Treatment of Peptic Ulcer Disease and Nonulcer Dyspepsia

In the June 2001 issue of The Journal of Family Practice, the diagnostic approach to the patient with dyspepsia was presented.¹ In that analysis, gastroesophageal reflux disease (GERD), gastric ulcers, and duodenal ulcers emerged as the most common identifiable causes of dyspepsia. However, most patients with dyspepsia do not have one of these conditions, and are considered to have functional or nonulcer dyspepsia. The diagnosis and management of adults with GERD was recently described in detail.² Therefore, this paper reviews the treatment of undifferentiated dyspepsia, gastric ulcer caused by nonsteroidal anti-inflammatory drugs (NSAIDs), peptic ulcer disease not associated with NSAID use, and nonulcer dyspepsia (dyspepsia in a patient who has no evidence of ulcer or GERD on endoscopy). An algorithm for the management of the patient with known ulcer disease is shown in the Figure 1. (J Fam Pract 2001; 50:614-619)

Undifferentiated dyspepsia

In primary care, the typical patient presenting with dyspepsia will not have had endoscopy. Therefore, the presence of an underlying lesion will be unknown, a situation known as undifferentiated dyspepsia. As described by Smucny,² randomized trials and economic analyses have demonstrated the cost-effectiveness of a test-and-treat strategy^3,4 in which patients with dyspepsia are tested for Helicobacter pylori and treated with eradication therapy if positive. This strategy would reserve endoscopy for those patients with alarm signs Table1 or those who have persistent symptoms despite appropriate empiric therapy. Certainly a physician must weigh the potential for complications during endoscopy with the risks of adverse reactions to eradication therapy, including the development of antimicrobial resistant organisms. All patients with dyspepsia should be counseled to avoid factors that exacerbate their symptoms or disrupt the integrity of the mucosal lining of the stomach, such as NSAID use and cigarette smoking. Both of these have been identified as risk factors for the development of peptic ulcers and delayed ulcer healing.^5,6

NSAID-related gastric ulcers

NSAIDs are associated with a 5- to 7-fold increased risk of gastric ulceration in the first 3 months of use. In a meta-analysis of observational studies of gastrointestinal bleeding risk due to various NSAIDs, a 4-fold increased risk associated with NSAID-use persisted throughout therapy and fell to baseline within 2 months of discontinuation of the NSAID.⁷ This study demonstrated a clear dose-response relationship; the difference between NSAIDs, however, was minimal.

Table 2 summarizes treatments for NSAID-related gastric ulcers. Misoprostol is an effective prophylaxis against ulcers when used with NSAIDs, but is associated with diarrhea, even at lower than optimal doses.⁸ Standard doses of H2-receptor agonists (H2RAs) are ineffective at preventing NSAID-related gastric ulcers, but double doses of H2RAs (eg, ranitidine 300 mg twice daily) and standard doses of proton-pump inhibitors (PPIs; eg, omeprazole 20 mg 4 times per day) are effective prophylactic agents for the duration of NSAID use according to the results of endoscopic studies. New COX-2 specific anti-inflammatory agents are associated with a significantly lower risk of ulcers as seen by endoscopy (4.7% with rofecoxib vs 27.7% with ibuprofen).⁹ The benefit in terms of actual adverse clinical outcomes and ulcer complications, however, is much smaller; the risk of symptomatic ulcer, perforation, symptomatic ulcer, and clinically significant bleeding was 1.3% for rofecoxib (Vioxx) and 1.8% with ibuprofen, diclofenac, or nabumetone taken for 1 year (P <.05).¹⁰ Thus, one would have to treat 200 patients for 1 year to prevent 1 adverse outcome. Similarly, the annual risk of bleeding, perforation, or gastric outlet obstruction was lower with celecoxib (Celebrex) than with naproxen, diclofenac, or ibuprofen (0.2% vs. 1.68%; P <.002; number needed to treat=60).11 COX-2 inhibitors have not been studied as an alternative therapy in patients with a history of NSAID-induced ulcers. While no trial data are available, a consensus-based recommendation has been made for dyspeptic patients with no alarm symptoms who are regular NSAID-users: The first step in management is to stop the NSAID use if possible, and determine if the symptoms improve.¹² If symptoms persist after NSAID use is discontinued, the patient should be managed as others with undifferentiated dyspepsia.

Non–NSAID-related duodenal and gastric ulcers

Helicobacter pylori is now well recognized as a major risk factor for the development of peptic ulcer disease. Although most H pylori–infected patients do not develop an ulcer, as many as 95% of patients with duodenal ulcers and 80% of those with gastric ulcers are infected. These rates may be lower in the United States because of greater use of NSAIDs and a lower rate of H pylori infection than in other parts of the world.¹³ Successful eradication of the organism following treatment heals ulcers and reduces the risk of recurrence from 67% to 6% for patients with duodenal ulcers, and from 59% to 4% for patients with gastric ulcers.¹⁴ A meta-analysis of North American randomized controlled trials of H pylori eradication for duodenal ulcer found that one ulcer recurrence (by endoscopy) would be prevented for every 2.8 patients successfully treated.¹⁵

However, a multi-drug regimen and an adequate length of treatment are required for eradication. Because antimicrobial resistance and incomplete treatment are major reasons for treatment failure,^16,17 convenience and tolerability become important considerations in choosing a treatment plan.

Effective H pylori treatment regimens include a combination of two antibiotics and acid suppression therapy, with or without a bismuth compound, taken for 10 to 14 days. Choosing a combination that can be taken in two daily doses may be easier for patients, although this option is also more costly.Table 3 provides a practical list of selected effective drug combinations used for treating H pylori infection. Triple therapies with reported eradication rates approaching 90% or more are included, though resistant strains may continue to emerge. Single and dual therapies, though FDA-approved for treatment of H pylori, have unacceptably low cure rates and are not recommended. PPI quadruple therapy or a regimen including furazolidone (a monoamine oxidase inhibitor) may serve as second-line treatment for eradication of initial failures and in case of metronidazole resistance.^18,19 Studies using ranitidine bismuth citrate in place of PPIs have also shown comparable results.

Patient education regarding the need for effective eradication therapy and to encourage adherence to the drug regimen is critical. Patients should have adequate follow-up, since further diagnostic testing may be needed to ensure eradication of the H pylori organism, particularly in the case of treatment failure or relapse. Because eradication usually cures peptic ulcer disease, chronic acid suppression therapy should not be needed in most patients who have cleared the H pylori infection and who are not taking NSAIDs. Among primary care patients with a history of peptic ulcer disease taking chronic acid suppressive therapy, 78% of those treated for H pylori were able to discontinue their therapy.²¹

Persistent or recurrent ulcers in patients treated for H pylori are strongly associated with persistent or recurrent infection.²² Because symptom relief is not always correlated with eradication, testing for cure following eradication therapy should be considered, particularly in high-risk patients, such as those with a history of bleeding or perforation.¹³ However, no randomized studies have been done to assess the outcomes associated with the decision to test for cure. If desired, noninvasive tests (eg stool antigen test or urea breath test) may be used for patients who become symptom-free following eradication therapy or for patients with persistent symptoms and a previously documented duodenal ulcer. Acid suppressive therapy with a PPI can result in false negative results, so this should be withheld for at least 2 weeks prior to testing for persistent infection.²³ Because of the risk of cancer associated with gastric ulcers, endoscopic documentation of gastric ulcer healing might be preferable to noninvasive testing, particularly in high-risk patients. In either case, patients with persistent or recurrent symptoms following eradication should have endoscopy to document ulcer healing and to obtain biopsies if necessary.

Nonulcer Dyspepsia

The pathophysiology of nonulcer dyspepsia (NUD) is not definitively known, though factors implicated include gastric acid secretion, gastro-duodenal dysmotility, visceral hypersensitivity, stress, and psychological factors.²⁴ A relationship with H pylori has not been established, though there is some evidence of an association.²⁵

Studies of drug treatments for NUD are limited by small samples, short duration of follow-up and the use of unvalidated outcome measures.²⁶ A recent meta-analysis found no significant benefit from the use of antacids or sucralfate for NUD, defined as dyspeptic symptoms with negative endoscopic or barium studies, excluding other organic (eg pancreato-biliary disease, oesophagitis) and drug-induced (NSAIDs) disease. The same study reported statistically significant benefits with the prokinetic drug cisapride, but there was evidence of a publication bias in these comparisons, making interpretation difficult (“positive” studies were more likely to be published than “negative” trials). Cisapride was also recently taken off the market, so is no longer available as a treatment option. No placebo-controlled studies of metoclopramide were identified in this systematic review.

Antisecretory treatments were more effective than placebo in the treatment of NUD, with a number needed to treat of 6 for H2-RAs and 11 for PPIs (ie, for every 11 patients who received a PPI instead of placebo, one benefited).²⁴ Thus, PPIs were actually less effective than H2-RAs in this meta-analysis of placebo-controlled trials. Long-term use of antisecretory therapy may be associated with hypergastrinemia, increased gastrointestinal bacterial counts, and altered absorption of nutrients, though the clinical significance of this is unclear.²⁷ For patients who do not respond to acid-suppressive therapy it might be necessary to entertain alternative diagnoses, and if no explanation for the symptoms can be identified, consider counseling or pain management strategies to help the patient cope with the discomfort. Studies of the use of antidepressants, though small and of questionable quality, consistently show improvement in symptoms of patients with functional gastrointestinal disorders, including NUD and irritable bowel syndrome.²⁸ Treatment recommendations for NUD are shown inTable 4.

To date there is no convincing evidence that empiric eradication of H pylori in patients with NUD improves symptoms. One recent meta-analysis of randomized controlled trials revealed no improvement with H pylori eradication for the symptoms of NUD,²⁹ while 2 others^25,30 showed a modest but statistically significant benefit, with 1 patient cured for every 19 treated (number needed to treat = 19).

Prognosis

Without treatment, peptic ulcer disease can lead to serious complications such as gastrointestinal bleeding and cancer. Acid suppression achieves ulcer healing rates of approximately 90%, but is associated with a 10% recurrence rate even with long-term treatment.²² Successful eradication of H pylori in the absence of NSAID use cures ulcer disease in 95% of cases; the recurrence rate is 33% to 41% if eradication is not achieved.¹⁴ Reinfection is rare once eradication has been accomplished, with a rate of about 1% per year,³¹ though rates can be much higher in endemic areas.³² Persistent infection requires re-treatment, ideally with a regimen not previously used, in case of antimicrobial resistance. Persistent gastric ulcers can harbor malignancy and therefore, evaluation with endoscopy might be prudent. H pylori itself is associated with a 2- to 6-fold increase in risk of gastric cancer.³³ Widespread screening or treatment to prevent cancer has not been recommended to date. A cost-benefit analysis suggests that the development and distribution of a vaccination for H pylori would be highly cost-effective,³⁴ but such a vaccine is not available as yet.

The prognosis of NUD is more discouraging. NUD is a chronic relapsing and remitting disorder, and treatment responses difficult to measure. For example, one systematic review of the literature found 56% of patients experienced an improvement in symptom scores when given placebo (range 5% - 90%). As with other functional gastrointestinal disorders, underlying psychosocial and lifestyle factors may be involved and must be addressed. Further research is needed in this area, particularly in the primary care setting.

Each Applied Evidence review article considers a common presenting complaint or disease and summarizes the best available evidence for clinicians. The collected reviews are published online at www.jfponline.com. Explanations of the Levels of Evidence can be found at http://cebm.jr2.ox.ac.uk/docs/levels.html.

In the June 2001 issue of The Journal of Family Practice, the diagnostic approach to the patient with dyspepsia was presented.¹ In that analysis, gastroesophageal reflux disease (GERD), gastric ulcers, and duodenal ulcers emerged as the most common identifiable causes of dyspepsia. However, most patients with dyspepsia do not have one of these conditions, and are considered to have functional or nonulcer dyspepsia. The diagnosis and management of adults with GERD was recently described in detail.² Therefore, this paper reviews the treatment of undifferentiated dyspepsia, gastric ulcer caused by nonsteroidal anti-inflammatory drugs (NSAIDs), peptic ulcer disease not associated with NSAID use, and nonulcer dyspepsia (dyspepsia in a patient who has no evidence of ulcer or GERD on endoscopy). An algorithm for the management of the patient with known ulcer disease is shown in the Figure 1. (J Fam Pract 2001; 50:614-619)

Undifferentiated dyspepsia

In primary care, the typical patient presenting with dyspepsia will not have had endoscopy. Therefore, the presence of an underlying lesion will be unknown, a situation known as undifferentiated dyspepsia. As described by Smucny,² randomized trials and economic analyses have demonstrated the cost-effectiveness of a test-and-treat strategy^3,4 in which patients with dyspepsia are tested for Helicobacter pylori and treated with eradication therapy if positive. This strategy would reserve endoscopy for those patients with alarm signs Table1 or those who have persistent symptoms despite appropriate empiric therapy. Certainly a physician must weigh the potential for complications during endoscopy with the risks of adverse reactions to eradication therapy, including the development of antimicrobial resistant organisms. All patients with dyspepsia should be counseled to avoid factors that exacerbate their symptoms or disrupt the integrity of the mucosal lining of the stomach, such as NSAID use and cigarette smoking. Both of these have been identified as risk factors for the development of peptic ulcers and delayed ulcer healing.^5,6

NSAID-related gastric ulcers

NSAIDs are associated with a 5- to 7-fold increased risk of gastric ulceration in the first 3 months of use. In a meta-analysis of observational studies of gastrointestinal bleeding risk due to various NSAIDs, a 4-fold increased risk associated with NSAID-use persisted throughout therapy and fell to baseline within 2 months of discontinuation of the NSAID.⁷ This study demonstrated a clear dose-response relationship; the difference between NSAIDs, however, was minimal.

Table 2 summarizes treatments for NSAID-related gastric ulcers. Misoprostol is an effective prophylaxis against ulcers when used with NSAIDs, but is associated with diarrhea, even at lower than optimal doses.⁸ Standard doses of H2-receptor agonists (H2RAs) are ineffective at preventing NSAID-related gastric ulcers, but double doses of H2RAs (eg, ranitidine 300 mg twice daily) and standard doses of proton-pump inhibitors (PPIs; eg, omeprazole 20 mg 4 times per day) are effective prophylactic agents for the duration of NSAID use according to the results of endoscopic studies. New COX-2 specific anti-inflammatory agents are associated with a significantly lower risk of ulcers as seen by endoscopy (4.7% with rofecoxib vs 27.7% with ibuprofen).⁹ The benefit in terms of actual adverse clinical outcomes and ulcer complications, however, is much smaller; the risk of symptomatic ulcer, perforation, symptomatic ulcer, and clinically significant bleeding was 1.3% for rofecoxib (Vioxx) and 1.8% with ibuprofen, diclofenac, or nabumetone taken for 1 year (P <.05).¹⁰ Thus, one would have to treat 200 patients for 1 year to prevent 1 adverse outcome. Similarly, the annual risk of bleeding, perforation, or gastric outlet obstruction was lower with celecoxib (Celebrex) than with naproxen, diclofenac, or ibuprofen (0.2% vs. 1.68%; P <.002; number needed to treat=60).11 COX-2 inhibitors have not been studied as an alternative therapy in patients with a history of NSAID-induced ulcers. While no trial data are available, a consensus-based recommendation has been made for dyspeptic patients with no alarm symptoms who are regular NSAID-users: The first step in management is to stop the NSAID use if possible, and determine if the symptoms improve.¹² If symptoms persist after NSAID use is discontinued, the patient should be managed as others with undifferentiated dyspepsia.

Non–NSAID-related duodenal and gastric ulcers

Helicobacter pylori is now well recognized as a major risk factor for the development of peptic ulcer disease. Although most H pylori–infected patients do not develop an ulcer, as many as 95% of patients with duodenal ulcers and 80% of those with gastric ulcers are infected. These rates may be lower in the United States because of greater use of NSAIDs and a lower rate of H pylori infection than in other parts of the world.¹³ Successful eradication of the organism following treatment heals ulcers and reduces the risk of recurrence from 67% to 6% for patients with duodenal ulcers, and from 59% to 4% for patients with gastric ulcers.¹⁴ A meta-analysis of North American randomized controlled trials of H pylori eradication for duodenal ulcer found that one ulcer recurrence (by endoscopy) would be prevented for every 2.8 patients successfully treated.¹⁵

However, a multi-drug regimen and an adequate length of treatment are required for eradication. Because antimicrobial resistance and incomplete treatment are major reasons for treatment failure,^16,17 convenience and tolerability become important considerations in choosing a treatment plan.

Effective H pylori treatment regimens include a combination of two antibiotics and acid suppression therapy, with or without a bismuth compound, taken for 10 to 14 days. Choosing a combination that can be taken in two daily doses may be easier for patients, although this option is also more costly.Table 3 provides a practical list of selected effective drug combinations used for treating H pylori infection. Triple therapies with reported eradication rates approaching 90% or more are included, though resistant strains may continue to emerge. Single and dual therapies, though FDA-approved for treatment of H pylori, have unacceptably low cure rates and are not recommended. PPI quadruple therapy or a regimen including furazolidone (a monoamine oxidase inhibitor) may serve as second-line treatment for eradication of initial failures and in case of metronidazole resistance.^18,19 Studies using ranitidine bismuth citrate in place of PPIs have also shown comparable results.

Patient education regarding the need for effective eradication therapy and to encourage adherence to the drug regimen is critical. Patients should have adequate follow-up, since further diagnostic testing may be needed to ensure eradication of the H pylori organism, particularly in the case of treatment failure or relapse. Because eradication usually cures peptic ulcer disease, chronic acid suppression therapy should not be needed in most patients who have cleared the H pylori infection and who are not taking NSAIDs. Among primary care patients with a history of peptic ulcer disease taking chronic acid suppressive therapy, 78% of those treated for H pylori were able to discontinue their therapy.²¹

Persistent or recurrent ulcers in patients treated for H pylori are strongly associated with persistent or recurrent infection.²² Because symptom relief is not always correlated with eradication, testing for cure following eradication therapy should be considered, particularly in high-risk patients, such as those with a history of bleeding or perforation.¹³ However, no randomized studies have been done to assess the outcomes associated with the decision to test for cure. If desired, noninvasive tests (eg stool antigen test or urea breath test) may be used for patients who become symptom-free following eradication therapy or for patients with persistent symptoms and a previously documented duodenal ulcer. Acid suppressive therapy with a PPI can result in false negative results, so this should be withheld for at least 2 weeks prior to testing for persistent infection.²³ Because of the risk of cancer associated with gastric ulcers, endoscopic documentation of gastric ulcer healing might be preferable to noninvasive testing, particularly in high-risk patients. In either case, patients with persistent or recurrent symptoms following eradication should have endoscopy to document ulcer healing and to obtain biopsies if necessary.

Nonulcer Dyspepsia

The pathophysiology of nonulcer dyspepsia (NUD) is not definitively known, though factors implicated include gastric acid secretion, gastro-duodenal dysmotility, visceral hypersensitivity, stress, and psychological factors.²⁴ A relationship with H pylori has not been established, though there is some evidence of an association.²⁵

Studies of drug treatments for NUD are limited by small samples, short duration of follow-up and the use of unvalidated outcome measures.²⁶ A recent meta-analysis found no significant benefit from the use of antacids or sucralfate for NUD, defined as dyspeptic symptoms with negative endoscopic or barium studies, excluding other organic (eg pancreato-biliary disease, oesophagitis) and drug-induced (NSAIDs) disease. The same study reported statistically significant benefits with the prokinetic drug cisapride, but there was evidence of a publication bias in these comparisons, making interpretation difficult (“positive” studies were more likely to be published than “negative” trials). Cisapride was also recently taken off the market, so is no longer available as a treatment option. No placebo-controlled studies of metoclopramide were identified in this systematic review.

Antisecretory treatments were more effective than placebo in the treatment of NUD, with a number needed to treat of 6 for H2-RAs and 11 for PPIs (ie, for every 11 patients who received a PPI instead of placebo, one benefited).²⁴ Thus, PPIs were actually less effective than H2-RAs in this meta-analysis of placebo-controlled trials. Long-term use of antisecretory therapy may be associated with hypergastrinemia, increased gastrointestinal bacterial counts, and altered absorption of nutrients, though the clinical significance of this is unclear.²⁷ For patients who do not respond to acid-suppressive therapy it might be necessary to entertain alternative diagnoses, and if no explanation for the symptoms can be identified, consider counseling or pain management strategies to help the patient cope with the discomfort. Studies of the use of antidepressants, though small and of questionable quality, consistently show improvement in symptoms of patients with functional gastrointestinal disorders, including NUD and irritable bowel syndrome.²⁸ Treatment recommendations for NUD are shown inTable 4.

To date there is no convincing evidence that empiric eradication of H pylori in patients with NUD improves symptoms. One recent meta-analysis of randomized controlled trials revealed no improvement with H pylori eradication for the symptoms of NUD,²⁹ while 2 others^25,30 showed a modest but statistically significant benefit, with 1 patient cured for every 19 treated (number needed to treat = 19).

Prognosis

Without treatment, peptic ulcer disease can lead to serious complications such as gastrointestinal bleeding and cancer. Acid suppression achieves ulcer healing rates of approximately 90%, but is associated with a 10% recurrence rate even with long-term treatment.²² Successful eradication of H pylori in the absence of NSAID use cures ulcer disease in 95% of cases; the recurrence rate is 33% to 41% if eradication is not achieved.¹⁴ Reinfection is rare once eradication has been accomplished, with a rate of about 1% per year,³¹ though rates can be much higher in endemic areas.³² Persistent infection requires re-treatment, ideally with a regimen not previously used, in case of antimicrobial resistance. Persistent gastric ulcers can harbor malignancy and therefore, evaluation with endoscopy might be prudent. H pylori itself is associated with a 2- to 6-fold increase in risk of gastric cancer.³³ Widespread screening or treatment to prevent cancer has not been recommended to date. A cost-benefit analysis suggests that the development and distribution of a vaccination for H pylori would be highly cost-effective,³⁴ but such a vaccine is not available as yet.

The prognosis of NUD is more discouraging. NUD is a chronic relapsing and remitting disorder, and treatment responses difficult to measure. For example, one systematic review of the literature found 56% of patients experienced an improvement in symptom scores when given placebo (range 5% - 90%). As with other functional gastrointestinal disorders, underlying psychosocial and lifestyle factors may be involved and must be addressed. Further research is needed in this area, particularly in the primary care setting.

Each Applied Evidence review article considers a common presenting complaint or disease and summarizes the best available evidence for clinicians. The collected reviews are published online at www.jfponline.com. Explanations of the Levels of Evidence can be found at http://cebm.jr2.ox.ac.uk/docs/levels.html.

In the June 2001 issue of The Journal of Family Practice, the diagnostic approach to the patient with dyspepsia was presented.¹ In that analysis, gastroesophageal reflux disease (GERD), gastric ulcers, and duodenal ulcers emerged as the most common identifiable causes of dyspepsia. However, most patients with dyspepsia do not have one of these conditions, and are considered to have functional or nonulcer dyspepsia. The diagnosis and management of adults with GERD was recently described in detail.² Therefore, this paper reviews the treatment of undifferentiated dyspepsia, gastric ulcer caused by nonsteroidal anti-inflammatory drugs (NSAIDs), peptic ulcer disease not associated with NSAID use, and nonulcer dyspepsia (dyspepsia in a patient who has no evidence of ulcer or GERD on endoscopy). An algorithm for the management of the patient with known ulcer disease is shown in the Figure 1. (J Fam Pract 2001; 50:614-619)

Undifferentiated dyspepsia

In primary care, the typical patient presenting with dyspepsia will not have had endoscopy. Therefore, the presence of an underlying lesion will be unknown, a situation known as undifferentiated dyspepsia. As described by Smucny,² randomized trials and economic analyses have demonstrated the cost-effectiveness of a test-and-treat strategy^3,4 in which patients with dyspepsia are tested for Helicobacter pylori and treated with eradication therapy if positive. This strategy would reserve endoscopy for those patients with alarm signs Table1 or those who have persistent symptoms despite appropriate empiric therapy. Certainly a physician must weigh the potential for complications during endoscopy with the risks of adverse reactions to eradication therapy, including the development of antimicrobial resistant organisms. All patients with dyspepsia should be counseled to avoid factors that exacerbate their symptoms or disrupt the integrity of the mucosal lining of the stomach, such as NSAID use and cigarette smoking. Both of these have been identified as risk factors for the development of peptic ulcers and delayed ulcer healing.^5,6

NSAID-related gastric ulcers

NSAIDs are associated with a 5- to 7-fold increased risk of gastric ulceration in the first 3 months of use. In a meta-analysis of observational studies of gastrointestinal bleeding risk due to various NSAIDs, a 4-fold increased risk associated with NSAID-use persisted throughout therapy and fell to baseline within 2 months of discontinuation of the NSAID.⁷ This study demonstrated a clear dose-response relationship; the difference between NSAIDs, however, was minimal.

Table 2 summarizes treatments for NSAID-related gastric ulcers. Misoprostol is an effective prophylaxis against ulcers when used with NSAIDs, but is associated with diarrhea, even at lower than optimal doses.⁸ Standard doses of H2-receptor agonists (H2RAs) are ineffective at preventing NSAID-related gastric ulcers, but double doses of H2RAs (eg, ranitidine 300 mg twice daily) and standard doses of proton-pump inhibitors (PPIs; eg, omeprazole 20 mg 4 times per day) are effective prophylactic agents for the duration of NSAID use according to the results of endoscopic studies. New COX-2 specific anti-inflammatory agents are associated with a significantly lower risk of ulcers as seen by endoscopy (4.7% with rofecoxib vs 27.7% with ibuprofen).⁹ The benefit in terms of actual adverse clinical outcomes and ulcer complications, however, is much smaller; the risk of symptomatic ulcer, perforation, symptomatic ulcer, and clinically significant bleeding was 1.3% for rofecoxib (Vioxx) and 1.8% with ibuprofen, diclofenac, or nabumetone taken for 1 year (P <.05).¹⁰ Thus, one would have to treat 200 patients for 1 year to prevent 1 adverse outcome. Similarly, the annual risk of bleeding, perforation, or gastric outlet obstruction was lower with celecoxib (Celebrex) than with naproxen, diclofenac, or ibuprofen (0.2% vs. 1.68%; P <.002; number needed to treat=60).11 COX-2 inhibitors have not been studied as an alternative therapy in patients with a history of NSAID-induced ulcers. While no trial data are available, a consensus-based recommendation has been made for dyspeptic patients with no alarm symptoms who are regular NSAID-users: The first step in management is to stop the NSAID use if possible, and determine if the symptoms improve.¹² If symptoms persist after NSAID use is discontinued, the patient should be managed as others with undifferentiated dyspepsia.

Non–NSAID-related duodenal and gastric ulcers

Helicobacter pylori is now well recognized as a major risk factor for the development of peptic ulcer disease. Although most H pylori–infected patients do not develop an ulcer, as many as 95% of patients with duodenal ulcers and 80% of those with gastric ulcers are infected. These rates may be lower in the United States because of greater use of NSAIDs and a lower rate of H pylori infection than in other parts of the world.¹³ Successful eradication of the organism following treatment heals ulcers and reduces the risk of recurrence from 67% to 6% for patients with duodenal ulcers, and from 59% to 4% for patients with gastric ulcers.¹⁴ A meta-analysis of North American randomized controlled trials of H pylori eradication for duodenal ulcer found that one ulcer recurrence (by endoscopy) would be prevented for every 2.8 patients successfully treated.¹⁵

However, a multi-drug regimen and an adequate length of treatment are required for eradication. Because antimicrobial resistance and incomplete treatment are major reasons for treatment failure,^16,17 convenience and tolerability become important considerations in choosing a treatment plan.

Effective H pylori treatment regimens include a combination of two antibiotics and acid suppression therapy, with or without a bismuth compound, taken for 10 to 14 days. Choosing a combination that can be taken in two daily doses may be easier for patients, although this option is also more costly.Table 3 provides a practical list of selected effective drug combinations used for treating H pylori infection. Triple therapies with reported eradication rates approaching 90% or more are included, though resistant strains may continue to emerge. Single and dual therapies, though FDA-approved for treatment of H pylori, have unacceptably low cure rates and are not recommended. PPI quadruple therapy or a regimen including furazolidone (a monoamine oxidase inhibitor) may serve as second-line treatment for eradication of initial failures and in case of metronidazole resistance.^18,19 Studies using ranitidine bismuth citrate in place of PPIs have also shown comparable results.

Patient education regarding the need for effective eradication therapy and to encourage adherence to the drug regimen is critical. Patients should have adequate follow-up, since further diagnostic testing may be needed to ensure eradication of the H pylori organism, particularly in the case of treatment failure or relapse. Because eradication usually cures peptic ulcer disease, chronic acid suppression therapy should not be needed in most patients who have cleared the H pylori infection and who are not taking NSAIDs. Among primary care patients with a history of peptic ulcer disease taking chronic acid suppressive therapy, 78% of those treated for H pylori were able to discontinue their therapy.²¹

Persistent or recurrent ulcers in patients treated for H pylori are strongly associated with persistent or recurrent infection.²² Because symptom relief is not always correlated with eradication, testing for cure following eradication therapy should be considered, particularly in high-risk patients, such as those with a history of bleeding or perforation.¹³ However, no randomized studies have been done to assess the outcomes associated with the decision to test for cure. If desired, noninvasive tests (eg stool antigen test or urea breath test) may be used for patients who become symptom-free following eradication therapy or for patients with persistent symptoms and a previously documented duodenal ulcer. Acid suppressive therapy with a PPI can result in false negative results, so this should be withheld for at least 2 weeks prior to testing for persistent infection.²³ Because of the risk of cancer associated with gastric ulcers, endoscopic documentation of gastric ulcer healing might be preferable to noninvasive testing, particularly in high-risk patients. In either case, patients with persistent or recurrent symptoms following eradication should have endoscopy to document ulcer healing and to obtain biopsies if necessary.

Nonulcer Dyspepsia

The pathophysiology of nonulcer dyspepsia (NUD) is not definitively known, though factors implicated include gastric acid secretion, gastro-duodenal dysmotility, visceral hypersensitivity, stress, and psychological factors.²⁴ A relationship with H pylori has not been established, though there is some evidence of an association.²⁵

Studies of drug treatments for NUD are limited by small samples, short duration of follow-up and the use of unvalidated outcome measures.²⁶ A recent meta-analysis found no significant benefit from the use of antacids or sucralfate for NUD, defined as dyspeptic symptoms with negative endoscopic or barium studies, excluding other organic (eg pancreato-biliary disease, oesophagitis) and drug-induced (NSAIDs) disease. The same study reported statistically significant benefits with the prokinetic drug cisapride, but there was evidence of a publication bias in these comparisons, making interpretation difficult (“positive” studies were more likely to be published than “negative” trials). Cisapride was also recently taken off the market, so is no longer available as a treatment option. No placebo-controlled studies of metoclopramide were identified in this systematic review.

Antisecretory treatments were more effective than placebo in the treatment of NUD, with a number needed to treat of 6 for H2-RAs and 11 for PPIs (ie, for every 11 patients who received a PPI instead of placebo, one benefited).²⁴ Thus, PPIs were actually less effective than H2-RAs in this meta-analysis of placebo-controlled trials. Long-term use of antisecretory therapy may be associated with hypergastrinemia, increased gastrointestinal bacterial counts, and altered absorption of nutrients, though the clinical significance of this is unclear.²⁷ For patients who do not respond to acid-suppressive therapy it might be necessary to entertain alternative diagnoses, and if no explanation for the symptoms can be identified, consider counseling or pain management strategies to help the patient cope with the discomfort. Studies of the use of antidepressants, though small and of questionable quality, consistently show improvement in symptoms of patients with functional gastrointestinal disorders, including NUD and irritable bowel syndrome.²⁸ Treatment recommendations for NUD are shown inTable 4.

To date there is no convincing evidence that empiric eradication of H pylori in patients with NUD improves symptoms. One recent meta-analysis of randomized controlled trials revealed no improvement with H pylori eradication for the symptoms of NUD,²⁹ while 2 others^25,30 showed a modest but statistically significant benefit, with 1 patient cured for every 19 treated (number needed to treat = 19).

Prognosis

Without treatment, peptic ulcer disease can lead to serious complications such as gastrointestinal bleeding and cancer. Acid suppression achieves ulcer healing rates of approximately 90%, but is associated with a 10% recurrence rate even with long-term treatment.²² Successful eradication of H pylori in the absence of NSAID use cures ulcer disease in 95% of cases; the recurrence rate is 33% to 41% if eradication is not achieved.¹⁴ Reinfection is rare once eradication has been accomplished, with a rate of about 1% per year,³¹ though rates can be much higher in endemic areas.³² Persistent infection requires re-treatment, ideally with a regimen not previously used, in case of antimicrobial resistance. Persistent gastric ulcers can harbor malignancy and therefore, evaluation with endoscopy might be prudent. H pylori itself is associated with a 2- to 6-fold increase in risk of gastric cancer.³³ Widespread screening or treatment to prevent cancer has not been recommended to date. A cost-benefit analysis suggests that the development and distribution of a vaccination for H pylori would be highly cost-effective,³⁴ but such a vaccine is not available as yet.

The prognosis of NUD is more discouraging. NUD is a chronic relapsing and remitting disorder, and treatment responses difficult to measure. For example, one systematic review of the literature found 56% of patients experienced an improvement in symptom scores when given placebo (range 5% - 90%). As with other functional gastrointestinal disorders, underlying psychosocial and lifestyle factors may be involved and must be addressed. Further research is needed in this area, particularly in the primary care setting.

Each Applied Evidence review article considers a common presenting complaint or disease and summarizes the best available evidence for clinicians. The collected reviews are published online at www.jfponline.com. Explanations of the Levels of Evidence can be found at http://cebm.jr2.ox.ac.uk/docs/levels.html.