Healing of benign gastric and prepyloric ulcers : A prospective , endoscopy-- controlled , randolllized , double-- blind , Canadian lllulticentre study of omeprazole 20 and 40 mg daily and ranitidine 150 mg twice a day

A double-blind, randomized study was conducted in 118 patients 
with benign gastric or prepyloric ulcers to compare the efficacy of omeprazole 20 
or 40 mg daily with ranitidine 150 mg twice daily. The healing rates at four weeks 
were 67, 79 and 54% and at eight weeks increased to 90, 97 and 71 % for the 
omeprazole 20 and 40 mg groups and the ranitidine group, respectively (P< 0.03 
for the differences between each of the omeprazole groups and the ranitidine 
group at eight weeks). Multivariate analysis showed influence on healing rate for 
ulcer size but not for smoking status, sex or ulcer site. Symptomatic relief was 
excellent and similar in the three groups. Ulcer recurrence during the six month 
follow-up off treatment after initial ulcer healing did not differ between the three 
groups. No serious adverse events could be attributed to the drugs. The authors 
conclude that treatment with omeprazole 20 or 40 mg daily for a period of four 
to eight weeks is safe and significantly more effective in ulcer healing than a 
standard dose of ranitidine.

O MEPRAZOLE.A SUBSTITUTED benzimiJ::izo le, is a gastri c proton pump inhihitor prod ucing potent inhibition of gasrric ac id secret ion ( 1,2).!rs efficacy in healing duoJenal ulcers has been shown in comparison to c ime tidine (3.4) anJ ranitiJine (5) .In patients with gastri c ulcers, irs efficacy has also been evaluated with Joses of 20 and 30 mg (6,7 ).The efficacy of a drug regimen producing marked ac id suppression, as can be ac hi eveJ with omcprazole 40 mg daily, was the objec t of this investigation.
Th b study compares two J oscs of omeprazole, 20 anJ 40 mg daily, produc ing Jiffcrent Jegrees of suppression of intragastric ac idity, with a srnndarJ dose of ranit iJine in the t reatmentofbe nign gastric ulce r.The re port deals with 118 patients enrolleJ by six C anadian centres accorJing to the protocol used in an international multicentre randomized stud y, and is the only North American contribution to

PATIENTS AND METHODS
Patients included 111 the study had o ne o r more active ulcers m the stomac h with a greater ax is of at least 5 mm, verified by endoscopy within four days of bcgi1ming treatme nt with the study drug.Prepyloric.ulcers were included and classified as such to a llow separate analysis, but no t stratified into the various treatment groups.Ulcers localized within 3 cm of the pyloric ring were defined prepyloric, an<l those localized mo re than 3 cm from the pylorus were considered gastric.
Patients were excluded from participation in the study for the fo llowing reasons: age below 18 o r above 80 years; pregnant or breascfee<lmg; treatment with therapeutic doses of anti-ulcer drugs fo r more than three days during the 14 Jays preceding endoscopy; presence of erosive reflux esophagitis, pyloric stenosis, concurrent duodenal ulcer or bleeding ulcers; history of gastric surgery except for simple ulcer closure; o ther signifi cant concurrent disease; and c hronic alcoholism.Patients on treatment with non-steroidal anti-inflammatory drugs (NSAIDs) were mcluded but advised not to change dose; these patients were classified but not stratified into the vario us treatment groups.
The study was conducted in accordance with the Declaration of Hclsmki and was approved by each inst itution's research and cthtCs review comnrntee.Written informed consent was obtained from all patient:,.
Six centres participated 111 the :study.The patients were randomly assigned to double-blind treatment with omeprazole 20 or 40 mg m the morning, or rani tidine 150 mg bid.A ll patients received the medicauon for four weeb.Paucnts whose ulcers had not healed after four weeks received a furrhcr four week treatment course.Ulcer healmg was assessed by endoscopy at four an<l, if necessary, eight weeks m patients with gastric ulcers and ar two and, if necessary, four and e ight weeks in patients with prepylonc ulcers.Complete rc-epithelialization of the ulcer site o r sites was used as the criterion of ulcer healing.Gastric biopsies were taken from the ulcer area in order to exclude malignancy.
Refore treatment, the ulcer and genera l medical histnnes were recorded for each patient, along with a history of drug, tobacco and alcohol use.During t reatment, pmients were assessed after two an<l four weeks a nd if necessary after e ight weeks with an ulcer symptom severity rating, physical ex• amination, r1Jvcrse event eval uation, drug al.countabil ity, and lahoratori tests (hemoglobin, wh ite hloo<l cell count, plr1cclets, seru m creatininc, electrolytes, bilirubin, a lkalme pho~ pharase, a lanme transaminase, aspar, tare transaminase, a n<l urine protein and sugar).
After completion of the ulcer he~ling study phase, healed patients were fo llowed for a pencxl of six month,.Patients were reviewed at three r1nd six months and an endoscopy performed at six months o r earlier if ulcer symptoms recurred.
The healing data were analyzed with standa rd x 2 tests and confidence inter vab for the heal mg probabilities.In ,iJ.d ition, r1 multivariate analysis of treatme nt regimen, ulcer size and site, smo king status, sex and NSAID use, was conducted to adj ust for possible 11n balances in one or more of these factor, in the comparison between treatments.Unknown healing status was classificJ as 'not healed'.Safety analysis mcluJcJ data from a ll patients who received study drugs.The laboratory tests were performed in each individual centre, and thus varying units an<l reference ranges were used for vmiables.

RESULTS
A total of 11 8 patients entered the study.Table 1 shows the pre-entl) characteristics of the patients in the Lhrce groups.None of the differences observed between the treatment group, were statistically significant.Five patients were withdrawn <luring the study (Table 2).
The cumul ative proportions of patients healed at four and eight weeks are shown in Tahle 3. The omeprazolc 20 mg regimen achieved significantly higher healing rates than ranitidine 150 mg hid afte r eight weeks of treatment ( P=0.03).With omeprazole 40 mg daily, the healing rates at four anJ eight weeks were significantly higher than the rates observed with raniti<linc (P=0.02 and P=0.001 ).Ana lysis of pm-  In the sma ll subgroup of patients ta king NSAIDs, there was a significantly higher h ealing rate in omeprazole-versus ranitidine-treated patients; after eight weeks of treatment, CAN J GASTROENTEROL VOL 4 N O I JANUARY/FEBRUARY 1990 six of nine, e ight of eight, and one of four patients were healed in the omeprazole 20 mg, omeprazole 40 mg and ranitidine groups, respecti vely (P=0.02).Healing rate~ afteTl wo weeb in prepyloric ulcer patients were six of l l, four of 14 and fo ur of 13 in the omeprazole 20 mg, omeprazole 40 mg and rani tid ine groups, respectively ( not significant) .
A t pre-entry, a ll patients reported symptoms.After two weeb, the proportion of patie nts free from symptoms or mildl y symptom.:itic was 86% in the omeprazole 20 mg group, 83% in the omcp razole 40 mg group and 86% in the ranitidine group; after four weeks, the corresponding figures were 97, 9 1 and 94%, respectively.
Complaints during the study were reported by l O patients in the o meprazole 20 mg group, three patients in the omeprazole 40 mg gro up and 14 patients in the ranitidine group (Table 4) .None of the compla ints was clinica ll y important.However, the adverse events caused the withdrawal of two patients in the rani tidine grour (headache, sleepi ness and unstead y gait in one patient who.e side effects disappeared completely on open treatment with the sa me d rug; nausea and d izzi, ness which disappeared one Jay after stopping the trial medication in the ocher patient).In addition, one pmient in the ranitidinc group was wi thdrawn when a concurrent disease (pancreatic ca rc inoma) was clinicall y diagnosed.During the st udy, there was no significant change in the mean values for the laboratory tests.
The cumulative proportions of the initially enrolled patients who were still in remission verified hy endoscopy, after the follow-up of six months, were 62% in the t)meprazole 20 mg group, 62% in the omeprazole 40 mg group and 4 7% in the ranitidine group (Figure l ); no statistically significant difference was observed between the three groups.

DISCUSSION
Healing of benign gastric ulcers is a slow process; indeed, o n placebo, reported h ealing rates are in the range of 25 to 54% after observation fo r six to eight weeks (8-l 2).For the same period  of treatment, clinica l trials using drugs acting to con tro l gast ric acid report healing rates in th e range of 67 LO 70% with antacids (8, I I ), 65 to 89'}-o with c ime tidine (6,8,9, 13, 14) 78 to 94% P atie n ts in rem issio n ( %) 100 wiLh rani t id ine (1 3-16) a nd 65 to 80% with fa motidinc ( 12).H owever, compa risons of these results may noL be appropriaLc mce the ini t ia l ulcer size, a maio r factor in determining the healing time course ( 17), was not taken into cons1derat ion .Furth ermore, t he st rong relation ship shown bcl\veen d egree of acid suppression and healing rate in duod enal ulce rs ( l 8) b no t o bserved in clinical t ria ls of acid-inh ibiting d rugs in the trea unent of gastric ulcer ( 19).
T his study shows that treatment with omeprazole at a J ose of 40 mg tin il y, a very powerful ac1<l-suppressm g regimen , is more effec ti ve than a standard Jose of ran 1tidme in healing gaMric ulcers.O meprazole 20 mg da 1ly was nlso more effecti ve than rani t i<line 150 mg t wice a J ay afte r e ight weeks of trentmenr.In comparison to c imeti<linc l g da il y, o meprazole 30 mg da ily was shown to he more effective in healing ulcers located in the a n Lrum o f the stomach (6), while ran it idme 150 mg bid a nd o mcprazole 20 mg daily in a previo us study resulted m similar healmg rates (7).The present results arc nor due to differences between the treatm ent gro ups in J e mographic J am, smo king status, ulce r locatio ns or sex.In stud ying the tota l populmion of subjects e nro lled in the three groups, ulcer :,ize wa:. a progn osLic facto r in the healing rate.However, the slightly smaller ulcer :.1ze in th e omep razole grour1 could no t explai n the differe nces in results since the log1t stallst1cal analysis that rhe autho rs used took into considerat ion rhc imba lance m ulcer size at t he hcg mnm g nf the study.It is genera II y assumed that p rcpylon c ul ------------, Omep r azole 20 mg cers behave physiologically similarly to duodenal ulcers (20), for which omeprazole has been shown to be superiorto 150 mg ranicidine twice daily in its healing capacity (5).Although ,tudies of gastric acid secretion were not performed, the relative numbers of prepyloric over gascric ulce rs d id not differ between the three groups, and in each treatment group, about two-thirds of ulcers were gastric ulcers.
Even if Lhe pmhogenesis of gastric ulcers appears L o be related more to abnormalities in mucosa l defence mechanisms than to aggressive factors (21 ,22), this study nonetheless suggests that control of gastric ac id is likely to he very important in the he,1 ling of gastric ulcers.Adding placebo data to the analysis of the effect of H z receptor antagonists in the treatment of gastric ulcers, Howden ct al ( 19) found correlations between acid suppression antl ulcer hea ling stronger with suppression of total 24 h versus nocturnal acidi ty, contrasting with duodena l ulcers fo r which suppression of nocturnal acidity 1 s the most relev<1nt c1cting factor with Hz receptor antagonists ( 18).In asymptomatic duode nal ulcer subjec ts, ran itidine 150 mg twice daily reduced 24 h intragastric acidity by 63 to 69% (24)(25)

TABLE 4
Complaints during the study (number of patients) (27)peated doses of omeprazole 40 mg daily have been reported to No I JANUARY/FEBRUARY l 990 circumstances of moderate acid inh ibition.Higher rates of ulcer healing achieved with omeprazole 20 antl 40 mg Jaily versus ranitidinc 150 mg twice daily a re consistent with the hypothesis that the degree of suppression of intragastric acitl is very important in the healing phase of gastric ulcers; there are furthermore no data suggesting that omeprazole has inherent cytoprotcctivc activity.However, pronounced inhibition of acid may abo promote liberation of gastrin and favour its trophic effect on the gastric mucosa.The authors conclude that omeprazole 20 mg dail y and omeprazole 40 mg tlaily resulted in a significantly more rapid healing rate than ranititline 150 mg twice daily in benign prepyloric and gastric ulcers.These results a rc similar to the data from the international study(27)and tlo not indicate a difference between North America and Europe in the reporting of hea ling rates.For a period of four co eight weeks, treatment with omeprazole was shown to be safe.Ulcer recurrence during the six month fo llow-up period was not different with e ither treatmenc usctl to heal the acute ulcer.