Dear Editor

With interest did we read the editorial of Dr Ken McColl [1] pertaining to our publication in the January issue on cure of Helicobacter pylori infection in patients with reflux oesophagitis treated with omeprazole.

We highly regard Dr McColl as an experienced clinical scientist and good friend. Nevertheless, we consider some of his editorial comments inaccurate and asking for a response.

The editorial stated that, contrary to our conclusion, our study provides further evidence that eradicating H. pylori infection makes it more difficult to achieve long term control of reflux disease with proton pump inhibitor therapy. It based this statement on two claims:
[a] that after two years the prevalence of reflux symptoms on omeprazole was 32 percent in those eradicated, versus 24 percent in those not eradicated, and
[b] that an analysis of changes in symptoms over the two-year follow- up period showed a minus 14 percent difference in the prevalence of symptoms in favour of the group not eradicated.

Unfortunately, both claims are wrong, leaving no ground for Dr McColl’s conclusion. First, we like to repeat that we presented our data by treatment arm, and not by H. pylori status at the end of follow-up. This is the correct statistical approach as it compares the two randomized groups. As a result, we did not analyse the prevalence of symptoms by H. pylori status, nor can these data be derived from our paper. Second, before any discussion, it might be helpful to correct the numbers calculated by Dr McColl. Simple math based on our Figure 4 shows that the prevalence of symptoms at the end of follow-up was 25 percent in the OM- only group, and the difference between groups in the decrease of any degree of reflux symptoms was 11%, based on a decrease of -4% in the OM- triple group, and -15% in the OM-only group. More importantly, in contrast to Dr McColl’s suggestion, these differences were not significant, nor did both groups differ in symptom severity (being moderate or severe in only 11 and 9% after two years in the OM-only and OM-triple group respectively), or the prevalence of oesophagitis (less than 8% at one and two year follow-up in both groups). In addition, the mean dose of omeprazole needed to control symptoms was almost identical in the two groups both at start of treatment and at the end. The decreasing prevalence of symptoms over time following eradication therapy and the absence of any difference with the control group completely refutes the statement that the present study gives evidence that eradicating H. pylori infection makes it more difficult to achieve long term control of reflux disease with proton pump inhibitor therapy. This is in agreement with previous literature in Caucasian populations as we pointed out in our Discussion. Dr McColl refers to one study which showed a minimal advantage in the healing of oesophagitis with PPI therapy for H. pylori-positives after 4 weeks treatment.[2] This advantage was no longer present after 8 weeks treatment. Our patients had on average a 5- year history of omeprazole use.

Dr McColl also addressed the issue of the effect of PPI therapy on H. pylori gastritis and concludes that our study is in line with others that contradict that PPI therapy accelerates corpus atrophy in H. pylori- infected subjects. Again, numbers are at stake, as we explained in our Discussion. In order to demonstrate that something does not occur, or is a slow process, a study requires large numbers of patients and many years of follow-up. Our current study, in contrast to a previous study,[3] did not have that strength, nor was it designed as such to investigate whether long term PPI treatment accelerates development of atrophy. The same was true for two other studies that Dr McColl referred to. Previous studies have suggested that H. pylori should be eradicated in long-term PPI users because of the persistent corpus-predominant pangastritis in these patients. This suggestion was approved by the European guidelines on H. pylori treatment, and also by dr. McColl.[4] It was however unknown whether H. pylori eradication would lead to improvement of advanced gastritis, and would not aggravate GORD. Our study aimed at these two issues, and we showed that corpus gastritis indeed improved, without significant impact on reflux disease.

We thus fully support the previous advice of Dr McColl that we should be recommending routine eradication of H. pylori infection in patients who need maintenance PPI therapy.[4]

References

(1) McColl KEL. Helicobacter pylori infection and long term proton pump inhibitor therapy. Gut 2004;53:5-7.

(2) Holtmann G, Cain C, Malfertheiner P. Gastric Helicobacter pylori infection accelerates healing of reflux esophagitis during treatment with the proton pump inhibitor pantoprazole. Gastroenterology 1999;117:11-6.

(3) Kuipers EJ, Lundell L, Klinkenberg-Knol EC, Havu N, Festen HPM, Liedman B, et al. Atrophic gastritis and Helicobacter pylori infection in patients with reflux esophagitis treated with omeprazole or fundoplication. N Engl J Med 1996;334:1018-22.

(4) McColl KEL, Murray LS, Gillen D. Omeprazole and accelerated onset of atrophic gastritis (Letter). Gastroenterology 2000;118:239.

Dear Editor

We read with interest the commentaries by McColl: Helicobacter pylori infection and long term proton pump inhibitor therapy.[1]

It is remarkable that he did not mention gastrin at all although hypergastrinemia is a result of reduced gastric acidity [2] as well as Helicobacter pylori (HP) infection,[3] and that patients with HP infection treated with proton pump inhibitors (PPI) have additive hypergastrinemia.[4] Hypergastrinemia predisposes to gastric carcinoids in animals [5,6] and man [7,8] as well as malignant ECL cell derived tumours (gastric carcinomas) in animals [9] and man.[10, 11]

Interestingly, the carcinogenic effect of Helicobacter pylori infection may be completely explained by its hypergastrinemic effect,[12] a work where McColl was one of the authors. Furthermore, the increased gastric cancer frequency in moderate hypergastrinemic INS-GAS mice concomitantly infected by HP infection [13] may also be caused by the increased hypergastrinemia in the infected mice.[14]

To conclude, it is odd that gastrin was not taken into consideration when discussing the risk of gastric cancer in the treatment with PPI in patients infected with HP. Animal as well as human studies linking gastrin to gastric cancer give supports for a strategy where HP is eradicated in patients on long-term PPI treatment.

References

1. McColl KEL. Helicobacter pylori infection and long term proton pump inhibitor therapy. Gut 2004;53:5-7.

2. Korman MG, Strickland RG, Hansky J. Serum gastrin in chronic gastritis. BMJ 1971;2:16-8.

3. Moss SF, Calam J. Acid secretion and sensitivity to gastrin in patients with duodenal lulcer: effect of eradication of Helicobacter pylori. Gut 1993:34:888-92.

4. Schenk BE, Kuipers EJ, Klinkenberg-Knol EC, et al. Hypergastrinaemia during long-term omeprazole therapy: influences of vagal nerve function, gastric emptying and Helicobacter pylori infection. Aliment Pharmacol Ther 1998;12:605-12.

5. Havu N. Enterochromaffin-like cell carcinoids of gastric mucosa in rats after life-long inhibition of gastric acid secretion. Digestion 1986;35 (suppl.):42-55.

6. Hirayama F, Takagi S, Iwao E, et al. Development of poorly differentiated adenocarcinoma and dcarcinoid due to long-term Helicobacter pylori colonization in Mongolian gerbils. J Gastroenterol 1999;24:450-4.

7. Sjöblom S-M, Sipponen P, Karonen S-L, et al. Argyrophil cell hyperplasia and carcinoid tumours in oxyntic mucosa of the stomach. Dependence on duration of pernicious anaemia. Eur J Gastroenterol Hepatol 1991;31:153-7.

8. Cadiot G, Vissuzaine C, Potet F, et al. Fundic argyrophic carcinoid tumor in a patient with sporadic-type Zollinger-Ellison syndrome. Dig Dis Sci 1995;40:1275-8.

9. Waldum HL, Rørvik H, Falkmer S, et al. Neuroendocrine (ECL-cell) differentiation of spontaneous gastric carcinomas of cotton rats (Sigmodon hispidus). Lab Anim Sci 1999;49:241-7.

10. Qvigstad G, Qvigstad T, Westre B, et al. Neuroendocrine differentiation in gastric adenocarcinomas associated with severe hypergastrinemia and/or pernicious anemia. APMIS 2002;110:132-9.

11. Qvigstad G, Falkmer S, Westre B, et al. Clinical and histopathological tumour progression in ECL cell carcinoids (“ECLomas�). APMIS 1999;107:1085-93.

12. Hansen S, Vollset SE, Ardill JES, et al. Hypergastrinemia is a strong predictor of distal gastric adenocarcinoma among Helicobacter pylori infected persons. Gastroenterology 1997;112:A575.

13. Wang TC, Dangler CA, Chen D, et al. Synergistic interaction between hypergastrinemia and Helicobacter infection in a mouse model of gastrin carcinoma. Gastroenterology 2000;118:36-47.

14. Waldum HL, Brenna E, Martinsen TC. Safety of proton pump inhibitors. Aliment Pharmacol Ther 2000;14: 1537-8.

Dear Editor

We read with considerable interest Dr Kuipers' paper [1], that found no significant adverse impact on the severity of reflux disease or its control after two-year omeprazole therapy following H pylori (Hp) eradication, during which time gastritis largely healed. In the commentary by Dr McColl [2] it is stated that, although published data are conflicting, Schwizer et al. reported improvement in reflux symptoms following Hp treatment.[3] Our preliminary published data,[4] which are in accordance with those by Dr Schwizer,[3] consisted of a small cohort of 69 patients with gastro-oesophageal reflux disease (GORD) and irritable bowel syndrome (IBS). Forty patients were treated with omeprazole (20 mg/day) plus trimebutine (600 mg/day) for three months (Group A), and 29 were treated with omeprazole (20 mg/day) as monotherapy for equal period of time (Group B). Inclusion and exclusion criteria matched those in Dr Kuipers' paper. Upper and lower GI endoscopic, histolog ical and clinical evaluations were made at baseline. Furthermore, upper GI evaluation was repeated three months post-treatment. At baseline, oesophagitis, confirmed by histology, and histological presence of Hp were observed in 67.5% and 62.1% and in 80% and 82.8% of Group A and B of patients, respectively. All Hp Hp positive patients received eradication treatment, as analysed in Dr Kuipers' paper. Eradication rate was observed in 84% of Hp+ve patients in group A and in 83% of Hp+ve patients in group B. Three months post-treatment, in group A there was a significant improvement in GORD (p=0.003), IBS symptoms (p <_0.0001 and="and" oesophagitis="oesophagitis" p="0.029)," compared="compared" with="with" group="group" b.="b." at="at" baseline="baseline" all="all" _24="_24" i="i">Hp+ve patients received omeprazole and eradication regimen had GORD symptoms and 15 (62.5%) had histologically proven oesophagitis. Three months post-treatment, GORD was present in 12 (50%) patients (p <_0.0001 and="and" oesophagitis="oesophagitis" in="in" _9="_9" _37.5="_37.5" patients="patients" p="p">0 .05). Since improvement in oesophagitis did not reach a statistically significant level, our study was continued and results are shown in the table (unpublished data). All 45 Hp+ve patients, who received omeprazole and eradication regimen, had GORD and 29 (64.4%) had oesophagitis, at baseline. Three months post-treatment, 22 (48.9%) had GORD symptoms (p <_0.00001 and="and" _18="_18" _40="_40" oesophagitis="oesophagitis" p="p" wilcoxons="wilcoxons" rank="rank" sum="sum" test="test" two-tailed="two-tailed" values="values" data="data" not="not" shown="shown" in="in" table.="table." there="there" was="was" statistically="statistically" better="better" response="response" the="the" patients="patients" who="who" also="also" received="received" trimebutine.="trimebutine."> In our preliminary study, Barrett's oesophagus was observed in 8 (20%) out of 40 and in 5 (17.2%) out of 29 patients in Groups A and B, respectively, similar to that (24.3%) observed in Dr Kuipers' paper. Since there is increased prevalence of IBS in patients with GORD,[5] it would be interesting to know how many of Dr Kuipers' patients had symptoms suggesting IBS, and if their regimen had similar results with our paper.

Our data show that Hp is frequent in GORD and may contribute to GORD pathogenesis by several mechanisms.[4] Besides, we propose that the increasing prevalence of GORD may be partially explained not just by the decrease in prevalence of Hp infection, as Dr McColl suggested,[2] but rather by healing of Hp-associated peptic ulcer disease, which coexists with GORD.[6,7] Thus, eliminating peptic ulcer disease unmasks GORD.[7]

In our latest unpublished data, 18 (48.6%) out of 37 patients, in whom Hp eradicated, had reflux symptoms on omeprazole compared with 4 (50%) out of 8 of Hp unsuccessful eradication. Although the latter group was too small to draw definite conclusions, it seems that eradicating Hp did not make GORD more difficult to control. While the editorial advocates that Hp eradication makes it more difficult to achieve long-term control of GORD with omeprazole therapy, we suggest that Hp eradication leads to better control of GORD symptoms and improves oesophagitis.

Table 1 Patients' profile prior and after therapeutic regimen and comparisons between them

	Group A (n=92) [Omeprazole plus Trimebutine]					Group B (n=56) [Omeprazole]					Significance (p)
	Prior		After 3 months			Prior		After 3 months			A v B		A1b v B1b	A2 v B2
	A1 (n=73)	A2 (n=19)	A1a (n=12)	A1b (n=61)	A2 (n=19)	B1 (n=45)	B2 (n=11)	B1a (n=8)	B1b (n=37)	B2 (n=11)
											Prior	After	After	After
GORD symptoms	73	19	5	10	3	45	11	4	18	7	NS		05	25
Oesophagitis	51	12	5	10	3	29	7	2	16	5	NS			NS
IBS symptoms	73	19	2	7	3	45	11	4	23	7	NS

p values between the two groups were calculated by chi-square test with Yate's correction (two-tailed p values) NS: not significant Group A: patients assigned to 3-month omeprazole plus trimebutine treatment Group B: patients assigned to 3-month omeprazole monotherapy Groups A1, B1: H pylori positive patients prior to treatment assigned to omeprazole plus trimebutine or omeprazole alone, respectively Groups A1a, B1a: patients from groups A1 and B1, respectively, who remained H pylori positive after eradication regimen Groups A1b, B1b: patients from groups A1 and B1, respectively, in whom H pylori infection eradicated Groups A2, B2: H pylori negative patients prior and after treatment with omeprazole plus trimebutine or omeprazole alone, respectively

References

1. Kuipers EJ, Nelis GF, Klinkenberg-Knol EC, et al.. Cure of Helicobacter pylori infection in patients with reflux oesophagitis treated with long term omeprazole reverses gastritis without exacerbation of reflux disease: results of a randomised controlled trial. Gut 2004;53:12-20.

2. McColl KEL. Helicobacter pylori infection and long term proton pump inhibitor therapy. Gut 2004;53:5-7.

3. Schwizer W, Thumshirn M, Dent J, et al.. Helicobacter pylori and symptomatic relapse of gastro-oesophageal reflux disease: a randomised controlled trial. Lancet 2001;357:1738-42.

4. Kountouras J, Chatzopoulos D, Zavos C, et al.. Efficacy of trimebutine therapy in patients with gastroesophageal reflux disease and irritable bowel syndrome. Hepatogastroenterology 2002;49:193-7.

5. Pimentel M, Rossi F, Chow EJ, et al.. Increased prevalence of irritable bowel syndrome in patients with gastroesophageal reflux. J Clin Gastroenterol 2002;34:221-4.

6. O'Connor HJO, McGee C, Mehana N, et al.. Prevalence of gastroesophageal reflux disease (GERD) in H. pylori-positive peptic ulcer disease and the impact of eradication therapy. Gastroenterology 1998;114:G1001 [abstract].

7. Ecclissato C, Carvalho AF, Ferraz JG, et al.. Prevalence of peptic lesions in asymptomatic, healthy volunteers. Dig Liver Dis 2001;33:403-6.