The equivalence of two open therapeutic regitnens of citnetidine in the treattnent of acute duodenal ulcer disease : A Canadian tnulticentre trial

One hundred and four patients completed a multicentre study 
comparing the standard cimetidine regimen of 300 mg qid with cimetidine 600 
mg bid in the treatment of acute duodenal ulcer. Both dosage regimens were 
effective in alleviating symptoms. At the two- and four-week assessments a 
significantly greater decrease in frequency, duration and severity of night time 
pain was recorded in the 600 mg bid group (P<0.05). The healing rates were 
equivalent in both treatment groups. After eight weeks of treatment, 96% of the 
patients had healed in each treatment group. Cimetidine 600 mg bid may 
represent a useful alternative therapeutic regimen to the standard 300 mg qid 
dosage in patients with symptomatic acute duodenal ulcer disease.

C IMETIOINE SUPPRESSES BOTH basal and stimulated gastric acid secretion (1,21) by competitively blocking the action of histamine at the H z-receptors of t he parietal cells.Con, sequently, cimetidine has become a first-line medical t herapy for the treatment of peptic ulcer d isease.
In North America the standard cimetidine regimen is 300 mg qid.This dosing schedule was established to maximize the pharmacological effect of cimetidine t hrough administration of the drug at times of peak acid produc, tion (3 ).A 24 h acid suppression study conducted by Mahachai and coworkers ( 4) indicated that cimetidine 600 mg bid was superior to cimetidine 300 mg q id in suppressing 24 h intragastric acidity and tended to be bet• ter after breakfast and overnight.This Equivalence de deux regimes therapeutiques ouverts de cimetidine dans le traitement de l'ulcere duodenal evolutif: Essai canadien multicentrique RESUME: Cent quatre patients atteints d'ulceres duodenaux evolutifs ont fait l'objet d'une etude multicentrique comparant un regime standard de cimetidine administree a la dose de 300 mg qid, d'une part et de 600 mg bid, d'autre part.
suggested that the regimen of cimetidine 600 mg twice daily should provide adequate gastric acid suppression to promote ulcer healing.Accordingly, a multicentre study in duodenal ulcer patients was undertaken in six Canadian centres to determine if healing rates and symptomatic improvements were equivalent with both therapeutic regimens of cimetidine, 300 mg qid and 600 mg bid.

PAT i£NTS AND METHODS
This was an open parallel multicentre trial of two regimens of cimetidine (Tagamet; Smith Kline & French Canada Ltd), 300 mg qid and 600 mg bid, in the treatment of acute duodenal ulcer.Patients entering the clinics with a suspected diagnosis of duodenal ulcer were considered as potential cand idates for entry into the study.
For inclusion, patients had to have an endoscopically confirmed duodenal ulcer measuring between 0.5 and 3.0 cm at the largest diameter.All subjects had to be between 18 and 70 years of age, and were required to given written informed consent.Protocols were approved by each centre's Ethics Committee on Human Research.
Excluded from the study were patients with concomitant gastric ulcer or esophageal erosions, previous gastric surgery and/or vagotomy, and treatment with cimetidine or other antiulcer drugs for more than one week immediately prior to screening.More specifically, patients who required the continued use of anticoagulants, meto-clopramide, anticholinergics, phenothiazines, anti-inflammatory or salicylate-containing drugs, thiourea derivatives, anti neoplastic agents, or systemic corticosteroids were excluded, as were patients who had taken an investigational drug within the past 30 days, pregnant o r lactating women, patients with a history of alcohol abuse, and patients with concomitant disease whose symptoms could impair the evaluation of their ulcer disease.Patients who had experienced gastrointestinal bleeding prior to admission had to have been free from frank bleeding for at least 24 h.Documented nonresponders to cimetidine were not to be admitted to the study, nor were patients suffering from severe systemic disease.
Screening assessments included complete medical history, physical examination, endoscopy, laboratory evaluations (hematology, clinical chemistry, urinalysis) and symptomatic assessment.No specific diet was recommended.However, patients were advised to take regular meals and avoid offending foods.Restraint in the use of tobacco, coffee and alcohol was also suggested.Antacid use was permitted but not recommended, and no standard antacids were given to patients; however, antacid consumption was recorded.
Symptomatic assessments were done by interview.Patients completed duplicate visual analogue scales of 100 mm on which they evaluated by a single pencil stroke the number of episodes, the duration, and the severity of the worst episode for both day and night pain.The scales were marked with figures from O (absence of symptoms) to 10 (severe or very long) at each centimetre.Thus, at every assessment, under the direction of the clinical monitor who was unaware of individual treatment, patients completed 12 cards presented to them at random.Patients graded their symptoms of the previous day by marking the symptomatic scale appropriately.They were also questioned about abdominal discomfort, flatulence, heartburn, nausea, vomiting and overall well being.
Assessment of compliance was made by means of a tablet count at each visit.All missing tablets were presumed to have been ingested by the patient.Patients were unaware of these verifications.
All pretreatment assessments were completed within five days prior to commencing therapy.Whenever possible therapy was started on the same day as the first endoscopy.During the screening period, all ulcer therapy was discontinued with the exception of antacids, which could be used for relief of pain.Patients were randomized to receive either the twice daily regimen or the standard 300 mg qid regimen of cimetid ine.
As often as possible, the same endoscopist, unaware of the patient's treatment, evaluated the therapeutic effect.
Patients returned at two weeks for a symptomatic assessment, physical examination and compliance check.At four weeks each patient underwent a symptomatic assessment, physical examination, endoscopy, compliance check and repeat laboratory tests.If the ulcer was healed (defined as a return to continuity of epithelium as determined by endoscopy), the study was complete.However, if the ulcer had not healed, the patient was issued an additional four-week supply of medication.At the end of eight weeks patients still in the study returned, and all assessments done at four weeks were repeated.Then, regardless of the endoscopic results, the study was terminated.Comparability of centres and groups at baseline: Centres were assessed at base-I ine for comparability by use of a variety entry into the study; three had what were judged to be drug-induced ulcers; one was a documented cimecidine nonresponder; and one patient was underage.Thus, 111 patients were eligible for inclusion in the study, provided the demographic data (Table 1) and were evaluated for side effects.
All comparisons were not statistically significant, ie, P>0.05 by Student's t test (two-tailed); 'x2 test Of these 111 patients, 53 had been randomized to the cimetidine 600 mg bid group and 58 to the 300 mg qid group.Seven patients did not complete the study.Four did not return for follow-up visits; one bled 24 h after starting the protocol; one was remove<l at four weeks because of exacerbation of symptoms; and in one, the participat• ing physician was unable co complete the final endoscopy.Thus, the Jara from 104 patients were available for of tests as appropriate.ln addition, before pooling results, centres were compared for the respective healing rates of their patients.To verify that the population was appropriately randomized, a x2 test was used co compare sex distribution and family history.The Student's t test was used to compare the groups' age, length of illness and duration of current episode.A ll tests were two-tailed.Analysis of healing rates: For the purpose of analysis, patients who had healed at four weeks were considered to be healed at eight weeks.Analysi, of symptomatic assessments: The marks on analogue scales were measured to the nearest millimetre and the mean value obtained from the two readings; the duration of the worst episode, severity of the worst episoJe and incidence of episodes both for day anJ night symptoms were logged.The Mann-Whitney test assessed any difference in visual analogue evaluation of the rwo treatment regimens at baseline and during therapy.Analysis of compliance: Compliance figures as determined by pill counts for each group was assessed for differences by the Student's t test for the duration of the study.

RESULTS
A total of 131 patients were considered for analysis in the study.Of these, 111 met the entry criteria.Twenty patients were excluded: L 2 from the 600 mg bid group and eight from the 300 mg qid group.They were excluded from the study analysis for the following reasons: seven had severe concomitant disease or previous gastric 56 surgery; in four the ulcer size was outside the allowable range or could not be determined accurately from available records; four had received an investigational drug during the 30 days prior to 100 (I) :::::: :::::: ::::: :,:,:    2 and 3.
A separate analysis of the treatment results in both groups of patients (c1metidine 300 mg q id and 600 mg bid) failed co show any appreciable difference between them and the study population, either in endoscopic healing races, symptomatic relief, or antacid consumption.Their exclusion does not introduce a bias into the study.Baseline dara from the six participating centres were evaluated and found to he statistically comparable.No difference in healing races was found between the centres, thus permitting the pooling of data.There were no significant differences between the treatment groups at baseline for age, sex, family history, length of illness or duration of current episode (Table 1).A lthough there appear to be differences in the pretreatment median values for some of the visual analogue symptomatic assess-ments, these differences were not statistically significant (Tables 2 and 3 ).The type and severity of ocher gastrointestinal symptoms were similar for both groups at pretreatment.Compliance assessment: No difference in compliance was apparent between groups, both having used 97 .5% of the pills distributed.Symptomatic assessme nt: Median values for the visua l analogue scales at the various rating times are presented in Tables 2 and 3. Boch groups showed marked decreases in number of episodes as well as duration and severity of worst episode for both day and night pain at two weeks, with a further dim inution of pain as therapy continued.Statistically significant differences in favour of the 600 mg bid regimen were seen at two and four weeks for all night pain values (Mann-Whitney, P<0.05 ).The frequency and severity of ocher gastrointestinal symptoms also decreased as therapy progressed.Because of the small number of patients at eight weeks, no comparison could be made between t he two groups at that time period.Assessment of healing rates: At week 4, 82% of the patients on the 600 mg bid regimen had healed their ulcers, as compared with 75% of those on the standard regimen.Ar week 8 a healing rate of 96% was recorded for both groups.Thus, there were no d ifferences between the two groups in the population of patients healing on the two treatment regimens.Consumption of antacids: Antacid consumption was minimal.Only four patients on cimetidine 600 mg used antacids during the study, compared co nine patients on the standard regimen.Side effects: Of the 111 patients who met entry criteria, seven were reported by the investigators to have experienced adverse reactions which, in their opinion, were 'related' or 'questionahly related' to cimecidine.There was one report each ofleg cramps and headaches in the 300 mg qid group.The remaining reports were from patients taking the twice daily regimen and included headaches and forgetfulness.Additionally, three patients experienced alterations in laboratory parameters which were assessed as 'not clinically significant': a slight le uko penia was noted in two patients at the conclusion of therapy, and e lev ated aspartate aminotransferase a nd alkaline phosphatase were noted in one patient at the end of the study.No patient had to be withdrawn from the study because of any adverse reaction.58 entry rate into the study as well as funding was diminishing, the study was terminated before recruitment of the targeted 200 patients.Nonetheless, the rate of ulcer healing was comparable in the two treatment groups at fo ur and eight weeks, with similar use of antacids a nd similar rates of compliance.

DISCUSSION
Most studies using cimetidine in duodenal ulcer have not shown a clear correlation between symptomatic improvement and healing rates (5-7).
Figure 1) Duodenal ulcer healing rates in two therapeutic regimens of cimecidine.n Number of patients Analysis of findings in this open parallel equivalence study indicate that both dosage regimens of cimetidine 300 mg qid and cimetidine 600 mg bid are effective in promoting healing and alleviating symptoms in duodenal ulcer.It must be emphasized that this study was designed as an equivalence trial and not an efficacy trial, which would have required approximately 2000 patients at an unaffordable cost.A s ACKNOWLEDGEMENTS: Contributing physicians were JN Amar, RJ Bailey, R Beaudry, A Farley, H Haddad, L Halperin, P Lefort, P Leroux, DB Menard, TL Moore, G Pilon, LM Price, EA Shaffer, RW Sherbaniuk, LR Sutherland, RE Warren, RH Wensel.Tlie authors express their gratitude to the numerous clinical and paraclinical personnel from the various centres who made chis study possible.In particular they wane co acknowledge the assistance and support of the Smith Kline & French Company, as well as the expertise of Dr M Grace, PhD Eng, of the Department of Medic ine, University of Alberta, for his advice on statistics.

TABLE 2
Visual analogue assessment of day pain

TABLE 3
Visual analogue assessment of night pain et al.Comparative effects of two 14 days) between visits to investigators, the present assessment focused specifically on the day prior to each v isit.This approach was used o n the assumption that patients had a more precise recol, lection of immediate events and on the observation t ha t diary ca rds are seldom filled out in a consistent manner by patients.Ir is a lso likely that the im, provement in nigh t symptoms with a higher bedtime dose of c imet idine (600