American Journal of Gastroenterology ISSN 0002-9270

C 2005 by Am. Coll. of Gastroenterology doi: 10.1111/j.1572-0241.2005.00225.x
Published by Blackwell Publishing

PRACTICE GUIDELINES

Guidelines for the Management of Dyspepsia

Nicholas J. Talley, M.D., Ph.D., F.A.C.G.,1 Nimish Vakil, M.D., F.A.C.G.,2 and the Practice Parameters
Committee of the American College of Gastroenterology
1
Division of Gastroenterology and Hepatology, Mayo Clinic, Clinical Enteric Neuroscience Translational and
Epidemiological Research Program, Mayo Clinic, Rochester, Minnesota; and 2 University of Wisconsin Medical
School and Marquette University College of Health Sciences, Milwaukee, Wisconsin

Dyspepsia is a chronic or recurrent pain or discomfort centered in the upper abdomen; patients with predominant
or frequent (more than once a week) heartburn or acid regurgitation, should be considered to have
gastroesophageal reflux disease (GERD) until proven otherwise. Dyspeptic patients over 55 yr of age, or those with
alarm features should undergo prompt esophagogastroduodenoscopy (EGD). In all other patients, there are two
approximately equivalent options: (i) test and treat for Helicobacter pylori (H. pylori) using a validated noninvasive
test and a trial of acid suppression if eradication is successful but symptoms do not resolve or (ii) an empiric trial
of acid suppression with a proton pump inhibitor (PPI) for 48 wk. The test-and-treat option is preferable in
populations with a moderate to high prevalence of H. pylori infection (10%); empirical PPI is an initial option in
low prevalence situations. If initial acid suppression fails after 24 wk, it is reasonable to consider changing drug
class or dosing. If the patient fails to respond or relapses rapidly on stopping antisecretory therapy, then the
test-and-treat strategy is best applied before consideration of referral for EGD. Prokinetics are not currently
recommended as first-line therapy for uninvestigated dyspepsia. EGD is not mandatory in those who remain
symptomatic as the yield is low; the decision to endoscope or not must be based on clinical judgement. In patients
who do respond to initial therapy, stop treatment after 48 wk; if symptoms recur, another course of the same
treatment is justified. The management of functional dyspepsia is challenging when initial antisecretory therapy
and H. pylori eradication fails. There are very limited data to support the use of low-dose tricyclic antidepressants
or psychological treatments in functional dyspepsia.
(Am J Gastroenterol 2005;100:23242337)

INTRODUCTION as a subjective negative feeling that is nonpainful, and can

incorporate a variety of symptoms including early satiety
These and the previous guidelines were developed under or upper abdominal fullness. Patients presenting with pre-
the auspices of the American College of Gastroenterol- dominant or frequent (more than once a week) heartburn
ogy and its Practice Parameters Committee and approved or acid regurgitation should be considered to have gastro-
by the Board of Trustees. The world literature was re- esophageal reflux disease (GERD) until proven otherwise.
viewed extensively using the National Library of Medicine Dyspepsia is a common complaint in clinical practice;
database. Appropriate studies were reviewed and any ad- therefore, its management should be based on the best ev-
ditional studies found in the reference list of these papers idence. Dyspepsia has often been loosely defined; the most
were obtained and reviewed. Evidence was evaluated along a widely applied definition of dyspepsia is the Rome Work-
hierarchy, with randomized, controlled trials given the great- ing Teams formulation, namely chronic or recurrent pain or
est weight. Abstracts presented at national and international discomfort centered in the upper abdomen (1). Predominant
meetings were only used when unique data from ongoing epigastric pain or discomfort helps to distinguish dyspepsia
trials were presented. When scientific data were lacking, from GERD; in the latter the dominant complaint is typically
recommendations were based on expert consensus obtained heartburn or acid regurgitation but there may be a distinct
from both the literature and the experience of the authors epigastric component that is confusing (2). Frequent reflux
and the Practice Parameters Committee. Each guideline was symptoms (twice a week or more) probably impair quality
evaluated by the committee and the strength of evidence to of life and are generally considered to identify GERD until
guide clinical practice was assessed using established criteria proven otherwise (36). Clinical trials in dyspepsia have used
(Table 1). various definitions and have often not distinguished obvious
GERD from dyspepsia, making interpretation of treatment
DEFINITIONS responses problematic.
Discomfort has been defined by the Rome Working Teams
Dyspepsia is defined as chronic or recurrent pain or discom- as a subjective negative feeling that is nonpainful, and
fort centered in the upper abdomen. Discomfort is defined has been considered to incorporate a variety of symptoms

2324
 Guidelines for the Management of Dyspepsia 2325

Table 1. Levels of Evidence symptoms, explaining the observation that the prevalence re-
Level mains stable.
I Evidence from RCTs with low false positive rates (i.e.,
significant p values), adequate sample sizes (low likelihood
of type II errors) and appropriate methodology (low NATURAL HISTORY AND COSTS OF DYSPEPSIA
likelihood of type I errors)
II Evidence from RCTs with high false positive rates, Dyspepsia is usually a chronic condition in primary and sec-
inadequate sample sizes, or inappropriate methodology ondary care. The costs in the United States remain poorly doc-
III Evidence from nonrandomized trials using a umented, but in Sweden a total societal cost of $63 per adult
contemporaneous cohort of controls was calculated for dyspepsia (including reflux disease) (11).
IV Evidence from nonrandomized trials using a historical cohort
In another study, 288 adult primary care patients with dys-
of controls
V Evidence from case series without controls pepsia were followed up for 1 yr; dyspepsia patients tended
to remain symptomatic with 61% using drugs and 43% hav-
Note: Adapted from Cook D et al. Chest 1992;102:305S.
ing gastrointestinal procedures, indicating intensive use of
medical resources (12).

including early satiety, bloating, upper abdominal fullness, or

nausea (1). However, bloating is most typically a symptom of DIAGNOSTIC TESTING
IBS and may not be located in the upper abdomen exclusively.
Nausea can be secondary to a variety of nonabdominal con- Dyspeptic patients more than 55 yr old, or those with
ditions. Hence, neither bloating nor nausea alone should be alarm features (bleeding, anemia, early satiety, unexplained
considered to identify dyspepsia. Belching alone is also an weight loss (>10% body weight), progressive dysphagia,
insufficient symptom to identify dyspepsia and can be sec- odynophagia, persistent vomiting, a family history of gas-
ondary to air swallowing, although it is commonly present trointestinal cancer, previous esophagogastric malignancy,
with epigastric pain or discomfort. Acute self-limited dys- previous documented peptic ulcer, lymphadenopathy, or an
pepsia generally requires no investigation and will not be abdominal mass) should undergo prompt endoscopy to rule
further considered here in these management guidelines. out peptic ulcer disease, esophagogastric malignancy, and
other rare upper gastrointestinal tract disease.
In patients aged 55 yr or younger with no alarm features,
the clinician may consider two approximately equivalent
EPIDEMIOLOGY OF DYSPEPSIA management options: (i) test and treat for H. pylori using
a validated noninvasive test and a trial of acid suppression
It is established that dyspepsia is a common problem world- if eradication is successful but symptoms do not resolve
wide. In the United States, the point prevalence is approxi- or (ii) an empiric trial of acid suppression with a proton
mately 25%, excluding those people who have typical GERD pump inhibitor (PPI) for 48 wk. The test-and-treat option
symptoms (7). The prevalence is lower if patients with any is preferable in populations with a moderate to high preva-
symptoms of heartburn and regurgitation are excluded (8). lence of H. pylori infection (10%), whereas the empirical
The incidence is more poorly documented. In the United PPI strategy is preferable in low prevalence situations.
States, approximately 9% of people who had no symptoms of Some anxious patients may need the reassurance af-
dyspepsia anually in the prior year reported new symptoms forded by endoscopy. On the other hand, repeat EGD is
on follow-up; however, those with a past history of dyspepsia not recommended once a firm diagnosis of functional dys-
or peptic ulcer were not excluded and hence the onset-rate pepsia has been made, unless completely new symptoms or
may be exaggerated (9). In Scandinavia, an incidence rate of alarm features develop. Repeat EGD is otherwise unlikely
less than 1% over 3 months has been reported (10). Whatever to ever be cost-effective.
the incidence, the number of subjects who develop dyspepsia Grades of evidence:
is matched by a similar number of subjects who lose their Early endoscopy for alarm symptoms: C
Test-and-treat strategy for H. pylori: A
Acid suppression therapy: A
Table 2. Graded Recommendations for Clinical Practice
Reassurance after endoscopy: C
Grade Strength of Evidence to Guide Clinical Practice Very few studies have investigated dyspepsia subjects from
A Supported by two or more level I studies without the community by esophagogastroduodenoscopy (EGD) and
conflicting evidence from other level I studies other tests, to determine the underlying causes of the symp-
B Supported by two or more level I studies with conflicting
toms. In a population-based study from northern Norway,
evidence from other level I studies or supported by only
one level I or two or more level II studies amongst those with epigastric pain only 9% had a peptic
C Supported by level IIIV evidence ulcer and 14% had reflux esophagitis, but how many had en-
Note: Adapted from Guyatt GH et al. JAMA 1995;274:18001804; Users Guides to
doscopy negative reflux disease is uncertain (13). In a com-
the Medical Literature, JAMA Press 2001; and Cook D et al. Chest 1992;102:305S. parable study from northern Sweden, a similar proportion of
2326 Talley et al.

Dyspepsia (uninvestigated)
uncertain; hence, these patients are not excluded from the
functional dyspepsia diagnosis category (26).

Age < 55 PATHOPHYSIOLOGICAL DISTURBANCES IN

Age > 55 or alarm features No alarm features
ENDOSCOPY-NEGATIVE (FUNCTIONAL) DYSPEPSIA
Approximately 40% of patients with functional dyspepsia
HP prevalence HP prevalence > 10%
<10%
have delayed gastric emptying (27). However, it is controver-
EGD sial whether a specific symptom profile is associated with
Test and treat
PPI trial for H. pylori
delayed gastric emptying, and whether changes in gastric
Fails
emptying can predict symptom improvement in functional
Fails
dyspepsia. Stanghellini et al. in 343 Italian patients reported
Test and treat
for H pylori
PPI trial that delayed gastric emptying was significantly more frequent
Fails Fails in patients characterized by female sex, low body weight,
Consider EGD Consider EGD presence of relevant and severe postprandial fullness, nausea,
vomiting, and absence of severe epigastric pain; female sex,
Figure 1. Algorithm for the management of uninvestigated
dyspepsia relevant and severe postprandial fullness, and severe vomiting
were independently associated with delayed gastric emptying
of solids (28). In a separate study of 483 patients, the same
subjects had peptic ulcer or esophagitis, although 32% with Italian group identified distinct subgroups based on predom-
esophagitis were asymptomatic (14). Many people with dys- inant symptoms and gastric emptying; one was character-
pepsia presenting to primary care have no obvious cause for ized by predominant epigastric pain, male gender and normal
their symptoms based on EGD. The most common finding in gastric emptying, and a second by predominant nonpainful
North America is probably esophagitis; in a Canadian study symptoms, female gender, and a high frequency of associ-
of uninvestigated dyspepsia in primary care, 43% of 1,040 pa- ated irritable bowel syndrome and delayed gastric emptying
tients had erosive esophagitis and only 5% a peptic ulcer, but (29). Sarnelli et al. also reported that delayed gastric empty-
this study did include patients with heartburn (15). Studies ing was associated with postprandial fullness and vomiting
from open-access endoscopy practices and outpatient series (30). Other studies, however, have failed to identify a definite
support the view that only a minority of patients presenting symptom profile associated with delayed gastric emptying
with dyspepsia have peptic ulcer disease or reflux esophagi- suggesting there is not a simple association (31). Moreover,
tis, and gastric cancer is relatively rare in western populations evidence that a gastric emptying test cost-effectively alters
(16, 17). management is not available.
Additional diagnostic testing over and above EGD has a There is evidence that the stomach and other regions of
low yield in dyspepsia, at least in primary care. Studies apply- the gut including the duodenum and esophagus are hyper-
ing abdominal ultrasonography in dyspepsia have reported sensitive to distention in functional dyspepsia, although this
few abnormalities aside from asymptomatic cholelithiasis applies only in a subgroup (3236). Tack et al. recently re-
that needs no intervention (18, 19). Endoscopic ultrasonog- ported in 160 patients with functional dyspepsia that one third
raphy (EUS) has been reported to have a higher yield of had gastric hypersensitivity and this abnormality was associ-
identifying pancreatico-biliary pathology but selection bias ated with increased postprandial pain as well as belching and
may explain the observation and much of the pathology iden- weight loss, but confirmatory data are needed on the symptom
tified is of questionable significance (20, 21). Twenty-four associations (35).
hour esophageal pH testing can identify pathological acid In a barostat study, Tack et al. studied patients with func-
reflux in approximately 20% of patients with a clinical and tional dyspepsia; impaired gastric accommodation to a meal
endoscopic diagnosis of functional dyspepsia (2225). How- (a stiff fundus) was found in 40%, and this abnormality
ever, the symptom criteria used to define functional dyspep- was associated with early satiety and weight loss but not with
sia in these studies have generally been broader than recom- hypersensitivity to gastric distention, presence of H. pylori,
mended by the Rome Committees, and hence patients with or delayed gastric emptying (37). However, Boeckxstaens
typical reflux symptoms contaminated the studies. Klauser et et al. failed to replicate these findings; while postprandial
al. extensively evaluated a group of patients with functional symptoms were more often evoked with a meal in functional
dyspepsia; they reported that 47% had abnormal findings on dyspepsia, there was no clear symptom profile that was asso-
additional testing but the significance of the various abnor- ciated with a failure of fundic relaxation (38). Noninvasive
malities identified, including minor delays in gastric empty- testing is available to assess abnormal fundic accommoda-
ing and lactose intolerance remains questionable (22). De- tion including gastric ultrasound, SPECT, and MRI, but the
pending on the background prevalence of H. pylori, this in- clinical relevance of identifying this abnormality remains in
fection will be identified in 2060% of patients with func- some dispute in terms of defining therapeutic interventions
tional dyspepsia, but the clinical relevance in most cases is (39).
 Guidelines for the Management of Dyspepsia 2327

New clinical tests of gastric function are under evaluation. abnormalities, but the results have been inconsistent (45, 46).
The water-load test and nutrient-load test may help identify The optimal age threshold for endoscopy is unclear but 55 yr
gastric dysfunction in clinical practice (40, 41). These rep- (rather than 45 yr) seems a reasonable cut-off because cancer
resent simple tests of the ability of a patient to drink water is rare in younger patients in the United States, but no age
or a nutrient load such as Ensure
until they feel completely threshold is absolute (47).
full. Dyspepsia patients tolerate lower volumes than controls Several other alarm features have been traditionally ap-
for example, and have more symptoms 30 min after reach- plied to try and identify serious underlying disease in dyspep-
ing satiation. Hence, this is a stomach stress test and can sia, especially malignancy. These include unexplained weight
objectively quantify postprandial distress. However, normal loss, anorexia, early satiety, vomiting, progressive dyspha-
cutoffs vary by laboratory (as do test protocols), and the rate gia, odynophagia, bleeding, anemia, jaundice, an abdominal
of gastric emptying of the nutrient meal as well as relaxation mass, lymphadenopathy, a family history of upper gastroin-
of the fundus secondary to meal ingestion can potentially testinal tract cancer, or a history of peptic ulcer, previous
modulate the test results. Some have found that the drink gastric surgery or malignancy. Upper gastrointestinal ma-
tests correlate with fundic dysaccommodation rather than lignancy is rarely present in young patients without alarm
visceral hypersensitivity (42). Others have failed to demon- features, but the positive predictive value of alarm features
strate a relationship to gastric dysfunction while some data remains very poor (47, 48). A long history of symptoms in
suggest these tests correlate with psychological disturbances patients should make cancer unlikely but a symptom dura-
(40, 41). Currently, patients with gastroduodenal motility dis- tion threshold has not been defined in the literature. Use of
turbances, gastroduodenal hypersensitivity, or other patho- antisecretory therapy can mask a cancer at endoscopy (49)
physiological abnormalities of uncertain relevance are not but does not appear to alter the outcome (50).
excluded from the functional dyspepsia umbrella. Although alarm symptoms are not specific for a serious
underlying disorder, few patients younger than 55 yr of age
SYMPTOMS AND SYMPTOM SUBGROUPS with an upper gastrointestinal malignancy present without
alarm symptoms. In patients with alarm features, and in older
There is convincing evidence that a patients symptoms cannot patients >55 yr of age with new symptoms, prompt EGD is
be used to identify structural disease in uninvestigated dys- considered the gold standard to ensure that malignancy has
pepsia (15, 43). Working teams have suggested subdividing not been missed. There are regions in the United States of
dyspepsia into ulcer-like or dysmotility-like dyspepsia based high cancer incidence where lower age thresholds may need
on symptom patterns or predominance; it was postulated that to be considered such as Alaska (51). On the basis of expert
symptom subgroups could identify more homogenous pop- opinion, if an EGD has already been done recently, repeating
ulations that would respond to targeted medical therapy (1, this test is highly unlikely to alter management.
7). However, individual symptoms, symptom subgroups, and The patient who presents with new onset dyspepsia or be-
scoring systems have all failed to be useful in identifying un- cause of chronic symptoms needs an appropriate, evidence-
derlying peptic ulcer disease, or distinguishing organic from based clinical evaluation. The physician generally wishes to
functional dyspepsia. A study from Canada reported that the ascertain the likely cause of the symptoms and exclude under-
patients dominant symptom (including heartburn) failed to lying serious structural disease. However, the patient may ac-
predict endoscopic findings in a primary care population (15). tually be presenting not necessarily because of the symptoms
It is thus controversial whether subdividing dyspepsia into per se but because of a fear of serious disease or recent psy-
symptom subgroups aids management in documented func- chological distress. It is reasonable that the physician identify
tional dyspepsia. and address such issues as fear of cancer or underlying heart
disease in order to optimize management (52).
ALARM FEATURES AND IDENTIFICATION OF STRUCTURAL The patient requiring major reassurance needs to be dif-
ferently managed than one who does not have such concerns,
DISEASE IN UNINVESTIGATED DYSPEPSIA
but fear of serious disease probably explains only some health
The risk of malignancy increases with age and therefore care seeking behavior (53). The physician also needs to de-
empirical therapy is not currently recommended in individ- cide whether pharmacological therapy is required, includ-
uals over 55 yr of age who develop new dyspeptic symptoms. ing which drug and for how long. This in turn depends on
Grade of evidence: C the underlying provisional diagnosis, which may need to be
New-onset dyspepsia in older age is an alarm feature or red refined after the patient has initially had a trial of therapy.
flag. The American College of Physicians in 1985 published
a guideline recommending that patients who were over the MANAGEMENT OPTIONS IN YOUNGER PATIENTS WITH NO
age of 45 deserved referral for prompt endoscopy to rule
ALARM FEATURES
out underlying malignancy, as gastric cancer is very rare in
the United States below the age of 45 yr although it increases A number of management options are available to the clin-
thereafter (44). Some studies have reported that older age is an ician in younger patients with no alarm features with un-
independent risk factor for identifying underlying structural investigated dyspepsia. A wait-and-see strategy of patient
2328 Talley et al.

reassurance and education, with use of over-the-counter gastric ulcer (GU) in 13%; those who were breath test nega-
antacids, H2 -blockers, or PPIs and reevaluation can be con- tive had a DU in 2% and GU in 3% (56). Other studies suggest
sidered, particularly in primary care. Another strategy worth that between 20% and 60% of patients with dyspepsia who
considering is prescription of empirical full-dose or high- are H. pylori infected will have underlying peptic ulcer dis-
dose antisecretory therapy, reserving further evaluation for ease, but this varies widely depending upon the background
those who are either unresponsive or have an early symp- incidence of peptic ulcer (57, 58). Cost-effectiveness stud-
tomatic relapse after ceasing medication. Empiric antisecre- ies in the United States suggest that when the prevalence of
tory therapy was the backbone of the guideline proposed by H. pylori infection in patients with functional dyspepsia is
the American College of Physicians and is still widely ap- less than 12% or when the prevalence of H. pylori infection
plied in practice (44). A third approach applies H. pylori test- in patients with peptic ulcer disease is less than 48%, ini-
and-treat as the initial strategy, currently most widely rec- tial empirical treatment with a PPI is preferable (59). Others
ommended around the world (54, 55). Here, young patients have suggested that when H. pylori infection decreases below
without alarm features are tested for H. pylori infection. If 20%, empiric PPI therapy starts to dominate test-and-treat in
H. pylori is detected, empiric antibiotic therapy is prescribed uninvestigated dyspepsia (60).
in an attempt to eradicate the infection; H. pylori-negative
patients are treated with empiric antisecretory therapy ini- Test-and-Treat H. pylori Versus Placebo in Dyspepsia in
tially. A modification of the H. pylori test-and-treat strategy the Community
is to either prescribe empiric antisecretory therapy first and There are data indicating a small benefit for treating H. pylori
reserve H. pylori testing later for failures, or apply empiric empirically in those with the infection in the community
antisecretory therapy after H. pylori eradication fails to re- (nonpatients). In a U.K. community trial, 32,929 individu-
lieve symptoms. A final approach is to perform prompt EGD als were invited and 8,455 attended and were eligible; 2,329
for all patients with dyspepsia. The best option remains un- were positive for H. pylori and were assigned active treatment
der debate, but new data are available to help guide a rational or placebo, with 1,773 (76%) returning at 2 yr (61). There
decision. was an absolute risk reduction of 5% for upper GI symptoms
on active therapy versus placebo, although quality of life was
unchanged. Presumably much of this benefit is explained by
the treatment of undiagnosed peptic ulcer disease.
TEST-AND-TREAT H. pylori
The application of a test-and-treat strategy for H. pylori Test-and-Treat H. pylori Versus Usual Management of
should be based on the practice setting (Fig. 1). High- Uninvestigated Dyspepsia in Primary Care
prevalence populations in the United States (e.g., recent Chiba et al. conducted a randomized placebo-controlled trial
immigrants from developing countries) should undergo in 36 family practices in Canada; they randomized 294
test-and-treat as the preferable nonendoscopic strategy. H. pylori-positive patients to omeprazole plus antibiotics or
Conversely, in communities where gastric or esophageal omeprazole plus placebo for 1 wk, and then arranged follow-
cancer has a high incidence, prompt endoscopy should be up by family physicians for usual care (62). They found eradi-
considered early but this would not apply to most of the cation resulted in no or minimal symptoms in 50% of patients
country. In low-prevalence populations (e.g., high socioe- compared to 36% in the placebo-therapy arm at the end of 12
conomic areas, where the background prevalence of ulcer months. It is of interest that this benefit was observed despite
or H. pylori infection is low), an alternative strategy is to including some GERD patients in this trial. The eradication
prescribe first a course of antisecretory therapy empirically therapy arm also reduced costs by Can$53 per patient. Al-
for 48 wk. If the patient fails to respond or relapses rapidly lison et al. in a study in primary care in the United States
on stopping antisecretory therapy, then the test-and-treat observed no cost benefit of test-and-treat over usual care al-
strategy is best applied before consideration of referral for though symptoms were significantly reduced in the test-and-
EGD. EGD is not mandatory in those who remain symp- treat arm (63). An underpowered U.S. study failed to detect
tomatic as the yield is low; the decision to endoscope or not a difference between test-and-treat and usual care (64).
must be based on clinical judgement.
Grade of evidence for test-and-treat or acid suppr- Test-and-Treat H. pylori Versus Prompt EGD in Primary
ession: A and Secondary Care
Grade of evidence for a H. pylori prevalence of less than There is consistent empiric evidence that a test-and-treat strat-
10% in the local community as the cutoff for deciding to egy is at least equivalent to prompt endoscopy in terms of
use empiric acid suppression rather than test-and-treat: C outcomes. Lassen et al. randomized 500 patients (including
The rationale for noninvasive H. pylori testing is the iden- older patients) in primary care with dyspepsia to either H.
tification of underlying peptic ulcer disease. For example, in pylori test-and-treat or prompt endoscopy (65). They found
Scotland where the incidence of peptic ulcer is high, McColl that there were no differences in symptomatic outcomes or
et al. showed that in patients with dyspepsia and a positive quality of life between the groups at 1 yr, although the en-
C13 urea breath test had a duodenal ulcer (DU) in 40% and doscopy group had a slightly higher patient satisfaction score
 Guidelines for the Management of Dyspepsia 2329

of questionable clinical significance. The authors also iden- lori is very uncommon, a positive test is more likely to be a
tified a reduction in the number of endoscopic procedures false positive. Where H. pylori infection is highly prevalent,
performed in the test-and-treat arm. Heaney et al. in Ireland a negative result is more likely to be a false negative (77).
evaluated dyspepsia patients less than 45 yr old referred to Cost-effectiveness studies suggest that the stool test and the
an open-access endoscopy unit who were H. pylori-positive urea breath test that detect active infection are preferable to
on noninvasive testing (66). Patients here were randomized serological tests in the United States (78, 79).
to either empiric H. pylori therapy or immediate EGD. They The current treatment of choice for H. pylori infected pa-
found that more patients became symptom free in the H. tients is a combination of PPI (standard dose twice daily)
pylori eradication arm than in the prompt endoscopy arm. with amoxicillin (1 g twice daily) and clarithromycin (500
McColl et al. evaluated 708 patients under age 55 yr referred mg twice daily) administered for 710 days (7-day therapy
for endoscopy; these patients were randomized to either H. is approved with rabeprazole; 10-day therapy is approved
pylori test-and-treat or endoscopy including H. pylori testing with lansoprazole, omeprazole, pantoprazole, and esomepra-
(67). They found no significant difference in dyspepsia score zole). Metronidazole (400 mg twice daily) may be substi-
at the 12 months follow-up in the two groups. Furthermore, tuted for amoxicillin in this regimen if the patient is allergic
only 8% of patients who had testing and treatment eventually to penicillin. An alternative strategy is the combination of
underwent endoscopy; overall patient satisfaction and quality Bismuth, metronidazole, and tetracycline (Bismuth subsal-
of life was similar in both groups. Jones et al. evaluated 232 icylate [Pepto Bismol
] 525 mg QID + metronidazole 250
patients in primary care, of whom 141 underwent testing and mg QID + tetracycline 500 mg QID) combined with a PPI
treatment for H. pylori; 91 who had previously undergone for 14 days (80, 81).
endoscopy comprised the control group (68). Although not A final issue relates to potential complications of therapy.
a randomized controlled trial, they identified similar clini- Antibiotic allergies and super-infection can occur. It is con-
cal outcomes but lower costs in the test-and-treat group at troversial whether eradication of H. pylori infection increases
1 yr. Because this was a retrospective, unmatched noncon- the risk of development of reflux esophagitis or reflux symp-
secutive controlled study, the results are difficult to interpret. toms (82, 83). However, it appears likely that this risk is
Additional randomized trial data (69) and a Cochrane meta- only present in those with a predisposition to GERD who
analysis (70) suggest overall that prompt EGD and test-and- also have severe gastritis in the body or fundus that impairs
treat have similar efficacy. acid secretion, which is reversed with H. pylori eradication;
Other evidence supports the view that H. pylori testing this is likely to be uncommon in most of the United States
may provide adequate patient reassurance. Patel et al. eval- (84). Hence, this issue while much discussed should not be
uated 193 dyspepsia patients under the age of 45 yr (71). a major clinical concern when contemplating test-and-treat,
Seventy of these patients were H. pylori-seronegative with- unless convincing data to the contrary arise. Progression of
out alarm features, 90 were seropositive for H. pylori and 23 H. pylori gastritis may occur on acid suppression, and some
had alarm features; the H. pylori-positive patients and those have suggested H. pylori eradication should be considered
with alarm features underwent prompt endoscopy. No dif- for all patients requiring long-term acid suppression, which
ference in outcome or satisfaction was detected between the seems reasonable (85, 86). An unresolved issue is whether
groups in follow up after referral back to their primary care test-and-treat will widen the problem of community acquired
physician. antibiotic resistance.

Disadvantages of Test-and-Treat
PROMPT ENDOSCOPY
A notable disadvantage of test-and-treat is that cure of H.
pylori infection will only lead to a minority reporting symp- Advantages of Prompt Endoscopy
tom improvement, as demonstrated in the above management There is empiric evidence from a management trial of prompt
trials, and this can be confusing to the clinician (6065). How- endoscopy in older patients that this is the strategy of first
ever, endoscopy and targeted medical therapy does no better. choice. Delaney et al. evaluated the cost-effectiveness of an
Indeed, eradication of H. pylori infection does not relieve initial endoscopy compared with usual management in pa-
symptoms in all patients with peptic ulcer disease, with at tients with dyspepsia over the age of 50 presenting in primary
least one third continuing to be symptomatic (72, 73). care (87). A total of 422 patients were randomly assigned to
The choice of the H. pylori test is critical. Many serologi- either usual care or initial endoscopy; the initial endoscopy
cal tests have not been locally validated, and have suboptimal arm showed significant improvement in symptom scores and
sensitivity and specificity in practice (74). The urea breath test quality of life as well as a 48% reduction in the use of PPIs.
and stool antigen test are currently the most accurate nonin- Hence, initial endoscopy in older patients with dyspepsia at
vasive diagnostic tools and can be used with confidence (75, least in this U.K. study was potentially cost-effective provided
76). The value of noninvasive H. pylori testing, even if a the cost of EGD was low. The cost-effectiveness of endoscopy
local evaluated test is applied, still depends on the positive in older people in the U.S. setting needs investigation.
and negative predictive value, which in turn is related to the There is only limited and unconvincing evidence that
background prevalence of H. pylori infection. When H. py- endoscopy leads to improved patient satisfaction scores in
2330 Talley et al.

dyspepsia. Bytzer el al. conducted a randomized trial com- EMPIRIC ANTISECRETORY THERAPY IN UNINVESTIGATED
paring prompt endoscopy with empiric H2 -receptive blocker DYSPEPSIA
therapy in dyspepsia (88). They found there was significant
improvement in satisfaction scores at one month after en- In H. pylori-negative cases with uninvestigated dyspepsia
doscopy compared to the empiric antisecretory therapy arm. and no alarm features, an empiric trial of acid suppression
In addition, 66% of the patients in the empiric therapy arm for 48 wk is recommended first-line therapy (Fig. 1).
eventually underwent endoscopy during the 12 months of Grade of evidence: A
follow-up. However, this unblinded study may have been bi- If initial acid suppression fails after 24 wk, it is rea-
ased by patient and physician expectation that endoscopy is sonable to step up therapy, although this is based on ex-
the preferred management strategy, and H. pylori status was pert opinion only; this may require changing drug class or
not considered. Other studies have suggested that patients dosing. In the absence of established prokinetic drugs for
with dyspepsia are reassured by EGD and may require fewer dyspepsia in the United States, this drug class is not cur-
prescriptions, although the duration of reassurance is not es- rently recommended as first-line therapy for dyspepsia in
tablished (8991). the United States.
Dyspeptic patients who seek medical attention are more Grade of evidence: C
concerned about the possible seriousness of their symptoms In patients who do respond to initial therapy, it is recom-
and are more likely to be concerned about underlying cancer mended that treatment be stopped after 48 wk and if symp-
(92). Health anxiety has been shown to lead to a cycle of toms recur, another course of the same treatment is justified.
repeated medical consultations. In a study of primary care There are no data on long-term self-directed therapy in this
patients undergoing open-access endoscopy, Hungin et al. condition, although this may be worth considering in some
demonstrated that consultations for dyspepsia fell by 57% patients.
in patients with normal endoscopy and by 37% in patients Grade of evidence: C
with minor abnormalities at endoscopy. In 60% of patients The American College of Physicians in 1985 recom-
with normal endoscopy, medication use was terminated or mended an empiric trial of an H2 receptor antagonist for 68
decreased (93). Quadri and Vakil demonstrated that one third wk; those who relapsed after therapy or those who failed
of patients referred for open-access endoscopy for dyspepsia to respond to therapy in 710 days were to be referred for
in the United States had high levels of health related anxiety; endoscopy (44). The widespread availability of PPIs has re-
following a normal endoscopy or the demonstration of mi- sulted in this class of agents frequently being prescribed as
nor abnormalities, and reassurance by the endoscopist, scales initial empiric therapy in uninvestigated dyspepsia in place
for preoccupation with health and fear of illness and death of H2 receptor antagonists (98).
showed significant improvement after endoscopy, and the ef- A meta-analysis of several large studies has demonstrated
fects were preserved for 6 months (86). a short course of PPI therapy compared with a H2 -receptor
antagonist, alginate, or placebo in primary care provides
better symptomatic outcomes (70). However, these stud-
Disadvantages of Endoscopy ies frequently included patients with symptomatic reflux
There are several potential disadvantages of prompt en- disease and did not exclude peptic ulcer. It is unknown
doscopy for all dyspeptic patients that need to be carefully whether GERD or ulcer disease, or both, accounts for the
considered. Endoscopy is invasive and although the risks of apparent short-term benefits of empiric therapy in these
this procedure in relatively healthy patients are very low, the reports.
issue of the risk-benefit ratio needs careful weighing, par- There are limited data that prokinetic therapy employed as
ticularly as the procedure is very unlikely to identify an un- an empiric strategy may be efficacious in uninvestigated dys-
expected structural cause in a young patient with no alarm pepsia. Kearney et al. noted no significant difference in the
features. Finding esophagitis, the most likely structural ab- severity of dyspeptic symptoms among 60 patients random-
normality, may often not lead to a change in management ized to receive cisapride as compared to placebo in the setting
(94, 95). Moreover, the high prevalence of dyspepsia means of uninvestigated dyspepsia and negative H. pylori-serology
that a general recommendation to perform endoscopies on (99). Quartero et al. conducted a trial in primary care of
all patients would be very costly and would overwhelm en- 563 patients who were randomized to ranitidine or cisapride;
doscopy services. Furthermore, it is contentious that prompt treatment success was similar in both groups but was un-
EGD provides any direct benefits despite some positive stud- der 50%, and the relapse-free periods were also similar with
ies quoted above. One study evaluated management strategies both drugs (100). A randomized trial in H. pylori-negative
in 326 primary care patients with dyspepsia; endoscopy was dyspepsia from Canada demonstrated that cisapride had low
not superior to any of the empirical treatment strategies uti- efficacy and was inferior to acid suppression (101). More-
lized in this study (96). A systematic review concluded that over, cisapride is no longer available because of rare toxicity
most data failed to support the view that endoscopy alone from QTC prolongation and sudden death. There have been
improves patient outcome in dyspepsia compared with other no trials of metoclopramide, tegaserod or domperidone in the
empiric strategies (97). management of uninvestigated dyspepsia.
 Guidelines for the Management of Dyspepsia 2331

Obvious disadvantages of empiric antisecretory therapy this model, the most costly approach was test-and-treat fol-
include the concern that peptic ulcer disease will be inappro- lowed by endoscopy for failures. This model also suggested
priately and inadequately treated, and patients subsequently that empirical PPI therapy became cost-effective if the preva-
may present with complicated ulcer disease if for any reason lence of H. pylori infection was 12% or less in the dyspeptic
the therapy is ceased. Antisecretory therapy can also lead to population. Ladabaum et al. observed that as the likelihood
misdiagnosis of peptic ulcer disease at subsequent endoscopy, of H. pylori (and ulcer disease) decreases below 20%, em-
as the ulcer will more likely heal and be missed. The impact piric PPI therapy starts to dominate test-and-treat in unin-
of acid rebound in dyspepsia remains unclear (102). Em- vestigated dyspepsia (60). Therefore, recommendations for
piric antisecretory therapy may lead to long-term inappropri- the test-and-treat strategy may need to be modified when the
ate maintenance therapy that the patient does not require. It prevalence of H. pylori infection is low, and we recommend
is unclear whether antisecretory therapy postpones eventual on the basis of expert opinion considering a PPI in the setting
investigation or not, which in turn impacts on its potential of a H. pylori prevalence below 10% in the local commu-
cost-effectiveness. nity. A recent systematic review and economic analysis using
generic/over-the-counter costs for PPIs found that they were
H. pylori TEST-AND-TREAT VERSUS EMPIRIC cost-effective in the United States provided generic costs of a
ANTISECRETORY THERAPY PPI were used in the analysis (108). Upper GI radiology was
not a cost-effective alternative to H. pylori test-and-treat in
There are only very limited data comparing empiric H. pylori
another U.S. model (109).
treatment versus empiric PPI therapy. Manes et al. compared
test-and-treat with PPI therapy for a month with 12 months
of follow-up in a secondary care setting in Italy (103). In the WEIGHING THE OPTIONS
test-and-treat arm, 56% were eventually endoscoped because
of poor symptom control, but none had a peptic ulcer; in A Cochrane review has been conducted of available manage-
the PPI arm, 88% were endoscoped and 17% had a peptic ment strategies for dyspepsia (70). They identified 18 pub-
ulcer, but most (88%) were infected with H. pylori. More lished papers that had 20 comparisons included. In a pooled
studies are needed, but these data suggest that in H. pylori- analysis, PPIs were significantly more effective than both
positive dyspeptic patients, empiric PPI therapy is not the H2 receptor antagonists and antacids in uninvestigated dys-
management option of choice in areas where the prevalence pepsia. A significant limitation of the studies is that they
of H. pylori is high. included broad groups of patients including those with obvi-
ous reflux disease. There was insufficient data to determine
ECONOMIC MODELS OF DYSPEPSIA MANAGEMENT whether empiric prokinetic therapy was beneficial. They also
concluded a H. pylori test-and-treat strategy may be as effec-
Fendrick et al. undertook economic modeling of manage- tive as endoscopy-based management with reduced costs be-
ment strategies in patient with suspected peptic ulcer disease, cause of the decreased numbers of patients that subsequently
which presumably applies to the majority of patients with require EGD, but it was unclear whether test-and-treat com-
uninvestigated dyspepsia (104). They found that an initial pared to empirical acid suppression was equivalent or not
strategy of H. pylori testing and treatment was cost-effective, because of the lack of data.
unless the cost of endoscopy fell to less than $500 when
prompt endoscopy became more cost-effective. Sonnenberg ENDOSCOPY-NEGATIVE DYSPEPSIA (FUNCTIONAL
noted that if the ulcer disease prevalence rate exceeded 10%
DYSPEPSIA, NONULCER DYSPEPSIA)
in H. pylori-infected subjects, then a noninvasive strategy
based on serological testing became cost-effective (105). Sil- The management of endoscopy-proven functional dyspep-
verstein el al. concluded that there was a toss up between sia is particularly challenging when initial antisecretory
H. pylori test-and-treat compared with other strategies, but therapy and H. pylori eradication fails. Patients who fail
reevaluation of this model applying the assumptions made by to respond to simple measures need to have their diagnosis
Fendrick et al. confirmed their results, supporting test-and- reconsidered. Dietary therapy has no established efficacy
treat (106). Ofman et al. concluded that test-and-treat was but may help some individuals. There are very limited data
cost-saving; the cost of endoscopy would need to drop from to support the use of herbal preparations, simethicone, and
$740 by 96% for an initial endoscopy strategy to become low-dose tricyclic antidepressants in functional dyspepsia.
equally cost-effective in their model (107). Bismuth, sucralfate, and antispasmodics are not established
Spiegel et al. tested four different management strategies in to be of benefit over placebo in functional dyspepsia. Hyp-
decision analysis (59). This analysis was confined to patients notherapy, psychotherapy, and cognitive-behavioral ther-
younger than 45 yr of age presenting in primary care. They apy are supported by limited studies but cannot be generally
identified initial antisecretory therapy followed by endoscopy recommended at the present time.
as the least costly therapy per patient treated. However, this Grades of evidence:
rendered fewer patients symptom-free at 1 yr than strategies Dietary modification: C
which combined empiric PPI therapy with test-and-treat. In Simethicone: B
2332 Talley et al.

Hypnotherapy, Psychotherapy, Cognitive-behavioral ication in functional dyspepsia, with the number needed to
therapy: B treat being 15 (118). While longer than 1-yr follow-up data
are generally lacking, one 5-yr study suggests any benefit
MANAGEMENT OF DOCUMENTED will persist (119). On the basis of the evidence, it is accept-
able to offer H. pylori eradication therapy to infected pa-
FUNCTIONAL DYSPEPSIA
tients with functional dyspepsia. The results also imply that
Once a diagnosis of functional dyspepsia is confirmed by a offering H. pylori eradication therapy empirically to those
negative endoscopy, an empiric trial of therapy is commonly with otherwise uninvestigated dyspepsia who are infected
prescribed. However, the benefits of all therapies in this con- is reasonable even if ulcer disease is unlikely. Moreover,
dition have been questioned. H. pylori eradication in those with documented functional
Many patients do not require medication for dyspepsia dyspepsia may help prevent ulcer disease, although convinc-
after they have had reassurance and education. It is there- ing evidence is not available. Hsu et al. observed during 1
fore important for the clinician to explain the meaning of yr of follow-up in a randomized controlled trial compris-
the symptoms and their benign nature. Ascertaining why a ing 161 patients with functional dyspepsia, 2 patients in the
patient with long-standing symptoms has presented on this H. pylori eradication treatment group (3%) and 6 patients
occasion for care can be helpful, as this may identify those in the placebo group (8%) developed peptic ulcers at repeat
who have fears of an underlying serious disease or specific endoscopy (120).
psychological distress that can be addressed. Potential precip- The benefit of other treatments remains uncertain. A
itating factors in dyspepsia remain poorly defined. High-fat Cochrane review included 12 trials with prokinetics com-
meals should be avoided; eating frequent and smaller meals prising 829 patients and showed that there was a relative
throughout the day can sometimes be helpful. Specific foods risk reduction of 50%, compared with placebo, but most of
that precipitate symptoms can be avoided. Food intolerance is the studies were with cisapride (98). Moreover, analysis of
uncommon, however, and food allergy very rare. Follow-up the studies suggested that publication bias at least partly ex-
of the patient helps determine the natural history and allows plains the apparent benefits of prokinetic therapy. Prokinetics
further correction of faulty ideas and provides reassurance should be reserved for difficult cases as options in the United
that can be very helpful in long-term management. States are few and current agents (e.g., metoclopramide, ery-
Antacids and sucralfate were not superior to placebo in thromycin, tegaserod) have limited or poorly established ef-
functional dyspepsia based on a Cochrane review (98). How- ficacy, or side-effects are common (121). Routine use of gas-
ever, a recent trial of simethicone has suggested potential tric emptying studies is not recommended as improvements
benefit compared with placebo, and in another study equiv- in gastric emptying do not correlate well with symptom im-
alence with cisapride (110, 111). A Cochrane review of 8 provement (31, 122). Drugs that relax the gastric fundus (e.g.,
trials of H2 receptor antagonists with 1,125 patients showed tegaserod, cisapride, sumatriptan, buspirone, clonidine, some
a relative risk reduction of 30% but the quality of the trials SSRIs, nitric oxide donors) may theoretically improve some
was generally poor (98). PPIs in this review also produced a dysmotility-like dyspepsia (e.g., early satiety) but adequate
relative risk reduction of approximately 30% and the quality randomized controlled trials are lacking (123). Antidepres-
of the trials was better (98). An economic model suggested sants are also of uncertain efficacy in functional dyspepsia but
that PPI therapy was cost-effective for functional dyspepsia in are often prescribed (121, 124). There are insufficient data on
the United States (108). However, in a recent randomized trial the use of tricyclic antidepressants such as amitryptyline in
of 453 patients from Hong Kong, the proportion of patients dyspepsia, but small studies have suggested benefit; how-
achieving complete relief of dyspepsia with lansoprazole 30 ever, the beneficial effect of low-dose amitryptyline seen in
and 60 mg was 23% and 23%, respectively, compared with functional dyspepsia was not related to changes in perception
30% on placebo (112). In contrast, another recent trial re- of gastric distension (125). An increased tolerance to aver-
ported significant benefit with lansoprazole in a U.S. popula- sive visceral sensations may play a role in the therapeutic
tion (113). H. pylori status is unlikely to affect the therapeutic effect. There are limited data with the SSRIs. Psychological
outcome of acid suppression therapy in functional dyspepsia therapies are promising, particularly hypnotherapy, but more
(108). Large trials have failed to identify any difference in data are needed in larger patient populations before these can
therapeutic outcome in H. pylori-positive versus negative pa- be recommended for routine use (126, 127). Other alterna-
tients, although Blum et al. did identify a superior response tive therapies such as herbal preparations remain of unproven
to PPI therapy in H. pylori-positive patients (114, 115). value (128, 129).
Eradication of H. pylori in functional dyspepsia is contro-
versial. Two high-quality meta-analyses have reached differ-
ent conclusions but this may be likely explained by which ADDITIONAL DIAGNOSES AND TESTING
trials were included and excluded in each systematic review
IN REFRACTORY CASES
(116, 117). Updating these meta-analyses now suggests that
when all appropriate trials are considered, there is a small In patients with resistant symptoms, it is worth reevaluating
but significant therapeutic gain achieved with H. pylori erad- the diagnosis.
 Guidelines for the Management of Dyspepsia 2333

Grade of evidence: C
Abdominal wall pain can be confused with functional dys- Functional Bowel and
pepsia; physical examination here is diagnostic (increased Lower Gut GI Motility
rather than reduced tenderness on tensing the abdominal wall Chair: Tim Koch, M.D. Chair: Henry Parkman, M.D.
muscles) (130). Biliary pain is characteristic and different Sub-Chair: Steven Sub-Chair: Lin Chang, M.D.
from dyspepsia; ultrasound usually is unhelpful in the ab- Edmundowicz, M.D.
sence of typical biliary pain. Exclusion of atypical GERD Alvin Zfass, M.D. Charlene Prather, M.D.
Darren Baroni, M.D. Adil E. Bharucha, M.D.
with esophageal pH testing may alter management; at least Subbaramiah Sridhar, M.D.
20% of patients with diagnosed functional dyspepsia clini-
cally turn out to have GERD on esophageal pH studies (23
25, 131). Thus, even if a trial of PPI therapy has failed, pH Reprint requests and correspondence: Nicholas J. Talley, M.D.,
Ph.D., Mayo Clinic College of Medicine, 200 First Street S.W., PL-
testing may be considered off therapy, although the yield in 656, Rochester, MN 55905.
this particular setting is not defined. Abdominal imaging to Received February 18, 2005; accepted May 23, 2005.
rule out chronic pancreatitis or small bowel pathology may
be worth considering too but usually has a low yield; cap-
sule endoscopy does not yet have an established role here. REFERENCES
Gastric function testing (gastric emptying; gastric accommo-
1. Drossman DA, Corrazziari E, Talley NJ, et al., Rome II:
dation; response to a nutrient or water load) may not change The functional gastrointestinal disorders. 2nd Ed. McLean:
management even if abnormalities are detected, although, if Degnon, 2000.
there is gastric dysaccomodation, trials of various drugs to 2. Bytzer P, Talley NJ. Dyspepsia. Ann Intern Med
relax the fundus may be worth trying empirically (123, 132). 2001;134:81522.
Questions about symptoms consistent with IBS may lead to 3. Anonymous. An evidence-based appraisal of reflux disease
management-the Genval Workshop Report. Gut 1999;44
a change in the diagnosis. Colonic evaluation may be consid- (Suppl 2):S16.
ered even if there are no bowel disturbances because disease 4. Veldhuyzen van Zanten S, Flook N, Chiba N, et al. An
in the transverse colon or elsewhere can occasionally present evidence-based approach to the management of uninvesti-
with referred symptoms labeled dyspepsia. A drug history gated dyspepsia in the era of Helicobacter pylori. CMAJ
is helpful but aside from NSAIDs, drugs are rarely major 2000;162(Suppl 12):S323.
5. Moayyedi P, Axon AT. The usefulness of the likelihood
contributors to chronic dyspepsia according to the available ratio in the diagnosis of dyspepsia and gastroesophageal
evidence (133). Diabetic radiculopathy can cause upper ab- reflux disease. Am J Gastroenterol 1999;94:31225.
dominal pain and EMG is diagnostic. Evaluation for referred 6. Ofman JJ, Shaw M, Sadik K, et al. Identifying patients with
pain from the chest or back should be considered in diffi- gastroesophageal reflux disease: Validation of a practical
cult cases. Finally, consider looking for rare metabolic or screening tool. Dig Dis Sci 2002;47:18639.
7. Talley NJ, Zinsmeister AR, Schleck CD, et al. Dyspep-
other causes of upper abdominal pain including thyroid dis- sia and dyspepsia subgroups: A population-based study.
ease, electrolyte abnormalities, hypercalcemia, heavy metals, Gastroenterology 1992;102(4 Pt 1):125968.
acute intermittent porphyria, angioneurotic edema, famil- 8. Moayyedi P, Forman D, Braunholtz D, et al. The propor-
ial mediterranean fever, chronic intestinal angina, superior tion of upper gastrointestinal symptoms in the community
mesenteric artery syndrome, liver disease (hepatoma, steato- associated with Helicobacter pylori, lifestyle factors, and
nonsteroidal anti-inflammatory drugs. Leeds HELP Study
hepatitis), eosinophilic gastroenteritis, or connective tissue Group. Am J Gastroenterol 2000;95:144855.
disease. 9. Talley N, Weaver A, Zinsmeister A, et al. Onset and disap-
pearance of gastrointestinal symptoms and functional gas-
trointestinal disorders. Am J Epidemiol 1992;136:16577.
10. Agreus L, Svardsudd K, Nyren O, et al. Irritable bowel
APPENDIX syndrome and dyspepsia in theh general population:
Overlap and lack of stability over time. Gastroenterology
ACG Practice Parameter Committee 1995;109:67180.
Chair: Ronnie Fass, M.D. 11. Agreus L, Borgquist L. The cost of gastro-oesophageal re-
flux disease, dyspepsia and peptic ulcer disease in Sweden.
Pharmacoeconomics 2002;20:34755.
12. Quartero AO, Numans ME, Post MWM, et al. One-year
Upper Gut Liver & Pancreas prognosis of primary care dyspepsia: Predictive value of
symptom pattern, Helicobacter pylori and GP manage-
Chair: William D. Chey, M.D. Chair: Daniel Pratt, M.D. ment. Eur J Gastroenterol Hepatol 2002;14:5560.
Sub-Chair: Richard Sub-Chair: John 13. Johnsen R, Bernersen B, Straume B, et al. Prevalences of
Sampliner, M.D. Cunningham, M.D. endoscopic and histological findings in subjects with and
Ece A. Mutlu, M.D. William Brugge, M.D. without dyspepsia. BMJ 1991;302:74952.
Nimish Vakil, M.D. William Carey, M.D. 14. Aro P, Ronkainen J, Storskrubb T, et al. Findings at upper
Miguel A. Valdovinos, M.D. Matthew Cohen, M.D. endoscopy in a random adult population. Gastroenterology
Benjamin Wong, M.D. David Bernstein, M.D. 2002;122(Suppl 1):A568.
2334 Talley et al.

15. Thomson A, Barkun A, Armstrong D, et al. The preva- 33. Holtmann G, Goebell H, Jockenhoevel F, et al. Altered
lence of clinically significant endoscopic findings in pri- vagal and intestinal mechanosensory function in chronic
mary care patients with uninvestigated dyspepsia: The unexplained dyspepsia. Gut 1998;42(4):5016.
Canadian Adult Dyspepsia Empiric treatment-prompt en- 34. Caldarella MP, Azpiroz F, Malagelada JR. Antro-fundic
doscopy (CADET-PE) study. Aliment Pharmacol Ther dysfunctions in functional dyspepsia. Gastroenterology
2003;17:148191. 2003;124(5):12209.
16. Voutilainen M, Mantynen T, Kunnamo I, et al. Impact of 35. Tack J, Caenepeel P, Fischler B, et al. Symptoms associ-
clinical symptoms and referral volume on endoscopy for ated with hypersensitivity to gastric distention in functional
detecting peptic ulcer and gastric neoplasma. Scand J Gas- dyspepsia. Gastroenterology 2001;121:52635.
troenterol 2003;38:10913. 36. Trimble KC, Farouk R, Pryde A, et al. Heightened vis-
17. Westbrook JI, Talley NJ. Diagnostic investigation rates ceral sensation in functional gastrointestinal disease is not
and use of prescription and non-prescription medica- site-specific. Evidence for a generalized disorder of gut
tions amongst dyspeptics: A population-based study of sensitivity. Dig Dis Sci 1995;40:160713.
2300 Australians. Aliment Pharmacol Ther 2003;17:1171 37. Tack J, Piessevaux H, Coulie B, et al. Role of impaired
8. gastric accommodation to a meal in functional dyspepsia.
18. Heikkinen MT, Pikkarainen PH, Takala JK, et al. Diagnos- Gastroenterology 1998;115:134652.
tic methods in dyspepsia: the usefulness of upper abdomi- 38. Boeckxstaens G, Hirsch D, Kuiken S, et al. The proximal
nal ultrasound and gastroscopy. Scand J Prim Health Care stomach and postprandial symptoms in functional dyspep-
1997;15:826. tics. Am J Gastroenterol 2002;97:408.
19. Berstad A, Hausken T, Gilja OH, et al. Imaging studies in 39. Bennink RJ, van den Elzen BD, Kuiken SD, et al. Nonin-
dyspepsia. Eur J Surg Suppl 1998;(582):429. vasive measurement of gastric accommodation by means
20. Lee YT, Lai AC, Hui Y, et al. EUS in the manage- of pertechnetate SPECT: Limiting radiation dose without
ment of uninvestigated dyspepsia. Gastrointest Endosc losing image quality. J Nucl Med 2004;45:14752.
2002;56:8428. 40. Jones MP, Hoffman S, Shah D, et al. The water load test:
21. Sahai AV, Mishra G, Penman ID, et al. EUS to de- Observations from healthy controls and patients with func-
tect evidence of pancreatic disease in patients with per- tional dyspepsia. Am J Physiol Gastrointest Liver Physiol
sistent or nonspecific dyspepsia. Gastrointest Endosc 2003;284:G896904.
2000;52(2):1539. 41. Boeckxstaens GE, Hirsch DP, van den Elzen BD, et al.
22. Klauser AG, Voderholzer WA, Knesewitsch PA, et al. What Impaired drinking capacity in patients with functional dys-
is behind dyspepsia? Dig Dis Sci 1993;38:14754. pepsia: Relationship with proximal stomach function. Gas-
23. Farup PG, Hovde O, Torp R, et al. Patients with functional troenterology 2001;121(5):105463.
dyspepsia responding to omeprazole have a characteristic 42. Tack J, Caenepeel P, Piessevaux H, et al. Assessment of
gastro-oesophageal reflux pattern. Scand J Gastroenterol meal induced gastric accommodation by a satiety drink-
1999;34:5759. ing test in health and in severe functional dyspepsia. Gut
24. Small PK, Loudon MA, Waldron B, et al. Impor- 2003;52(9):12717.
tance of reflux symptoms in functional dyspepsia. Gut 43. Bytzer P, Hansen JM, Schaffalitzky de Muckadell OB, et
1995;36(2):18992. al. Predicting endoscopic diagnosis in the dyspeptic pa-
25. Quigley EM. Non-erosive reflux disease: Part of the spec- tient. The value of predictive score models. Scand J Gas-
trum of gastro-oesophageal reflux disease, a component of troenterol 1997;32:11825.
functional dyspepsia, or both? Eur J Gastroenterol Hepatol 44. Anonymous. Endoscopy in the evaluation of dyspepsia.
2001;13(Suppl 1):S138. Health and Public Policy Committee, American College of
26. Malfertheiner P, Megraud F, OMorain C, et al. Cur- Physicians. Ann Intern Med 1985;102:2669.
rent concepts in the management of Helicobacter pylori 45. Gillen D, McColl KE. Does concern about missing ma-
infectionThe Maastricht 2-2000 Consensus Report. Al- lignancy justify endoscopy in uncomplicated dyspep-
iment Pharmacol Ther 2002;16:16780. sia in patients aged less than 55? Am J Gastroenterol
27. Quartero AO, de Wit NJ, Lodder AC, et al. Disturbed solid- 1999;94(1):759.
phase gastric emptying in functional dyspepsia: A meta- 46. Breslin NP, Thomson A, Bailey R, et al. Gastric cancer
analysis. Dig Dis Sci 1998;43(9):202833. and other endoscopic diagnoses in patients with benign
28. Stanghellini V, Tosetti C, Paternico A, et al. Risk indica- dyspepsia. Gut 2000;46:937.
tors of delayed gastric emptying of solids in patients with 47. Canga Cr, Vakil N. Upper GI malignancy, uncomplicated
functional dyspepsia. Gastroenterology 1996;110:1036 dyspepsia, and the age threshold for early endoscopy. Am
42. J Gastroenterol 2002;97(3):6003.
29. Stanghellini V, Tosetti C, Paternico A, et al. Predominant 48. Hammer J, Eslick G, Howell S, et al. Diagnostic yield of
symptoms identify different subgroups in functional dys- alarm features in irritable bowel syndrome and functional
pepsia. Am J Gastroenterol 1999;94(8):20805. dyspepsia. Gut 2004;53:66672.
30. Sarnelli G, Caenepeel P, Geypens B, et al. Symp- 49. Bramble MG, Suvakovic Z, Hungin AP. Detection of up-
toms associated with impaired gastric emptying of solids per gastrointestinal cancer in patients taking antisecretory
and liquids in functional dyspepsia. Am J Gastroenterol therapy prior to gastroscopy. Gut 2000;6:4647.
2003;98(4):7838. 50. Panter SJ, OFlanagan H, Bramble MG, et al. Empirical use
31. Talley NJ, Verlinden M, Jones M. Can symptoms discrim- of antisecretory drug therapy delays diagnosis of upper gas-
inate among those with delayed or normal gastric emp- trointestinal adenocarcinoma but does not effect outcome.
tying in dysmotility-like dyspepsia? Am J Gastroenterol Aliment Pharmacol Ther 2004;19:9818.
2001;96(5):14228. 51. Paltoo DN, Chu KC. Patterns in cancer incidence among
32. Holtmann G, Gschossmann J, Neufang-Huber J, et al. Dif- American Indians/Alaska Natives, United States, 1992
ferences in gastric mechanosensory function after repeated 1999. Public Health Rep 2004;119:44351.
ramp distensions in non-consulters with dyspepsia and 52. Howell S, Talley NJ. Does fear of serious disease
healthy controls. Gut. 2000;47(3):3326. predict consulting behaviour amongst patients with
 Guidelines for the Management of Dyspepsia 2335

dyspepsia in general practice? Eur J Gastroenterol Hep- 69. Arents NL, Thijs JC, van Zwet AA, et al. Approach to
atol 1999;11:8816. treatment of dyspepsia in primary care: A randomized trial
53. Koloski NA, Talley NJ, Huskic SS, et al. Predictors of con- comparing test-and-treat with prompt endoscopy. Arch
ventional and alternative health care seeking for irritable Intern Med 2003;163:160612.
bowel syndrome and functional dyspepsia. Aliment Phar- 70. Delaney BC, Moayyedi P, Forman D. Initial management
macol Ther 2003;17:84151. strategies for dyspepsia. Cochrane Database of Syst Rev
54. Talley NJ. Dyspepsia management in the millen- 2003;2:CD001961.
nium: The death of test and treat? Gastroenterology 71. Patel P, Khulusi S, Mendall MA, et al. Prospective screen-
2002;122(5):15215. ing of dyspeptic patients by Helicobacter pylori serology.
55. Talley NJ, Axon AT, Bytzer P, et al. Management of unin- Lancet 1995;346:13158.
vestigated and functional dyspepsia: A Working Party re- 72. Forbes GM, Glaser ME, Cullen DJ, et al. Duodenal ulcer
port for the World Congresses of Gastroenterology 1998. treated with Helicobacter pylori eradication: Seven-year
Aliment Pharmacol Ther 1999;13(9):113548. follow-up. Lancet 1994;343:258.
56. McColl KE, el-Nujumi A, Murray L, et al. The Helicobac- 73. Laine L, Hopkins RJ, Girardi LS. Has the impact of Heli-
ter pylori breath test: A surrogate marker for peptic ulcer cobacter pylori therapy on ulcer recurrence in the United
disease in dyspeptic patients. Gut 1997;40(3):3026. States been overstated? A meta-analysis of rigorously de-
57. Chiorean MV, Locke GR, Zinsmeister AR, et al. Chang- signed trials. Am J Gastroenterol 1998;98:140915.
ing rates of Helicobacter pylori testing and treatment in 74. Loy CT, Irwig LM, Katelaris PH, et al. Do commer-
patients with peptic ulcer disease. Am J Gastroenterol cial serological kits for Helicobacter pylori infection dif-
2002;97:301522. fer in accuracy? A meta-analysis. Am J Gastroenterol
58. Ciociola AA, McSorley DJ, Turner K, et al. Helicobac- 1996;91:113844.
ter pylori infection rates in duodenal ulcer patients in the 75. Vaira D, Vakil N, Menegatti M, et al. The stool antigen
United States may be lower than previously estimated. Am test for detection of Helicobacter pylori after eradication
J Gastroenterol 1999;94:183440. therapy. Ann Intern Med 2002;136:2807.
59. Spiegel BM, Vakil NB, Ofman JJ. Dyspepsia management 76. Vaira D, Vakil N. Blood, urine, stool, breath, money, and
in primary care: A decision analysis of competing strate- Helicobacter pylori. Gut 2001;48:2879.
gies. Gastroenterology 2002;122(5):127085. 77. Fletcher RH, Fletcher SW, Wagner EH, Clinical epidemi-
60. Ladabaum U, Chey WD, Scheiman JM, et al. Reappraisal ology: The essentials. 3rd Ed. Baltimore: Williams &
of non-invasive management strategies for uninvestigated Wilkins, 1996.
dyspepsia: A cost-minimization analysis. Aliment Phar- 78. Vakil N, Rhew D, Soll A, et al. The cost-effectiveness of
macol Ther 2002;16:1491501. diagnostic testing strategies for Helicobacter pylori. Am J
61. Moayyedi P, Feltbower R, Brown J, et al. Effect of pop- Gastroenterol 2000;95:16918.
ulation screening and treatment for Helicobacter pylori 79. Chey WD, Fendrick AM. Noninvasive Helicobacter pylori
on dyspepsia and quality of life in the community: A ran- testing for the test-and-treat strategy: A decision analysis
domised controlled trial. Leeds HELP Study Group. Lancet to assess the effect of past infection on test choice. Arch
2000;355:16659. Intern Med 2001;161:212932.
62. Chiba N, Van Zanten SJ, Sinclair P, et al. Treating He- 80. Vakil N, Lanza F, Schwartz H, et al. Seven-day therapy for
licobacter pylori infection in primary care patients with Helicobacter pylori in the United States. Aliment Pharma-
uninvestigated dyspepsia: The Canadian adult dyspepsia col Ther 2004;20:99107.
empiric treatmentHelicobacter pylori positive (CADET- 81. Laine L, Hunt R, El-Zimaity H, et al. Bismuth-based
Hp) randomised controlled trial. BMJ 2002;324:10126. quadruple therapy using a single capsule of bismuth
63. Allison JE, Hurley LB, Hiatt RA, et al. A randomized con- biskalcitrate, metronidazole, and tetracycline given with
trolled trial of test-and-treat strategy for Helicobacter py- omeprazole versus omeprazole, amoxicillin, and clar-
lori: Clinical outcomes and health care costs in a man- ithromycin for eradication of Helicobacter pylori in duode-
aged care population receiving long-term acid suppression nal ulcer patients: A prospective, randomized, multicenter,
therapy for physician-diagnosed peptic ulcer disease. Arch North American trial. Am J Gastroenterol 2003;98:562
Intern Med 2003;163:116571. 7.
64. Ladabaum U, Fendrick AM, Glidden D, et al. Helicobacter 82. Sharma P, Vakil N. Helicobacter pylori and reflux disease.
pylori test-and-treat intervention compared to usual care in Aliment Pharmacol Ther 2003;17:297305.
primary care patients with suspected peptic ulcer disease in 83. Dent J. Review article: Is Helicobacter pylori relevant in
the United States. Am J Gastroenterol 2002;97:300714. the management of reflux disease? Aliment Pharmacol
65. Lassen AT, Pedersen FM, Bytzer P, et al. Helicobacter py- Ther 2001;15(Suppl 1):1621.
lori test-and-eradicate versus prompt endoscopy for man- 84. Talley NJ, Vakil N, Ballard C, et al. Effect of eradicating
agement of dyspeptic patients: A randomised trial. Lancet Helicobacter pylori in patients with non-ulcer dyspepsia.
2000;356:45560. N Engl J Med 1999;341:110611.
66. Heaney A, Collins JS, Watson RG, et al. A prospective ran- 85. Kuipers EJ, Uyterlinde AM, Pena AS, et al. Increase of
domised trial of a test and treat policy versus endoscopy Helicobacter pylori-associated corpus gastritis during acid
based management in young Helicobacter pylori positive suppressive therapy: Implications for long-term safety. Am
patients with ulcer-like dyspepsia, referred to a hospital J Gastroenterol 1995;90:14016.
clinic. Gut 1999;45:18690. 86. Graham DY, Opekun AR, Yamaoka Y, et al. Early events
67. McColl KE, Murray LS, Gillen D, et al. Randomised trial of in proton pump inhibitor-associated exacerbation of corpus
endoscopy with testing for Helicobacter pylori compared gastritis. Aliment Pharmacol Ther 2003;17:193200.
with non-invasive H. pylori testing alone in the manage- 87. Delaney BC, Wilson S, Roalfe A, et al. Cost effective-
ment of dyspepsia. BMJ 2002;324:9991002. ness of initial endoscopy for dyspepsia in patients over age
68. Jones RJ, Tait C, Sladen G, et al. A trial of a test-and-treat 50 years: A randomised controlled trial in primary care.
strategy for Helicobacter pylori positive dyspepsia patients Lancet 2000;356:19659.
in general practice. Int J Clin Pract 1999;53:4136. 88. Bytzer P, Hansen JM, Schaffalitzky de Muckadell OB.
2336 Talley et al.

Empirical H2-blocker therapy or prompt endoscopy in with dyspepsia: Clinical and economic consequences. Ann
management of dyspepsia. Lancet 1994;343:8116. Intern Med 1997;126:28091.
89. Wiklund I, Glise H, Jerndal P, et al. Does endoscopy have a 108. Moayyedi P, Delaney B, Vakil N, et al. The efficacy of
positive impact on quality of life in dyspepsia? Gastrointest proton pump inhibitors in non-ulcer dyspepsia: A sys-
Endosc 1998;47:44954. tematic review and economic analysis. Gastroenterology
90. Quadri A, Vakil N. Health-related anxiety and the effect 2004;127:132937.
of open-access endoscopy in US patients with dyspepsia. 109. Rich M, Scheiman JM, Tierney W, et al. Is upper gas-
Aliment Pharmacol Ther 2003;17:83540. trointestinal radiography a cost-effective alternative to a
91. Rabeneck L, Wristers K, Souchek J, et al. Impact of upper Helicobacter pylori test and treat strategy for patients
endoscopy on satisfaction in patients with previously unin- with suspected peptic ulcer disease? Am J Gastroenterol
vestigated dyspepsia. Gastrointest Endosc 2003;57:2959. 2000;95:6518.
92. Lydeard S, Jones R. Factors affecting the decision to con- 110. Holtmann G, Gschossmann J, Mayr P, et al. A randomised
sult with dyspepsia: Comparison of consulters and non- placebo-controlled trial of simethicone and cisapride for
consulters. J R Coll Gen Pract 1989;39:4958. the treatment of patients with functional dyspepsia. Ali-
93. Hungin A, Thomas P, Bramble M, et al. What happens to ment Pharmacol Ther 2002;16:16418.
patients following open access gastroscopy? An outcone 111. Holtmann G, Gschossmann J, Karaus M, et al. Ran-
study from general practice. Br J Gen Pract 1994;44:519 domised double-blind comparison of simethicone with cis-
21. apride in functional dyspepsia. Aliment Pharmacol Ther
94. Blustein PK, Beck PL, Meddings JB, et al. The util- 1999;13:145965.
ity of endoscopy in the management of patients with 112. Wong WM, Wong BC, Hung WK, et al. Double blind, ran-
gastroesophageal reflux symptoms. Am J Gastroenterol domised, placebo controlled study of four weeks of lanso-
1998;93:250812. prazole for the treatment of functional dyspepsia in Chinese
95. Talley NJ. Yield of endoscopy in dyspepsia and concurrent patients. Gut 2002;51:5026.
treatment with proton pump inhibitors: The blind leading 113. Peura DA, Kovacs TOG, Metz DC, et al. Lansoprazole
the blind? Gastrointest Endosc 2003;58:8992. in the treatment of functional dyspepsia: Two double
96. Lewin van den Broek NT, Numans ME, Buskens E, et al. A blind, randomized, placebo-controlled trials. Am J Med
randomised controlled trial of four management strategies 2004;116:7408.
for dyspepsia: Relationships between symptom subgroups 114. Talley NJ, Lauritsen K. The potential role of acid suppres-
and strategy outcome. Br J Gen Pract 2001;51:61924. sion in functional dyspepsia: The BOND, OPERA, PILOT,
97. Ofman JJ, Radbeneck L. The effectiveness of endoscopy and ENCORE studies. Gut 2002;50(Suppl 4):iv3641.
in the management of dyspepsia: A qualitative systematic 115. Blum AL, Arnold R, Stolte M, et al. Short course acid
review. Am J Medicine 1999;106:33546. suppressive treatment for patients with functional dyspep-
98. Moayyedi P, Soo S, Deeks J, et al. Pharmacological inter- sia: Results depend on Helicobacter pylori status. Gut
ventions for non-ulcer dyspepsia. Cochrane Database of 2000;47:47380.
Syst Rev 2003;1:CD001960. 116. Moayyedi P, Soo S, Deeks J, et al. Eradication of
99. Kearney DJ, Avins AL, McQuaid KR. Treatment of unin- Helicobacter pylori for non-ulcer dyspepsia. Cochrane
vestigated dyspepsia with cisapride for patients with neg- Database of Syst Rev 2003;1:CD002096.
ative Helicobacter pylori serologies. Am J Gastroenterol 117. Laine L, Schoenfeld P, Fennerty MB. Therapy for He-
2000;95:22127. licobacter pylori in patients with nonulcer dyspepsia. A
100. Quartero AO, Numans ME, de Melker RA, et al. Dyspepsia meta-analysis of randomized, controlled trials. Ann Intern
in primary care: Acid suppression as effective as prokinetic Med 2001;134:3619.
therapy. A randomized clinical trial. Scand J Gastroenterol 118. Moayyedi P, Deeks J, Talley NJ, et al. An update of the
2001;36:9427. Cochrane systematic review of Helicobacter pylori erad-
101. Van Zanten SJ, Chiba N, Armstrong D, et al. A random- ication therapy in nonulcer dyspepsia: Resolving the dis-
ized trial comparing omeprazole, ranitidine, cisapride, or crepany between systematic reviews. Am J Gastroenterol
placebo in Helicobacter pylori negative, primary care pa- 2003;98:26216.
tients with dyspepsia: The Cadet-HN Study. Am J Gas- 119. McNamara D, Buckley M, Gilvarry J, et al. Does He-
troenterol. doi:10.1111/j.1572-0241.2005.50332.x. licobacter pylori eradication affect symptoms in nonul-
102. Gillen D, McColl KE. Problems related to acid rebound cer dyspepsia: A 5-year follow-up study. Helicobacter
and tachyphylaxis. Best Pract Res Clin Gastroenterol 2002;7:31721.
2001;15:48795. 120. Hsu PI, Lai KH, Lo GH, et al. Risk factors for ulcer devel-
103. Manes G, Menchise A, de Nucci C, et al. Empiri- opment in patients with non-ulcer dyspepsia: A prospective
cal prescribing for dyspepsia: Randomised controlled two year follow up study of 209 patients. Gut 2002;51:15
trial of test and treat versus omeprazole treatment. BMJ 20.
2003;326:111824. 121. Talley NJ. Therapeutic options in nonulcer dyspepsia. J
104. Fendrick AM, Chernew ME, Hirth RA, et al. Alternative Clin Gastroenterol 2001;32:28693.
management strategies for patients with suspected peptic 122. Dhir R, Richter JE. Erythromycin in the short- and long-
ulcer disease. Ann Intern Med 1995;123:2608. term control of dyspepsia symptoms in patients with gas-
105. Sonnenberg A. Cost-benefit analysis of testing for Heli- troparesis. J Clin Gastroenterol 2004;38:23742.
cobacter pylori in dyspeptic subjects. Am J Gastroenterol 123. Tack J, Bisschops R, DeMarchi B. Causes and treatment of
1996;91:17737. functional dyspepsia. Curr Gastroenterol Rep 2001;3:503
106. Silverstein MD, Petterson T, Talley NJ. Initial endoscopy or 8.
empirical therapy with or without testing for Helicobacter 124. Tanum L, Malt UF. A new pharmacologic treatment
pylori for dyspepsia: A decision analysis. Gastroenterology of functional gastrointestinal disorder. A double-blind
1996;110:7283. placebo-controlled study with Mianserin. Scand J Gas-
107. Ofman JJ, Etchason J, Fullerton S, et al. Management troenterol 1996;31:31825.
strategies for Helicobacter pylori-seropositive patients 125. Mertz H, Fass R, Kodner A, et al. Effect of amitriptyline on
 Guidelines for the Management of Dyspepsia 2337

symptoms, sleep, and visceral perception in patients with 130. Srinivasan R, Greenbaum DS. Chronic abdominal wall
functional dyspepsia. Am J Gastroenterol 1998;93:1605. pain: A frequently overlooked problem. Practical ap-
126. Calvert EL, Houghton LA, Cooper P, et al. Long-term proach to diagnosis and management. Am J Gastroenterol
improvement in functional dyspepsia using hypnotherapy. 2002;97(4):82430.
Gastroenterology 2002;123:177885. 131. Wayman J, Griffin SM, Campbell FC. Is functional
127. Soo S, Moayyedi P, Deeks J, et al. Psychological interven- dyspepsia largely explained by gastro-oesophageal re-
tions for non-ulcer dyspepsia. Cochrane Database of Syst flux disease? Baillieres Clin Gastroenterol 1998;12:463
Rev 2005;2:CD002301. 76.
128. Bortolotti M, Coccia G, Grossi G, et al. The treatment of 132. Camilleri M, Talley NJ. Pathophysiology as a basis
functional dyspepsia with red pepper. Aliment Pharmacol for understanding symptom complexes and therapeu-
Ther 2002;16:107582. tic targets. Neurogastroenterol Motil 2004;16(2):135
129. May B, Kohler S, Schneider B. Efficacy and tolerability 42.
of a fixed combination of peppermint oil and caraway oil 133. Hallas J, Bytzer P. Screening for drug related dyspepsia:
in patients suffering from functional dyspepsia. Aliment An analysis of prescription symmetry. Eur J Gastroenterol
Pharmacol Ther 2000;14:16717. Hepatol 1998;10(1):2732.