One-week therapy for Helicobacter pylori. A randomized trial of two treatment regimens.

In clinical practice, eradication of Helicobacter pylori infection may be difficult due to medication side effects and the need for 2 weeks of therapy. Because therapies of shorter duration may improve patient compliance and reduce treatment side effects, we compared the efficacy and tolerability of two anti-H. pylori treatments of 1 week's duration. Patients with H. pylori infection were randomized to treatment with either (a) short-course triple therapy, composed-of bismuth subsalicylate (Pepto-Bismol, Procter & Gamble, Cincinnati, OH, U.S.A.) two tablets four times daily, amoxicillin 1 g (two 500-mg tablets) twice daily, and metronidazole 500 mg four times daily on days 5-7 or (b) omeprazole 40 mg twice a day with amoxicillin 1 g twice a day for 1 week. At least 4 weeks posttreatment, efficacy was evaluated with either histological evaluation of antral biopsies for H. pylori or 14C urea breath testing. Patients who failed initial therapy were allowed to cross over to the alternative treatment regimen after a minimum "wash-out" period of 5 weeks. Patients completed a diary during therapy to monitor both compliance and side effects. Thirty-four patients completed the study, 10 receiving both treatment regimens. Treatment with the shortcourse triple therapy eradicated H. pylori in 78.3% of treatments compared with 38% with the high-dose omeprazole/ amoxicillin combination (p < 0.05). Patients were highly compliant with both treatments, and mild side effects, such as transient loose stools or abdominal pain, were common in both groups. This is the first report from North America confirming the success of the short-course triple therapy for the eradication of H. pylori. The high-dose omeprazole/ amoxicillin regimen's eradication rate was markedly inferior to that achieved by the short-course triple therapy regimen and should not be used. Comparative studies of the short-course triple therapy regimen with other 7-day anti-H. pylori treatment regimen therapies are indicated.

Single-day, divided-dose oral sodium phosphate laxative versus intestinal lavage as preparation for colonoscopy: efficacy and patient tolerance.

Polyethylene glycol-electrolyte lavage solutions are widely used to prepare the colon for colonoscopy. Unfortunately, some patients find this preparation difficult to complete. Recent studies of a sodium phosphate-based laxative have shown both good patient tolerance and good bowel preparation. In these studies, the laxative has generally been prescribed in two doses, with the second dose taken early the morning of colonoscopy. Because the morning dose is inconvenient for many patients, we compared giving a common polyethylene glycol-based electrolyte lavage solution the day before colonoscopy with our method of giving both doses of sodium phosphate-based laxative the day before colonoscopy: one dose at 4 PM and the second dose at 8 PM. We judged efficacy by an assessment of residual liquid and fecal matter in the colon and judged tolerance by the results of a symptom questionnaire completed by each patient immediately before the procedure. Our results in more than 200 patients showed similar efficacy ratings and similar symptom scores for both preparations, but patients rated the sodium phosphate-based preparation as easier to tolerate. In conclusion, in selected patients this new dosing method for sodium phosphate is preferable to large-volume, whole-gut lavage solutions.

Long-term follow-up of Helicobacter pylori treatment in non-ulcer dyspepsia patients.

BACKGROUND/AIMS: It remains controversial whether Helicobacter pylori infection causes symptoms in non-ulcer dyspepsia. One hundred non-ulcer dyspepsia patients were screened for H. pylori infection between November 1989 and February 1994. Forty patients entered a trial where both infected and uninfected patients were treated with H. pylori therapy, with the uninfected group serving as controls. METHODS: Non-ulcer dyspepsia was defined as unexplained epigastric discomfort lasting for at least 4 wk. From November 1989 until February 1992, all patients, regardless of H. pylori status, were treated with bismuth subsalicylate tablets (524 mg q.i.d.) for 4 wk and metronidazole (250 mg q.i.d.) for the first 2 of the 4 wk. From March 1992 until February 1994, only infected patients were treated in an attempt to obtain equal numbers in each group. H. pylori infection was diagnosed histologically at the index endoscopy and 1 month after completion of therapy. Symptoms were scored on a 0-5 scale for both frequency and severity. RESULTS: Of 100 patients screened, 33 were infected with H. pylori (mean age, 42; 10 men, 23 women), and 67 were uninfected (mean age, 38; 16 men, 51 women). Thirty-six uninfected patients were not offered treatment during the latter part of the trial. Of the remaining 31 uninfected patients, 10 dropped out; of the 33 infected patients, 14 dropped out. Twenty-one uninfected patients and 19 H. pylori-infected patients completed treatment; in 13 of 19 patients (68%), H. pylori was eradicated. Symptoms improved in eight of 13 (61%) H. pylori-eradicated patients and in four of six (66%) H. pylori-persistent patients, compared with 14 of 21 (66%) uninfected patients. Long-term follow-up (mean, 34 months) showed similar symptom outcome in the two treatment groups. CONCLUSIONS: Thirty-three percent of our non-ulcer dyspepsia patients were infected with H. pylori, a number similar to the percentage of infected age-matched controls in the U.S. Treatment with bismuth subsalicylate and metronidazole resulted in symptomatic improvement in 61-66% of non-ulcer dyspepsia patients regardless of initial or post-treatment H. pylori status. Long-term symptom follow-up in both the control and infected groups gave similar results. H. pylori infection is not related to the symptoms of non-ulcer dyspepsia.

Omeprazole ameliorates aspirin-induced gastroduodenal injury.

Aspirin and nonsteroidal antiinflammatory drugs (NSAIDs) damage the gastroduodenal epithelium by two mechanisms: direct toxic effects and effects related to the depletion of endogenous prostaglandins. The prostaglandin-depleted mucosa has increased susceptibility to luminal aggressive factors, yet the role of acid in the pathogenesis of the NSAID ulcer is controversial. In humans, standard doses of H2-receptor antagonists prevent only duodenal injury and provide no protection for the gastric mucosa. It is not known whether more potent suppression of acid can prevent NSAID damage. Twenty healthy volunteers were randomized to a double-blind, placebo-controlled, crossover study to determine if omeprazole, 40 mg/day prevents gastroduodenal injury due to two weeks of aspirin administration (650 mg four times a day). The severity of mucosal injury was quantitated by endoscopy and stratified by a scale from 0 (normal) to 4 (ulcer). Fourteen of the 20 subjects had less gastric injury during cotherapy with omeprazole. All six with no difference received aspirin plus omeprazole in the first treatment period. Omeprazole significantly decreased aspirin-induced gastric mucosal injury (P < 0.001, Wilcoxon signed-rank test). Omeprazole protected 85% of subjects from extensive gastric erosions (often associated with evidence of intraluminal bleeding) or ulceration, whereas 70% of the subjects developed aspirin-induced grades 3 and 4 gastric injury on placebo (P < 0.01 by chi 2). No subject taking omeprazole developed duodenal injury of any grade, while 50% taking placebo developed erosions and 15% had ulcer (P < 0.001). Medication side effects were mild in the majority of subjects.(ABSTRACT TRUNCATED AT 250 WORDS)

Comparison of coffee intake and coffee-induced symptoms in patients with duodenal ulcer, nonulcer dyspepsia, and normal controls.

Coffee and decaffeinated coffee stimulate acid secretion. In addition, many patients experience dyspepsia after coffee ingestion. Therefore, coffee is often prohibited by physicians in patients with peptic diseases. However, the association between peptic disease and symptoms remains unclear. This study compares coffee intake and the induction of symptoms by coffee in patients with duodenal ulcer disease, nonulcer dyspepsia, and normal controls. We have studied the coffee drinking habits of 58 duodenal ulcer patients, 55 nonulcer dyspepsia patients, and 55 normal controls. The use of coffee on a daily basis was not significantly different between duodenal ulcer patients (64%) and controls (56%), or between nonulcer dyspepsia patients (55%) and controls. There was also no difference between the three groups in the use of decaffeinated coffee, the number of cups per day, the method of preparation, the length of time of coffee use, or any change in coffee intake in the previous year. The intake of tea, caffeinated carbonated beverages, and aspirin or nonsteroidal anti-inflammatory drugs was also similar in the three groups. The duodenal ulcer patients were more likely to be cigarette smokers (45%) than either the controls (16%) or the nonulcer dyspepsia patients (24%). Daily alcohol intake was not significantly different in the three groups. The prevalence of coffee induction of dyspeptic symptoms was similar in duodenal ulcer patients (29%) and controls (22%), but was much more common in nonulcer dyspepsia patients (53%) than in controls (22%), p = 0.0036. In conclusion, there was no difference in coffee intake between patients with duodenal ulcer, nonulcer dyspepsia, or normal controls. However, patients with nonulcer dyspepsia, but not duodenal ulcer, were more likely to experience dyspeptic symptoms after coffee ingestion.

Campylobacter pylori is not associated with gastroparesis.

There is a high incidence of Campylobacter pylori in the gastric mucosa of patients with duodenal ulcer, gastric ulcer, and nonulcer dyspepsia. Factors that lead to development of this infection are unknown. We hypothesized that delayed solid-phase gastric emptying, a condition characterized by antral stasis, might predispose to Campylobacter pylori infection. We prospectively studied 51 patients with symptoms of gastroparesis using a solid-phase gastric emptying study and upper endoscopy. Patients were excluded if they had predominant symptoms of epigastric pain or an abnormal endoscopy. Three biopsies were obtained from the antrum and stained with H&E. When any inflammation was present, a Warthin-Starry stain was also performed. These were blindly examined for chronic inflammation, activity, and presence of Campylobacter pylori. Campylobacter pylori was not more common in patients with gastroparesis, documented by delayed gastric emptying, than in patients with a normal emptying study. On the contrary, there was a significantly lower incidence of Campylobacter pylori in those with delayed emptying compared to those with normal emptying (5% vs 31%, P less than 0.05). Gastritis activity correlated closely with Campylobacter presence. Inactive chronic gastritis with Campylobacter was equally common in those with delayed or normal gastric emptying. Diabetics were no more likely to harbor Campylobacter pylori than nondiabetics (16% vs 25%). The 5% incidence of Campylobacter in the gastroparesis group is less than, but approaches, that previously reported in asymptomatic controls. The 31% incidence of Campylobacter in the group with symptoms of gastroparesis but normal gastric emptying approaches that reported for nonulcer dyspepsia. Our data suggest that gastroparesis does not predispose to Campylobacter pylori infection or histologic chronic gastritis.

Campylobacter pylori in patients with dyspeptic symptoms and endoscopic evidence of erosion(s).

The relationship between Campylobacter pylori (CP), histologic gastritis, and dyspeptic symptoms is becoming gradually clearer, but there is still a lack of knowledge of the natural history of treated or untreated gastritis. We examined serial biopsies from the gastric fundus, body, and antrum, and from the duodenum in 16 dyspeptic patients. Patients with concomitant peptic ulcers, alcoholism, or nonsteroidal anti-inflammatory drug use were excluded. CP was present in the biopsies of 50% of patients at presentation. When CP was present, the antrum was always infected, and often had the highest density of organisms. In the duodenum, CP was found only in areas of gastric metaplasia. The presence of CP was highly correlated with gastritis activity (neutrophilic infiltrate). A 4-yr follow-up study of symptoms, endoscopic appearance, and histologic findings including the presence of CP was performed in 10 of the original 16 patients. After 4 yr, both the severity and frequency of epigastric pain remained the same in seven patients, worsened in one, and improved in two. All patients who had CP at initial presentation retained the organism (5/10), whereas none of the previously noninfected patients acquired the infection (5/10). Both CP-positive and -negative patients were treated for 3 wk with 524 mg bismuth subsalicylate qid, and for the first 2 of 3 wk with 250 mg metronidazole qid. One patient who was CP positive was lost to follow-up. In three of the remaining four patients on this regimen, the organism was eradicated. Of the nine patients who completed the treatment program, two had no change in symptoms and seven improved. CP was present in three of seven with improved symptoms and in one of two with no change in symptoms. After treatment, the only change in histology was the disappearance of activity in the CP-positive patients who lost the organism. In conclusion, CP was present in 50% of dyspeptic patients with endoscopic evidence of at least one erosion. Both the symptoms and CP persisted for 4 yr. Dyspeptic symptoms improved after bismuth subsalicylate/metronidazole therapy, regardless of the presence or absence of CP, although the regimen did succeed in eradicating the organism in three of the four CP-positive patients who completed the study.

Salicylsalicylic acid causes less gastroduodenal mucosal damage than enteric-coated aspirin. An endoscopic comparison.

The gastroduodenal mucosal damage caused by aspirin and nonsteroidal antiinflammatory drugs is a common clinical problem. We compared two medications designed to diminish mucosal damage: enteric-coated aspirin and salicylsalicylic acid (salsalate). Ten healthy volunteers were randomized to receive either 1.5 g salsalate twice a day or 650 mg enteric-coated aspirin four times a day for six days and were then crossed over to the other drug after a one-week medication-free period. Endoscopic inspection of gastroduodenal mucosa was performed at entry and again after six days of drug therapy for each medicine. Mean serum salicylate concentrations taken before the morning drug dose were 11.2 mg/dl for enteric-coated aspirin and 18.1 mg/dl for salsalate. Only one of 10 subjects receiving salsalate developed mild (grade 1) mucosal damage while six of 10 receiving enteric-coated aspirin developed moderate to severe damage (grade 2-3) (P = 0.01). Symptoms were mild in both groups. We conclude that salsalate causes less gastroduodenal mucosal damage than enteric-coated aspirin.

A study of the correlation between endoscopic and histological diagnoses in gastroduodenitis.

Serial histological specimens from 14 patients with the endoscopic diagnosis of erosive gastritis and/or duodenitis were examined for correlation between endoscopic and histological findings. All patients were symptomatic outpatients without history of alcoholism or usage of aspirin or nonsteroidal antiinflammatory drugs. After the initial diagnosis, the patients underwent follow-up endoscopy until healing of erosions at 1, 4, and 8 wk. Pairs of biopsies from the gastric fundus, body, and antrum, and the duodenum were obtained at each endoscopy. Agreement between histological and endoscopic findings occurred in only 56% of the 161 sites studied. The best correlation occurred in the duodenum when there was endoscopic disease (89%) and was worst in the stomach at all sites regardless of endoscopic findings (46%). A normal histology in the face of abnormal endoscopic changes was seen in only 16% of all biopsies. Histological inflammation occurred in 27% of all biopsies with a normal endoscopic appearance and in 55% of the normal endoscopic areas in the stomach. Histological appearances at each biopsy site remained constant in individual patients throughout the study. The specific histological findings, such as activity and severity, did not correlate with the endoscopic severity of inflammation or with any specific endoscopic appearances, such as erosions, petechiae, or nodules. In conclusion, the histological and endoscopic findings in the stomach from patients with symptomatic erosive gastroduodenitis correlate poorly while good correlation occurs in the duodenum.