Role of Helicobacter pylori eradication in aspirin or non-steroidal anti-inflammatory drug users

Abstract

Helicobacter pylori (H pylori) infection and the use of non-steroidal anti-inflammatory drugs (NSAIDs) including aspirin at any dosage and formulation represent well-established risk factors for the development of uncomplicated and complicated peptic ulcer disease accounting for the majority of such cases. Although the interaction between H pylori and NSAID/aspirin use in the same individuals was questioned in some epidemiological studies, it has now become widely accepted that they are at least independent risk factors for peptic ulcer disease. According to data from randomized intervention trials, naive NSAID users certainly benefit from testing for H pylori infection and, if positive, H pylori eradication therapy prior to the initiation of NSAID. A similar strategy is also suggested for naive aspirin users, although the efficacy of such an approach has not been evaluated yet. Strong data also support that chronic aspirin users with a recent ulcer complication should be tested for H pylori infection and, if positive, receive H pylori eradication therapy after ulcer healing, while they appear to benefit from additional long-term therapy with a proton pump inhibitor (PPI). A similar approach is often recommended to chronic aspirin users at a high risk of ulcer complication. H pylori eradication alone does not efficiently protect chronic NSAID users with a recent ulcer complication or those at a high-risk, who certainly should be treated with long-term PPI therapy, but H pylori eradication may be additionally offered even in this setting. In contrast, testing for H pylori or PPI therapy is not recommended for chronic NSAID/aspirin users with no ulcer complications or those at a low risk of complications.

INTRODUCTION

Salicylates have been used in therapeutic medicine since the Hippocrates’ era and their use is still growing. During the last 50 years, there is a continuously increasing consumption of aspirin for cardioprotection and for secondary prophylaxis of recurrent stroke or other vascular occlusion, while the drug seems to have a possible role in chemoprevention of cancer and Alzheimer’s disease[1-4]. Non-steroidal anti-inflammatory drugs (NSAIDs) are also widely used agents[5]. In the USA, it is estimated that more than 50% of the population over 65 years take aspirin or NSAIDs frequently[6].

The increasing widespread consumption of aspirin/ NSAIDs, however, is associated with an increasing incidence of their well-known gastrointestinal complications, which include dyspepsia, gastric and/or duodenal erosions and ulcers and peptic ulcer complications. Peptic ulcer complications, usually bleeding, represent the most frequent serious adverse events of the use of aspirin/NSAIDs[1,7]. Peptic ulcer(s) may be found at endoscopy in up to 20-25% and ulcer complications requiring hospital admission develop in 2-5% of chronic users of NSAIDs[7-12]. Use of NSAIDs has also been shown to increase the risk of lower gastrointestinal bleeding[13]. The damaging effect of aspirin on the gastric mucosa may be less potent than the effect of NSAIDs[14]. Thus, it is estimated that the chronic use of aspirin increases the absolute annual risk of gastrointestinal bleeding by 0.04% (absolute annual risk of bleeding with and without aspirin: 0.09% and 0.05% respectively)[15]. Nevertheless, despite the relatively low absolute risk of bleeding in aspirin users, the numbers of aspirin related acute gastrointestinal bleeding episodes are rather high probably due to the huge numbers of individuals who take the drug regularly for long periods often having additional factors with increased risk for bleeding, such as old age and history of peptic ulcer disease. The use of selective NSAIDs, such as selective cyclo-oxygenase 2 (COX-2) inhibitors, significantly reduces but does not completely eliminate the risk of gastrointestinal complications[11,12,16], while their gastrointestinal benefit appears to be significantly restricted in cases of concomitant use of aspirin, even at low doses[11,17].

Helicobacter pylori (H pylori) is undoubtedly associated with the development of gastritis and uncomplicated and complicated peptic ulcer diseases[18]. Although the presence of two factors that can damage the gastric mucosa, such as H pylori and aspirin/NSAIDs, would be reasonably considered to increase the risk for development of uncomplicated and complicated peptic ulcer, data from several, mainly epidem-iological, studies appear to be controversial and do not always confirm such an assumption[19]. This review focuses on the role of H pylori infection and the need for its eradication for prevention of gastrointestinal complications among aspirin/ NSAIDs users by evaluating the relevant pathophysiological and epidemiological data as well as the results of the randomized, controlled clinical trials of therapeutic intervention.

PATHOPHYSIOLOGY

Aspirin or NSAIDs use is associated with the development of peptic erosions or ulcers through several mechanisms. First, aspirin acts locally through the release of salicylic acid in the stomach, which is not ionized by the gastric acid. Salicylic acid enters and accumulates within the gastric epithelial cells, is ionized intracellularly and disrupts cell metabolic functions increasing mucosal permeability and permitting the back diffusion of H⁺ ions[20]. Moreover, aspirin promotes topical inflammation by inducing recruitment of leukocytes, which eventually results in capillary constriction and topical ischemia. The topical gastrotoxic effect of aspirin, however, does not seem to be particularly important, since it is associated only with superficial ulcerations that often resolve spontaneously despite the continued aspirin use[20]. The systemic gastrotoxic effect of aspirin is related to the inhibition of cyclo-oxygenase-1 (COX-1) and the subsequent disruption of prostaglandin synthesis and to the antiplatelet function that promotes bleeding complications[20]. The key role of the systemic effects of aspirin in the development of gastrointestinal complications is strongly supported by the data showing that the risk of such complications is independent of the drug formulation[17,21,22]. Even low doses of aspirin, such as 75 mg/d, have been shown to increase the risk of gastroduodenal ulcerations[21,22]. Inhibition of COX-1 with disruption of prostaglandin production is also the main mechanism of NSAIDs induced gastroduodenal complications[11-13].

H pylori infection induces a substantial inflammatory reaction in the gastric mucosa with recruitment of leukocytes and production of several inflammatory cytokines, which eventually result in attenuation of mucosal defense mecha-nisms[18]. Thus, H pylori infection and aspirin/NSAID use impaired gastric mucosal defense by different mechanisms and therefore an interaction between these two factors is biologically plausible.

EPIDEMIOLOGICAL STUDIES

The interaction between H pylori infection and aspirin/NSAIDs use in the development of ulcer and ulcer complications has been initially evaluated in several cohort or case-control studies. The findings of these studies, however, have been controversial, since some studies suggested an independent or additive role of H pylori infection and aspirin/NSAIDs use in gastrointestinal complications[21-30] and others proposed no association or even a protective role of H pylori infection in users of aspirin/NSAIDs[31-33]. Moreover, in one study, H pylori infection was found to increase the risk of gastric but not of duodenal ulceration in this setting[34]. The heterogeneity in study design and methodology, definitions, power, outcome, and selection of controls have been suggested to be responsible for such conflicting results[19].

In a systemic review published in 2002, the combined analysis of the data available up to October 2000 showed that there is synergism for the development of peptic ulcer and ulcer bleeding between H pylori infection and aspirin/ NSAID use[35]. In particular, the presence of H pylori infection was found to increase 3-5-fold the risk of peptic ulcer in aspirin/NSAID users (prevalence of peptic ulcer in H pylori positive: 53% and H pylori negative: 21%, OR: 3.5) and 18-fold in subjects not taking aspirin/NSAID (prevalence of peptic ulcer in H pylori positive: 18% and H pylori negative: 0%, OR: 18.1)[35]. Thus, the risk of peptic ulcer is approximately 60-fold higher in H pylori positive aspirin/NSAID users compared with H pylori negative subjects not taking aspirin/ NSAID[35]. Moreover, H pylori infection was shown to increase the risk of ulcer bleeding 1.8-fold, aspirin/NSAID use 4.85-fold, and the presence of both factors 6.1-fold compared with the risk of bleeding among H pylori negative subjects not taking aspirin/NSAID[35]. H pylori infection has also been found to increase the risk of upper gastrointestinal bleeding even in chronic users of low dose aspirin[27]. In a more recent case-control study from our group, H pylori infection was again found to increase the risk for upper gastrointestinal bleeding in aspirin/NSAID users 2.9-fold, or 1.7-fold when adjustment for other risk factors for bleeding was performed[28]. Taking all together, it seems that aspirin/NSAID use and presence of H pylori infection are at least independent risk factors for peptic ulcer and bleeding from peptic ulcer.

RANDOMIZED CLINICAL TRIALS

H pylori eradication in naive aspirin/NSAID users

If H pylori gastritis does enhance the risk for ulcer bleeding in aspirin/NSAID users, then H pylori eradication should substantially reduce such a risk in this setting. Since the risk of bleeding in aspirin/NSAID users is strongly related to the duration of drug use, being higher in subjects with new or recent drug onset (< 1-3 mo) than in chronic drug users (> 1-3 mo)[36-38], the possible beneficial effect of H pylori eradication on naive aspirin/NSAID users was initially evaluated. In fact, only naive users of non-aspirin NSAIDs have been included in the relevant clinical trials to date, while the possible benefit of H pylori eradication in naive users of aspirin has not been evaluated yet.

H pylori eradication before NSAID use was found to significantly reduce the occurrence of peptic ulcers in 92 H pylori positive, NSAID naive patients with musculoskeletal pain treated with an 8-wk course of naproxen at a daily dose of 750 mg (peptic ulcers: 3/45 or 7% of patients in the H pylori eradication group vs 12/47 or 26% of patients in the placebo group, P = 0.01)[39]. In a longer trial with a similar design, H pylori eradication before NSAID use was again found to significantly reduce the risk of peptic ulcers in 100 H pylori positive, NSAID naive, patients with arthritis and a history of peptic ulcer or dyspepsia treated with a 6-mo course of diclofenac slow release at a daily dose of 100 mg (peptic ulcers: 5/51 or 12% vs 15/49 or 34%, P = 0.0085)[40]. In the latter trial, H pylori eradication was also found to significantly reduce the risk of ulcer complications as well [6-mo probability: 4.2% (1.3-9.7) vs 27.1% (14.7-39.5), P = 0.0026][40].

H pylori eradication in chronic aspirin/NSAID users without a history of peptic ulcer complications

The results of the first large clinical trial of H pylori eradication in chronic NSAID users raised several questions for the benefit of such an intervention. In this trial[41], 285 H pylori positive chronic NSAID users with past or current peptic ulcers or NSAID-associated dyspepsia who continued a minimum dosage of NSAID for at least 6 mo were randomized to receive H pylori eradication therapy with omeprazole, amoxycillin and clarithromycin (n = 142) or omeprazole plus placebo antibiotics (n = 143) for 1 wk. Subsequently, all patients received omeprazole 20 mg daily for 3 wk followed by an additional 4-wk omeprazole course in cases with endoscopically detected peptic ulcers at 4^th wk. The probability of being peptic ulcer free at 6^th mo was similar in the H pylori eradication [0.56 (95%CI: 0.47-0.65)] and the omeprazole-control group [0.53 (95%CI: 0.44-0.62)], while healing of gastric ulcers was significantly impaired in the H pylori eradication group (gastric ulcers healed at 8^th wk: 72% in the H pylori eradication group vs 100% in the omeprazole-control group, P = 0.006)[41].

The design of the latter trial, however, was different from the design of the trials in naive NSAID users, since H pylori eradication therapy was given to subjects with ulcers or at high-risk of ulcers, who had already been on long-term NSAID consumption. Moreover, both the H pylori eradication and control groups were treated with 4-8 wk of omeprazole for ulcer healing. The lower probability of gastric ulcer healing at 8^th wk in the H pylori eradication group should be associated with the more potent antisecretory activity of the PPIs including omeprazole in the presence than absence of H pylori infection[42]. Similar findings have also been observed in another large study including 692 chronic NSAID users, in which gastric ulcer healing with ranitidine or lansoprazole was shown to be significantly enhanced in the presence of H pylori infection (healing of gastric ulcers at 8^th wk: 70% in H pylori positive vs 61% in H pylori negative, P < 0.05)[43]. According to these data, it has been reasonably suggested that any attempt to eradicate H pylori infection should follow ulcer healing in the management of chronic NSAID users, although the efficacy of such an approach remains to be tested. The efficacy of H pylori erad-ication in chronic aspirin users has not been evaluated yet.

H pylori eradication in chronic aspirin/NSAID users with a recent peptic ulcer complication

Subjects with a history of upper gastrointestinal bleeding or other peptic ulcer complications represent a particular subgroup of aspirin/NSAID users who are at a high risk for recurrent bleeding during continued aspirin/NSAID use[44,45]. Strategies that may prevent bleeding in this setting include concurrent therapy with a PPI or eradication of H pylori infection in H pylori positive subjects. The efficacy of these two strategies was evaluated in a large clinical trial including 400 H pylori positive aspirin/NSAID users with a history of upper gastrointestinal bleeding[46]. All patients initially discontinued aspirin or NSAID therapy and were treated with omeprazole 20 mg daily for at least 8 wk to promote ulcer healing. Once the healing of ulcer was confirmed, 250 patients who were given 80 mg of aspirin daily for heart disease or stroke and 150 patients who were given 500 mg of naproxen twice daily for arthritis, both for at least 6 mo, were separately randomized to receive 20 mg of omeprazole daily for 6 mo or a 7-d course of H pylori eradication therapy followed by placebo once daily for 6 mo. In patients taking aspirin, no significant difference in the probability of recurrent bleeding during the 6-mo follow-up period was observed between those who received H pylori eradication therapy (1.9%) and those who received omeprazole (0.9%) (absolute difference: 1%, 95%CI: -1.9-3.9%). In contrast, in patients taking naproxen, the 6-mo probability of recurrent bleeding was significantly lower in the omeprazole (4.4%) than in the H pylori eradication group (18.8%) (absolute difference: 14.4%, 95%CI: 4.4-24.4%, P = 0.005)[46]. According to these data, it seems that, after ulcer healing, H pylori eradication may be effective in preventing recurrence of upper gastrointestinal bleeding in chronic aspirin users, but not in chronic NSAID users, who require long-term potent antisecretory therapy with a PPI.

Whether the combination of H pylori eradication and long-term use of PPIs may further decrease the risk of recurrence of peptic ulcer complications in chronic aspirin users was evaluated in a recent clinical trial[47]. Thus, 123 H pylori positive patients with a history of an aspirin-related peptic ulcer complication and current peptic ulcer were initially treated with a 7-d H pylori eradication therapy followed by 40 mg of famotidine daily for 5 or 13 additional weeks until ulcer healing. Then, they all restarted taking 100 mg of aspirin daily and randomized to receive 30 mg of lansoprazole daily or placebo. During a median follow-up of 12 mo, recurrence of ulcer complications was observed in 9 (14.8%) of 61 patients in the placebo group and in only 1 (1.6%) of 62 patients in the lansoprazole group (adjusted hazard ratio: 9.6, 95%CI: 1.2-76.1, P = 0.008)[47]. It should be noted, however, that four of the nine placebo treated patients with a recurrence of ulcer complications were reinfected with H pylori and an additional two patients of this group took other NSAIDs. Despite these problems in the latter trial, it is becoming widely accepted that long-term therapy with a PPI after H pylori eradication offers additional benefit in preventing peptic ulcer complications in high risk H pylori positive chronic aspirin users[19].

DISCUSSION

All existing data suggest that the presence of H pylori infection represents an additional risk factor for peptic ulcer compli-cations in aspirin/NSAID users[19,48]. However, in current clinical practice which should be guided by the evidence-based medicine and should take into account the cost/benefit analysis of any major intervention, the management of H pylori infection and generally the gastrointestinal prevention in aspirin/NSAID users should probably be individualized (Table 1). Thus, the optimal management of such subjects appears to depend on the main factors affecting the risk of ulcer complications, which are: (1) whether the subject is a naive aspirin/NSAID user or already on long-term (> 1-3 mo) or chronic drug use[36-38]; (2) whether the subject is at high risk for bleeding or other complication of peptic ulcer (history of complicated or uncomplicated peptic ulcer, age older than 60-65 years, recent dyspepsia, treatment with anticoagulants)[6,36-38,49]; and (3) perhaps whether they take aspirin or non-aspirin NSAID[19,48].

Table 1.

Recommendations[19,48] and evidence for the H pylori test-and-treat approach and/or the long-term therapy with a proton pump inhibitor (PPI) in users of aspirin or non-steroidal anti-inflammatory drugs (NSAIDs)

	H pylori test-and-treat approach	Long-term PPI therapy
Naive aspirin users	Recommendation	No
Naive NSAIDs users	Recommendation-evidence[39,40]	No
Chronic aspirin users
With a recent ulcer complication	Recommendation1-evidence[46,47]	Recommendation-evidence[47]
At high risk for ulcer complication	Recommendation1	Recommendation
At low risk for ulcer complication	No	No
Chronic NSAIDs users
With a recent ulcer complication	Potential benefit12	Recommendation-evidence[46]
At high risk for ulcer complication	Potential benefit12	Recommendation
At low risk for ulcer complication	No	No

¹H pylori eradication therapy in chronic users of aspirin or NSAIDs with a recent ulcer complication or those at a high-risk should be administered after confirmation of ulcer healing.

²H pylori eradication therapy in chronic NSAIDs users with a recent ulcer complication or those at a high-risk may be given as a potentially beneficial intervention in addition to the long-term PPI therapy.

In naive NSAID users, it is well accepted and supported by strong data[39,40] that they should be tested for the presence of H pylori infection and, if positive, receive H pylori eradication therapy before NSAID use[19-48]. A similar strategy is also suggested for naive aspirin users[19], although the efficacy of such an approach has not been evaluated yet.

In chronic aspirin/NSAIDs users, the recommendations may depend on the risk for peptic ulcer complications[6] and the type of drug. The indication for use of aspirin or NSAIDs should be first evaluated in all such users at high risk for peptic ulcer complications. Moreover, the probability and the cost/benefit of replacement of aspirin or NSAID with a less gastrotoxic antiplatelet agent or a selective COX-2 inhibitor respectively may be considered[6,11,12].

All individuals, who should continue taking aspirin after development of a peptic ulcer complication, should be tested for the presence of H pylori infection and, if positive, receive H pylori eradication therapy after peptic ulcer healing. In addition, they should subsequently receive long-term therapy with a PPI[19,47]. A similar approach may be recommended in chronic aspirin users without a recent ulcer complication but at high risk for ulcer complication, such as those with a history of peptic ulcer[19]. It should be noted, however, that there are no strong data to support the combined prophylactic approach with both H pylori eradication and long-term PPI therapy in this setting.

All individuals, who should continue taking NSAIDs being at high-risk for peptic ulcer complication, certainly benefit from long-term therapy with a PPI[19,46,48]. The risk of relapse of ulcer complication in chronic NSAIDs users taking PPI, however, is higher than the risk of such a relapse in aspirin users irrespective of the type of gastroprotection[46]. Thus, given that H pylori infection represents an independent risk factor for gastrointestinal bleeding in chronic NSAIDs users[35], it is often recommended that testing for H pylori infection and, if positive, H pylori eradication therapy should be offered to high-risk chronic NSAIDs users in addition to the long-term PPI therapy[19], despite that there are no strong data to support such an approach.

Finally, testing for H pylori infection or PPI therapy is not recommended for chronic users of aspirin or NSAIDs with no peptic ulcer or complication or those at a low risk of the same[19].