Aliment Pharmacol Ther 2003; 17: 211 216. doi: 10.1046/j.0269-2813.2003.01416.x Effect of oral omeprazole in reducing re-bleeding in bleeding peptic ulcers: a prospective, double-blind, randomized, clinical trial M. J. KAVIANI*, M. R. HASHEMI, A. R. KAZEMIFAR*, S. ROOZITALAB*, A.-A. MOSTAGHNIà, S. MERAT, M. ALIZADEH-NAINI* & H. YARMOHAMMADI* *Gastroenterohepatology Research Center, Nemazee Hospital, Shiraz University of Medical Sciences, Shiraz, Iran; Bandar-Abbas University of Medical Sciences, Bandar-Abbas, Iran; àboushehr University of Medical Sciences, Boushehr, Iran; Digestive Diseases Research Center, Shariati Hospital, Tehran University of Medical Sciences, Tehran, Iran Accepted for publication 29 September 2002 SUMMARY Background: Endoscopic therapies and continuous intravenous omeprazole can decrease the morbidity and duration of hospital stay of patients with high-risk peptic ulcer. Aim: To evaluate the role of oral omeprazole in highrisk bleeders. Methods: After injection therapy of 160 patients with high-risk peptic ulcer, 80 received oral omeprazole and 80 received placebo, and all were followed up. Results: One hundred and forty-nine patients (71 omeprazole and 78 placebo) completed the study. Eleven patients were excluded from the study. Thirtyseven (25%) patients had gastric ulcer and 112 (75%) had duodenal ulcer. Fifty-seven (38%) ulcers showed visible vessels, 80 (54%) showed oozing of blood and 12 (8%) showed a spurting artery. Only one patient died (placebo group). The mean hospital stays were 62.8 ± 28.6 h and 75 ± 39 h in the omeprazole and placebo groups, respectively (P ¼ 0.032). The mean amounts of blood transfused were 1.13 ± 1.36 and 1.68 ± 1.68 bags in the omeprazole and placebo groups, respectively (P ¼ 0.029). The re-bleeding rate was lower in the omeprazole group than in the placebo group (12 vs. 26, respectively; P ¼ 0.022). Conclusion: Oral omeprazole is effective in decreasing the hospital stay, re-bleeding rate and the need for blood transfusion in high-risk ulcer bleeders treated with endoscopic injection therapy. INTRODUCTION Upper gastrointestinal bleeding is one of the most common emergencies in internal medicine, and is frequently accompanied by high morbidity and mortality. It affects about 0.1% of the general population. 1 The most common cause of upper gastrointestinal bleeding in adult patients is peptic ulcer disease, which accounts for about 50% of episodes. 2 Correspondence to: Dr M. J. Kaviani, Gastroenterohepatology Research Center, Nemazee Hospital, Shiraz University of Medical Sciences, P.O. Box 71935-1311, Shiraz, Iran. E-mail: kavianim@sums.ac.ir Upper gastrointestinal bleeding may occur in more than 20% of patients with peptic ulcer disease. 3 Most haemorrhages stop spontaneously and are not harmful, but in about 35% of cases there is an increased chance of re-bleeding and mortality. 4 The characteristics of the ulcer at endoscopy may provide the most helpful prognostic information. 4 Ulcers with oozing of blood from the ulcer bed, ulcers with a visible vessel and ulcers with spurting arteries harbour the risk of re-bleeding in 33%, 50% and 80 100% of occasions, respectively. 5, 6 Endoscopic treatment of high-risk ulcers can reduce re-bleeding, morbidity and even mortality. 7 9 Ó 2003 Blackwell Publishing Ltd 211
212 M. J. KAVIANI et al. Pepsin is active in acidic gastric juice (ph < 4) 10 and causes clot lysis. 11 Active pepsin also destroys the mucosal gel layers over the gastric mucosa. 12 An acid milieu not only disturbs primary haemostasis and clot formation (by the induction of platelet dysfunction) but, by destroying coagulation factors, can also disturb clot stabilization. 13 Therefore, the reduction of gastric acidity may be helpful in the prevention of re-bleeding. With the development of tolerance, histamine-2 receptor antagonists (H 2 RAs) cannot maintain their acid-suppressing activity; there is no evidence that such treatment reduces re-bleeding from duodenal ulcer, and only very weak evidence that it reduces re-bleeding from gastric ulcer. 14 Recently, it has been shown that continuous intravenous high doses of omeprazole reduce the re-bleeding and morbidity rates in high-risk patients with non-variceal upper gastrointestinal bleeding. 15 The role of oral omeprazole in reducing the re-bleeding rate, the need for surgery and the duration of hospital stay was studied by Khuroo et al. 5 and Javid et al. 16 However, problems were encountered in both studies. Khuroo did not select endoscopically high-risk patients and endoscopic therapy was not performed. 5 Javid et al. considered a large number of patients with adherent clots (37% of patients), but only applied gentle washing. A number of possible non-bleeding visible vessels might have been hidden behind these clots and might have caused bias in the study results. 16 In order to achieve a realistic outlook, we designed a prospective, double-blind, randomized, controlled study to compare oral high-dose omeprazole vs. placebo in high-risk peptic ulcer bleeders after endoscopic therapy. MATERIALS AND METHODS Study design Between April 1999 and May 2000, all adult patients with signs or symptoms of upper gastrointestinal bleeding (patients presenting with haematemesis, melaena, bloody nasogastric aspirate or melaena on rectal examination), who were admitted to the medical emergency rooms of Shiraz University of Medical Sciences Hospitals (Faghihi Hospital or Nemazee Hospital), were considered for inclusion in the study. Oesophago-gastro-duodenoscopy was performed within 12 h after admission. Inclusion criteria Patients older than 15 years in whom endoscopic treatment of actively bleeding ulcers or ulcers with non-bleeding visible vessels had been successful were eligible for the study. Exclusion criteria Low-risk bleeders (white-based ulcers, ulcers with a simple clot after washing with 500 cm 3 of tap water), cases in which the bleeding site was uncertain (for example, cases with oesophageal varices and peptic ulcers), cases with unknown or other sources of bleeding, patients currently taking antisecretory drugs (H 2 RAs or proton pump inhibitors) and patients with highly probable gastric malignancies (huge, dirty, undefined crater-like gastric ulcers or ulcers associated with a mass) were excluded from the study. Patients in whom endoscopic treatment was unsuccessful were not enrolled and instead underwent immediate surgery. Endoscopy and haemostasis Patients with high-risk peptic ulcers (ulcers with nonbleeding visible vessels, with oozing haemorrhage or with spurting haemorrhage) were treated endoscopically by injecting 5 20 cm 3 of adrenaline (epinephrine) (diluted 1 : 10 000) around the ulcer crater and on the ulcer bed (for large ulcers) to stop the bleeding. Haemostasis was considered to be established if bleeding vessels were flattened or cavitated. In patients with nonbleeding visible vessels, haemostasis was considered to be established when the vessel disappeared. Afterwards, a questionnaire, including information on demography, history of present and previous upper gastrointestinal bleeding, family history of upper gastrointestinal bleeding, concomitant disease, history of consumption of non-steroidal anti-inflammatory drugs, warfarin or corticosteroids, presence of Helicobacter pylori infection or eradication, physical examination, basic laboratory data, ongoing intravenous and blood transfusion and urine output, was completed for all high-risk patients. Therapy and randomization After completing the questionnaire, the patients were randomly divided into two groups. A pharmacist from Babak Pharmacy (M. E. Kaviani) filled the placebo
ORAL OMEPRAZOLE IN HIGH-RISK PEPTIC ULCERS 213 capsules with dextrose. The placebo capsules were identical in appearance to the omeprazole capsules and were provided by Exir Pharmaceutical Co. The capsules were then placed in sealed envelopes and coded based on random table numbers. Only the pharmacist knew the codes. The other medical personnel, endoscopists and patients were blind to this information. The study was conducted in a double-blind manner as all treatment assignments were revealed at the end of the study. The omeprazole group received oral omeprazole [Exir Pharmaceutical Co. (former name Lorestan Pharmaceutical Co.), Lorestan, Iran; copied from UQUIFA Pharmaceutical Co., Spain], 20 mg every 6 h. From the studies of Khuroo et al. and Javid et al., in which a dose 5, 16 of 40 mg twice daily was used, we hypothesized that four separate doses (20 mg every 6 h) would achieve better pharmacokinetics. The placebo group received placebo capsules every 6 h. position) or a decrease of more than 1 g dl in haemoglobin despite blood transfusion developed after the first endoscopic treatment. Patients who were suspected of re-bleeding were urgently re-endoscoped, which was usually performed 24 48 h after the first endoscopy. If a visible vessel, oozing of blood, spurting artery or non-bleeding visible vessel was seen, recurrent bleeding was confirmed and the patient was defined as high risk. Endoscopic therapy by injection of adrenaline (epinephrine) (diluted 1:10 000) was then repeated. Patients with no signs or symptoms of re-bleeding, who had been haemodynamically stable for the first 72 h and who had tolerated oral food were discharged. On the sixth day of admission or on discharge of the patient, oral omeprazole, 20 mg b.d., and antibiotics for the eradication of H. pylori (if positive for H. pylori) were started. All patients were followed up to the end of the third week. Follow-up Vital signs (blood pressure and pulse rate in the supine and sitting positions), intake of intravenous fluids and blood transfusions were monitored. The criteria for blood transfusion were as follows: for patients older than 60 years, blood was transfused if the haemoglobin was lower than 9.5 g dl or if the patient had suffered from two or more episodes of haematemesis; in younger patients, blood was transfused when the haemoglobin was lower than 8 g dl or if the patient had suffered from three or more episodes of haematemesis; if a state of shock existed, in both groups, blood was transfused independent of the haemoglobin level. However, the haemoglobin level was the most important criterion for blood transfusion. Urine output was monitored every hour in the first 48 h of treatment with oral omeprazole and placebo, and every 6 h thereafter until the patients were discharged. Haemoglobin was checked every 8 h until discharge. If a patient had received H 2 RAs, he or she was excluded from the study. Re-bleeding was considered to have occurred if haematemesis, shock (systolic blood pressure of less than 90 mmhg in the supine position or a pulse rate of more than 110 beats min), orthostatic hypotension (decrease of more than 20 mmhg of systolic or 10 mmhg of diastolic blood pressure, 3 min after changing from the supine to the sitting position, respectively), orthostatic tachycardia (increase of 10 beats min in the pulse rate, 3 min after changing from the supine to the sitting Statistical analysis In order to detect a 20% difference in the re-bleeding rate (30% vs. 10%), and with a power of 80% and a confidence interval of 95%, 160 patients were entered into the study (80 patients in the case group and 80 in the control group). All statistical analyses were performed using the software SPSSWIN-9. Statistical differences between quantitative and qualitative variables (i.e. proportions) were calculated by Student s t-test and the chi-squared test. A P value of less than 5% was considered to be statistically significant. The data were represented as the mean ± standard deviation (s.d.). We estimated the relative risk (RR) and 95% confidence interval (CI) associated with proportions. RESULTS Over an 8-month period, 149 patients (71 in the case group and 78 in the control group) completed the study (intravenous H 2 RAs were started in 11 patients and so they were excluded from the study). The mean age of the patients was 52.13 years (range, 15 92 years). Fifty-seven (38%) had visible vessels, 80 (54%) had oozing of blood and 12 (8%) had a spurting artery. The basic data of all the patients are shown in Table 1. Only one patient died (placebo group on the first day of admission), and two patients were operated on due to peptic ulcer bleeding
214 M. J. KAVIANI et al. Table 1. Basic data of the 149 patients studied Omeprazole group (n ¼ 71) Placebo group (n ¼ 78) Sex (male female) 57 14 64 14 Age (years) (mean ± s.d.) 52.79 ± 18.42 51.51 ± 18.96 NSAIDs, corticosteroids, warfarin use (%) 19 (26) 20 (26) Previous history of upper gastrointestinal bleeding (%) 13 (18) 9 (12) Previous gastric operation (%) 1 (1) 1 (1) History of concomitant disease (%) 16 (23) 27 (35) Types of concomitant disease Cerebral stroke (%) 1 (1) 2 (3) Ischaemic heart disease (%) 6 (8) 6 (8) Diabetes mellitus (%) 3 (4) 1 (1) Hypertension (%) 8 (11) 7 (9) Malignant diseases (%) 0 (0) 2 (3) Other diseases (%) 4 (6) 7 (9) Shock state (%) 3 (5) 6 (8) Orthostatic hypotension (%) 25 (35) 29 (37) Haemoglobin level (g dl) (mean ± s.d.) 10.2 ± 2.9 10.5 ± 3 Blood transfusion in the first 6 h (%) 17 (24) 18 (23) Units of blood transfused in the first 6 h (bag) (mean ± s.d.) 1.7 ± 0.7 1.6 ± 0.7 Gastric ulcer as the source of bleeding (%) 16 (23) 21 (27) Gastric ulcer with visible vessel (%) 6 (8) 10 (13) Gastric ulcer with oozing of blood (%) 10 (14) 9 (12) Gastric ulcer with spurting artery (%) 0 2 (3) Duodenal ulcer as the source of bleeding (%) 55 (77) 57 (73) Duodenal ulcer with visible vessel (%) 22 (31) 19 (24) Duodenal ulcer with oozing of blood (%) 29 (40) 32 (41) Duodenal ulcer with spurting artery (%) 4 (6) 6 (8) NSAIDs, non-steroidal anti-inflammatory drugs. (one from each group). The re-bleeding rate was lower in the omeprazole group than in the placebo group (12 vs. 26 patients, respectively; P ¼ 0.022; RR ¼ 0.51; CI ¼ 0.28 0.93). Re-endoscopy was performed in 49 (33%) patients (23 and 26 patients in the omeprazole and placebo groups, respectively). According to the protocol, only patients with re-bleeding were re-endoscoped. However, as the number of patients with re-bleeding was slightly lower than expected (according to previous studies), we re-endoscoped 11 more patients to achieve a higher degree of accuracy. These 11 patients did not have the criteria for re-bleeding, but were highly suspicious of having the condition. All of these patients belonged to the omeprazole group, a result achieved only by chance. The hospital stay was longer in the placebo group than in the omeprazole group (75 ± 39 h vs. 62.8 ± 28.6 h, respectively; P ¼ 0.032). The need for blood transfusion was higher in the placebo group than in the omeprazole group (56 vs. 40 patients, respectively; P ¼ 0.049; RR ¼ 0.78; CI ¼ 0.61 1.01). Other primary end-point data are shown in Table 2. No mortality or morbidity was observed during the 3-week follow-up. DISCUSSION The use of proton pump inhibitors in patients with bleeding peptic ulcers has been evaluated in several trials. Most of the previous studies on omeprazole have been conducted on intravenous omeprazole and not on the oral drug. In this study, we analysed the effect of oral omeprazole on patients with bleeding peptic ulcer, and showed that oral omeprazole reduced the re-bleeding rate, hospital stay and need for blood transfusion even in high-risk patients after endoscopic therapy. Lau et al. applied an injection of adrenaline (epinephrine) and a heater probe for the control of bleeding, with an intravenous bolus followed by continuous intravenous infusion of omeprazole for the prevention of
ORAL OMEPRAZOLE IN HIGH-RISK PEPTIC ULCERS 215 Table 2. Primary end-points Omeprazole group (n ¼ 71) Placebo group (n ¼ 78) P value Death (%) 0 1 (1) N.S. Operation (%) 1 (1) 1 (1) N.S. High-risk ulcers on re-endoscopy* (%) 2 23 (9) 9 26 (35) 0.039 Ulcer with visible vessel 0 2 2 9 Ulcer with oozing of blood 2 2 4 9 Ulcer with spurting artery 0 2 3 9 Clinical re-bleeding (%) 12 (17; 95% CI, 12.7 39) 26 (33; 95% CI, 29.6 57.6) 0.022 Hospital stay (h) (mean ± s.d.) 62.8 ± 28.6 75 ± 39 0.032 Hospital stay > 5 days (%) 1 (1) 8 (10) 0.034 Need for blood transfusion (%) 40 (56) 56 (72) 0.049 Amount of blood transfused (bag) (mean ± s.d.) 1.13 ± 1.36 1.68 ± 1.68 0.029 Hb on discharge (g dl) (mean ± s.d.) 10.31 ± 1.88 10.29 ± 1.78 N.S. Hb on discharge ) Hb 6 h after admission + 0.63 ± 2.2 + 0.36 ± 1.9 N.S. (g dl) (mean ± s.d.) Need for 3 bags of blood (%) 14 (20) 20 (26) N.S. 95% CI, 95% confidence interval; Hb, haemoglobin; N.S., statistically not significant. * Re-endoscopy was performed in 49 patients. re-bleeding. 17 They reported a 6.7% re-bleeding rate in the omeprazole group and 22.5% in the placebo group. The mortality rates in these groups were 4.2% and 10%, respectively. Schaffalitzky de Muckadell et al. applied injection therapy, with or without thermal or electrocoagulation, and an intravenous bolus followed by continuous intravenous infusion of omeprazole. 15 In this study, successful endoscopic therapy was not a criterion for inclusion of the patients. Successful treatment within 72 h of admission (no death, operation or additional endoscopic treatment) was reported in 91% and 79.7% in the omeprazole and placebo groups, respectively. The mortality rates in these groups were 7.7% and 8.5%, respectively. 15 In comparison with the above-mentioned studies, we achieved a very low mortality rate despite the use of injection therapy as the sole endoscopic treatment. The most important factors with regard to this achievement were as follows. (a) Our patients had a lower mean age and therefore a lower rate of serious co-morbidities. The mean age of our patients was 53 years but, in the studies of Lau et al. 17 and Schaffalitzky de Muckadell et al., 15 the mean ages were 64 years and 66.3 years, respectively. It appears that age may be a major determinant in the prognosis of peptic ulcer bleeding. (b) Our patients had a higher percentage of duodenal ulcers as the main source of gastrointestinal bleeding (gastric ulcer to duodenal ulcer ratio: 25% 75% in our study vs. 51% 49% 15 and 46% 54% 17 ). A possible reason for the abovementioned observation may be the lower mean age of our patients; H. pylori was the most probable cause of the ulcers in our study vs. non-steroidal antiinflammatory drugs in the other studies. This finding is comparable to that observed by Khuroo et al.: only 17% of patients had gastric ulcer. 5 In another study by Javid et al., oral omeprazole was used with similar aims to our study. 16 The patients participating in this study had an almost identical age range to our patients. The incidence of duodenal ulcer was higher in the study of Javid et al. (90% vs. 57% in our study). In addition, the mortality rate (three patients vs. one in our study) and the need for operation (9 166 patients vs. 2 149 in our study) were higher in the study of Javid et al. This observation may be due to two factors: (i) Javid et al. re-endoscoped all of their patients, however didn t perform endoscopic therapy (in contrast to our study); (ii) Javid et al. used oral omeprazole every 12 h (in contrast with our study, in which oral omeprazole was given every 6 h to achieve better plasma levels). All episodes of gastrointestinal bleeding occurred during hospital admission, and there was no recurrence of gastrointestinal bleeding during follow-up after discharge from hospital.
216 M. J. KAVIANI et al. In Asian patients, the parietal cell mass is smaller than that in white patients. On the basis of this fact, together with the studies of Khuroo et al. 5 (who reported the efficacy of oral high-dose omeprazole in acute nonvariceal bleeding) and Li et al., 18 we concluded that measurement of the ph of the stomach post-drug consumption is not essential. 19 In conclusion, oral high-dose omeprazole is effective in reducing the hospital stay, need for blood transfusion and re-bleeding rate in patients with high-risk peptic ulcer bleeding after endoscopic treatment. Therefore, it can be used as an adjunctive treatment after endoscopic therapy in these patients. ACKNOWLEDGEMENTS The authors are grateful to Shiraz University of Medical Sciences who provided financial support. The authors would also like to thank Dr M. E. Kaviani and Dr Zohrab-beyg for their kind and effective co-operation in preparing the drugs and placebo capsules. REFERENCES 1 Longstreth GF. Epidemiology of hospitalization for acute upper gastrointestinal hemorrhage: a population-based study. Am J Gastroenterol 1995; 90: 206 10. 2 Christensen A, Bousfield R, Christansen J. Incidence of perforated and bleeding peptic ulcers before and after the introduction of H2-receptor antagonists. Ann Surg 1988; 207: 4 6. 3 Chinn AB, Weckesser EC. Acute hemorrhage from peptic ulceration: an analysis of 322 cases. Ann Intern Med 1951; 34: 339 51. 4 Laine L, Peterson WL. Bleeding peptic ulcer. N Engl J Med 1994; 331: 717 27. 5 Khuroo MS, Yattoo GN, Javid G, et al. A comparison of omeprazole and placebo for bleeding peptic ulcer. N Engl J Med 1997; 336: 1054 8. 6 Jensen DM. Heat probe for hemostasis of bleeding peptic ulcers: techniques and results of randomized controlled trials. Gastrointest Endosc 1990; 36(Suppl.): S42 S49. 7 Panes J, Viver J, Forne M, et al. Controlled trial of endoscopic sclerosis in bleeding peptic ulcers. Lancet 1987; 2(8571): 1292 4. 8 Cook DJ, Guyatt GH, Salena BJ, Laine LA. Endoscopic therapy for acute nonvariceal upper gastrointestinal hemorrhage: a meta-analysis. Gastroenterology 1992; 102: 139 48. 9 Jiranek GC, Kozarek RA. A cost-effective approach to the patient with peptic ulcer bleeding. Surg Clin North Am 1996; 76: 83 103. 10 Barkun AN, Cockeram AW, Plourde V, Fedorak RN. Review article: acid suppression in non-variceal acute upper gastrointestinal bleeding. Aliment Pharmacol Ther 1999; 13: 1565 84. 11 Berstad A. Does profound acid inhibition improve haemostasis in peptic ulcer bleeding? Scand J Gastroenterol 1997; 32: 396 8. 12 Pearsson JP, Ward R, Allen A, Roberts NB, Taylor WH. Mucous degradation by pepsin: comparison of mucolytic activity of human pepsin 1 and pepsin 3: implications in peptic ulceration. Gut 1986; 27: 243 8. 13 Green FW Jr, Kaplan MM, Curtis LE, Levine PH. Effect of acid and pepsin on blood coagulation and platelet aggregation. A possible contributor to prolonged gastroduodenal mucosal hemorrhage. Gastroenterology 1978; 74: 38 43. 14 Smith JTL, Gavey C, Nwokolo CU, Pounder RE. Tolerance during 8 days of high-dose H2-blockade: placebo-controlled studies of 24-hour acidity and gastrin. Aliment Pharmacol Ther 1990; 4(Suppl. 1): 47 63. 15 Schaffalitzky de Muckadell OB, Havelund T, Harling H, et al. Effect of omeprazole on the outcome of endoscopically treated bleeding peptic ulcers. Randomized double-blind placebocontrolled multicenter study. Scand J Gastroenterol 1997; 32: 320 7. 16 Javid G, Masoodi I, Zargar SA, et al. Omeprazole as adjuvant therapy to endoscopic combination injection sclerotherapy for treating bleeding peptic ulcer. Am J Med 2001; 111(4): 280 4. 17 Lau JYW, Sung JJY, Lee KKC, et al. Effect of intravenous omeprazole on recurrent bleeding after endoscopic treatment of bleeding peptic ulcers. N Engl J Med 2000; 343: 310 6. 18 Li Y, Sha W, Nie Y, et al. Effect of intragastric ph on control of peptic ulcer bleeding. J Gastroenterol Hepatol 2000; 15: 148 54. 19 Lam SK, Hasan M, Sircus W, et al. Comparison of maximal acid output and gastrin response to meals in Chinese and Scottish normal and duodenal ulcer subjects. Gut 1980; 21: 324 8.