural PGEz affords mucosal barrier protection, from both aspirin and indomethacin, as measured by gastric transmucosal potential difference (6). Protection from this aspirin damage increases with the dose of prostaglandin (5). Prostaglandin Ez reduces the aspirin-induced rise in fecal blood loss (4) and this protection too is dose dependent (14). Prostaglandin Ez has also been shown to reduce gastric bleeding induced by aspirin as determined by the measurement of hemoglobin in gastric washings (15,16) and increased cell shedding (16,17). Prostaglandins have been shown to be effective against ethanol and bile salt injury in humans, and synthetic analogs of PGE have also been shown to protect the human stomach. These studies all used indirect methods to assess the protection afforded by orally administered prostaglandins. Aspirin consistently causes visible gastric mucosal damage in hu- mans (18-20). Gilbert and his associates (8) showed that 15-R-15-methyl PGEz given for 24 h before a single dose of 1300 mg of aspirin caused a significant reduction in injury as assessed endoscopically. Synthetic analogs of PGEz have been shown to be effective in the treatment of patients with peptic ulcer (21-23) and there have been anecdotal reports of their use in the treatment of resistant upper gastrointestinal bleeding (24--26). In this study we chose aspirin injury because it is well documented and reproducible. We selected a 5-day course of 650 mg q.i.d. as this is a dose level often used therapeutically. Furthermore, we had previously shown that exactly this regimen caused a significant increase in fecal blood loss in normal subjects (14). The data in that study (14) can therefore be compared with the data reported here. Gastroscopy is well tolerated by most subjects, often without any sedation, and is essentially free of risk. It is likely to become the standard technique for the assessment of gastric mucosal damage in humans replacing such indirect measurements as deoxyribonucleic acid output, hemoglobin content of gastric aspirate, potential difference, and fecal blood loss. We have now confirmed endoscopically our previous finding that concurrent prostaglandin administration protects human gastric mucosa from aspirin damage (5,14). There is some evidence that, as in the rat, the human gastric mucosa adapts to chronic aspirin administration. Serial endoscopy performed during a 7 -day course of aspirin showed that the peak injury probably was reached on the third day and that by the seventh day injury was less although not significantly so (27). It was noted in that study that erosions appeared only after 24 h of aspirin ingestion. Damage, however, has been observed as early as 1-2 h after a single dose of aspirin (8,19). Endoscopy after a 2-wk course of aspirin showed antral erosions in all subjects (19). Guth and his colleagues (28) were unable to demonstrate adaptation to aspirin by human gastric mucosa. It is possible that by choosing to submit our subjects to repeat endoscopy after 5 days of aspirin ingestion, we may have missed the peak injury. Nonetheless, in the placebo group there was significant injury to all three anatomic areas and this choice of timing did not prevent us from observing significant protection. Whether the protection we observed was due to increased adaptation is unknown. The clinical usefulness of natural PGEz is limited by its instability and the fact that doses that afford substantial protection are associated with diarrhea and nausea. Intense efforts are being made to develop potent prostaglandin analogs with specific gastrointestinal protective activity and minimal adverse

4 January 1985 PROSTAGLANDIN AND ASPIRIN INJURY 385 effects. Such analogs may have considerable potential in the prevention and treatment of peptic ulcer disease. Enprostil is a potent synthetic dihydro-pge z with gastric acid-inhibitory properties which have been defined in both animals and humans. It is stable for 2 yr at room temperature. In humans, a 1- /Lg/kg oral dose results in profound and prolonged gastric acid inhibition (29). A dose of 0.1 /Lg/kg does not cause clinically significant acid reduction (29). Both of these doses are free from severe side effects and appear to be safe. In this study, we used two doses of a new synthetic prostaglandin analog, one of which profoundly inhibits gastric secretion and the other of which has a weak effect on secretion. Both doses of enprostil protected the mucosa against aspirin damage. The reduction in damage to the fundal mucosa did not achieve statistical significance. Both doses significantly and equally protected the antral mucosa and in the duodenum protection was significant and dose-related. It is interesting that the higher dose of enprostil was more effective in the duodenum where presumably the acid load is more important. It is not clear why the fundal mucosa was not significantly protected and this regional variation, if confirmed by others, will require explanation. In the rat, the fundus is the major site of injury by most agents and the antrum is virtually spared. In humans, the fundus typically is less severely injured by aspirin than is the antrum (20,30-33). This wellrecognized clinical observation was confirmed in the present study. This difference in susceptibility to injury could perhaps be related to regional differences in the mucosal capacity to produce endogenous prostaglandins. The different capacities of the human fundal and antral mucosae have not been systematically studied, but there is evidence that the antrum synthesizes relatively more PGE z (34,35). The mechanism by which prostaglandin prevents damage remains unclear. Mechanisms suggested include a direct trophic effect, stimulation of mucus and bicarbonate, stimulation of cyclic adenosine monophosphate, an alteration of mucosal blood flow, an effect on the gastric mucosal barrier, a stabilizing effect on the cell membrane, and an effect on surface-active lipids (3). This trial confirms that acid inhibition is not an important component of the mechanism of prostaglandin protection of the antral mucosa in humans. Acid reduction may be important for protection of the duodenum. This study adds further support to the use of the stable synthetic prostaglandin analogs in the management of gastric and duodenal mucosal disease. These new agents should be tried not only in patients with peptic ulcer but also in the prevention and treatment of gastritis and acute mucosal lesions. References 1. Cohen MM. Gastric mucosal protection with prostaglandins. In: Pfeiffer q, ed. Drugs and peptic ulcers. Vol. 1. Boca Raton, Fla.: CRC Press, 1982: Robert A. Antisecretory, antiulcer, cytoprotective and diarrhoegenic properties of prostaglandins. Adv Prostaglandin Thromboxane Leukotriene Res 1976;2: Miller TA. Protective effects of prostaglandins against gastric mucosal damage: current knowledge and proposed mechanisms. Am J Physiol 1983;245:G Cohen MM, Cheung G, Lyster DM. Prevention of aspirin induced fecal blood loss in men with prostaglandin E z. Gut 1980;21: Cohen MM, Clark L. Preventing acetylsalicylic acid damage to human gastric mucosa by use of prostaglandin Ez. Can J Surg 1983;26: Cohen MM, Pollett JM. Prostaglandin E2 prevents aspirin and indomethacin damage to human gastric mucosa. Surg Forum 1976;27: Cohen MM, Clark L, Armstrong L, D'Souza J. Reduced aspirin-induced fecal blood loss with simultaneous administration of Misoprostil (PGE, methyl analogue) (abstr). Gastroenterology 1983;84: Gilbert DA, Surawicz CM, Silverstein FE, et al. Prevention of acute aspirin-induced gastric mucosal injury by 15-R-15- methyl prostaglandin Ez: an endoscopic study. Gastroenterology 1984;86: Lanza FL, Royer GL, Nelson RS. Endoscopic evaluation of the effect of aspirin, buffered aspirin, and enteric-coated aspirin on gastric and duodenal mucosa. N Engl J Med 1980;303: Robert A, Nezamis JE, Phillips JP. Inhibition of gastric secretion by prostaglandins. Am J Dig Dis 1967;12: Wilson DE, Phillips C, Levine RA. Inhibition of gastric secretion in man by prostaglandin A,. Gastroenterology 1971; 61: Cohen MM. Prostaglandin Ez prevents gastric mucosal barrier damage. Gastroenterology 1975;68: Johansson C. Bergstrom S. Prostaglandins and protection of the gastroduodenal mucosa. Scand J Gastroenterol 1982; 17(Suppl 77): Cohen MM. Prevention of aspirin-induced fecal blood loss with oral prostaglandin E2: dose-response studies in man. Prostaglandins 1981;21(Suppl): Hunt IN, Franz DR. Effect of prostaglandin E2 on gastric mucosal bleeding caused by aspirin. Dig Dis Sci 1981;26: Konturek SJ, Kwiecien N, Obtulowicz W, Polanski M, Kopp B, Oleksy J. Comparison of prostaglandin Ez and ranitidine in prevention of gastric bleeding by aspirin in man. Gut 1983;24: Ruppin H. Person B, Robert A, Domschke W. Gastric cytoprotection in man by prostaglandin Ez. Scand J Gastroenterol 1981;16: Cohen MM, MacDonald We. Mechanism of aspirin injury to human gastroduodenal mucosa. Prostaglandins Leukotrienes Med 1982;9: O'Laughlin JC, Hoftiezer JW, Ivey KJ. Effect of aspirin on the human stomach in normals: endoscopic comparison of damage produced one hour, 24 hours and 2 weeks after administration. Scand J GastroenteroI1981;16(Suppl 67): Hoftiezer JW, O'Laughlin IC. [vey KI. Effects of 24 hours of aspirin, Bufferin, paracetamol and placebo on normal human gastroduodenal mucosa. Gut 1982;23: Fung WP, Karim SMM, Tye CY. Effect of 15(R)-15-methyl

5 386 COHEN ET AL. GASTROENTEROLOGY Vol. 88, No.1, Part 2 prostaglandin E2 methyl ester on healing of gastric ulcers. Controlled endoscopic study. Lancet 1974;ii: Gibinski K, Rybicka J, Mikos E, Nowack A. Double-blind clinical trial on gastroduodenal ulcer healing with prostaglandin E2 analogues. Gut 1977;18: Vantrappen G, Janssens J, Popiela T, et al. Effect of 15(R)-15- methyl prostaglandin E2 (Arbaprostil) on the healing of duodenal ulcer. A double-blind multicenter study. Gastroenterology 1982;83: Groeger JS, Dazza S1, Carlon GC, Turnbull AS, Pierri MK, Howland WS. Prostaglandin therapy in a case of refractory stress ulcer bleeding. Crit Care Med 1982;10: Giannikopoulous I, Ringel p, Zai D, et al. Use of 15(R)-15- methyl prostaglandin E2 for the treatment of massive upper gastrointestinal bleeding in a critically ill patient. Gastrointest Endosc 1982;28: Weiss 1B, Peskin GW, Isenberg JI. Treatment of hemorrhagic gastritis with 15(R)-15-methyl prostaglandin E2: report of a case. Gastroenterology 1982;82: Graham DY, Smith 1L, Dobbs SM. Gastric adaptation occurs with aspirin administration in man. Dig Dis Sci 1983;28: Metzger WH, McAdam L, Bluestone R, Guth PH. Acute gastric mucosal injury during continuous or interrupted aspirin ingestion in humans. Dig Dis Sci 1976;21: Davis GR, Morawski SF, Fordtran 1S. Effect of synthetic PGE 2 on food stimulated gastric acid secretion. Clin Pharmaco l Therap 1982;31: Lanza FL, Royer GL, Nelson RS, Chen 'ft, Seckman CE, Rack MF. The effects of ibuprofen, indomethacin, aspirin, naproxen, and placebo on the gastric mucosa of normal volunteers. A gastroscopic and photographic study. Dig Dis Sci 1979;24: Silvoso GR, Ivey KJ, Butt JH, et al. Incidence of gastric lesions in patients with rheumatic disease on chronic aspirin therapy. Ann Intern Med 1979;91: Gillberg R, Korsan-Bergsten K, Magnusson B, Nyberg G. Gastrointestinal blood loss, gastroscopy and coagulation factors in normal volunteers during administration of acetylsalicylic acid and fluproquazone. Scand 1 Rheumatol 1981;10: Kilander A, Dotevall G. Endoscopic evaluation of the comparative effects of acetylsalicylic acid and choline magnesium trisalicylate on human gastric and duodenal mucosa. Br 1 Rheumatol 1983;22: Ligumsky M, Sharon p, Karmeli F, Rachmilewitz D. Prostaglandins and the pathogenesis of duodenal ulcer: no correlation with gastric mucosal PGE2 content. Isr J Med Sci 1979;15: Konturek S1, Kwiecien N, Obtulowicz W, et al. Effect of carprofen and indomethacin on gastric function, mucosal integrity and generation of prostaglandins in men. Hepatogastroenterology 1982;29: