Aliment. Pharmacol. Therap. (1991) 5 , 631-643.
Famotidine (20 mg) b.d. relieves gastrooesophageal reflux symptoms in patients without erosive oesophagitis
M. ROBINSON*, D. L.DECKTOR*, R. C S T O N E t , M.PEVELERY*, P. BARDENS, R. MOYERg, S. HOLT**, J. R O O T t t , K. HUFNAGELtt, T. J. H U M P H R I E S t t t , R. D. BERLINtt and the Famotidine/GERD Investigation Group.*** * Oklahoma Foundation for Digestive Research, Oklahoma City, OK, t Dallas Medical and Surgical Clinic, Dallas, TX, Louisville, KY, 5 Albuquerque, N M , g Hibbing, MN, ** Springfield, IL, tt Merck Sharp & Dohme Research Laboratories, West Point, PA, (** Current address : Clinical Science Research, Inc. Baltimore, M D ) & the Farnotidine/ GERD Investigators Group***
*
Accepted for publication 28 July 1991
SUMMARY
Previous clinical trials have evaluated a large number of symptomatic individuals with heartburn. Most studies have documented the need for multiple daily dosing with H,-antagonists to achieve clinical and statistical efficacy for symptom relief. The purpose of this study was to compare the safety profile and efficacy of farnotidine oral dosing regimens, 40 mg nocte and 20 mg b.d. with placebo in the relief of symptoms in patients suffering from frequent heartburn found to have endoscopically normal oesophageal mucosa or mild non-erosive oesophagitis. Famotidine (20 mg) b.d. reduced and eventually completely Correspondence to : Dr D. L. Decktor, Oklahoma Foundation for Digestive Research, Oklahoma City, OK 73104, USA. ‘I* The remainder of the Famotidine/GERD Investigators Group include: Drs G. Ahtaridis, M. M. Berenson, M. Breeden, M. J. Collen, B. I. Hirschowitz, A. F. Ippoliti, D. A. Johnson, F. L. Lanza, D. T. Lyon, D. J. Pambianco, J. E. Richter, H. M. Serfer, S. J. Sontag, & Z. R. Vlahcevic. 631
632
M. ROBINSON ef al.
relieved gastro-oesophageal reflux disease symptoms in most patients during the 6-week trial. Global assessment of improvement at 2 and 6 weeks indicated significantly greater improvement with a b.d. treatment regimen than with either a 40 mg nocte or placebo treatment. No statistically significant differences between famotidine and placebo in the number of patients who experienced clinical adverse experiences were noted and no serious adverse events attributable to famotidine occurred. Based upon these findings, patients with gastro-oesophageal reflux symptoms experience good relief of their complaints with twice daily famotidine in standard doses.
INTRODUCTION Although the accurate epidemiology of gastro-oesophageal reflux disease has been very difficult to characterize, symptomatic reflux is clearly an extremely common problem. It is estimated that at least 7 % of the Western adult population experiences frequent heartburn, the cardinal manifestation of gastro-oesophageal reflux disease. Careful endoscopic assessment of patients with heartburn demonstrates no more than two-thirds of patients with any visible mucosal damage.’ Even with erosive or ulcerative changes noted initially on endoscopy, treatment for gastro-oesophageal reflux disease should always begin with Phase I manoeuvres including dietary manipulation, antacids, head of the bed elevation, and other lifestyle changes. However, many patients either are not compliant with such recommendations or fail to respond. When these conservative treatment approaches fail, the customary therapeutic approach has been to use a gastric antisecretory HIreceptor antagonist and/or a promotility agent. Large numbers of symptomatic individuals with heartburn have been evaluated in a number of clinical trials assessing the response of this acid-mediated disorder to antisecretory drugs. Most studies document the need for multiple daily dosing to achieve clinical and statistical effi~acy,~”as opposed to the well-accepted usefulness of once daily evening dosing of antisecretory agents for active and maintenance therapy of peptic ulcer disease.6 Ambulatory pH studies have yielded objective support for multiple daily dosing by demonstrating that a b.d. antisecretory treatment regimen can normalize acid contact time of patients with severe symptomatic gastro-oesophageal reflux d i ~ e a s e while ,~ bedtime dosing fails to achieve this endpoint. Such physiological results suggest that b.d. dosing would be most effective in relieving symptoms in reflux patients. Famotidine has been in use for the treatment of duodenal ulcer, gastric ulcer, and other acid-related diseases for several years. It is a competitive H,-receptor antagonist and its dose related suppression of basal and stimulated gastric acid secretion has been demonstrated with oral doses from 5-40 mg.’ The purpose of this study was to investigate the safety profile and efficacy of famotidine oral dosing regimens, 40 mg nocte and 20 mg b.d. vs. placebo in the
B.D. FAMOTIDINE IN GASTRO-OESOPHAGEAL REFLUX
633
relief of symptoms in patients suffering from frequent heartburn found to have normal oesophagus or endoscopically-defined mild non-erosive oesophagitis. PATIENTS A N D M E T H O D S For entry into the study, all patients had to be 18 years of age (legal age of consent) or older. Criteria for enrollment into the study included heartburn, characterized by retrosternal burning pain present for approximately 15 out of 30 days prior to entry and occurring daily at least 5 of 7 days during the single-blind placebo period. A positive Bernstein test and endoscopic evidence of a lack of any break in the oesophageal mucosa (either normal oesophageal mucosa or erythema) were also required prior to entry. Patients with erosive oesophagitis were excluded. Patients were selected who had no history of taking therapeutic doses of H,-receptor antagonists for more than 5 successive days within the 2 weeks prior to beginning the placebo baseline period. Reasons for exclusion from the study included erosive oesophagitis (grade 2 or worse-see Endoscopy), manometric evidence of a primary oesophageal motility disorder, a history of Zollinger-Ellison syndrome, oesophageal disease other than oesophagitis or oesophageal stricture. Oesophageal and/or gastric varices, active peptic ulcer disease involving the stomach, pyloric channel or duodenum, paraoesophageal hernia or non-reducible sliding hiatal hernia and other standard exclusionary criteria were applied. Standard clinical laboratory tests were performed. Patients with abnormal hepatic enzymes (any value 25 % above the upper limit of normal value range) or with other severe chronic medical disorders were also excluded as were patients requiring concurrent medication that might affect acid secretion or mucosal healing such as corticosteroids, non-steroidal anti-inflammatory drugs (NSAIDs), antidepressants, salicylates, anticholinergics or antineoplastic agents. Twenty-two centres took part in this study and the protocol was approved by the appropriate institutional review board committee at every site. All patients who entered the study gave written informed consent.
Study protocol Patients who satisfied the inclusion/exclusion criteria, as determined by medical history, physical examination and clinical laboratory evaluation, were entered into the single-blind 1-week baseline period. At the time of entry, routine endoscopic evaluation and standard Bernstein testing were performed. Patients with endoscopic evidence of normal oesophageal mucosa or mild oesophagitis (grade 0 or I, see Endoscopy) and a positive Bernstein test’ were issued a diary card and given placebo to be taken before breakfast, before dinner and at bedtime. Antacid tablets (Gelusil) were dispensed at each visit throughout the study, to be taken for heartburn (not to exceed 30 tablets/week). An instruction sheet recommending conservative measures for the treatment 43
BAP 5
634
M. R O B I N S O N et al.
of gastro-oesophageal reflux disease was distributed to each patient upon entry into the study. The recommended measures included head of bed elevation, diet modification (that is, avoid fats, chocolate, alcohol, peppermint and spearmint ; avoidance of eating for 2-3 h before bedtime), and decreased or discontinued smoking. This double-blind multicentre trial randomized patients, 2 : 2 :1by design, into 40 mg famotidine nocte (n = 155), 20 mg famotidine b.d. (n = 158) and matching placebo ( n = 76). Dosing was before bedtime for the 40 mg once daily treatment and before both breakfast and dinner for the 20 mg b.d. regimen. Doses taken were indicated by the patient on the diary card as were the number of antacid tablets consumed daily. Patients also recorded daytime heartburn, nighttime heartburn and a global assessment (see Symptom assessment). Physical examination was performed, laboratory parameters were re-evaluated, and diary cards were reviewed at the time of Week 2 and Week 6. Symptom assessment Each patient kept a daily record of symptoms, including the number and severity of episodes of daytime and nighttime heartburn. Heartburn, defined as retrosternal warmth or burning was assessed using scores ranging from no heartburn to disabling heartburn. The scoring was defined in the following manner; (0) no heartburn, (I) mild; present but causing little or no discomfort, (2) moderate; annoying but not interfering with daily routine or sleep, ( 3 )severe; causing marked discomfort and some interference with daily routine or sleep, (4) disabling; interferes considerably with daily activities and/or sleep. A global assessment relative to how the patient felt during the placebo baseline week was obtained from each patient at the end of Weeks 2 and 6. The rating scale included (0) no improvement, (I)slight improvement, (2) moderate improvement or ( 3 )excellent improvement.
Endoscopy Oesophago-gastro-duodenoscopywas performed using standard flexible fiberoptic instruments. Endoscopic fmdings of the oesophageal mucosa were classified as follows : (0)normal mucosa, no abnormalities noted, (I)no macroscopic lesions but erythema or hyperaemia may be present ( 2 ) superficial ulceration or erosions involving less than 10%of the mucosal surface of the last 5 cm of oesophageal squamous mucosa, ( 3 ) superficial ulceration or erosions involving 10-50% of the mucosal surface of the last 5 cm of oesophageal squamous mucosa (4) deep ulceration anywhere in the oesophagus or confluent erosion of more than 50% of the mucosal surface of the last 5 cm of oesophageal squamous mucosa. Statistical analysis The analysis of efficacy at each time point was done as the endpoint analysis. An ‘all patients treated’ analysis used data from all patients who had baseline data.
B.D. FAMOTIDINE IN GASTRO-OESOPHAGEAL REFLUX
635
The distribution of patients across the 4 categories of global response was assessed using the Cochran-Mantel-Haenszel statistic." Global assessments were also analyzed by collapsing the 4 categories into 2: success (moderate or excellent improvement) or failure (none or slight improvement). The distribution of patients across the categories of success/failure was assessed using Fisher's exact test. Daytime heartburn and nighttime heartburn were assessed in several ways. The time until complete relief without recurrence was analysed using the MantelHaenszel life-table methods, stratifying on baseline scores. The distribution of patients across the categories of relief/no relief at the end of the study was analysed using Fisher's exact test. These scores were also assessed by computing a change in score from baseline of Days 1-7 and Weeks 2-6 and analysing these changes using analysis of variance on the ranked change from baseline. All tests were two-tailed with an alpha level of 0.05 considered significant. The study was designed to detect a 30 % difference in the proportion of heartburn-free patients between the famotidine groups and the placebo group with a 95 % power and a 15 % difference between the famotidine groups with an 80 % power. RESULTS
Pafienf demographics Three hundred and eighty-nine patients were randomized into this study following a one week single-blind baseline period. Patient characteristics at randomization are presented in Table I. Pairwise comparisons of all demographic characteristics indicated marginally significantly differences (0.05 < P < 0.10) in daily alcohol consumption between 40 mg famotidine nocte and 20 mg b.d. and between 20 mg famotidine b.d. and placebo. Most notable among the pairwise comparisons of baseline characteristics, illustrated in Table 2, was a significantly more severe daytime heartburn exhibited by patients in the 40 mg nocte group relative to the placebo group at baseline. In general, however, the treatment groups were similar with respect to baseline characteristics. Seventeen and 13 of the 389 initially randomized patients had no global evaluation (the primary efficacy variable) at Weeks 2 and 6, respectively, and were thus excluded from the intent-to-treat analysis,
Eficacy Global assessment. Although patient global assessment of symptom relief was recorded at Weeks 2 and 6, daytime and nighttime heartburn were rated daily throughout the single-blind baseline week and study period. Between group comparisons were based both upon the time until complete relief of symptoms (with no recurrence) and upon the change in heartburn score at each of Days 1-7 of active treatment and at Weeks 2-6. Global evaluations were analysed and distributed over four assessment categories (no improvement, slight, moderate or excellent improvement). Between 43-2
636
M. R O B I N S O N et al.
Table 1. Randomized patient demographic characteristics Famotidine 40 mg nocte ( M = 155)
20 mg b.d. (a = 158)
Placebo
Total
(n = 76)
(n = 389)
197 192
-
Gender Female Male Age (Years) Mean Standard deviation Race Caucasian Negro Other Smoking No Yes Alcohol (daily) No Yes Caffeine No Yes History of dysphagia No Yes History of acid regurgitation No Yes
81 74
85 73
31* 45
44.2 14.7
45.2 14.2
44.5 13.0
44.7 14.2
132 17 6
135 21 2
67 8 1
334 46 9
108 47
116 42
49 27
2 73 116
140** 15
151 7
68‘ 8
359 30
41 114
51 107
23 53
115 274
85 70
101 57
49 27
235 154
22 133
21 137
15 61
58 33 1
“Marginally significant difference found between 20 mg famotidine b.d. and placebo (0.05) < P < 0.10). ** Marginally significant difference found between 40 mg famotidine nocte and 20 mg famotidine b.d. (0.05 < P < 0.010).
group comparisons of the distributions of global assessments indicate that at Weeks 2 and 6, both farnotidine groups fared better than placebo. These findings are illustrated in Figure I, where the four assessment categories were collapsed into two : successful evaluation (moderate or excellent improvement) or unsuccessful evaluation (no or slight improvement) At Week 2, 69.9% of the patients in the 20 mg b.d. group achieved moderate or excellent improvement. This was significantly better than the 56.5 % of patients achieving similar improvement within the 40 rng nocte ( P d 0.05) group or the 50% in the placebo group (P d 0.01). Among the symptom assessment categories included in the global assessment, only ’excellent improvement’ included patients who were symptom free. At Week
B.D. FAMOTIDINE IN GASTRO-OESOPHAGEAL R E F L U X
637
Table 2. Patient baseline characteristics Famotidine 40 mg nocte 20 mg b.d. ( n = 155) (n = 158)
_____ Past history of heartburn frequency 134 Daily Weekly 21 Daytime heartburn severity" 3 None Mild 61 Moderate 86 Severe 4 Disabling 0 Missing data 1 Baseline nighttime Heartburn severity 28 None Mild 65 Moderate 55 Severe 5 Disabling 0 Missing data 2
(n = 76)
Total (n = 389)
127 31
64 12
325 64
3
0 32 34
6 164 194 21
71 74 8 2 0
32 65 47 12 2 0
Placebo
9 0
2
1
2
14 35 19 7 0 1
74 165 121 24 2 3
* Famotidine 40 mg nocte. Experienced significantly greater daytime heartburn severity at baseline than placebo, P < 0.05.
2, 22 % of patients within the 20 mg b.d., 25 Yo of patients within the 40 mg nocte group, and 14 % of patients within the placebo group were symptom free. By Week 6,45 % of patients receiving 20 mg b.d. famotidine were symptom free. In contrast, only 25% of patients within the 40 mg nocte group or placebo groups were symptom free. Famotidine 40 mg Fomotidine 20 mg b.d
ttxx
80 70
8
60
0
5
50
!$c
40 30
20 10
0 Week 2 x
Week 6
Farnotidine 20 mg b.d. significantly better than placebo P
< 0 01
t Famotidine 20mg b d significantly better than fomotidine 40 mg P (0.05 t i Farnotidine ZOrng b.d. significantly better than fornotidine
40mg P
Famotidine (20 mg) b.d. relieves gastrooesophageal reflux symptoms in patients without erosive oesophagitis
M. ROBINSON*, D. L.DECKTOR*, R. C S T O N E t , M.PEVELERY*, P. BARDENS, R. MOYERg, S. HOLT**, J. R O O T t t , K. HUFNAGELtt, T. J. H U M P H R I E S t t t , R. D. BERLINtt and the Famotidine/GERD Investigation Group.*** * Oklahoma Foundation for Digestive Research, Oklahoma City, OK, t Dallas Medical and Surgical Clinic, Dallas, TX, Louisville, KY, 5 Albuquerque, N M , g Hibbing, MN, ** Springfield, IL, tt Merck Sharp & Dohme Research Laboratories, West Point, PA, (** Current address : Clinical Science Research, Inc. Baltimore, M D ) & the Farnotidine/ GERD Investigators Group***
*
Accepted for publication 28 July 1991
SUMMARY
Previous clinical trials have evaluated a large number of symptomatic individuals with heartburn. Most studies have documented the need for multiple daily dosing with H,-antagonists to achieve clinical and statistical efficacy for symptom relief. The purpose of this study was to compare the safety profile and efficacy of farnotidine oral dosing regimens, 40 mg nocte and 20 mg b.d. with placebo in the relief of symptoms in patients suffering from frequent heartburn found to have endoscopically normal oesophageal mucosa or mild non-erosive oesophagitis. Famotidine (20 mg) b.d. reduced and eventually completely Correspondence to : Dr D. L. Decktor, Oklahoma Foundation for Digestive Research, Oklahoma City, OK 73104, USA. ‘I* The remainder of the Famotidine/GERD Investigators Group include: Drs G. Ahtaridis, M. M. Berenson, M. Breeden, M. J. Collen, B. I. Hirschowitz, A. F. Ippoliti, D. A. Johnson, F. L. Lanza, D. T. Lyon, D. J. Pambianco, J. E. Richter, H. M. Serfer, S. J. Sontag, & Z. R. Vlahcevic. 631
632
M. ROBINSON ef al.
relieved gastro-oesophageal reflux disease symptoms in most patients during the 6-week trial. Global assessment of improvement at 2 and 6 weeks indicated significantly greater improvement with a b.d. treatment regimen than with either a 40 mg nocte or placebo treatment. No statistically significant differences between famotidine and placebo in the number of patients who experienced clinical adverse experiences were noted and no serious adverse events attributable to famotidine occurred. Based upon these findings, patients with gastro-oesophageal reflux symptoms experience good relief of their complaints with twice daily famotidine in standard doses.
INTRODUCTION Although the accurate epidemiology of gastro-oesophageal reflux disease has been very difficult to characterize, symptomatic reflux is clearly an extremely common problem. It is estimated that at least 7 % of the Western adult population experiences frequent heartburn, the cardinal manifestation of gastro-oesophageal reflux disease. Careful endoscopic assessment of patients with heartburn demonstrates no more than two-thirds of patients with any visible mucosal damage.’ Even with erosive or ulcerative changes noted initially on endoscopy, treatment for gastro-oesophageal reflux disease should always begin with Phase I manoeuvres including dietary manipulation, antacids, head of the bed elevation, and other lifestyle changes. However, many patients either are not compliant with such recommendations or fail to respond. When these conservative treatment approaches fail, the customary therapeutic approach has been to use a gastric antisecretory HIreceptor antagonist and/or a promotility agent. Large numbers of symptomatic individuals with heartburn have been evaluated in a number of clinical trials assessing the response of this acid-mediated disorder to antisecretory drugs. Most studies document the need for multiple daily dosing to achieve clinical and statistical effi~acy,~”as opposed to the well-accepted usefulness of once daily evening dosing of antisecretory agents for active and maintenance therapy of peptic ulcer disease.6 Ambulatory pH studies have yielded objective support for multiple daily dosing by demonstrating that a b.d. antisecretory treatment regimen can normalize acid contact time of patients with severe symptomatic gastro-oesophageal reflux d i ~ e a s e while ,~ bedtime dosing fails to achieve this endpoint. Such physiological results suggest that b.d. dosing would be most effective in relieving symptoms in reflux patients. Famotidine has been in use for the treatment of duodenal ulcer, gastric ulcer, and other acid-related diseases for several years. It is a competitive H,-receptor antagonist and its dose related suppression of basal and stimulated gastric acid secretion has been demonstrated with oral doses from 5-40 mg.’ The purpose of this study was to investigate the safety profile and efficacy of famotidine oral dosing regimens, 40 mg nocte and 20 mg b.d. vs. placebo in the
B.D. FAMOTIDINE IN GASTRO-OESOPHAGEAL REFLUX
633
relief of symptoms in patients suffering from frequent heartburn found to have normal oesophagus or endoscopically-defined mild non-erosive oesophagitis. PATIENTS A N D M E T H O D S For entry into the study, all patients had to be 18 years of age (legal age of consent) or older. Criteria for enrollment into the study included heartburn, characterized by retrosternal burning pain present for approximately 15 out of 30 days prior to entry and occurring daily at least 5 of 7 days during the single-blind placebo period. A positive Bernstein test and endoscopic evidence of a lack of any break in the oesophageal mucosa (either normal oesophageal mucosa or erythema) were also required prior to entry. Patients with erosive oesophagitis were excluded. Patients were selected who had no history of taking therapeutic doses of H,-receptor antagonists for more than 5 successive days within the 2 weeks prior to beginning the placebo baseline period. Reasons for exclusion from the study included erosive oesophagitis (grade 2 or worse-see Endoscopy), manometric evidence of a primary oesophageal motility disorder, a history of Zollinger-Ellison syndrome, oesophageal disease other than oesophagitis or oesophageal stricture. Oesophageal and/or gastric varices, active peptic ulcer disease involving the stomach, pyloric channel or duodenum, paraoesophageal hernia or non-reducible sliding hiatal hernia and other standard exclusionary criteria were applied. Standard clinical laboratory tests were performed. Patients with abnormal hepatic enzymes (any value 25 % above the upper limit of normal value range) or with other severe chronic medical disorders were also excluded as were patients requiring concurrent medication that might affect acid secretion or mucosal healing such as corticosteroids, non-steroidal anti-inflammatory drugs (NSAIDs), antidepressants, salicylates, anticholinergics or antineoplastic agents. Twenty-two centres took part in this study and the protocol was approved by the appropriate institutional review board committee at every site. All patients who entered the study gave written informed consent.
Study protocol Patients who satisfied the inclusion/exclusion criteria, as determined by medical history, physical examination and clinical laboratory evaluation, were entered into the single-blind 1-week baseline period. At the time of entry, routine endoscopic evaluation and standard Bernstein testing were performed. Patients with endoscopic evidence of normal oesophageal mucosa or mild oesophagitis (grade 0 or I, see Endoscopy) and a positive Bernstein test’ were issued a diary card and given placebo to be taken before breakfast, before dinner and at bedtime. Antacid tablets (Gelusil) were dispensed at each visit throughout the study, to be taken for heartburn (not to exceed 30 tablets/week). An instruction sheet recommending conservative measures for the treatment 43
BAP 5
634
M. R O B I N S O N et al.
of gastro-oesophageal reflux disease was distributed to each patient upon entry into the study. The recommended measures included head of bed elevation, diet modification (that is, avoid fats, chocolate, alcohol, peppermint and spearmint ; avoidance of eating for 2-3 h before bedtime), and decreased or discontinued smoking. This double-blind multicentre trial randomized patients, 2 : 2 :1by design, into 40 mg famotidine nocte (n = 155), 20 mg famotidine b.d. (n = 158) and matching placebo ( n = 76). Dosing was before bedtime for the 40 mg once daily treatment and before both breakfast and dinner for the 20 mg b.d. regimen. Doses taken were indicated by the patient on the diary card as were the number of antacid tablets consumed daily. Patients also recorded daytime heartburn, nighttime heartburn and a global assessment (see Symptom assessment). Physical examination was performed, laboratory parameters were re-evaluated, and diary cards were reviewed at the time of Week 2 and Week 6. Symptom assessment Each patient kept a daily record of symptoms, including the number and severity of episodes of daytime and nighttime heartburn. Heartburn, defined as retrosternal warmth or burning was assessed using scores ranging from no heartburn to disabling heartburn. The scoring was defined in the following manner; (0) no heartburn, (I) mild; present but causing little or no discomfort, (2) moderate; annoying but not interfering with daily routine or sleep, ( 3 )severe; causing marked discomfort and some interference with daily routine or sleep, (4) disabling; interferes considerably with daily activities and/or sleep. A global assessment relative to how the patient felt during the placebo baseline week was obtained from each patient at the end of Weeks 2 and 6. The rating scale included (0) no improvement, (I)slight improvement, (2) moderate improvement or ( 3 )excellent improvement.
Endoscopy Oesophago-gastro-duodenoscopywas performed using standard flexible fiberoptic instruments. Endoscopic fmdings of the oesophageal mucosa were classified as follows : (0)normal mucosa, no abnormalities noted, (I)no macroscopic lesions but erythema or hyperaemia may be present ( 2 ) superficial ulceration or erosions involving less than 10%of the mucosal surface of the last 5 cm of oesophageal squamous mucosa, ( 3 ) superficial ulceration or erosions involving 10-50% of the mucosal surface of the last 5 cm of oesophageal squamous mucosa (4) deep ulceration anywhere in the oesophagus or confluent erosion of more than 50% of the mucosal surface of the last 5 cm of oesophageal squamous mucosa. Statistical analysis The analysis of efficacy at each time point was done as the endpoint analysis. An ‘all patients treated’ analysis used data from all patients who had baseline data.
B.D. FAMOTIDINE IN GASTRO-OESOPHAGEAL REFLUX
635
The distribution of patients across the 4 categories of global response was assessed using the Cochran-Mantel-Haenszel statistic." Global assessments were also analyzed by collapsing the 4 categories into 2: success (moderate or excellent improvement) or failure (none or slight improvement). The distribution of patients across the categories of success/failure was assessed using Fisher's exact test. Daytime heartburn and nighttime heartburn were assessed in several ways. The time until complete relief without recurrence was analysed using the MantelHaenszel life-table methods, stratifying on baseline scores. The distribution of patients across the categories of relief/no relief at the end of the study was analysed using Fisher's exact test. These scores were also assessed by computing a change in score from baseline of Days 1-7 and Weeks 2-6 and analysing these changes using analysis of variance on the ranked change from baseline. All tests were two-tailed with an alpha level of 0.05 considered significant. The study was designed to detect a 30 % difference in the proportion of heartburn-free patients between the famotidine groups and the placebo group with a 95 % power and a 15 % difference between the famotidine groups with an 80 % power. RESULTS
Pafienf demographics Three hundred and eighty-nine patients were randomized into this study following a one week single-blind baseline period. Patient characteristics at randomization are presented in Table I. Pairwise comparisons of all demographic characteristics indicated marginally significantly differences (0.05 < P < 0.10) in daily alcohol consumption between 40 mg famotidine nocte and 20 mg b.d. and between 20 mg famotidine b.d. and placebo. Most notable among the pairwise comparisons of baseline characteristics, illustrated in Table 2, was a significantly more severe daytime heartburn exhibited by patients in the 40 mg nocte group relative to the placebo group at baseline. In general, however, the treatment groups were similar with respect to baseline characteristics. Seventeen and 13 of the 389 initially randomized patients had no global evaluation (the primary efficacy variable) at Weeks 2 and 6, respectively, and were thus excluded from the intent-to-treat analysis,
Eficacy Global assessment. Although patient global assessment of symptom relief was recorded at Weeks 2 and 6, daytime and nighttime heartburn were rated daily throughout the single-blind baseline week and study period. Between group comparisons were based both upon the time until complete relief of symptoms (with no recurrence) and upon the change in heartburn score at each of Days 1-7 of active treatment and at Weeks 2-6. Global evaluations were analysed and distributed over four assessment categories (no improvement, slight, moderate or excellent improvement). Between 43-2
636
M. R O B I N S O N et al.
Table 1. Randomized patient demographic characteristics Famotidine 40 mg nocte ( M = 155)
20 mg b.d. (a = 158)
Placebo
Total
(n = 76)
(n = 389)
197 192
-
Gender Female Male Age (Years) Mean Standard deviation Race Caucasian Negro Other Smoking No Yes Alcohol (daily) No Yes Caffeine No Yes History of dysphagia No Yes History of acid regurgitation No Yes
81 74
85 73
31* 45
44.2 14.7
45.2 14.2
44.5 13.0
44.7 14.2
132 17 6
135 21 2
67 8 1
334 46 9
108 47
116 42
49 27
2 73 116
140** 15
151 7
68‘ 8
359 30
41 114
51 107
23 53
115 274
85 70
101 57
49 27
235 154
22 133
21 137
15 61
58 33 1
“Marginally significant difference found between 20 mg famotidine b.d. and placebo (0.05) < P < 0.10). ** Marginally significant difference found between 40 mg famotidine nocte and 20 mg famotidine b.d. (0.05 < P < 0.010).
group comparisons of the distributions of global assessments indicate that at Weeks 2 and 6, both farnotidine groups fared better than placebo. These findings are illustrated in Figure I, where the four assessment categories were collapsed into two : successful evaluation (moderate or excellent improvement) or unsuccessful evaluation (no or slight improvement) At Week 2, 69.9% of the patients in the 20 mg b.d. group achieved moderate or excellent improvement. This was significantly better than the 56.5 % of patients achieving similar improvement within the 40 rng nocte ( P d 0.05) group or the 50% in the placebo group (P d 0.01). Among the symptom assessment categories included in the global assessment, only ’excellent improvement’ included patients who were symptom free. At Week
B.D. FAMOTIDINE IN GASTRO-OESOPHAGEAL R E F L U X
637
Table 2. Patient baseline characteristics Famotidine 40 mg nocte 20 mg b.d. ( n = 155) (n = 158)
_____ Past history of heartburn frequency 134 Daily Weekly 21 Daytime heartburn severity" 3 None Mild 61 Moderate 86 Severe 4 Disabling 0 Missing data 1 Baseline nighttime Heartburn severity 28 None Mild 65 Moderate 55 Severe 5 Disabling 0 Missing data 2
(n = 76)
Total (n = 389)
127 31
64 12
325 64
3
0 32 34
6 164 194 21
71 74 8 2 0
32 65 47 12 2 0
Placebo
9 0
2
1
2
14 35 19 7 0 1
74 165 121 24 2 3
* Famotidine 40 mg nocte. Experienced significantly greater daytime heartburn severity at baseline than placebo, P < 0.05.
2, 22 % of patients within the 20 mg b.d., 25 Yo of patients within the 40 mg nocte group, and 14 % of patients within the placebo group were symptom free. By Week 6,45 % of patients receiving 20 mg b.d. famotidine were symptom free. In contrast, only 25% of patients within the 40 mg nocte group or placebo groups were symptom free. Famotidine 40 mg Fomotidine 20 mg b.d
ttxx
80 70
8
60
0
5
50
!$c
40 30
20 10
0 Week 2 x
Week 6
Farnotidine 20 mg b.d. significantly better than placebo P
< 0 01
t Famotidine 20mg b d significantly better than fomotidine 40 mg P (0.05 t i Farnotidine ZOrng b.d. significantly better than fornotidine
40mg P
