Best Practice & Research Clinical Gastroenterology 24 (2010) 109–120

Contents lists available at ScienceDirect

Best Practice & Research Clinical

Gastroenterology

Dyspepsia as an adverse effect of drugs

Peter Bytzer, MD, PhD, AGAF *
Department of Gastroenterology, Køge University Hospital, Copenhagen University, DK-4600 Køge, Denmark

Keywords:
Drugs are frequently implicated as a possible cause in new onset
Dyspepsia dyspeptic symptoms and few drugs are free of this suspicion.
Adverse events Nausea, anorexia, abdominal pain and dyspepsia make up between
Drugs one-tenth and one-third of reported adverse reactions but they are
Pharmaco-epidemiology all so common, both in the background population and among
Side effects patients, that they are frequently attributed to an illness rather
Medications than to medications. No symptom or clinical sign is pathogno-
monic for adverse drug effects, maybe with the exception of
vomiting. Dyspepsia is a common reporting in placebo-arms
of treatment trials. Owing to the high background incidence of
dyspepsia, it is difficult to discern between spontaneous and true
drug-related dyspepsia. The mechanisms by which a drug causes
dyspepsia are often unknown even though some drugs are known
to cause direct mucosal injury. Non-steroidal anti-inflammatory
drugs and antibiotics are common causes of drug-related
dyspepsia.
Ó 2009 Elsevier Ltd. All rights reserved.

Dyspepsia, defined as pain or discomfort centred in the upper part of the abdomen is a very
common complaint, even among healthy individuals in the background population.
Dyspeptic symptoms are reported by 11% of healthy blood donors [1] and approximately 1.5% of the
UK population consults a general practitioner (GP) for stomach problems each year.
Potential underlying causes include peptic ulcer disease, gastro-oesophageal reflux disease, func-
tional dyspepsia and gastric cancer. Drugs are also frequently implicated as a possible cause in new
onset dyspeptic symptoms and few drugs are free of the suspicion of causing abdominal complaints.
Gastrointestinal (GI) symptoms such as nausea, anorexia, abdominal pain and dyspepsia make up
between one-tenth and one-third of reported adverse reactions but they are all so common, both in the
background population and among patients, that they are all too easily attributed to the illness rather

* Tel.: þ45 4732 2414.

E-mail address: peter.bytzer@dadlnet.dk

1521-6918/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bpg.2009.11.006
110 P. Bytzer / Best Practice & Research Clinical Gastroenterology 24 (2010) 109–120

than to the drug. Relationship to drug use is also obscured by variations in the terminology used to
report GI symptoms. Dyspepsia is often defined as any symptom thought by the physician to originate
from the GI tract, and may thus embrace a variety of symptoms such as nausea, early satiety, vomiting,
epigastric pain, heartburn and even bloating. Others reserve the term dyspepsia for epigastric pain or
discomfort [2]. Many reports of dyspepsia as an adverse effect of drug use do not define the term. At
least in part as a consequence of this terminological confusion, epidemiological studies have produced
differing estimates of the incidence and prevalence of dyspepsia, and of its association with potential
risk factors, such as drug use (Table 1).
Estimating the relative frequency of dyspepsia as an adverse drug reaction is not easy. Dyspepsia is
a common reporting in placebo-arms of treatment trials ranging from 2.3% to 4.2% in some studies
[2]. On the top 20 adverse events list, by number of reports in the period 1969–2002, nausea is
number four, vomiting is number nine and abdominal pain is number 13. The term dyspepsia is not on
the list [3].
The newest version of the international encyclopaedia of adverse drug reactions Meyler’s Side Effects
of Drugs, published in 2006 [4] lists dyspepsia as a side effect to 46 drugs or drug classes. Additional
information from the annual reportings from the same source from the years 2007–2008 adds another
eight drugs to the list.
Keeping track with adverse drug reactions is a difficult task. In the period 1999–2003 6576 articles
related to adverse drug reactions were published [4]. A search in PubMed using the MeSH headings
‘dyspepsia’ and ‘chemically induced’ produced 272 citations. 128 of these were published within the last
10 years and more than half (66/128) related to NSAID or aspirin use.
This review summarises current knowledge about dyspepsia, defined as epigastric pain or
discomfort, as a side effect to drug use.

Methodological challenges in side effects research

The spontaneous reporting rate for adverse drug reactions that require hospital admission is in the
order of 1% [5] and for less severe reactions the reporting rate is undoubtedly lower.
No symptom or clinical sign is pathognomonic for adverse drug effects, maybe with the exception of
promptly occurring or persistent vomiting [6]. Even serious events, such as upper GI bleeding, may be
missed for a long time as a possible drug-related effect, because the individual physician will see only
few such cases. Interestingly, it took 39 years before the association between aspirin use and upper GI
bleeding was recognised.
Incriminating a drug as a cause of a specific symptom, such as dyspepsia, requires use of the drug to
be temporarily associated with the symptom, for the symptom to resolve when the drug is withdrawn
and for it to reappear when a patient is re-challenged with the drug.
The temporal relationship between a drug and the side effect may not be very close, which is one
reason why iatrogenic drug-induced disease – or symptoms – may be missed. Furthermore, patients
are frequently on several drugs at the same time and the requirements to prove a causal relationship
are thus seldom met. In everyday clinical practise the suspect drug is withdrawn but the subsequent
challenge rarely pursued.

Table 1
Drugs that commonly cause dyspepsia.

NSAIDs
Cox2-inhibitors
Acetylsalicylic acid
Proton pump inhibitors
Bisphosphonates
Erythromycin
Tetracyclines
Sildenafil and tadalafil
Theophylline
 P. Bytzer / Best Practice & Research Clinical Gastroenterology 24 (2010) 109–120 111

Uncovering adverse effects to drugs is often published via sporadic reporting of patient cases [7] or
through retrospective chart reviews [8,9]. Unpublished data summarised in the US Food and Drug
Administration (FDA) reviews have also contributed and may be an important supplement in some
cases [10].

Data from randomised trials

Relying on data from randomised placebo-controlled trials (RCTs) will produce the least biased
method of estimating the risk of dyspepsia associated with drug use. However, that approach will
almost certainly underestimate the incidence of reported dyspepsia in the general population of drug
users. This is because clinical trials tend to enrol healthier subjects than the general population, and the
duration of exposure is usually relatively short. On the other hand, this method is biased against new
drugs because any symptom reported more frequently compared to placebo-treated patients will be
notified. Randomised controlled trials were not used prior to marketing for older drugs and new drugs
may therefore appear to have more side effects than traditional drugs, which have been on the market
for many years – even though their side effects profiles may not differ if compared directly. Further-
more, the controlled trials face the difficulty of differentiating between true drug-related dyspepsia
and the underlying spontaneous incidence, which is illustrated by the frequent occurrence of
dyspepsia as an alleged adverse effect of placebo.

Pharmacovigilance

A recent study evaluated 1418 cases of spontaneously reported adverse drug events and the clinical
features and causative drugs. The most common side effect was skin manifestations (42% of the
patients), followed by neurological manifestations (14%) and GI symptoms (12.9%). Antibiotics and
NSAIDs were the most common causative drugs for adverse events [7].
Prescription databases are crucial in drug safety research and surveillance. For suspected rare
serious adverse drug reactions, the confirmation and quantitative risk assessment requires proper
epidemiological studies. Often prescription databases provide the information on drug exposure for
such studies.

Prescription sequence symmetry analysis

Owing to the high background incidence of dyspepsia, a potential reporting physician finds it
difficult to discern between spontaneous and true drug-related dyspepsia. Using prescription data on
drugs used to treat dyspepsia has been suggested as a solution to this problem [11,12]. Most patients,
who consult because of newly developed dyspepsia, are given empirical symptomatic drug treatment
either as acid inhibitors or as prokinetics. Therefore, if a drug causes dyspepsia as an adverse effect, one
might expect to see an excess of patients who are first prescribed this drug and subsequently, after an
appropriate time interval, a drug to treat dyspepsia, e.g. a proton pump inhibitor [11] (Fig. 1). By using
large pharmaco-epidemiologic databases that capture all prescriptions in a well defined population
a symmetry analysis can be performed that might reveal hitherto unknown dyspepsinogenic medi-
cations. Unfortunately, using drug prescriptions as a surrogate endpoint raises several serious prob-
lems with confounding and a critical analysis of any signals must be performed. It can be shown,
however, that the symmetry principle effectively controls confounders that are stable over time [13].
The prescription symmetry analysis works best in conditions where the drug-induced symptom, in this
case dyspepsia, is unknown to the prescribing physician. If dyspepsia is a well-known side effect to
a drug, e.g. an NSAID, physicians who encounter a drug user with new onset dyspepsia will probably
respond by discontinuing the offensive drug instead of prescribing an ulcer drug.
A recent analysis was done using a Danish database comprising 1.5 million prescriptions drawn
from a study base of 2.0 million person-years [11]. Using prescriptions for acid-suppressive medica-
tions as a marker for new onset dyspepsia five drugs or drug classes showed an asymmetry suggesting
a dyspepsinogenic effect: NSAIDs, angiotensin converting enzyme inhibitors (ACEIs), calcium channel
blockers, systemic steroids and methyl- xanthines (Fig. 2). A detailed analysis of the data suggested that
112 P. Bytzer / Best Practice & Research Clinical Gastroenterology 24 (2010) 109–120

Prescription Sequence Symmetry Analysis

Index drug
e.g. penicillin Ulcer drug
e.g. omeprazole Index drug
e.g. penicillin

100 days
Fig. 1. Prescription sequence symmetry analysis presupposes that the development of a drug-induced side effect (e.g. dyspepsia
induced by penicillin) is followed by a prescription of a drug that is used to treat the side effect (e.g. a proton pump inhibitor).
By analysing the distribution of prescription orders in patients prescribed both drugs within a fixed time period (e.g. 100 days) side
effects to index drugs can be uncovered [11].

two of these signals were false. The signal for corticosteroids was confined to those who had used
NSAIDs before their first ulcer drug prescription and the non-symmetry for ACEIs was mainly seen in
patients who also used loop diuretics. This use is highly suggestive of congestive heart failure, where
dyspepsia is a well-known manifestation [11].
The prescription sequence symmetry principle was also used to screen for dyspepsia as a drug-
related side effect by analysing prescriptions for prokinetics (cisapride) and anti-emetics (metoclo-
pramide) as these drugs are recommended for patients with symptoms compatible with functional
dyspepsia [12]. In the cisapride analysis, which consisted of data from 1825 patients, no single drug
could be shown to induce symptoms resembling functional dyspepsia. On the other hand, the meto-
clopramide analysis of 6126 patients showed positive signals for 14 drugs, all well-known for causing
nausea as a side effect (Fig. 3) [12].
The strength of the method of prescription sequence symmetry in identifying drug-related side-
effects has been demonstrated by the results relating to ulcer drug and metoclopramide prescriptions
and even with antiepileptic and cardiovascular drugs [13,14].

Estrogen/gestagen
Macrolides
Asthma drugs
Estrogenes
Nitrates
Oral beta-agonists
Thyroid hormones
Corticosteroids
Digoxin
Anti-rheumatics
Anti-epileptics
Lithium
ACE inhibitors
Calcium blockers
Methylxanthines
NSAIDs
0,1 1,0 10
Rate ratio
Fig. 2. Prescription orders expressed as rate ratios with 95% confidence intervals in patients who redeemed their first prescription of
an ulcer drug and an index drug. A sequence ratio above 1 indicates a possible dyspepsinogenic effect of the drug. Data from [11].
 P. Bytzer / Best Practice & Research Clinical Gastroenterology 24 (2010) 109–120 113

Thiazides
Oral antidiabetics
Calcium blockers
Corticosteroids
Penicillins
Asthma drugs
Potassium supplement
NSAIDs
Loop diuretics
Macrolides
Nitrates
Methylxanthines
ACE-inhibitors
Opioid analgesics
Digoxin
Insulin

0,1 1 10

Fig. 3. Distribution of prescription orders for 6126 incident users of metoclopramide. Results are rate ratios with 95% confidence
intervals. A sequence ratio above 1 indicates that the drug may induce nausea. Data from [12].

Mechanisms of action

Although some side effects are dose related, most depend on the sensitivity of the host [15].
Pre-existing disease, such as peptic ulcer, may predispose to side effects, as may intrinsic defects of
the GI tract such as gastric atrophy, coeliac disease and changes brought about by ageing [15].

Direct mucosal injury

NSAIDs cause a wide range of damage throughout the gut and serve as the primary example of
varied iatrogenic disease patterns in this account [16].
Erosive oesophagitis has been demonstrated in 20% or more of patients taking NSAIDs [17]. The
mechanism of action is uncertain but may include local chemical effects, systemic effects related to
inhibition of platelet aggregation and exacerbation of pre-existing gastro-oesophageal reflux disease.
Oesophageal injury can be caused directly by prolonged mucosal contact with tablets or capsules.
Studies have confirmed poor clearance of tablets from the oesophagus in the supine position, especially
if taken with inadequate (<100 mL) amounts of liquid [18]. The acidity or alkalinity of a medication is
unlikely to be the only factor; other chemical properties may also be important [19]. Factors such as
drug dissociation rate, osmolarity, and intrinsic chemical toxicity have also been implicated in the
pathogenesis. Sustained-release medications may be more injurious to the oesophagus than standard
preparations. Some drugs have been formulated to disintegrate rapidly on contact with water and the
hygroscopicity of the tablet can cause it to stick to the mucosa and reach a high ulcerogenic local
concentration [20]. For example, deposits of potassium chloride have been demonstrated in biopsies
from an oesophageal ulcer.
NSAIDs appear to contribute to esophageal pathology by direct cellular toxicity and disruption of
the mucosal barrier (local effect) rather than by inhibiting prostaglandin synthesis (systemic effect)
[21].
Even after one or two doses of aspirin, superficial gastric erosions have been described in over 50%
of healthy individuals. The local effects depend in part on the tablet particle size, solubility, and rate of
gastric absorption, while the most important variable appears to be gastric pH [4]. Aspirin formulation
does not seem to affect gastric toxicity. Data from pooled studies negate a significant protective effect of
the most frequently used aspirin formulations, such as enteric-coated or buffered formulations [22].
A direct toxic effect has also been demonstrated for various other drugs. Supporting this mechanism
of action are the results from conversion studies in patients intolerant to e.g. mycophenolate.
114 P. Bytzer / Best Practice & Research Clinical Gastroenterology 24 (2010) 109–120

Enteric-coated equimolar doses of the drug resulted in significant improvement in GI symptoms [23].
Crystalline iron deposition has been described in gastric mucosal biopsies. These deposits can cause or
exacerbate a distinctive histological pattern of erosive mucosal injury, especially in patients with
associated upper GI disorders [24].

Alterations in gastric motility

Erythromycin is a motilin receptor agonist. This mechanism may be at least partly responsible for
the GI side effects reported with that drug. The prominent gastrokinetic properties of erythromycin are
used clinically to treat patients with gastroparesis with repeated vomiting [25,26] and to improve
gastric cleansing in bleeding patients prior to endoscopy [27]. However, the clinical usefulness of
erythromycin for gastroparesis is hampered by tachyphylaxis.

Induction of gastric acid rebound

Both H2-receptor antagonists and PPIs may induce an increased capacity to gastric acid secretion
that sets off once the drugs are discontinued. The proposed mechanism for this is via changes in gastrin
induced by antral pH changes due to the prolonged acid inhibition. Gastrin has a well established
trophic effect on the oxyntic mucosa. This may induce hyperplasia or hypertrophy of the enter-
ochromafine-like cells and the parietal cells with an increased acid secretory capacity as a conse-
quence. This so-called rebound acid hypersecretion may induce dyspeptic and reflux symptoms in
some individuals when the acid secretory inhibitor drugs are stopped [28].

Provocation of gastro-oesophageal reflux

Drugs with anticholinergic properties may facilitate reflux oesophagitis by a relaxing effect of the
lower oesophageal sphincter. Calcium channel blockers relax oesophageal smooth muscle and may
provoke reflux symptoms in patients with oesophagitis and in normal subjects [29]. Other drugs with
a similar potential include nitrates, sildenafil, nicotine, theophylline and substances with anti-
muscarinic potential [30].

Individual susceptibility factors

Idiosyncratic mechanisms are probably very important in determining side effect profiles. Indi-
vidual susceptibility factors usually include age, sex, genetic and physiological factors and underlying
disease.

Drugs that often induce dyspeptic symptoms

Non-steroidal anti-inflammatory drugs (NSAIDs)

NSAIDs are used by more than 20 million people in the US and the point prevalence of prescription
NSAID use is 10–15% in persons older than 65 years [31].
Gastric mucosal damage by NSAIDs is a major health problem. Estimates from the UK have shown
that 20–30% of all hospitalised cases of ulcer complications in subjects aged over 60 are directly
attributable to NSAIDs and that some 10% will die from this [32]. On average one to two out of every one
hundred patients taking NSAIDs for one year are hospitalised for a GI event, most commonly an ulcer.
Even though NSAIDs have a widespread reputation for poor gastric tolerance, epidemiologic data to
support the strength of the association to dyspepsia is not strong. Ofman and coworkers performed
a meta-analysis of dyspepsia and NSAID use based on the reportings from 37 placebo-controlled RCTs
and 11 FDA reviews encompassing nearly 12.000 patients [31]. In the NSAID-treated group 4.8% of the
patients had dyspepsia compared with 2.3% in the placebo group. High dosage of NSAIDs and the drugs
indomethacin, meclofenamate, and piroxicam were associated with an increased risk of dyspepsia [31].
A random sample survey of an elderly population found an odds ratio of 1.9 for dyspepsia among NSAID
 P. Bytzer / Best Practice & Research Clinical Gastroenterology 24 (2010) 109–120 115

users compared to non-users [33]. Uncontrolled surveys of arthritic NSAID users report daily dyspepsia
in 16% and as many as 62% reported having experienced an upper GI symptom within the last year [34].
Dyspepsia is the most common reason for discontinuation of NSAIDs [35].
Even though development of dyspepsia during NSAID therapy is a strong risk factor for an NSAID-
related ulcer complication (OR 8.7; 95% CI 4.0–18.9) [36] there is generally not a close correlation
between endoscopic findings and subjective intolerance to the drugs.
A recent German primary care study followed 104 patients, who were free of dyspepsia at start of
treatment with an NSAID. Under treatment, 44% of the patients developed troublesome dyspepsia and/
or ulcer; 35% of the patients developed troublesome dyspepsia that required treatment and 16%
developed an ulcer. The frequency of dyspepsia was independent of the duration of NSAID use.
Dyspepsia was reported more frequently by ulcer patients than those with no ulcer but their overall
symptom frequency was no higher than in the latter and patients who developed an ulcer were not
identifiable on the basis of symptoms or risk factors [37].
The propensity to GI intolerance may not differ between equipotent doses of the various NSAIDs
drugs as there seems to be a close correlation between efficacy and adverse effects.
Higher doses of NSAIDs have been associated with a greater risk for dyspepsia [31]. It is therefore
plausible that a reduction in NSAID dose may alleviate dyspeptic symptoms. There are, however, no
controlled trials to support this hypothesis.
Both Helicobacter pylori infection and NSAID use independently, synergistically and significantly
increase the risk of peptic ulcer and its complications. This is substantiated by experimental data which
has shown that NSAID-induced ulceration can be reduced by eradication of H. pylori before NSAID
administration [38]. It is still unclear, however, if synergism exists between NSAID use and H. pylori
infection in the development of dyspepsia symptoms [39].
H2-receptor antagonists and PPIs may be effective in the treatment of NSAID-associated dyspepsia.
In a randomised trial of arthritic NSAID users without peptic ulcer disease dyspeptic symptoms
disappeared in 26% of the ranitidine-treated patients which was significantly better than the 6%
placebo-treated patients [40]. H2-receptor antagonists when used at standard doses are not effective at
preventing gastric ulcers resulting from the use of NSAIDs.
Once-daily therapy with proton pump inhibitors has been documented to significantly decrease the
development of NSAID-associated ulcers in endoscopic studies, reduce the rate of NSAID-related ulcer
complications and reduce upper GI symptoms in NSAID users [41].
Misoprostol effectively decreases NSAID-induced ulcers and GI complications, but misoprostol
co-therapy does not appear to reduce the incidence of NSAID-associated dyspepsia and issues of
compliance (multiple daily doses) and side effects (e.g. diarrhoea and dyspepsia) limits its use [41].

Cox-2 inhibitors

Cox-2 selective agents (Coxibs) cause less damage to the upper GI mucosa than traditional NSAIDs
but the clinical relevance of this is still debated. In endoscopic comparisons of Coxibs with traditional
NSAIDs (t-NSAIDs) erosions and ulcers in the upper GI tract were fewer with Coxibs and were similar to
those found with placebo [42].
Evidence from clinical trials suggests an incidence of dyspepsia and other GI symptoms not very
different to that caused by t-NSAIDs. Even the withdrawal rate from clinical trials because of adverse
events was comparable in head-to-head comparison studies.
In the VIGOR study, 8076 patients with rheumatoid arthritis were randomised to rofecoxib or
naproxen [43]. The most common adverse events that lead to withdrawal of therapy were dyspepsia,
abdominal pain, epigastric discomfort, nausea and heartburn. Significantly fewer patients dis-
continued due to these side effects in the rofecoxib group compared with the naproxen group, but the
absolute difference was small (3.5% vs. 4.9%) [43].
A post-hoc analysis of two controlled studies suggested that valdecoxib is associated with improved
dyspepsia-related health compared with nonspecific NSAIDs in patients with osteoarthritis or rheu-
matoid arthritis [44]. This was also the conclusion from a randomised trial comparing celecoxib with
a combination of lansoprazole and naproxen [45]. A meta-analysis of 26 studies comparing dyspepsia
between Coxibs and t-NSAIDs revealed a 12% relative risk reduction for Coxibs with an absolute risk
116 P. Bytzer / Best Practice & Research Clinical Gastroenterology 24 (2010) 109–120

reduction of 3.7%. Meta-analysis of four studies comparing dyspepsia between an NSAID plus PPI
combination and NSAIDs alone showed a 66% relative risk reduction for NSAID plus PPI with an
absolute risk reduction of 9%. Because there are limited head-to-head data comparing Coxibs versus
NSAID plus PPI, these data provide the best indirect evidence that NSAID plus PPI may be superior to
Coxibs in minimising incident dyspepsia [46]. Even though the incidence of dyspepsia may be lower
with Coxibs compared to traditional NSAIDs dyspepsia remains a significant problem. A combined
analysis using data from nine randomised, double-blind, controlled, clinical trials with etoricoxib
showed a discontinuation rate because of dyspepsia of 1.5 per 100 patient-years and an overall rate of
new use of gastroprotective drugs of 9.1 per 100 patient-years. The corresponding figures for t-NSAIDs
were 2.7 and 13.0, respectively [47].

Acetylsalicylic acid

Aspirin is an independent risk factor for dyspepsia in community subjects in Australia (odds ratio
2.2; 95% CI, 1.3–3.7). Surprisingly, the same study was unable to show any effect of NSAID use on the
risk for dyspepsia [48]. Dyspepsia, nausea and vomiting occur in 2–6% of patients after aspirin
ingestion. Patients with rheumatoid arthritis are more sensitive and the frequency of dyspepsia in this
group is as high as 10–30% [4]. A recent study demonstrated the temporal course of dyspeptic
complaints in aspirin-naı̈ve patients who had not reported upper GI symptoms before start of therapy.
In these patients, the onset of dyspepsia symptoms was rapid; 13% of patients reported symptoms after
just one day of low-dose aspirin use, rising to 34% at one week [49].
Proton pump inhibitor therapy may reduce the incidence of dyspeptic symptoms associated with
aspirin use, but results from clinical trials are conflicting [50,51].

Corticosteroids

Dyspepsia occurs with all types of corticosteroids, is to some extent dose-related though still
dependent on individual sensitivity and is more likely in people with previous ulcer histories [15]. A
meta-analysis of whether corticosteroids caused peptic ulcers was negative and peptic ulcer disease
should not be considered a contraindication when steroid therapy is indicated [52]. However, the risk
of a fatal outcome due to ulcer complications is increased about fourfold in glucocorticoid-treated
patients according to case-control studies [4].

Antimicrobials

Macrolide antibiotics are generally well tolerated but nausea, vomiting, abdominal pain and
dyspepsia are relatively common and sometimes result in premature withdrawal. Newer macrolides,
such as azithromycin, clarithromycin and roxithromycin are better tolerated and cause fewer adverse
GI events (4–10%) compared with erythromycin (27%) [4].
Mild gastrointestinal disturbances are common with use of tetracyclines [4]. Nausea (reported by 8–
15% of patients), vomiting and epigastric burning are the most common adverse effects but oeso-
phageal ulcers are the most dramatic. Acute onset of substernal burning pain and dysphagia has been
described in cases where remaining parts of ingested tetracycline capsules have been identified by
endoscopy [4].
Other antimicrobials incriminated for provoking dyspepsia include itraconazole, terbinafine,
ribaverin, abacavir and oseltamivir [4].

Drugs that act on the central nervous system

Risperidone, lesopritone and the SSRIs all have the potential to induce dyspepsia [4].
 P. Bytzer / Best Practice & Research Clinical Gastroenterology 24 (2010) 109–120 117

Drugs that act on the immune system

Dyspepsia and other GI adverse effects are very common (up to 50%) after oral administration of
methotrexate and often require dosage adjustment. Folic acid supplementation reduces the incidence
of severe GI side effects. Dyspepsia is reported by 7.6% of patients treated with tacrolimus [53].
D-penicillamine induces dyspepsia in 11% of patients with rheumatoid arthritis [9]. Leflunomide and
mycophenolate sodium also induce dyspeptic symptoms relatively often [4].

Cardiovascular drugs

According to Meyler’s Side Effects of Drugs dyspepsia has been reported as an adverse effect of the
following drugs or drug classes used to treat cardiovascular diseases: atorvastatin, beta-receptor
agonists, losartan, calcium channel blockers and digoxine [4].

Drugs that act on the GI tract

The use of PPIs for acid-related symptoms and disorders is extensive and continuously escalating. In
2006 approximately 7% of the Danish population was treated with a PPI. While the incidence of new
treatments with PPIs remains stable the prevalence of long-term treatment is rising [54]. The reasons
for this increasing long-term use are not fully understood. Previous studies have established
a temporarily increased capacity to secrete acid after discontinuation of long-term PPI treatment [55].
If this rebound acid hypersecretion is of clinical relevance and induces acid-related symptoms this
might lead to reuptake of PPI therapy and thus PPI dependency. In a recent trial the clinical relevance of
the acid rebound phenomenon was demonstrated for the first time [28]. Healthy volunteers with no
history of acid-related symptoms were randomised to treatment with placebo for 12 weeks or eight
weeks of PPI therapy followed by a blinded shift to placebo for four weeks. Symptoms were registered
weekly. A significantly higher symptom score for acid-related symptoms as well as a significantly
higher proportion reporting clinically relevant heartburn, acid regurgitation or dyspepsia was
demonstrated in the actively treated group compared to the placebo group in the weeks after the
blinded discontinuation supporting the hypothesis that rebound acid hypersecretion is clinically
relevant. Thus, PPI therapy induces the symptoms it is used to treat [56].
PPIs may also induce dyspepsia during treatment. A recent prescription-event monitoring study
from the UK of 11.595 patients showed that dyspepsia was among the adverse events that occurred
significantly more often in the first month of treatment with esomeprazole compared with months 2–6
[57].
H2-receptor antagonists also induce acid rebound but the clinical implications are probably very
modest as the symptoms associated with this phenomenon are very short-lived [56].
Sulfasalazine is of proven value in chronic inflammatory colon disease. However, owing to adverse
effects the drug often has to be withdrawn. A Danish retrospective chart review of 704 patients showed
that dyspeptic complaints developed in 17% of the patients. Generally, dyspepsia occurred within the
first days after start of therapy. In 107 of 121 patients dyspepsia ceased after regular sulfasalazine had
been replaced by enteric-coated dragées [8].
Orlistat, an anti-obesity drug, is a specific inhibitor of intestinal lipases. Orlistat has been associated
with several mild to moderate GI adverse events, including abdominal pain and dyspepsia [58].

Various other drugs

Iron has a widespread reputation for poor gastric tolerance. However, in normal daily doses GI
adverse events are infrequent and usually related to constipation. Dyspepsia can be seen with higher
doses [4].
Acarbose may give abdominal pain and dyspepsia relatively often (1–10%).
Oral nicotinic acid may be associated with general GI symptoms, including heartburn, dyspepsia,
vomiting, nausea and hunger pains.
118 P. Bytzer / Best Practice & Research Clinical Gastroenterology 24 (2010) 109–120

Dyspepsia is reported as a side effect to sildenafil in 5–15% of subjects in placebo-controlled trials

and dyspepsia is among the most commonly reported adverse effects with sildenafil and tadalafil used
to treat erectile dysfunction or sexual dysfunction [59–61].
GI complaints are common with sulfonylureas. They comprise dyspepsia, nausea, vomiting and
abdominal pain and they are reportedly less troublesome when the drug is taken after meals.
The bisphosphonates cause severe GI adverse effects. Of greatest concern is their capacity to cause
oesophageal ulcerations. Symptoms are usually acute and can often be related to ingestion of the
offending drug. Patients present with chest pain, odynophagia, heartburn, globus, and sometimes
dysphagia. Diarrhoea, nausea and dyspepsia have also been reported. Severe oesophagitis or ulcera-
tions were first reported in patients taking alendronate [62] and is usually linked to improper ingestion
and inappropriate timing of drug intake [18]. Oesophageal lesions usually heal on withdrawal and acid-
suppressive medication is only supportive even though PPIs are commonly used to reduce gastric
irritation when bisphosphonates are prescribed [63]. A British primary care study found an increased
risk of dyspepsia among current users of bisphosphonates (OR 2.3, 95% CI: 1.3–4.2). However,
bisphosphonate use was a rare explanation for dyspepsia in the cohort as these drugs were used by
only 0.4% of all dyspeptic patients [64]. The most frequently reported event in the first month of
treatment with risedronate was dyspepsia, being also amongst the most frequently reported reasons
for stopping risedronate [65].
The commonest side effects of theophylline in adults are nausea and dyspepsia-like symptoms. Side
effects are related to drug levels in the blood, although some additional local irritation in the upper GI
tract may be induced by oral administration. Methylxanthines, mainly in the form of theophylline,
showed a moderate signal in a prescription symmetry analysis of ulcer drug prescribing [11]. A
regression analysis showed a strong positive effect modification by a short span between theophylline
and ulcer drug therapy suggesting a rapid onset of medication-related dyspepsia [11].
Potassium chloride is irritating to the GI tract and ulcers and perforations of the gut have been
reported. GI side effects are usually mild.

Conflicts of interest

None

References

[1] Holtmann G, Goebell H, Holtmann M, et al. Dyspepsia in healthy blood donors. Pattern of symptoms and association with
Helicobacter pylori. Dig Dis Sci 1994;39:1090–8.
[2] Straus WL, Ofman JJ, MacLean C, et al. Do NSAIDs cause dyspepsia? a meta-analysis evaluating alternative dyspepsia
definitions. Am J Gastroenterol 2002;97:1951–8.
[3] Aronson JK. Drug withdrawals because of adverse effects. In: Aronson JK, editor. Side effects of drugs. Oxford: Elsevier;
2008. p. xxxi–xxxv. Annual 30.
*[4] Meyler’s side effects of drugs. 15th ed. Oxford: Elsevier; 2006.
[5] Hallas J. Drug related hospital admissions in subspecialities of internal medicine. Dan Med Bull 1996;43:141–55.
*[6] Bruppacher R, Gyr N, Fisch T. Abdominal pain, indigestion, anorexia, nausea and vomiting. Baillieres Clin Gastroenterol
1988;2:275–92.
*[7] Shin YS, Lee YW, Choi YH, et al. Spontaneous reporting of adverse drug events by Korean regional pharmacovigilance
centers. Pharmacoepidemiol Drug Saf 2009.
[8] Nielsen OH. Sulfasalazine intolerance. A retrospective survey of the reasons for discontinuing treatment with sulfasa-
lazine in patients with chronic inflammatory bowel disease. Scand J Gastroenterol 1982;17:389–93.
[9] Weiss AS, Markenson JA, Weiss MS, et al. Toxicity of D-penicillamine in rheumatoid arthritis. A report of 63 patients
including two with aplastic anemia and one with the nephrotic syndrome. Am J Med 1978;64:114–20.
[10] Maclean CH, Morton SC, Ofman JJ, et al. How useful are unpublished data from the Food and Drug Administration in
meta-analysis? J Clin Epidemiol 2003;56:44–51.
*[11] Hallas J, Bytzer P. Screening for drug related dyspepsia: an analysis of prescription symmetry. Eur J Gastroenterol Hepatol
1998;10:27–32.
*[12] Bytzer P, Hallas J. Drug induced symptoms of functional dyspepsia and nausea. A symmetry analysis of one million
prescriptions. Aliment Pharmacol Ther 2000;14:1479–84.
[13] Hallas J. Evidence of depression provoked by cardiovascular medication: a prescription sequence symmetry analysis.
Epidemiology 1996;7:478–84.
[14] Tsiropoulos I, Andersen M, Hallas J. Adverse events with use of antiepileptic drugs: a prescription and event symmetry
analysis. Pharmacoepidemiol Drug Saf 2009;18:483–91.
[15] Paton A. Diseases of the alimentary system. Gastrointestinal reactions to drugs. Br Med J 1976;2:1179–80.
 P. Bytzer / Best Practice & Research Clinical Gastroenterology 24 (2010) 109–120 119

*[16] Price AB. Pathology of drug-associated gastrointestinal disease. Br J Clin Pharmacol 2003;56:477–82.
[17] Mantry P, Shah A, Sundaram U. Celecoxib associated esophagitis. Review of gastrointestinal side effects from Cox-2
inhibitors. J Clin Gastroenterol 2003;37:61–3.
[18] Hey H, Jørgensen F, Sørensen K, et al. Oesophageal transit of six commonly used tablets and capsules. BMJ 1982;285:1717–9.
*[19] Parfitt JR, Driman DK. Pathological effects of drugs on the gastrointestinal tract: a review. Hum Pathol 2007;38:527–36.
[20] Eng J, Sabanathan S. Drug-induced esophagitis. Am J Gastroenterol 1991;86:1127–33.
[21] Avidan B, Sonnenberg A, Schnell TG, et al. Risk factors for erosive reflux esophagitis: a case-control study. Am J Gas-
troenterol 2001;96:41–6.
[22] Walker J, Robinson J, Stewart J, et al. Does enteric-coated aspirin result in a lower incidence of gastrointestinal
complications compared to normal aspirin? Interact Cardiovasc Thorac Surg 2007;6:519–22.
[23] Darji P, Vijayaraghavan R, Thiagarajan CM, et al. Conversion from mycophenolate mofetil to enteric-coated mycophe-
nolate sodium in renal transplant recipients with gastrointestinal tract disorders. Transplant Proc 2008;40:2262–7.
[24] Eckstein RP, Symons P. Iron tablets cause histopathologically distinctive lesions in mucosal biopsies of the stomach and
esophagus. Pathology 1996;28:142–5.
[25] Camilleri M. Diabetic gastroparesis. N Engl J Med 2007;356:820–9.
[26] Tack J. The difficult patient with gastroparesis. Best Pract Res Clin Gastroenterol 2007;21:379–91.
[27] Carbonell N, Pauwels A, Srfaty L, et al. Erythromycin infusion prior to endoscopy for acute upper gastrointestinal
bleeding: a randomized, controlled, double-blind trial. Am J Gastroenterol 2006;101:1211–5.
[28] Reimer C, Sondergaard B, Hilsted L, et al. Proton-pump inhibitor therapy induces acid-related symptoms in healthy
volunteers after withdrawal of therapy. Gastroenterology 2009;137:80–7. 87.
[29] Mehta AJ, de Caestecker JS, Northfield TC. Effects of nifedipine and atenolol on gastro-oesophageal reflux. Eur J Gas-
troenterol Hepatol 1993;5:627–9.
[30] Petersen KU, Jaspersen D. Medication-induced oesophageal disorders. Expert Opin Drug Saf 2003;2:495–507.
*[31] Ofman JJ, Maclean CH, Straus WL, et al. Meta-analysis of dyspepsia and nonsteroidal antiinflammatory drugs. Arthritis
Rheum 2003;49:508–18.
[32] Henry DA, Johnston A, Dobson A, et al. Fatal peptic ulcer complications and the use of non-steroidal anti-inflammatory
drugs, aspirin, and corticosteroids. Br Med J (Clin Res Ed) 1987;295:1227–9.
[33] Talley NJ, Evans JM, Fleming KC, et al. Nonsteroidal antiinflammatory drugs and dyspepsia in the elderly. Dig Dis Sci 1995;
40:1345–50.
*[34] Larkai EN, Smith JL, Lidsky MD, et al. Dyspepsia in NSAID users: the size of the problem. J Clin Gastroenterol 1989;
11:158–62.
[35] Vakil N, Rhew D, Soll A, et al. The cost-effectiveness of diagnostic testing strategies for Helicobacter pylori. Am J Gas-
troenterol 2000;95:1691–8.
[36] Hansen JM, Hallas J, Lauritsen J, et al. Non-steroidal anti-inflammatory drugs and ulcer complications: a risk factor
analysis for clinical decision-making. Scand J Gastroenterol 1996;31:126–30.
[37] Hollenz M, Stolte M, Leodolter A, et al. NSAID-associated dyspepsia and ulcers: a prospective cohort study in primary
care. Dig Dis 2006;24:189–94.
[38] Chan FK, Sung JJ, Chung SC, et al. Randomised trial of eradication of Helicobacter pylori before non- steroidal anti-
inflammatory drug therapy to prevent peptic ulcers. Lancet 1997;350:975–9.
[39] Huang JQ, Sridhar S, Hunt RH. Role of Helicobacter pylori infection and non-steroidal anti-inflammatory drugs in peptic-
ulcer disease: a meta-analysis. Lancet 2002;359:14–22.
[40] Van Groenendael JH, Markusse HM, Dijkmans BA, et al. The effect of ranitidine on NSAID related dyspeptic symptoms
with and without peptic ulcer disease of patients with rheumatoid arthritis and osteoarthritis. Clin Rheumatol 1996;15:
450–6.
[41] Gupta S, McQuaid K. Management of nonsteroidal, anti-inflammatory, drug-associated dyspepsia. Gastroenterology
2005;129:1711–9.
[42] Feldman M, McMahon AT. Do cyclooxygenase-2 inhibitors provide benefits similar to those of traditional nonsteroidal
anti-inflammatory drugs, with less gastrointestinal toxicity? Ann Intern Med 2000;132:134–43.
[43] Bombardier C, Laine L, Reicin A, et al. Comparison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients
with rheumatoid arthritis. VIGOR Study Group. N Engl J Med 2000;343:1520–8. 2.
[44] Rabeneck L, Goldstein JL, Vu A, et al. Valdecoxib is associated with improved dyspepsia-related health compared with
nonspecific NSAIDs in patients with osteoarthritis or rheumatoid arthritis. Am J Gastroenterol 2005;100:1043–50.
[45] Lai KC, Chu KM, Hui WM, et al. Celecoxib compared with lansoprazole and naproxen to prevent gastrointestinal ulcer
complications. Am J Med 2005;118:1271–8.
*[46] Spiegel BM, Farid M, Dulai GS, et al. Comparing rates of dyspepsia with Coxibs vs NSAIDþPPI: a meta-analysis. Am J Med
2006;119:448 e27–36.
[47] Watson DJ, Bolognese JA, Yu C, et al. Use of gastroprotective agents and discontinuations due to dyspepsia with
the selective cyclooxygenase-2 inhibitor etoricoxib compared with non-selective NSAIDs. Curr Med Res Opin 2004;20:
1899–908.
[48] Nandurkar S, Talley NJ, Xia HHX, et al. Dyspepsia in the community is linked to smoking and aspirin use but not to
Helicobacter pylori infection. Arch Intern Med 1998;158:1427–33.
[49] Chappel C, Neuman D, Thompson VJ, et al. Dyspeptic symptoms and sleep disturbance: a report from a large observa-
tional study in low-dose acetylsalicylic acid users. Gastroenterology 2009;136(Suppl. 1):T1050.
[50] Yeomans N, Lanas A, Labenz J, et al. Efficacy of esomeprazole (20 mg once daily) for reducing the risk of gastroduodenal
ulcers associated with continuous use of low-dose aspirin. Am J Gastroenterol 2008;103:2465–73.
[51] Laheij RJ, van Rossum LG, Jansen JB, et al. Proton-pump inhibitor therapy for acetylsalicylic acid associated upper
gastrointestinal symptoms: a randomized placebo-controlled trial. Aliment Pharmacol Ther 2003;18:109–15.
[52] Conn HO, Poynard T. Corticosteroids and peptic ulcer: meta-analysis of adverse events during steroid therapy. J Intern
Med 1994;236:619–32.
120 P. Bytzer / Best Practice & Research Clinical Gastroenterology 24 (2010) 109–120

[53] Yocum DE, Furst DE, Bensen WG, et al. Safety of tacrolimus in patients with rheumatoid arthritis: long-term experience.
Rheumatology (Oxford) 2004;43:992–9.
[54] Lassen A, Hallas J, Schaffalitzky de Muckadell OB. Use of anti-secretory medication: a population-based cohort study.
Aliment Pharmacol Ther 2004;20:577–83.
[55] Hunfeld NG, Geus WP, Kuipers EJ. Systematic review: rebound acid hypersecretion after therapy with proton pump
inhibitors. Aliment Pharmacol Ther 2007;25:39–46.
[56] McColl KE, Gillen D. Evidence that proton-pump inhibitor therapy induces the symptoms it is used to treat. Gastroen-
terology 2009;137:20–2.
[57] Davies M, Wilton LV, Shakir SA. Safety profile of esomeprazole: results of a prescription-event monitoring study of 11 595
patients in England. Drug Saf 2008;31:313–23.
[58] Filippatos TD, Derdemezis CS, Gazi IF, et al. Orlistat-associated adverse effects and drug interactions: a critical review.
Drug Saf 2008;31:53–65.
[59] Tsertsvadze A, Yazdi F, Fink HA, et al. Oral sildenafil citrate (viagra) for erectile dysfunction: a systematic review and
meta-analysis of harms. Urology 2009.
[60] Hazell L, Cornelius V, Wilton LV, et al. The safety profile of tadalafil as prescribed in general practice in England: results
from a prescription-event monitoring study involving 16 129 patients. BJU Int 2009;103:506–14.
[61] Nurnberg HG, Hensley PL, Heiman JR, et al. Sildenafil treatment of women with antidepressant-associated sexual
dysfunction: a randomized controlled trial. JAMA 2008;300:395–404.
[62] de Groen PC, Lubbe DF, Hirsch LJ, et al. Esophagitis associated with the use of alendronate. N Engl J Med 1996;335:1016–
21.
[63] McGowan B, Bennett K, Barry M, et al. The utilisation and expenditure of medicines for the prophylaxis and treatment of
osteoporosis. Ir Med J 2008;101:38–41.
[64] Wallander MA, Johansson S, Ruigomez A, et al. Dyspepsia in general practice: incidence, risk factors, comorbidity and
mortality. Fam Pract 2007;24:403–11.
[65] Barrera BA, Wilton L, Harris S, et al. Prescription-event monitoring study on 13,164 patients prescribed risedronate in
primary care in England. Osteoporos Int 2005;16:1989–98.