Nonulcer Dyspepsia and Peptic Ulcer Disease

Transcription

1 GASTROENTEROLOGY BOARD REVIEW MANUAL PUBLISHING STAFF PRESIDENT, GROUP PUBLISHER Bruce M. White EXECUTIVE EDITOR Debra Dreger SENIOR EDITOR Becky Krumm, ELS ASSOCIATE EDITOR Lamont Williams ASSISTANT EDITOR Jennifer M. Vander Bush EDITORIAL ASSISTANT Nora H. Landon EXECUTIVE VICE PRESIDENT Barbara T. White, MBA PRODUCTION DIRECTOR Suzanne S. Banish PRODUCTION ASSOCIATES Tish Berchtold Klus Christie Grams Mary Beth Cunney ADVERTISING/PROJECT MANAGER Patricia Payne Castle NOTE FROM THE PUBLISHER: This publication has been developed without involvement of or review by the American Board of Internal Medicine. Endorsed by the Association for Hospital Medical Education The Association for Hospital Medical Education endorses HOSPITAL PHYSICIAN for the purpose of presenting the latest developments in medical education as they affect residency programs and clinical hospital practice. Nonulcer Dyspepsia and Peptic Ulcer Disease Series Editor: Robert M. Craig, MD Professor of Medicine Department of Medicine Division of Gastroenterology and Hepatology Northwestern University Medical School Chicago, IL Contributing Author: Colin W. Howden, MD Professor of Medicine Department of Medicine Division of Gastroenterology and Hepatology Northwestern University Medical School Chicago, IL Table of Contents Preface ii Introduction Patient with a Short History of Dyspeptic Symptoms Recurrent Duodenal Ulcer Gastric Ulcer in a Patient with Arthritis Intractable Peptic Ulcer Disease References Cover Illustration by Christine Schaar Copyright 2001, Turner White Communications, Inc., 125 Strafford Avenue, Suite 220, Wayne, PA , All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, mechanical, electronic, photocopying, recording, or otherwise, without the prior written permission of Turner White Communications, Inc. The editors are solely responsible for selecting content. Although the editors take great care to ensure accuracy, Turner White Communications, Inc., will not be liable for any errors of omission or inaccuracies in this publication. Opinions expressed are those of the authors and do not necessarily reflect those of Turner White Communications, Inc. Gastroenterology Volume 7, Part 4 i

2 GASTROENTEROLOGY BOARD REVIEW MANUAL Preface Board certification in gastroenterology and hepatology requires an understanding of all aspects of physiology, diagnosis, and therapy for gastrointestinal tract, liver, and pancreatic disorders. In addition to self-study of this discipline, candidates for board certification must have completed 2 years of fellowship, during which time the practical aspects of the diagnosis and treatment of gastroenterologic and hepatologic complications are learned through hands-on clinical investigation. The Hospital Physician Gastroenterology Board Review Manual is a study guide intended to help candidates prepare for the written and oral components of this examination. The manual consists of quarterly publications that address the following areas: Acid peptic disease Acute appendicitis Aging and the gastrointestinal tract Basic biology Chronic cholestatic syndromes Cirrhosis and portal hypertension Diverticulosis coli Drug-induced and alcoholic liver disease Endoscopy of upper gastrointestinal hemorrhage Esophageal disorders Gallstones Gastrointestinal manifestations of AIDS Inflammatory bowel diseases and other diarrheal diseases Intestinal obstruction Irritable bowel syndromes Ischemic bowel disease Liver transplantation Malabsorption syndromes Nutritional support Pancreatic disease Small intestinal, colonic, and other tumors Viral and chronic hepatitis Most of these topics are covered in the manual; however, some areas of interest are discussed in greater depth than others. The manual presents clinical scenarios using a casebased format; questions and answers relating to the clinical presentation are provided. The editors believe that this question-and-answer format is an effective teaching tool and allows for adequate self-assessment. Each quarterly publication addresses only a few of the topics mentioned; board certification candidates should review the entire list of topics to be appropriately prepared for the examination. The Hospital Physician Gastroenterology Board Review Manual is prepared by the Series Editor and contributing authors and not in collaboration with the American Board of Internal Medicine, Gastroenterology/Hepatology. Robert M. Craig, MD Professor of Medicine Department of Medicine Division of Gastroenterology and Hepatology Northwestern University Medical School Chicago, IL ii Hospital Physician Board Review Manual

3 GASTROENTEROLOGY BOARD REVIEW MANUAL Nonulcer Dyspepsia and Peptic Ulcer Disease Series Editor: Robert M. Craig, MD Professor of Medicine, Department of Medicine Division of Gastroenterology and Hepatology Northwestern University Medical School Chicago, IL Contributing Author: Colin W. Howden, MD Professor of Medicine Department of Medicine Division of Gastroenterology and Hepatology Northwestern University Medical School Chicago, IL INTRODUCTION Dyspepsia was defined by the experts at the Rome II Multinational Consensus on Functional Gastrointestinal Disorders as pain or discomfort centered in the upper abdomen. 1 Specific subtypes of dyspepsia recognized by the Rome II committee include ulcer-like dyspepsia, dysmotility-like dyspepsia, and unspecified or nonspecific dyspepsia. 1,2 (Abdominal pain that is mainly located in the left or right sides of the abdomen is not regarded as dyspepsia and will not be considered further in this manual.) Dyspepsia is a highly prevalent complaint and can result from a number of potentially identifiable causes. In 1997, approximately 2% of the adult population in the US received some form of outpatient or inpatient care because of dyspepsia. 3 Expenditures for this care totaled approximately $2.5 billion. Dyspepsia is, therefore, an extremely important topic. It is frequently a reason for consultations in gastroenterologists offices and in primary care facilities. It is important that the terms dyspepsia and nonulcer dyspepsia (NUD) not be confused; these terms are not interchangeable or synonymous. The term dyspepsia should be used to refer to patients who present before any investigation has been performed. For patients with dyspepsia, there are a number of potential causes of their symptoms including peptic ulcer disease (PUD) and gastric cancer. However, in clinical practice in the US, such patients are usually found to have NUD. The diagnosis of NUD can only be made once the condition has been investigated. In order for a patient to be diagnosed with NUD, he or she must have had a normal esophagogastroduodenoscopy (EGD) while being symptomatic and while not taking acid-suppressing or other ulcer-healing medicines. The Rome II Multinational Consensus on Functional Gastrointestinal Disorders developed additional specific diagnostic criteria for NUD. 1 According to the criteria, for a diagnosis of NUD to be made, a patient must have had persistent or recurrent dyspepsia for 12 weeks, which need not be consecutive, during the past 12 months. Furthermore, there should be no evidence of organic disease including at EGD that is likely to explain the symptoms. Some patients with irritable bowel syndrome may have pain that is predominantly or exclusively located in the upper abdomen and might erroneously be considered to have NUD. For a diagnosis of IBS to be made, there must be evidence that the abdominal pain can be relieved by defecation exclusively, and/or is associated with a change in stool frequency, and/or is associated with a change in stool appearance or form. 4 However, many patients may have diagnostic features Gastroenterology Volume 7, Part 4 1

4 of both IBS and NUD either concurrently or at different times. Although most patients who present with dyspepsia will ultimately be found, after appropriate investigation, to have NUD, other causes of dyspepsia must always be considered during the diagnostic evaluation. Patients with dyspepsia may harbor a fear that there is a serious underlying cause of their symptoms often a fear of cancer. For the year 2001, the American Cancer Society predicted 21,700 new cases of gastric cancer in the US 13,400 for men and 8,300 for women. 5 To attempt to put this into proportion, this represents less than one fifth of the predicted number of new cases of colorectal cancer. Clearly, only a very small proportion of patients with dyspepsia will have gastric cancer. However, patients usually do not know this; a fear of cancer may prompt them to seek medical attention. PATIENT WITH A SHORT HISTORY OF DYSPEPTIC SYMPTOMS CASE 1 PRESENTATION Patient 1 is a 32-year-old previously healthy white woman who works as a marketing executive. She presents to her primary care physician, complaining of abdominal pain that is mainly located in the epigastrium. She has never had a pain-free day since the problem started approximately 2 months ago. The pain does not bother her at night and is sometimes made worse by eating. She has had no vomiting episodes and has no signs of gastrointestinal (GI) bleeding. However, she does complain of nausea. She denies heartburn and regurgitation. Her appetite has been poor, and she has been reluctant to eat normally for fear that her pain would be made worse. She says that during the time she has been experiencing the pain, her weight has not changed, and she has been on no medications including aspirin and nonsteroidal anti-inflammatory drugs (NSAIDs). She has no relevant past medical history of note but had previously presented to her physician with headaches (that were later attributed to tension) and with lower back pain, for which no cause was found and which resolved spontaneously. Her family history is noncontributory, although she thinks that her grandmother may have died of stomach cancer. She smokes half a pack of cigarettes daily and drinks only an occasional glass of wine. Review of systems is noncontributory, although she expresses worry about her symptoms. The results of a physical examination are unremarkable. Can the cause of dyspepsia be discerned on the basis of a clinical history alone? DISCUSSION It is not possible to make a definitive diagnosis of the cause of dyspepsia from a clinical history. However, the chance of patient 1 having peptic ulcer disease (PUD) or a more serious diagnosis such as gastric cancer appears low. Her pain lacks the typical periodicity of PUD, does not trouble her at night, and is not even temporarily relieved by food intake indeed, the opposite seems to be the case. The history of her symptoms is quite short, and although she has a vague family history of stomach cancer, it is highly unlikely that she has such a diagnosis. The incidence of gastric cancer continues to fall in the US. Furthermore, it is predominantly a disease of later life, and the contribution of heredity is much lower for this disease than for other GI cancers notably colorectal cancer. 5 Also, patient 1 has no particularly worrisome features in her history such as vomiting, GI bleeding, or weight loss. It is likely that she has NUD, although this is only a presumptive diagnosis at this point because her condition has not been investigated. What are initial management strategies for dyspepsia? INITIAL MANAGEMENT STRATEGIES FOR DYSPEPSIA At this stage, the primary care physician has a number of management options from which to choose. He or she may choose to reassure the patient and offer some form of empiric treatment, test the patient noninvasively for Helicobacter pylori infection, schedule an EGD, or refer the patient to a gastroenterologist. Reassurance and Empiric Treatment Based on what is currently known about patient 1, reassurance and empiric treatment would seem a highly reasonable option. The history of the patient s symptoms is quite short, and there is no evidence from the history or physical examination of a serious underlying condition. The primary care physician might try to further characterize the nature of the patient s symptoms (ie, as ulcer-like dyspepsia, dysmotility-like dyspepsia, or unspecified or nonspecific dyspepsia 1,2 ). Because patient 1 s symptoms seem to fit best into the ulcer-like category, an empiric trial of an H 2 -receptor antagonist (H 2 RA) or a proton pump inhibitor (PPI) would be appropriate. There have been multiple randomized controlled trials (RCTs) comparing H 2 RA treatment with placebo for patients with uninvestigated dyspepsia or with proven NUD. Table 1 summarizes the principal results of 3 of 2 Hospital Physician Board Review Manual

5 the largest placebo-controlled RCTs of H 2 RA treatment for NUD. 6-8 The H 2 RA was associated with a response rate that was between 14% and 21% greater than that of the placebo. This fairly modest therapeutic gain is largely attributable to the consistently high placebo response rate that is observed in studies of dyspepsia or NUD and, indeed, other functional GI disorders. H 2 RAs, therefore, do seem to offer some therapeutic gain albeit fairly low. It is reasonable to assume that this is through suppression of gastric acid secretion, because this is their only important pharmacologic effect. PPIs exert a greater inhibitory effect on gastric acid secretion than do H 2 RAs. Therefore, if some of the symptoms of dyspepsia are related to gastric acid, it might be expected that the PPIs would be more effective than H 2 RAs. Two large multinational RCTs the BOND and OPERA studies compared omeprazole 10 mg and 20 mg with placebo in patients with NUD. The results of these 2 trials were combined for a single publication. 9 Omeprazole 20 mg was superior to placebo in patients with ulcer-like dyspepsia but was no different from placebo in patients with dysmotility-like or nonspecific dyspepsia. The principal results of these 2 studies are summarized in Table 2. Also included in these studies were patients with reflux-like dyspepsia. However, such a category is no longer recognized as a subtype of dyspepsia. Patients whose predominant or sole symptom is heartburn should be considered as having gastroesophageal reflux disease (GERD) and managed accordingly. Nevertheless, it is apparent from Table 2 that patients with uninvestigated reflux-like symptoms do as would be expected respond to empiric PPI treatment. Often, patients with upper GI symptoms do not fit neatly into any one specific category such as dyspepsia or GERD because they may have symptoms of more than one disorder. Patients with dyspepsia who also have some GERD symptoms have frequently been included in treatment trials for dyspepsia or NUD and may have confounded the interpretation of the results. Those patients who have some reflux-like symptoms are more likely to respond to treatment with an acid-suppressing drug and may have contributed to some of the positive results observed in some studies. Table 3 lists the principal results of other RCTs of PPI treatment for dyspeptic symptoms There is a generally consistent therapeutic gain of PPI over either H 2 RA or placebo. Table 1. Summary of the Results of Three Large, Randomized, Placebo-Controlled Trials of H 2 -Receptor Antagonist Treatment for Nonulcer Dyspepsia That Showed an Advantage of the H 2 -Receptor Antagonists Therapeutic Response Rate (%) Gain for Study H 2 RA Placebo H 2 RA (%) Delattre et al, 1985* (N = 414) Saunders et al, 1986 (N= 221) Muller et al, 1994 (N = 509) H 2 RA = H 2 -Receptor antagonist. *Data from Delattre M, Malesky M, Prinzie A. Symptomatic treatment of non-ulcer dyspepsia with cimetidine. Current Therapeutic Research 1985;37: Cimetidine Data from Saunders JH, Oliver RJ, Higson DL. Dyspepsia: incidence of a non-ulcer disease in a controlled trial of ranitidine in general practice. Br Med J (Clin Res Ed) 1986;292: Ranitidine Data from Muller P, Hotz J, Franz E, Simon B. Ranitidine in the treatment of non-ulcer dyspepsia. A placebo-controlled study in the Federal Republic of Germany. Arzneimittelforschung 1994;44: Noninvasive Testing for Helicobacter pylori Infection H. pylori infection may be the single most common chronic bacterial infection afflicting humans. It is estimated that approximately 60% of the world s population has this infection. However, most of these individuals do not live in the US; the infection is much more prevalent in developing countries than it is in highly developed countries. Patients with dyspepsia or NUD may or may not have H. pylori infection; most epidemiological studies do not show a statistically significant difference in the prevalence of the infection among patients with dyspepsia and matched controls. H. pylori infection has been conclusively linked to PUD; it is the single most common cause of peptic ulcers. H. pylori infection has also been conclusively linked to gastric adenocarcinoma and gastric mucosa associated lymphoid tissue (MALT) lymphoma; however, it has not been proven to be linked to dyspepsia or NUD. Although the dyspeptic symptoms of H. pylori related PUD usually respond to eradication of the infection, the response in patients with uninvestigated dyspepsia or NUD has been less impressive and less consistent. However, there has been considerable enthusiasm for the so-called test-and-treat strategy for patients with dyspepsia at the primary care level. In principle, the recommendation is to identify young patients (generally Gastroenterology Volume 7, Part 4 3

6 Table 2. Summary of the Combined Results of Two Randomized, Placebo-Controlled Trials of Omeprazole Therapy for Nonulcer Dyspepsia* Percentage of Patients with Complete Symptom Relief Ulcer-Like Dyspepsia Reflux-Like Dyspepsia Dysmotility-Like Dyspepsia Medication (N = 708) (N = 143) (N = 291) Omeprazole 20 mg Omeprazole 10 mg # Placebo *Data from Talley NJ, Meineche-Schmidt V, Paré P, et al. Efficacy of omeprazole in functional dyspepsia: double-blind, randomized, placebocontrolled trials (the Bond and Opera studies). Aliment Pharmacol Ther 1998;12: P = vs placebo P = vs placebo P = 0.92 vs placebo P = 0.08 vs placebo P = 0.02 vs placebo # P = 0.33 vs placebo defined as those under 45 or under 50 years of age) with dyspepsia, exclude those with worrisome or alarm features in their history or physical examination, exclude those with exclusive or predominant symptoms of GERD, and test the remainder for H. pylori infection. Those who have positive test results should be offered effective treatment for the infection. Patient 1 seems to fit these criteria, and this approach, like the option of empiric therapy, is, therefore, potentially appropriate. Appropriate methods of noninvasive testing are the tests of active infection including the 13 C- or 14 C- labeled urea breath test (UBT) and the H. pylori stool antigen test. Although serologic studies are still the most frequently used method of testing for H. pylori infection in the US, the operating characteristics of serologic tests are inferior to those of tests of active infection. As with any test, the positive predictive value (PPV) of serologic tests decreases as the population prevalence of the infection decreases. In parts of the US, the population prevalence particularly among younger adults who would be potential candidates for a test-and-treat approach is now so low that the PPV of serologic tests is equivalent or inferior to tossing a coin. Therefore, serologic tests should no longer be used to determine pretreatment H. pylori infection status (and should never have been used as a test of posttreatment H. pylori infection status). The rationale in support of the test-and-treat approach is that most of the patients who have PUD as the cause of their dyspepsia will have positive H. pylori test results. The patients with dyspepsia and positive test results (and who would thus potentially have PUD) would then be offered anti H. pylori treatment, which should result in the cure of the ulcer diathesis in most of them and end their dyspeptic symptoms. (The test-and-treat approach would never clearly identify which patients actually had PUD.) It has been suggested that this strategy would be cost saving because it would mean that fewer patients would be referred for endoscopy for the simple identification of an ulcer. A recent trial randomized patients to either this testand-treat strategy or to prompt endoscopy with subsequent management dictated according to endoscopic findings. 15 After 1 year of follow-up, the patients in the 2 groups had very similar outcomes, although the patients who had received an early endoscopic evaluation were significantly more satisfied with their standard of care. This further reinforces the observation that many patients who present with dyspepsia are afraid that they might have a serious underlying disorder and are likely to be reassured by negative results on EGD. A Canadian multicenter trial compared treatment for H. pylori infection with a short course of a PPI in patients with uninvestigated dyspepsia. 16 This trial known as the CADET-Hp (Canadian Adult Dyspepsia Empirical Treatment H. pylori -positive) study found that 50% of patients who had received treatment for H. pylori infection were asymptomatic 1 year later compared with 36% of those who had received only a short course of a PPI (Figure 1). Numerous RCTs have compared treatment for H. pylori infection with a short course of either placebo or 4 Hospital Physician Board Review Manual

7 Table 3. Summary of the Results of Several Randomized, Placebo-Controlled Trials of Proton Pump Inhibitor Therapy for Nonulcer Dyspepsia Response Rate Study PPI (%) Comparator (%) Therapeutic Gain for PPI (%) Talley et al a b 28 c 8 10 Goves et al d 41 e 16 f 25 Mason et al g 46 e 17 f h 40 i 21 Jones et al j 55 k 33 l m 44 n 25 Hengels o 62 p 44 c 18 Peura et al q p c r c PPI = proton pump inhibitor. a Data from Talley NJ, Meineche-Schmidt V, Paré P, et al. Efficacy of omeprazole in functional dyspepsia: double-blind, randomized, placebocontrolled trials (the Bond and Opera studies). Aliment Pharmacol Ther 1998;12: b Omeprazole 10 or 20 mg. c Placebo. d Data from Goves J, Oldring JK, Kerr D, et al. First line treatment with omeprazole provides an effective and superior alternative strategy in the management of dyspepsia compared to antacid/alginate liquid: a multicentre study in general practice. Aliment Pharmacol Ther 1998;12: e Omeprazole 10 mg. f Antacid/alginate. g Data from Mason I, Millar LJ, Sheikh RR, et al. The management of acid-related dyspepsia in general practice: a comparison of an omeprazole versus an antacid-alginate/ranitidine management strategy. Aliment Pharmacol Ther 1998;12: h Omeprazole 20 to 40 mg. i Ranitidine 150 mg twice daily. j Data from Jones RH, Baxter G. Lansoprazole 30 mg daily versus ranitidine 150 mg b.d. in the treatment of acid-related dyspepsia in general practice. Aliment Pharmacol Ther 1997;11: k Lansoprazole 30 mg for 2 weeks. l Ranitidine 150 mg twice daily for 2 weeks. m Lansoprazole 30 mg for 4 weeks. n Ranitidine 150 mg twice daily for 4 weeks. o Data from Hengels KJ. Therapeutic efficacy of 15 mg lansoprazole mane in 269 patients suffering from non-ulcer dyspepsia (NUD): A multicentre, randomised, double-blind study. Gut 1998;43(Suppl 2):A89. p Lansoprazole 15 mg for 8 weeks. q Data from Peura DA, Kovacs TO, Metz DA, et al. Low-dose lansoprazole: effective for non-ulcer dyspepsia. Gastroenterol 2000;118:A439. r Lansoprazole 30 mg for 8 weeks. Gastroenterology Volume 7, Part 4 5

8 No or minimal dyspepsia at 1 year (%) OMC 7 days (n = 149) Omeprazole 20 mg twice daily for 7 days (n = 149) 50* ITT analysis 36 *P = 0.02 Figure 1. Results of the CADET-Hp (Canadian Adult Dyspepsia Empirical Treatment H. pylori-positive) study. Anti Helicobacter pylori therapy was compared with a short course of a proton pump inhibitor for treatment of dyspepsia. ITT = intent-to-treat; OMC = omeprazole + metronidazole + clarithromycin. (Data from Chiba N, Veldhuyzen van Zanten SJO, Sinclair P, et al. Beneficial effect of Helicobacter pylori eradication therapy on long term symptom relief in primary care patients with uninvestigated dyspepsia: The CADET-Hp study. Gastroenterol 2000;118:A438.) a PPI in patients with proven NUD. Recently, there have been 2 meta-analyses published in this area. 17,18 The first 17 found a small but statistically significant advantage to treatment for H. pylori infection over control treatment. Treatment for H. pylori infection was associated with an 8% reduction in the relative risk of ongoing dyspeptic symptoms compared with control treatment, which was statistically significant (Figure 2). The second meta-analysis 18 included a slightly different set of RCTs and found no evidence of a statistically significant benefit. Even adopting the most optimistic interpretation, the eradication of H. pylori infection in patients with NUD offers only a small therapeutic benefit. However, it would be wrong to conclude from this that patients with NUD should not be tested for H. pylori. It is reasonable (although not mandatory) to test such patients for H. pylori provided that treatment will be offered to those who test positive and that the patient is given realistic expectations of the effect of treatment on his or her symptoms. There may be many intangible benefits of eradicating H. pylori infection from patients with NUD including the reversal of the gastric inflammatory process and the removal of one of the major risk factors for the development of PUD. It is possible although not proven that eradication of H. pylori infection from infected adults will also reduce the small risk that they may ultimately face of developing gastric cancer. EGD Evaluation Technically, evaluating a patient through EGD is the only way to make the diagnosis of NUD and is the best Active treatment (n = 1501) Placebo (n = 1140) means of excluding other possibilities. Although many patients particularly those who are especially anxious about the possibility of serious organic disease might be put at ease early in their care by a negative result on prompt EGD, this is an expensive management strategy. However, some primary care physicians would use this as an initial management strategy, perhaps because of fear of missing serious organic disease, even though they may acknowledge that the chance of that is remote. Primary care physicians may schedule an EGD procedure at centers where open endoscopy is practiced without first consulting a gastroenterologist. When a gastroenterologist sees a patient with dyspepsia, there may be considerable pressure from the patient and the primary care physician to do something. Ordinarily, this will mean EGD, although it is generally understood that this will have a low diagnostic yield as an initial management strategy. INITIAL MANAGEMENT OF PATIENT 1 The physician prescribes a 2-week course of omeprazole 20 mg once daily and schedules the patient for a return visit. The physician is unaware that the 13 C- UBT is available at a nearby hospital and so instead performs a serologic test for H. pylori infection, which yields a negative result. When the patient returns after 2 weeks, she reports only minimal symptom improvement on omeprazole, and she appears more anxious about her symptoms. The physician schedules an EGD, which is performed 1 week later with the patient off omeprazole; it yields normal results. What is the diagnosis for patient 1? 8% benefit RR reduction = 9% (95% CI = 4% 14%) Dyspeptic symptoms resolved or improved (%) (Mean ± 95% CI) 100 Figure 2. Results of a meta-analysis of randomized controlled trials of anti Helicobacter pylori therapy in patients with nonulcer dyspepsia. Patients received either a placebo or anti Helicobacter pylori therapy (active treatment). CI = confidence interval; RR = relative risk. (Data from Moayyedi P, Soo S, Deeks J, et al. Systematic review and economic evaluation of Helicobacter pylori eradication treatment for non-ulcer dyspepsia. BMJ 2000;321: ) 6 Hospital Physician Board Review Manual

9 DIAGNOSIS FOR PATIENT 1 Patient 1 almost certainly has NUD. Technically, however, such a diagnosis cannot be made with certainty at this stage because she has recently completed a 2-week course of a PPI. Therefore, it is possible that she could have had an ulcer that had recently healed because of the PPI. However, this seems unlikely because she had such a small symptomatic improvement while on the PPI and because she has no evidence of either of the 2 principle risk factors for PUD namely, H. pylori infection and the use of aspirin or NSAIDs. (It is important that a physician confirm that a patient is not taking aspirin or NSAIDs. Patient 1 denied use of these drugs at her first visit.) Although serologic tests for H. pylori infection are imprecise, more faith can still be put in a negative result being a true negative than in a positive result being a true positive. Therefore, it is reasonable to assume that patient 1 truly is H. pylori negative, and with such a determination, it would not be necessary for her physician to embark on further diagnostic evaluation for H. pylori infection. What is the next step in patient 1 s management? CONTINUED MANAGEMENT OF PATIENT 1 The next step in management is to reinforce the impact of the negative EGD results. The patient may derive substantial benefit from the knowledge that this test found no evidence of specific organic diseases including PUD and cancer. The physician should then attempt to provide her with some medication for symptom control. Although she did not initially have a substantial response to omeprazole, it may still be worthwhile persevering with PPI treatment. If her symptoms had not responded at all to the PPI, a therapeutic trial of a lowdose tricyclic antidepressant such as imipramine or amitriptyline might have been considered. A meta-analysis has suggested that these drugs are of benefit for NUD and other functional GI disorders. 19 However, there are relatively few high quality RCTs to reach a firm conclusion regarding this. 5 The selective serotonin re-uptake inhibitors have not been adequately examined in the management of NUD or other functional GI disorders. OUTCOME OF PATIENT 1 Two months later, the patient is feeling better. She continues to experience some mild, intermittent dyspepsia that is more likely to occur when she is under stress at work. She is using omeprazole 20 mg on an irregular basis. Her weight is stable, and she is eating normally. She feels less anxious about her symptoms and their possible cause. RECURRENT DUODENAL ULCER CASE 2 PRESENTATION Patient 2 is a 45-year-old white male construction worker who immigrated to the US from Eastern Europe 5 years ago. He presents to his primary care physician with a 2-week history of severe pain in the epigastrium. On closer questioning, it is apparent that he has had pain in this area for many years but that it has come and gone over time. It often disturbs him at night, frequently waking him from sleep. The pain is moderately relieved by eating or by taking liquid antacids. He has had no heartburn, nausea, vomiting, early satiety, weight loss, or appetite change. Approximately 10 years ago while still living in Eastern Europe, he had undergone EGD, during which a duodenal ulcer (DU) was discovered. He was treated with cimetidine at that time with good effect. After this, he used cimetidine or ranitidine for periods of a few weeks at a time as dictated by his symptoms. Another physician once prescribed him a short course of lansoprazole, which gave him excellent relief from his symptoms. However, his insurance company refused to cover a repeat prescription for that. He is currently on no prescribed medications, but is consuming many over-the-counter famotidine tablets. He has not used cimetidine or ranitidine in more than a year. He denies using aspirin or NSAIDs. He has no other relevant past medical history. His family history is noncontributory. He does not smoke and does not drink alcoholic beverages. There is some epigastric tenderness on physical examination, but the examination otherwise yields normal results. How should patient 2 initially be managed? Should he be tested for H. pylori infection? MANAGEMENT OF PATIENT 2 Patient 2 has had a previous endoscopic diagnosis of DU and presents with a typical recurrence of symptoms. Although he had responded in the past to H 2 RAs and to a PPI, he had never been tested for H. pylori infection. He came from Eastern Europe, where there is a higher prevalence of H. pylori infection than in the US. The primary care physician has to decide whether or not to (1) request an EGD, (2) prescribe him another course of acid-suppressing medicine, and (3) test him for H. pylori infection. EGD is probably not essential at this stage. Patient 2 had endoscopic confirmation of a DU in the past and seemed to be following a typical clinical course for that Gastroenterology Volume 7, Part 4 7

10 disorder. He has simply had a recurrence of his usual symptoms without any new or additional worrisome symptoms. It is, therefore, reasonable to assume that his pain is caused by a recurrence of his DU. A further course of an H 2 RA or a PPI would almost certainly have alleviated his symptoms and rehealed his ulcer. However, this would not have adequately addressed the underlying cause of his DU, which is probably H. pylori infection. Although most physicians test for H. pylori infection in a patient with a new diagnosis of DU, surprisingly few do this for patients with a previous diagnosis of DU. 20,21 This is highly regrettable, considering that these patients are just as likely to derive benefit from treating this infection as patients with a new ulcer diagnosis. The 1997 American Digestive Health Foundation International Update Conference on H. pylori 22 and the 1998 American College of Gastroenterology practice guidelines 23 both specifically recommend testing for H. pylori infection in patients with a previous diagnosis of ulcer. Appropriate methods of testing patient 2 for H. pylori infection would be the UBT or the stool antigen test. However, if the physician elects to pursue endoscopy for patient 2 (even though this is probably not absolutely necessary), his H. pylori status could be assessed by a biopsy urease test or histologic evaluation. CONTINUED MANAGEMENT OF PATIENT 2 The primary care physician gives patient 2 some sample 150-mg ranitidine tablets and tells him to take one tablet twice daily. He schedules the patient for an open EGD within the next week. At the time of EGD, the patient is still symptomatic although slightly improved. On EGD a grossly scarred and irregular duodenal bulb that contains a 1-cm ulcer is found. The EGD otherwise yields normal results; biopsy specimens from the gastric antrum are positive for H. pylori, as determined by a biopsy urease test. The patient is prescribed lansoprazole 30 mg, clarithromycin 500 mg, and amoxicillin 1000 mg (to be taken together) twice daily for 14 days. After the 14-day course, he takes no further medications. He sees his primary care physician 3 weeks later, at which time he reports that he is feeling well and has no residual abdominal pain. He noticed some diarrhea and taste disturbance during treatment with lansoprazole, clarithromycin, and amoxicillin, but these were mild and transient. Is patient 2 cured? Does he need to be retested for H. pylori infection? DISCUSSION As might have been expected, a DU was found on EGD. The scarring and deformity of the duodenal cap were consistent with a long history of recurrent DU. The patient was H. pylori positive and was treated with an appropriate eradication regimen. Assuming that the patient was fully compliant with the regimen, the probability that the infection has been successfully eradicated is 85% to 90%. Assuming for a moment that eradication has been successful, this should be associated with a cure of his ulcer diathesis. Unless he is exposed to another known ulcerogenic factor such as aspirin or an NSAID it is unlikely that he will have another recurrence of DU. Reinfection with H. pylori is extremely unusual in adults after successful eradication. Even if his spouse and close family members were H. pylori positive, it is highly unlikely that he would become reinfected. Most H. pylori infections are acquired in childhood 24 ; primary infection and reinfection are uncommon in adulthood. However, at this point in patient 2 s clinical course, it is not known whether or not the treatment regimen has been successful. Although current practice guidelines 23 do not endorse routine posttreatment testing, there is now increasing interest in and enthusiasm for this among both patients and physicians. At least one cost analysis 25 has found that routine posttreatment testing of H. pylori status among patients with uncomplicated PUD would save overall healthcare costs by allowing for the identification of those for whom the eradication regimen failed and the prompt retreatment of these individuals with an alternative regimen. This could prevent further recurrences and occasional complications that might otherwise require expensive inpatient care. OUTCOME OF PATIENT 2 Four weeks after completing his course of lansoprazole, clarithromycin, and amoxicillin, patient 2 has a negative test result on 13 C - UBT. He is asymptomatic at this time and is not taking any medication. He is reassured that his H. pylori infection has been eradicated. No routine follow-up is arranged. GASTRIC ULCER IN A PATIENT WITH ARTHRITIS CASE 3 PRESENTATION Patient 3 is a 72-year-old retired African American woman with a long history of fairly mild rheumatoid arthritis (RA) affecting her hands, wrists, and knees. Her internal medicine physician has referred her to a gastroenterology specialist because of abdominal pain. 8 Hospital Physician Board Review Manual

11 At the gastroenterologist s office, patient 3 complains of upper abdominal pain. She has been experiencing this pain intermittently for approximately 3 months. She has taken some 75-mg ranitidine tablets for this, but they have only given her some slight symptom relief. She is concerned about the chronicity of her symptoms and feels that she is not getting any better. She denies any nausea or vomiting but complains of a poor appetite. She is disinclined to eat and reports a 6-lb weight loss. She has no bowel habit upset and has not noticed any change in the color or form of her stools. Her past medical history is positive for mild hypertension, type 2 diabetes that is diet controlled, and hypothyroidism. The medications she is currently taking are atenolol, thyroxine, and naproxen. She has taken naproxen 250 mg twice daily for about 6 months; it has been very helpful for her arthritis, whereas previous NSAIDs had either not helped or had given her an upset stomach. She is a nonsmoker and does not drink alcoholic beverages. Her family history is noncontributory. The only positive findings on physical examination are rheumatoid changes in her hands and mild epigastric tenderness. Does patient 3 require early EGD? What role is naproxen playing in her problem? DISCUSSION Patient 3 has a longer history of dyspeptic complaints than did patient 1 and is also older. These factors increase the chances of finding a structural abnormality at EGD. Furthermore, she had been taking an NSAID. It is well known that NSAIDs can cause ulceration in the upper GI tract. However, many patients on NSAIDs who have dyspeptic symptoms are found not to have an ulcer; there is a very poor correlation between symptoms and endoscopic findings in patients taking these drugs. It is highly likely that the naproxen is a major contributor if not the sole cause of patient 3 s abdominal pain, but it cannot be predicted whether or not she will have an ulcer. For all of these reasons, EGD is necessary. INITIAL MANAGEMENT OF PATIENT 3 The gastroenterologist is able to work patient 3 into his endoscopy schedule later that week. On evaluation, patient 3 is found to have a 1.5-cm gastric ulcer (GU) in the antrum. The results of the remainder of the endoscopic evaluation are normal. Biopsy specimens are taken from the ulcer margin to investigate the possibility of the patient having a malignancy. Additional biopsy specimens are taken from multiple sites in the stomach to examine for evidence of H. pylori infection through histologic evaluation and biopsy urease testing. Results from the biopsy urease test are negative after 24 hours. The patient is started on lansoprazole 30 mg once daily, in the morning, to be taken before food. When the gastroenterologist sees her at his office 1 week later, she reports that she is feeling better. During this time, her biopsy specimens are reported to be negative for malignancy and for H. pylori infection. She is taking lansoprazole 30 mg once daily and continues to take naproxen 250 mg twice daily because her RA symptoms have been well controlled by it. Should patient 3 stop taking naproxen? How long will she require PPI treatment? Is further endoscopic evaluation required? DICUSSION Patient 3 has responded symptomatically to treatment with a PPI. However, she is still taking the NSAID, which is presumably the cause of the ulcer, considering that 2 different tests for H. pylori infection have yielded negative results. RCTs have shown that NSAID-related ulcers will heal on an H 2 RA, misoprostol, or a PPI even if patients continue to take NSAIDs However, healing is faster if the patient is able to stop taking the NSAID. PPIs are the most effective agents for healing NSAID-related ulcers whether NSAID treatment is continued or not Healing rates of individuals taking PPIs are higher than those of individuals taking H 2 RAs. Healing rates associated with PPIs are slightly higher than healing rates associated with misoprostol. PPI treatment is preferable to misoprostol treatment because adverse events notably diarrhea and abdominal discomfort are significantly more common with misoprostol. PPIs are also effective in the secondary prophylaxis of NSAID-related ulcer recurrence in patients with healed NSAID-related ulcers who will be continuing NSAID therapy. Therefore, it would be reasonable to treat patient 3 with a PPI for at least 8 weeks in order to heal the ulcer. After that time, PPI therapy should probably be continued as secondary prophylaxis if the NSAID is to be continued. In the case of patient 3, it is likely that NSAID therapy will be continued she has mild RA for which the NSAID is providing good symptom control. One option would be to keep her on naproxen and lansoprazole indefinitely. Another would be to discontinue the naproxen and treat her with a selective inhibitor of cyclooxygenase-2 (COX-2). Such inhibitors are associated with significantly fewer ulcers (as determined by Gastroenterology Volume 7, Part 4 9

12 endoscopic evaluation) than are conventional, nonselective NSAIDs. 29 Furthermore, in a large, prospective outcomes study, rofecoxib was associated with significantly fewer symptomatic ulcers, perforations, and bleeds than was naproxen. 30 However, switching patient 3 to a COX-2 selective inhibitor might not obviate the need for ulcer prophylaxis. The gastric biopsy specimens taken during the EGD evaluation showed no evidence of malignancy. Many physicians would repeat the EGD procedure after PPI treatment to ensure that the GU had healed and, perhaps, to repeat the biopsy procedures to be able to exclude gastric cancer more definitively. OUTCOME OF PATIENT 3 After 8 weeks of taking lansoprazole 30 mg once daily, the patient is seen again by the gastroenterologist and is asymptomatic. A repeat EGD shows healing of the GU. The patient is reluctant to take another medication in place of the naproxen because she has experienced good control of her joint pain with the medication. She is therefore kept on naproxen 250 mg twice daily and is given secondary prophylaxis with lansoprazole 30 mg once daily. INTRACTABLE PEPTIC ULCER DISEASE CASE 4 PRESENTATION Patient 4 is a 45-year-old white woman who works as a hospital pharmacy technician. She was referred to the gastroenterology outpatient clinic at a university medical center for a second opinion regarding intractable PUD. At the clinic, she complains of epigastric pain that has troubled her intermittently for over 10 years. She has not obtained relief from H 2 RAs over that time and is currently taking omeprazole 20 mg twice daily. Despite this, she has still had moderate epigastric pain. She denies nausea, vomiting, and symptoms of GI bleeding. Regarding her past history, she has had 3 documented episodes of hematemesis and/or melena in the past 10 years. She has been admitted to different hospitals for management and has brought some of the pertinent medical records with her. On one occasion, a prepyloric ulcer with duodenal scarring was found at EGD; on another occasion, a DU was found at EGD. Once, she required emergency laparotomy and oversewing of a perforated DU. All of the tests she has undergone for possible H. pylori infection including serologic tests, a 14 C - UBT, and biopsy urease tests at the time of EGD have yielded negative results. Within the past year, a fasting serum gastrin level, measured while she was temporarily off omeprazole, had been normal. Blood work from 1 month before this current consultation included a normal complete blood count and a normal serum calcium level. She denies use of aspirin and NSAIDs. She has no other relevant past medical history and is on no prescribed medications except for omeprazole. She smokes a pack of cigarettes daily but denies alcoholic-beverage consumption and drug abuse. Her family history is noncontributory. Results from a physical examination are unremarkable except for the presence of epigastric tenderness and the scar from her laparotomy. What is likely to be the cause of patient 4 s recurrent peptic ulceration? Are any additional investigations indicated at this time? What general advice should patient 4 be given? DISCUSSION This patient has a long and impressive history of recurrent and complicated PUD. Despite this, she appears to lack the 2 principal risk factors for PUD namely, H. pylori infection and aspirin/nsaid use. Other rare causes of PUD have been investigated, but she appears not to have either Zollinger-Ellison syndrome or hypercalcemia as judged by her recent laboratory investigations. Certainly, her past medical records including the negative test results for H. pylori infection must be carefully reviewed. If the 14 C - UBT had been performed while patient 4 was on PPI treatment, the result obtained might have been a false negative; PPIs reduce the sensitivity of the UBT 31 as well as biopsy urease tests and histologic evaluations. Furthermore, biopsy urease tests may have impaired sensitivity if there is blood in the gastric lumen after an upper GI bleed. However, because patient 4 also had a negative serologic test result, it seems likely that she truly is H. pylori negative. Consideration might be given to repeating the EGD, although it is unlikely that this would influence her immediate management. A repeat serum calcium measurement would cost very little and should probably be performed. It would also be worthwhile to consider remeasuring her fasting serum gastrin level, although this would require stopping the omeprazole for approximately 1 week. Her use of nonprescription drugs including aspirin, aspirin-containing medicines, and NSAIDs should be explored with tact and sensitivity. Patients 10 Hospital Physician Board Review Manual

13 may not report use of these products to their physician and may not consider them to be drugs or medicines. The surreptitious use of aspirin may be found among women who have some connection to healthcare work patient 4 works as a pharmacy technician. Such aspirin abuse has been linked to PUD that is unresponsive to conventional treatment and that has been associated with complications. 32 Lastly, some patients take aspirin or NSAIDs for unconventional reasons perhaps as a sleep aid or for relief of painful conditions for which they have not sought medical attention. All these possibilities should be explored. A random serum salicylate level measurement might be considered but should probably only be done with the patient s informed consent. However, this would only exclude aspirin use and would not detect use of nonaspirin NSAIDs. If real doubt persists, a test of platelet function might be considered. All currently available NSAIDs aside from selective COX-2 inhibitors inhibit platelet COX-1 activity. Regarding general advice, patient 4 should be advised to stop smoking both for the benefit of her general health and for her ulcer disease. Hopefully, she has previously been advised to avoid aspirin and other NSAIDs; this should be reemphasized. INITIAL MANAGEMENT OF PATIENT 4 A first-year GI fellow is assigned to discuss her medical care, and the results of her investigations, with the physicians who had previously examined and treated her. The physicians confirm the previous endoscopic findings and that her 14 C-UBT had been performed while she had been off omeprazole for 2 weeks. The GI fellow performs another EGD, and on this occasion, a scarred and deformed prepyloric area and duodenal bulb are found. However, the physician does not find any active ulceration. Patient 4 is asked to stop taking omeprazole for 1 week for a recheck of her fasting serum gastrin level, which is once again within the normal range. Compliance with medications is discussed in detail with the patient. She adamantly denies noncompliance with omeprazole and also further denies using aspirin, NSAIDs, or other nonprescription products. A serum salicylate level is undetectable. A careful review of the records of patient 4 s previous hospital admissions and relevant investigations confirms that she is uninfected with H. pylori and that there is no evidence of aspirin abuse. If doubt had persisted about NSAID use, an in vitro test of platelet function might have been considered. However, based on all the accumulated evidence, this seems unlikely. It is reasonable to assume, therefore, that this patient has H. pylori negative PUD that is idiopathic. This is an increasingly recognized phenomenon, particularly in the United States. 33,34 Although it was once believed that H. pylori infection accounted for almost all peptic ulcers observed in clinical practice, this is no longer considered to be true. In the US, H. pylori may only be responsible for approximately 70% of ulcers, although it remains the single most common cause. In Europe and Asia, the vast majority of ulcers are associated with H. pylori infection. The reason for the high incidence of H. pylori negative ulcers in the US is not clear. The background population prevalence of H. pylori infection is lower in the US than in Asia and most parts of Europe and may have been falling at a more rapid rate. Presumably, a proportion of ulcers have always been unrelated to H. pylori infection. In the US, this proportion is currently larger than elsewhere. Patients with H. pylori negative ulcers typically have a more aggressive clinical course than those with H. pylori positive ulcers. In an initial small study from Scotland, McColl et al identified 6 patients with an H. pylori negative, or idiopathic, DU. 35 They had a particularly severe form of ulcer disease with frequent relapses and complications despite treatment with H 2 RAs. OUTCOME OF PATIENT 4 Patient 4 is advised to continue omeprazole 20 mg twice daily indefinitely. Once again, she is advised to stop smoking and to avoid use of aspirin and NSAIDs. Acid-reducing surgery is considered, but she indicates a preference to stay on medication because she is confident that it will be safe and efficacious and because there are no RCTs of surgery for H. pylori - negative PUD. REFERENCES 1. Talley NJ, Stanghellini V, Heading RC, et al. Functional gastroduodenal disorders. Gut 1999;45 Suppl: II Talley NJ. Therapeutic options in nonulcer dyspepsia. J Clin Gastroenterol 2001;32: Rabeneck L, Menke T. Increased numbers of women, older individuals, and Blacks receive health care for dyspepsia in the United States. J Clin Gastroenterol 2001; 32: Thompson WG, Longstreth GF, Drossman DA, et al. Functional bowel disorders and functional abdominal pain. Gut 1999;45 Suppl 2:II Greenlee RT, Hill-Harmon MB, Murray T, Thun M. Cancer statistics, CA Cancer J Clin 2001;51: Gastroenterology Volume 7, Part 4 11