Download NSAID-Induced Gastrointestinal Damage

®
GASTROENTEROLOGY BOARD REVIEW MANUAL
STATEMENT OF
EDITORIAL PURPOSE
The Hospital Physician Gastroenterology Board
Review Manual is a study guide for fellows and
practicing physicians preparing for board
examinations in gastroenterology. Each quarterly manual reviews a topic essential to the
current practice of gastroenterology.
PUBLISHING STAFF
PRESIDENT, GROUP PUBLISHER
NSAID-Induced
Gastrointestinal Damage
Series Editor:
Robert M. Craig, MD
Professor of Medicine, Division of Gastroenterology, Department of
Medicine, The Feinberg School of Medicine, Northwestern University,
Chicago, IL
Bruce M. White
EDITORIAL DIRECTOR
Debra Dreger
SENIOR EDITOR
Becky Krumm
Contributor:
James M. Scheiman, MD
Professor of Medicine, Division of Gastroenterology, University of
Michigan Medical Center, Ann Arbor, MI
ASSISTANT EDITOR
Rita E. Gould
EXECUTIVE VICE PRESIDENT
Barbara T. White
EXECUTIVE DIRECTOR
OF OPERATIONS
Jean M. Gaul
PRODUCTION DIRECTOR
Suzanne S. Banish
PRODUCTION ASSISTANT
Kathryn K. Johnson
ADVERTISING/PROJECT MANAGER
Patricia Payne Castle
SALES & MARKETING MANAGER
Deborah D. Chavis
Table of Contents
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Assessing Risk. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Pathogenesis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Preventive Strategies. . . . . . . . . . . . . . . . . . . . . . . . . 6
NOTE FROM THE PUBLISHER:
This publication has been developed without involvement of or review by the
American Board of Internal Medicine.
NSAID-Induced Small Bowel and Colonic
Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Endorsed by the
Association for Hospital
Medical Education
Cover Illustration by Scott Holladay
Copyright 2004, Turner White Communications, Inc., 125 Strafford Avenue, Suite 220, Wayne, PA 19087-3391, www.turner-white.com. All
rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means,
mechanical, electronic, photocopying, recording, or otherwise, without the prior written permission of Turner White Communications, Inc.
The editors are solely responsible for selecting content. Although the editors take great care to ensure accuracy, Turner White
Communications, Inc., will not be liable for any errors of omission or inaccuracies in this publication. Opinions expressed are those of the
authors and do not necessarily reflect those of Turner White Communications, Inc.
Gastroenterology Volume 10, Part 2 1
GASTROENTEROLOGY BOARD REVIEW MANUAL
NSAID-Induced Gastrointestinal Damage
James M. Scheiman, MD
INTRODUCTION
CASE PRESENTATION
It’s 5:30 AM, and you’ve just left the intensive care
unit following successful endoscopic therapy to control
a bleeding gastric ulcer in a 55-year-old man. He was
generally healthy, but he was taking “one of those new,
safer, NSAIDs” for knee osteoarthritis. He also had been
taking an enteric-coated aspirin at night, because “all
his friends take one, too.” He takes no other medications. At the time of the endoscopy, biopsies were obtained for Helicobacter pylori status.
EPIDEMIOLOGY OF NSAID-INDUCED ULCERS
Each year, more than 40 billion aspirin tablets are
consumed and more than 110 million prescriptions for
nonsteroidal anti-inflammatory drugs (NSAIDs) are
filled in the United States. At least one third of the prescriptions are for cyclooxygenase (COX)-2–specific inhibitors (coxibs) at a cost of $5 billion dollars/yearly.1
These drugs can have significant adverse gastrointestinal
(GI) effects ranging from symptoms such as nausea and
dyspepsia (ie, persistent pain or discomfort in the upper
abdomen) to serious ulcer complications such as bleeding and perforation. Among chronic NSAID users (defined as daily use for more than 1 year), the risk of developing a symptomatic ulcer, gastrointestinal bleed, or
perforation is 1% to 4%.2
Although there is a low probability that any individual
NSAID user will experience a drug-related complication,
the huge number of chronic users translates NSAID toxicity into a major health care problem. It is estimated that
more than 100,000 hospitalizations and up to 10,000 to
20,000 deaths each year in the United States can be
attributed to NSAID complications.2
There is no consensus among clinicians on how best
to weigh the potential clinical benefits of various antiinflammatory agents against the possibility of adverse
events associated with their use. Choosing between a
generic traditional NSAID and the safer but more costly medications in the COX-2–specific inhibitor class requires a framework for informed decision making.
2 Hospital Physician Board Review Manual
Considerations include the patient’s risk for toxicity, the
need for concurrent aspirin therapy, and whether the
patient is already taking a gastroprotective agent such as
a proton pump inhibitor (PPI). Patient risk factors
should drive clinical decision making to target these
medications in cost-effective manner.
ASSESSING RISK
NSAID users have an approximately 3-fold elevated
risk of an ulcer complication (ie, bleeding, perforation,
or a clinical event requiring hospitalization or causing
death) compared to nonusers. Although 1% to 4% of
chronic NSAID users develop an adverse event, individual risk varies considerably across the population. When
chronic NSAID therapy is necessary, it is essential to
assess a patient’s individual risk for complications.
Recognized risk factors (in order of their relative importance) are listed in Table 1.
RECENT OUTCOME STUDIES
Until the recent CLASS3 and VIGOR4 outcome studies
were completed evaluating the safety of coxibs (celecoxib
and rofecoxib, respectively) compared to traditional
NSAIDs, the only prospective data regarding the risk of
serious complications due to NSAIDs was from the
MUCOSA5 trial, which compared the outcome of rheumatoid arthritis (RA) patients taking non-aspirin NSAIDs
plus either misoprostol or placebo. 0.95% of patients on
non-aspirin NSAIDs plus placebo developed serious GI
complications within 6 months as compared to 0.57% of
patients on NSAIDs plus misoprostol. These findings are
consistent with the frequently quoted 2% to 4% risk
included in the prescribing information for prescription
NSAIDs. The CLASS and VIGOR studies confirmed these
rates—approximately 2% annually for complicated ulcers, and 4% annually for symptomatic ulcers.3,4
RISK FACTORS
Patients with a history of ulcer complications and those
taking concomitant anticoagulant therapy have the highest risk of developing NSAID-associated serious GI complications. Moderate risk factors include advanced age,
NSAID-Induced Gastrointestinal Damage
concomitant corticosteroid use, underlying major organ
impairment, the use of high-dose or multiple NSAIDs,
and arthritis-related disability. Gender6 and a history of
GI symptoms7 do not appear to predict increased risk.
The past occurrence of an NSAID-associated serious GI
complication is unequivocally associated with a markedly
increased likelihood of another complication with recurrent NSAID use. Unsurprisingly, the concomitant use
of traditional NSAIDs and anticoagulants markedly exacerbates the risk of GI bleeding and results in a 2.2-fold
increased risk of hospitalization.8 Although corticosteroids alone do not increase the risk of peptic ulcer disease, their use with NSAIDs leads to nearly a 2-fold
increase in risk of complications, and an even greater
increase in the risk of death.
Given the increasing use of aspirin to prevent cardiovascular disease and colorectal cancer, attention to the
toxicity of low-dose aspirin is particularly relevant. When
aspirin is added to an ongoing NSAID treatment regimen,
the risk of toxicity rises substantially. Doses of aspirin as low
as 10 mg daily deplete prostaglandins and can cause acute
GI toxicity.9 Aspirin doses as low as 75 mg daily have been
associated with a significantly increased risk of bleeding
gastric and duodenal ulcers in epidemiologic studies from
both the United States and the United Kingdom.10 The
risk of upper GI bleeding with low-dose aspirin in these
studies was increased 3-fold, and enteric coating and
buffering provided no protection from bleeding. These
findings support the notion that the costs and benefits of
prophylactic aspirin use should be carefully weighed in
every patient. Physicians should warn NSAID users that
aspirin use increases the risk of ulcers and should encourage these patients to consult a health care provider if they
are considering taking low-dose aspirin for cardiovascular
health.11 For patients who are concerned about their cardiovascular risk, such as the one presented here, a risk calculator designed for patient use is available at the National
Cholesterol Education Program Web site (http://hin.
nhlbi.nih.gov/atpiii/calculator.asp).
Increasing age is an independent predictor of experiencing an NSAID-associated GI complication.12 Clinically
significant major organ impairment was identified as an
independent risk factor in the MUCOSA trial. Patients
with cardiovascular disease have been shown to be at a
nearly 2-fold increased risk for GI complications due to
NSAID therapy, independent of aspirin use.
Symptoms, or the lack thereof, are not good predictors of NSAID complication risk. In one study, 58% of
patients admitted with an NSAID complication had no
antecedent dyspeptic symptoms compared with the
presence of symptoms in 75% of those with non-NSAID
complicated ulcers.7
Table 1. Risk Factors for Aspirin- and NSAIDAssociated Ulcer Complications*
Personal history of complicated ulcer disease
Concurrent use of more than 1 NSAID (including aspirin)
Use of high doses of NSAIDs
Concurrent use of an anticoagulant
Personal history of uncomplicated peptic ulcer disease
Age greater than 70 years
Concurrent use of steroids
NSAID = nonsteroidal anti-inflammatory drug.
*Listed in order of relative importance.
Meta-analyses have assessed the effect of different
types and dosages of NSAIDs on serious GI complications, using ibuprofen as the reference medication.
NSAIDs with increasing potency and duration of COX-1
inhibition were associated with increasing risks of serious
GI complications. The relative risk was greater than
2-fold higher with high- versus low-dose NSAID therapy.13 The data indicated a trend for ibuprofen being less
likely to cause serious complications than naproxen or
indomethacin, likely due to the lower dosages of ibuprofen frequently used by patients.13 These results likely
explain ibuprofen’s apparent safety when used appropriately at over-the-counter dosages. Several studies have
evaluated agents that bypass gastric absorption (eg, salsalate, nabumetone) or agents that are less potent COX-1
inhibitors (eg, etodolac, nabumetone, meloxicam).14
These trials demonstrated a significant reduction in
endoscopic gastric ulcers and erosions with salsalate,
etodolac, and nabumetone.
Meloxicam and nabumetone also have been associated with a low rate of symptomatic ulcers in analyses of
their respective clinical trial programs.14 However, the
results of these studies should be interpreted cautiously
for several reasons, including variability in the relationship of NSAID-associated endoscopic damage to clinical
events. Most importantly, these agents have not been
subjected to large-scale outcome trials designed to examine the incidence of serious GI complications (eg, bleeding, perforation, hospitalization, or death).
The COX-2–specific inhibitors were developed with
the goal of reducing ulcers and complications by sparing
inhibition of COX-1, which mediates prostaglandin levels.
At usual clinical doses, these agents do not inhibit COX-1.
As predicted, these drugs, even at very high doses, do not
significantly reduce upper GI prostaglandin levels.15 Clinical trials have demonstrated equivalent anti-inflammatory
efficacy of COX-2 specific inhibitors to commonly used
Gastroenterology Volume 10, Part 2 3
NSAID-Induced Gastrointestinal Damage
NSAIDs in arthritis patients.16 When studied with endoscopy, both celecoxib and rofecoxib produce rates of injury nearly equivalent to placebo.17 Rofecoxib also has
been shown to be equivalent to placebo in fecal blood loss
studies in humans. The reduction in serious GI complications observed in the CLASS and VIGOR clinical outcome trials is discussed later in this paper.
PATHOGENESIS
UPPER GASTROINTESTINAL DEFENSE MECHANISMS
In general, the key causes of upper GI injury to consider are NSAIDs and H. pylori infection. The upper GI
mucosa has in place a remarkable barrier that, under
normal conditions, withstands the hostile acidic environment of the gastric lumen. Indeed, the fundamental question with regard to understanding the development of NSAID-related ulcers is not so much why they
occur, but why they do not occur more frequently.
The complex elements that defend the gastroduodenal mucosa from damage are largely dependent on
endogenous prostaglandins synthesized in the upper gastrointestinal mucosa. COX is the rate-limiting catalytic
enzyme in prostaglandin production. At least 2 isoforms
of COX have been identified. COX-1 is constitutively
expressed in the GI tract, and COX-1–derived prostaglandins play an important role in the maintenance of
normal gastric and duodenal physiology. COX-2 is an
inducible form that is upregulated in areas of injury,
including H. pylori gastritis.18 However, COX-2 also is regulated in response to physiologic stimuli in other tissues,
including the kidney, brain, and reproductive tract. The
nonselective NSAIDs nonspecifically inhibit COX isoforms, leading to both beneficial (anti-inflammatory)
and toxic (GI bleeding) outcomes.
The stomach and duodenum are covered by a mucusbicarbonate barrier that is regulated by prostaglandins
and provides the primary defense against the acidic gastric lumen. The surface epithelium provides the “second
line” of gastroduodenal defense. Regeneration, the process by which larger epithelial defects (eg, ulcers) heal,
requires cellular proliferation, which is at least partly
dependent upon prostaglandins and growth factors.19
Although very little COX-2 is present in the intact stomach, prostaglandins derived from COX-2 induced in the
damaged stomach play an important role in ulcer healing, in part related to angiogenesis stimulated by growth
factors.20 Another key factor preventing mucosal injury is
maintenance of microvascular blood flow, also regulated
by COX-1–derived prostaglandins.
4 Hospital Physician Board Review Manual
NSAID MECHANISMS OF INJURY
Although the mechanisms by which NSAIDs cause
ulcers remain incompletely understood, they are
known to involve both topical injury and systemic effects mediated by depletion of endogenous prostaglandins. Analogous to H. pylori –associated ulcer disease (ie,
a high exposure risk coupled with a low absolute risk),
the biologic basis of the factors that place certain individuals at an increased risk for NSAID-related ulcers
remains unknown.
Local Effects
Aspirin and most NSAIDs undergo “ion trapping”
within the proximal GI mucosa, causing direct cellular
injury. NSAIDs also directly attenuate the hydrophobic, or
“non-wettable,” properties of the mucous barrier independent of prostaglandin-mediated actions.21 Although
topical effects can largely be prevented by administering
enteric-coated NSAID formulations or prodrugs, the failure of these approaches to reduce the incidence of symptomatic NSAID-induced ulcers demonstrates that topical
injury is not the critical determinant of NSAID-induced
injury. For example, parenteral administration of an
NSAID such as ketorolac may lead to ulcer complications.
Certain NSAIDs, including indomethacin, piroxicam,
oxaprozin, and ketorolac, also undergo an extensive
enterohepatic recirculation, resulting in repeated exposure to the GI mucosa and increased toxicity.19
Systemic Effects
The clinically important adverse effects of NSAIDs—
ulcers with an increased risk of complications—appear
to be due largely to systemic actions. Nonselective COX
inhibition with a resultant decrease in the endogenous
prostaglandins that are critical to mucosal defense is
thought to be the most important mechanism of action.
In addition, platelet COX-1 is inhibited irreversibly by
aspirin and for as long as 18 hours by other NSAIDs. The
impaired platelet function may potentiate bleeding
from both the upper and lower GI tracts.
COX-1 inhibition leads not only to quantitative but
also to qualitative decreases in mucous barrier function.
Because prostaglandin deficiency impairs regenerative
responses, erosions created by direct topical injury are
exposed to acid with the mucosa in a vulnerable state.
Ulcerations tend to occur in areas of decreased blood
flow; NSAIDs induce microvascular ischemia, in part, by
promoting neutrophil adherence in the microcirculation, in addition to reduced prostaglandin production.
Nitric oxide (NO) has a role in the maintenance of
epithelial integrity that is related to its ability to maintain
mucosal blood flow. In animal models, inhibition of NO
NSAID-Induced Gastrointestinal Damage
synthesis exacerbates NSAID injury and thus NO donors
reduce NSAID toxicity.22 Nitric oxide–releasing aspirin
and NSAID derivatives, currently not available in clinical
practice, cause reduced damage despite marked inhibition of prostaglandin levels. This strategy may be of
greatest value when aspirin therapy is needed because it
would allow the utilization of a non-ulcerogenic antiplatelet agent; clinical trials supporting the animal data
are anxiously awaited.
The pathophysiology that formed the mechanistic
basis for the “COX-2 hypothesis” is that gastroduodenal
mucosal prostaglandin deficiency initiates ulceration,
and virtually all mucosal prostaglandins are derived from
COX-1. Recent animal data, however, challenge this
notion and indicate that for gastric ulceration to occur,
both COX-1 and COX-2 must be inhibited.23 In that study,
selective inhibition of COX-1 alone or COX-2 alone did
not cause gastric damage, whereas inhibition of both
COX isoforms produced considerable gastric ulceration.
Thus, the explanation for reduced GI toxicity for COX2–specific inhibitors may be their lack of dual COX inhibition rather than simply their COX-1–sparing effects. In
this theoretical framework, it may be anticipated that the
addition of low-dose aspirin (primarily a COX-1 inhibitor at a low dose) to a COX-2 inhibitor creates an ulcerogenic dual COX inhibitor, as in the case patient.
Finally, acid plays an important secondary role in
NSAID-induced ulceration. The role of acid is supported by recent data demonstrating the efficacy of PPIs in
the treatment and prevention of NSAID damage.24
These data suggest that topical injury is the first step in
NSAID ulceration, with secondary acid injury occurring
in the setting of prostaglandin depletion and impaired
repair mechanisms, all leading to the development of
clinically important ulceration (Figure 1).
ROLE OF HELICOBACTER PYLORI INFECTION
In the absence of NSAID use, H. pylori is accepted as
the cause of most ulcers; its role in the pathogenesis of
NSAID-associated ulcers, however, remains controversial.
The pathophysiologic mechanisms of NSAID-associated
and H. pylori –associated ulcers are distinct: NSAID ulcers
occur without gastritis (the injury associated with NSAIDs
occurs with little to no microscopic inflammation), whereas H. pylori ulcers occur in the setting of diffuse inflammation. The degree of inflammation is likely related to the
virulence of the H. pylori strain and host factors. Although
NSAIDs decrease prostaglandin synthesis and H. pylori
increases the synthesis of prostaglandins, there is little evidence that this increase provides protection from ulceration. Most importantly, it is clear from epidemiologic studies that H. pylori infection is not a required cofactor for
NSAID-associated ulcers because these ulcers may occur
in the absence of H. pylori. In patients who use NSAIDs
chronically, the prevalence of H. pylori infection appears
to be similar to those with or without ulcers.25
Whether eradication of H. pylori protects against
NSAID-associated ulcers is another area of controversy.
Although a study from Hong Kong found that NSAIDnaive patients who had successful H. pylori eradication had
fewer ulcers, follow-up studies in non–NSAID-naive patients failed to confirm these results, supporting the independence of these two ulcerogens.26,27 These same investigators demonstrated that eradication of H. pylori alone
was insufficient to prevent recurrence of a bleeding
NSAID-associated ulcer in patients with a history of NSAIDassociated ulcer bleeding.28 In another arm of that study,
eradication appeared as effective as omeprazole maintenance therapy for individuals taking low-dose aspirin for
cardiovascular protection.28 A study of similar design from
a different group of Hong Kong investigators supported
the value of additional prophylaxis with a PPI. However,
some of the ulcer recurrences were associated with reinfection, a common problem in endemic areas.29
In summary, because H. pylori and NSAIDs appear to
produce ulcers by different mechanisms, and in the
absence of sound evidence suggesting a therapeutic
advantage, testing for H. pylori does not appear to be costeffective for all patients starting on NSAID therapy.30
Patients with a history of peptic ulcer disease, such as the
case patient, should be tested for H. pylori and treated
with antibiotics if the test is positive to reduce recurrence
of H. pylori –associated ulcers. Eradication of the infection
does not sufficiently reduce NSAID ulcer risk to preclude
the need for additional therapies for reducing risk.
MANAGEMENT
ALTERNATIVES FOR PAIN RELIEF
The traditional approach to a chronic NSAID user who
bleeds from an ulcer is to stop the NSAID until one would
expect the ulcer to be healed (typically 8 weeks). Because
patients taking NSAIDs for chronic pain may suffer functional losses from increased pain when the NSAIDs are
stopped, an alternative pain reliever may be desirable. The
typical choice for pain relief in a patient off NSAID therapy is high doses of acetaminophen (not to exceed 4000 mg
daily, in divided doses), assuming that heavy alcohol use
or impaired liver or renal function does not make this
therapy hazardous. Small doses of a short-acting narcotic
medication (eg, propoxyphene, codeine, hydrocodone)
may be used temporarily for more severe pain. The
Gastroenterology Volume 10, Part 2 5
NSAID-Induced Gastrointestinal Damage
NSAID
Epithelial injury
Prostaglandinmediated effects
• ↓ Mucin
• ↓ Surface active phospholipids
• ↓ Bicarbonate secretion
• ↓ Mucosal proliferation
Direct effects
Microvascular injury
• Increased adhesion molecule expression
Figure 1. Pathogenesis of NSAID-induced
ulcers. NSAID = nonsteroidal antiinflammatory drug. (Adapted from Scheiman JM. NSAIDs, gastrointestinal injury,
and cytoprotection. Gastroenterol Clin
North Am 1996;25:279, with permission
from Elsevier.)
• Neutrophil adherence
• Stasis
• Microvascular ischemia
• Free radical formation
Erosion
Chronic ulcer
non-narcotic analgesic tramadol also may be tried. Other
approaches to pain management, such as topical medications, intra-articular corticosteroids or hyaluronic acid injection, can be used concurrently with oral analgesics.
ULCER HEALING
Several anti-ulcer therapies may be used for healing
NSAID-induced ulcers. When NSAIDs are continued, H2
receptor antagonists (H2RAs) have impaired effectiveness
for healing ulcers compared to when the NSAID is discontinued. In a study of ulcer patients treated with ranitidine 150 mg twice daily, 63% of gastric ulcer patients who
continued NSAID use were healed at 8 weeks compared
with 95% of those who discontinued NSAIDs. Duodenal
ulcers healed in 84% of patients continuing NSAIDs and
in 100% of those who discontinued NSAIDs.31
PPIs are superior to either H2RAs or misoprostol for
healing NSAID ulcers in the setting of continued NSAID
use. In a study of patients randomized to omeprazole
20 mg daily, omeprazole 40 mg daily, or ranitidine 150 mg
twice daily, the rates of healing at 8 weeks were higher in
those treated with omeprazole compared to those treated
with ranitidine. The higher PPI dose was not superior to
the lower dose.32 Similar data exist for other PPIs.33 Omeprazole 20 or 40 mg once daily for 8 weeks also has been
6 Hospital Physician Board Review Manual
demonstrated to be significantly more effective than misoprostol for gastric ulcer healing.24
When patients can discontinue NSAIDs, all forms of
anti-ulcer therapy are effective. If the patient has a large
or a complicated ulcer, many clinicians use a PPI once
or twice daily, based upon evidence that larger ulcers
heal faster with more potent acid suppression. In a patient with an uncomplicated NSAID-related ulcer who
is able to discontinue the NSAID, any anti-ulcer therapy will be effective.
It is the standard of care that all patients with peptic
ulcer disease, whether taking NSAIDs or not, undergo
testing for and treatment of H. pylori infection. Literature
suggests that because healing occurs more quickly with
PPIs when the organism is present due to improved acid
suppression, it is reasonable to defer eradication until
after a course of therapy to heal the ulcer.27 This may
improve compliance as it can be done electively in the
outpatient setting after the bleeding hospitalization.
PREVENTIVE STRATEGIES
At present, data from rigorously designed clinical
and economic trials are not available to make accurate
NSAID-Induced Gastrointestinal Damage
head-to-head comparisons between different strategies
to prevent ulcers and complications.
PROSTAGLANDIN REPLACEMENT
Prostaglandin depletion is central to the development
of NSAID-induced ulcers, and replacement therapy with
a synthetic prostaglandin reduces NSAID toxicity.
Placebo-controlled studies have demonstrated the efficacy of misoprostol for the prevention of endoscopic ulcers
in NSAID-using arthritis patients.34 Misoprostol is the only
agent currently approved by the US Food and Drug
Administration (FDA) for the prevention of NSAIDassociated serious GI complications (eg, bleeding, perforation, obstruction). This approval is based upon the results of the MUCOSA trial,5 in which a 40% relative risk
reduction was observed. Misoprostol appears to be costeffective in high-risk patients only, as the number needed
to treat is 264 for 6 months to prevent a single definite
upper GI complication.35 In certain subgroups with an increased risk of serious upper GI complications, the use of
misoprostol would be more cost-effective and associated
with a lower number needed to treat.36 Those who benefited the most from misoprostol were patients with a previous history of GI bleed (50% risk reduction), history of
previous peptic ulcer disease (52% risk reduction), significant cardiovascular disease (38% risk reduction), significant functional disability (87% risk reduction), and those
who required concomitant antacid use (48% risk reduction).36 Despite this agent being the only FDA-approved
regimen for prevention of NSAID ulcers and their complications, it rarely is used due to adverse effects of diarrhea and cramps, a consequence of the prostaglandin
analogue on small bowel motility and secretion.
ACID SUPPRESSION
Proton Pump Inhibitors
Omeprazole proved effective in primary prophylaxis
of NSAID-induced ulcers compared to placebo in a
study of 169 patients requiring continuous NSAID therapy.37 After 6 months, 78% of the omeprazole group
remained in remission compared with 53% in the
placebo group (P = 0.004). There were 3 gastric ulcers
and no duodenal ulcers in the omeprazole group compared with 8 and 3, respectively, in the placebo group.
In the ASTRONAUT study,32 patients requiring continuous NSAID therapy were randomized following
ulcer healing to ranitidine 150 mg twice daily or omeprazole 20 mg daily. Gastric ulcers recurred in 5.2% of
the omeprazole-treated patients versus 16.3% of the
ranitidine-treated group (P < 0.001). For duodenal
ulcers, there was a 0.5% versus a 4.2% (P = 0.02) rate of
recurrence in the two groups, respectively. In the
OMNIUM study,24 omeprazole 20 mg once daily was
compared to misoprostol 200 µg twice daily and placebo over 6 months in a cohort of patients with recently
healed ulcers. In this study, 32% of patients taking
placebo developed a gastric ulcer relapse compared to
10% in the misoprostol group and 13% in the omeprazole group. Duodenal ulcers relapsed in 12% of those
given placebos, 10% of those given misoprostol, and
3% of those given omeprazole. Omeprazole also was
superior in the maintenance of overall remission, largely because of its ability to improve NSAID-associated
dyspepsia and overall quality of life.
In a randomized, double-blind, placebo-controlled,
prospective study of 458 long-term NSAID users with
endoscopically documented, H. pylori –negative, recently
healed gastric ulcers, Graham et al38 found that patients in
the lansoprazole 15- or 30-mg groups remained free from
gastric ulcer significantly longer than those who received
placebo. Furthermore, whereas the misoprostol-treated
group remained free of gastric ulcer longer than the lansoprazole groups at either the 15-mg or 30-mg dose, 31%
in the misoprostol group prematurely discontinued use of
the study medication due to reported adverse drugrelated events, compared with only 10% and 7%, respectively, in the lansoprazole groups.
Recently completed studies with esomeprazole 20 mg
and 40 mg have demonstrated the efficacy of this agent
to reduce NSAID dyspepsia and associated reductions in
quality of life. In studies of H. pylori –negative patients taking NSAIDs who were at increased risk of developing
ulcers (age > 60 years or recent gastric ulcer or duodenal
ulcer), there were greater than 50% reductions in ulcer
recurrence. Strikingly, among coxib users taking concomitant esomeprazole therapy, very few ulcers were
seen, whereas coxib users taking concomitant placebo
experienced similar rates of ulcer recurrence as those
taking traditional NSAIDs.39 This finding emphasizes the
value of using PPI cotherapy even when taking coxibs.
H2 Receptor Antagonists
The level of acid suppression provided by traditional
doses of H2RAs does not prevent most NSAID ulcers;
only ulcer formation in the duodenum is significantly
reduced. In a single study, however, when H2RAs were
given at double the traditional dose (twice daily instead
of once daily), they were effective at reducing both gastric and duodenal ulcers.40 There are no studies comparing high doses of H2 blockers to misoprostol or PPIs
for the prevention of NSAID ulcers. Practically speaking, given the compliance concerns with twice-daily
dosing, PPI therapy is generally preferable to H2RAs in
this clinical setting.
Gastroenterology Volume 10, Part 2 7
NSAID-Induced Gastrointestinal Damage
Table 2. Incidence of GI Events in the VIGOR Study
Rate (per 100 patient-yr)
Rofecoxib
(n = 4047)
Naproxen
(n = 4029)
Relative Risk
(95% CI)
Relative Risk
Reduction, %
P Value
Clinical upper GI Event
2.1
4.5
0.46 (0.33–0.64)
54
< 0.001
Complicated upper GI Event
0.6
1.4
0.43 (0.24–0.78)
57
0.005
Any GI bleeding
1.2
3.0
0.38 (0.25–0.57)
62
< 0.001
Event
Data from 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.
CI = confidence interval; GI = gastrointestinal.
COX-2–SPECIFIC ANTI-INFLAMMATORY DRUGS
Coxibs versus Nonselective NSAIDs
Deeks and colleagues41 recently reviewed the results
of 4 trials that compared celecoxib with other NSAIDs
on the incidence of ulcers detected by endoscopy in
patients not receiving aspirin and those receiving aspirin
at doses up to 325 mg/day. While the reduction in ulcer
risk was significant with celecoxib versus other NSAIDs
among those individuals not receiving aspirin (relative
risk = 0.27; P = 0.001), no significant difference in ulcer
risk was observed between those individuals receiving
aspirin and celecoxib and those receiving aspirin with
another NSAID (relative risk = 0.49; P = 0.54).
Two 6-month, placebo-controlled, endoscopy studies have been performed with rofecoxib 25 mg and
50 mg compared to ibuprofen 800 mg 3 times daily. In
the 2 studies, a total of 1517 patients were randomized.
In a study of 742 patients,42 at 12 weeks, 7.3% of patients
on placebo developed an endoscopic ulcer compared
to 4.7% on 25 mg of rofecoxib, 8.1% on 50 mg of rofecoxib, and 28.5% in the ibuprofen group. In a second
study of identical design,17 similar safety with rofecoxib
was observed. When the two studies were combined,
the incidence of endoscopic ulcers in patients taking
rofecoxib met predefined criteria for equivalence to
placebo.17 However, similar to the combination of aspirin with celecoxib,41 a recent endoscopic study with
rofecoxib found that the benefits of the coxib were lost
with the addition of aspirin.43 When evaluated by endoscopy over a 12-week study period, there was no difference in the cumulative incidence of endoscopic gastroduodenal ulcers in patients taking low-dose (81 mg)
enteric-coated aspirin with rofecoxib 25 mg as compared to those taking ibuprofen 2400 mg daily.
Because the relationship between endoscopic injury
and serious GI complication rates among NSAID users
remains controversial, manufacturers of both leading
coxibs performed long-term safety trials. In the VIGOR
8 Hospital Physician Board Review Manual
study, the GI safety of rofecoxib 50 mg daily was compared to naproxen 500 mg twice daily in more than
7000 rheumatoid arthritis patients worldwide.4 The primary and secondary endpoints of this study were the
occurrence of symptomatic and complicated upper GI
events, respectively. The results of the trial are shown in
Table 2. The coxib was associated with a 50% to 60%
reduction in ulcers, bleeding, or both.
In the CLASS study, 8000 patients worldwide with
osteoarthritis (90%) or rheumatoid arthritis (10%) were
randomized to celecoxib 400 mg twice daily, Ibuprofen
800 mg 3 times daily, or diclofenac 75 mg twice daily for
up to 12 months.3 In this trial, which included patients
taking low-dose (325 mg or less) aspirin for cardiovascular prophylaxis, the primary endpoint was the development of complicated upper GI events, with symptomatic
ulcer development as the secondary endpoint. The primary endpoint of ulcer complications did not reach significance in the CLASS trial (Figure 2); this was largely
attributable to the impact of low-dose aspirin (Figure 3).
When the entire 12 months of the trial became available,
the primary endpoint did not reach significance even
when those taking aspirin were excluded, likely reflecting
numerous methodologic problems with the design of the
trial, including the high drop-out rate that was observed
in the study.44 Both endoscopic and clinical outcome
studies have demonstrated that for those patients receiving coxibs who have a need for an antiplatelet agent and
also must take low-dose aspirin, the coxib protection
from ulceration is markedly impaired.
Coxibs versus an NSAID with a PPI
In a trial from Hong Kong, Chan et al45 randomized
patients with a history of ulcer bleeding to either a PPI
(omeprazole 20 mg daily) plus a traditional NSAID
(diclofenac 75 mg twice daily) or celecoxib 200 mg twice
daily for 6 months. All patients had H. pylori eradicated or
were uninfected. In this large study, no significant
NSAID-Induced Gastrointestinal Damage
P = 0.02
Annualized incidence (%)
4
3.54
Conventional NSAIDs
3
P = 0.09
2
1
0
2.08
1.45
0.76
Complications
5
Incidence
(events/100 patient-yr)
Celecoxib
4
3
P = 0.618
Celecoxib
NSAIDs
P = 0.02
2
1
0
Complications and
symptomatic ulcers
Figure 2. Incidence of symptomatic upper gastrointestinal ulcers
and ulcer complications in the CLASS trial at 6 months (all patients).
(Adapted with permission from Silverstein FE, Faich G, Goldstein JL,
et al. Gastrointestinal toxicity with celecoxib vs nonsteroidal antiinflammatory drugs for osteoarthritis and rheumatoid arthritis: the
CLASS study. A randomized controlled trial. Celecoxib Long-term
Arthritis Safety Study. JAMA 2000;284:1251. Copyright © 2000,
American Medical Association. All rights reserved.)
difference was found in the number of patients presenting with an ulcer bleed: 4.9% on the coxib and 6.4% on
diclofenac with the PPI. Although the study was not powered to demonstrate equivalence, the results suggest both
that the strategies are similar in efficacy and that neither
approach is sufficient in this high-risk group of patients.
Most experts recommend that these highest-risk patients
receive cotherapy with a PPI (or misoprostol) in addition
to a coxib to provide multiple risk-reducing strategies.
Formulating a Cost-Effective Preventive Strategy
Key components of the overall cost of disease management with NSAIDs include physician visits, medication expenditure, and endoscopies related to the development of GI side effects such as dyspepsia. Coxibs
appear to have fewer nuisance GI side effects compared
to traditional NSAIDs, although such side effects are
more frequent than with placebo. Rofecoxib was noted
to have significantly fewer GI symptoms and need for
GI medication cotherapy in clinical trials of up to
6 months’ duration.46 In general, the incidence of these
adverse events and rates of anti-inflammatory drug
discontinuation with rofecoxib were intermediate
between those of traditional NSAIDs and those of placebo. In a trial specifically designed to address the question of GI tolerability, the rates of cumulative discontinuation due to GI adverse events were statistically
significantly lower in the rofecoxib group than in the
naproxen group (5.9% versus 8.1%), as were rates of
cumulative use of medication to treat GI symptoms
Aspirin
No aspirin
Figure 3. Ulcers and ulcer complications in the CLASS trial at
12 months, according to aspirin use. P value by log rank test.
(Adapted from Lefkowith JB. Safety profile of celecoxib. CLASS:
Celecoxib Long-Term Arthritis Safety Study. Slide no. 110. Available at www.fda.gov/ohrms/dockets/ac/01/slides/3677s1_01_
sponsor.pdf. Accessed 26 Mar 2004.)
(9.1% versus 11.20%).47 Although the superior GI tolerability was statistically significant, however, it was modest in absolute terms, limiting the economic value of the
coxib because of its higher cost.
Because coxibs are as effective as traditional NSAIDs
but safer, they represent a logical alternative on the basis
of both clinical and economic considerations if the traditional NSAID is being given with another medication,
such as an H2RA or PPI, that can be discontinued if the
patient is switched to a coxib. Patients with a history of
ulcer complications and concomitant anticoagulant
therapy have the highest risk of developing NSAIDassociated serious GI complications. These patients
would be expected to derive the greatest overall benefit
from the reduced GI risk associated with coxibs. Moderate risk factors include advanced age, corticosteroid use,
chronic major organ impairment (particularly cardiovascular disease), and the use of high-dose or multiple
NSAIDs. Coxibs in these patients who are at aboveaverage risk are likely to be cost effective, particularly if
more than 1 of these risk factors is present.48 Additional
economic benefits will be achieved if additional medications are not required, or if adequate symptom control
occurs with the addition of a low-cost generic antisecretory drug to the coxib therapy.
Management of the patient requiring low-dose aspirin is more complex. Testing and treatment of H. pylori
may prove beneficial in these patients.28 In high-risk patients, concomitant aspirin therapy may mandate antisecretory cotherapy to prevent bleeding, regardless of
whether a coxib is being used. An algorithm for preventing GI complications in patients who require aspirin therapy (with or without anti-inflammatory drugs) is shown
in Figure 4.
Gastroenterology Volume 10, Part 2 9
NSAID-Induced Gastrointestinal Damage
Patient requiring low-dose aspirin
Anti-inflammatory drug not required
Anti-inflammatory drug required
Prior history of ulcer or GI bleeding
Prior history of ulcer or GI bleeding
Yes
No
No
Yes
Eradicate H. pylori
if present
Monitor for signs and
symptoms of bleeding
No H. pylori testing
Eradicate H. pylori if present
GI prophylactic medication added based upon ulcer
risk independent of coxib or traditional NSAID choice
Figure 4. Algorithm for preventing gastrointestinal (GI) complications in patients requiring aspirin therapy. NSAID = nonsteroidal antiinflammatory drug.
NSAID-INDUCED SMALL BOWEL AND COLONIC
DISEASE
Clinically important NSAID-induced GI damage beyond the duodenum has been recognized. Autopsy
studies have confirmed a marked increase in nonspecific small bowel ulceration in NSAID users (8.4%)
compared to control subjects (0.6%).49 Inflammation
and increased blood and protein loss from the small
bowel occur in some chronic NSAID users and is known
as NSAID enteropathy. Rare reports suggest that intestinal ulcers can become chronic and lead to strictures.
Recent studies utilizing wireless capsule endoscopy have
confirmed a high incidence of endoscopic damage in
NSAID users. These studies have shown a surprising
background prevalence of lesions, which is increased
following administration of a traditional NSAID.50,51
Coxibs were shown to cause far fewer lesions, and cotherapy with PPI did not, as was expected, reduce this
damage. The clinical relevance of these lesions remain
uncertain; in the analysis of the VIGOR trial, however,
bleeding events attributed to the lower GI tract represented approximately one third of the events, with the
risk reduction for rofecoxib compared to naproxen
similar to that seen in the upper tract.52
NSAIDs also may have adverse effects on the colon,
including nonspecific ulcerations in the cecum and
transverse and sigmoid colon. NSAIDs may induce a
colitis whose appearance may be confused with inflammatory bowel disease (IBD); however, the biopsy
changes are nonspecific and are inconsistent with IBD.
Rectal administration of NSAIDs can cause proctitis. In
10 Hospital Physician Board Review Manual
case-control studies, preexisting lesions (eg, diverticula) have been reported to bleed or perforate with increased frequency in NSAID users. NSAIDs also have
been implicated in the relapse of quiescent IBD.
The pathogenesis of intestinal injury is thought to be
due to direct effects of NSAIDs to alter cellular metabolism in enterocyte mitochondria with subsequent loss of
cellular integrity. In this hypothesis, cyclooxygenase inhibition plays a secondary role by impairing prostaglandindependent healing mechanisms. Increased permeability
due to a damaged barrier then allows luminal aggressive
agents (eg, bacteria, bile) to further perpetuate injury.
Drugs such as indomethacin or piroxicam, which undergo enterohepatic recirculation, appear to cause the most
injury by virtue of their repeated exposure to the mucosa. It has been observed that metronidazole and
sulfasalazine, but not misoprostol, can decrease the
inflammation and blood loss due to NSAID enteropathy
without affecting intestinal permeability.
Clinically significant intestinal injury due to NSAIDs is
rare, and there is little information to guide patient management. Coxibs would be anticipated to cause less injury
in most patients,53 except those with pre-existing intestinal damage (eg, those with IBD). Although there have
been reports of the successful use of coxibs in this patient
population, these agents should be used with caution
owing to concern of exacerbating the existing disease.
REFERENCES
1. Laine L. Approaches to nonsteroidal anti-inflammatory
drug use in the high-risk patient. Gastroenterology
NSAID-Induced Gastrointestinal Damage
2001;120:594–606.
2. Wolfe MM, Lichtenstein DR, Singh G. Gastrointestinal
toxicity of nonsteroidal antiinflammatory drugs [published erratum appears in N Engl J Med 1999;341:548].
N Engl J Med 1999;340:1888–99.
3. Silverstein FE, Faich G, Goldstein JL, et al. Gastrointestinal
toxicity with celecoxib vs nonsteroidal anti-inflammatory
drugs for osteoarthritis and rheumatoid arthritis: the
CLASS study. A randomized controlled trial. Celecoxib
Long-term Arthritis Safety Study. JAMA 2000;284:1247–55.
4. 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.
5. Silverstein FE, Graham DY, Senior JR, et al. Misoprostol
reduces serious gastrointestinal complications in patients
with rheumatoid arthritis receiving nonsteroidal antiinflammatory drugs. A randomized, double-blind,
placebo-controlled trial. Ann Intern Med 1995;123:241–9.
6. Gabriel SE, Jaakkimainen L, Bombardier C. Risk for serious gastrointestinal complications related to use of nonsteroidal anti-inflammatory drugs. A meta-analysis. Ann
Intern Med 1991;115:787–96.
7. Armstrong CP, Blower AL. Non-steroidal anti-inflammatory drugs and life threatening complications of peptic
ulceration. Gut 1987;28:527–32.
8. Shorr RI, Ray WA, Daugherty JR, Griffin MR. Concurrent
use of nonsteroidal anti-inflammatory drugs and oral
anticoagulants places elderly persons at high risk for hemorrhagic peptic ulcer disease. Arch Intern Med 1993;
153:1665–70.
9. Cryer B, Feldman M. Effects of very low dose daily, longterm aspirin therapy on gastric, duodenal, and rectal
prostaglandin levels and on mucosal injury in healthy
humans. Gastroenterology 1999;117:17–25.
10. U.S. Preventative Services Task Force. Aspirin for the primary prevention of cardiovascular events: recommendation and rationale. Ann Intern Med 2002;136:157–60.
11. Hayden M, Pignone M, Phillips C, Mulrow C. Aspirin for
the primary prevention of cardiovascular events: a summary of the evidence for the U.S. Preventive Services
Task Force. Ann Intern Med 2002;136:161–72.
12. Griffin MR, Piper JM, Daugherty JR, et al. Nonsteroidal
anti-inflammatory drug use and increased risk for peptic
ulcer disease in elderly persons. Ann Intern Med 1991;
114:257–63.
13. Henry D, Lim LL, Garcia Rodriguez LA, et al. Variability
in risk of gastrointestinal complications with individual
non-steroidal anti-inflammatory drugs: results of a collaborative meta-analysis. BMJ 1996;312:1563–66.
14. Delvalle J, Chey WD, Scheiman JM. Acid peptic disorders. In: Textbook of gastroenterology, 4th ed. Yamada T,
editor. Philadelphia: Lippincott; 2003:1321–76.
15. Scheiman JM, Greenson JK, Lee J, Cryer B. Effect of
cyclooxygenase-2 inhibition on human Helicobacter pylori
gastritis: mechanisms underlying gastrointestinal safety
and implications for cancer chemoprevention. Aliment
Pharmacol Ther 2003;17:1535–44.
16. Feldman M, McMahon AT. Do cyclooxygenase-2 inhibitors provide benefits similar to those of traditional
nonsteroidal anti-inflammatory drugs with less gastrointestinal toxicity [published erratum Ann Intern Med
2000;132:1011]? Ann Intern Med 2000;132:134–43.
17. Hawkey C, Laine L, Simon T, et al. Comparison of the
effect of rofecoxib (a cyclooxygenase 2 inhibitor), ibuprofen, and placebo on the gastroduodenal mucosa of
patients with osteoarthritis: a randomized, double-blind,
placebo-controlled trial. The Rofecoxib Osteoarthritis
Endoscopy Multinational Study Group. Arthritis Rheum
2000;43:370–7.
18. McCarthy CJ, Crofford LJ, Greenson J, Scheiman JM.
Cyclooxygenase-2 expression in gastric antral mucosa
before and after eradication of Helicobacter pylori infection. Am J Gastroenterol 1999;94:1218–23.
19. Scheiman JM. NSAIDs, gastrointestinal injury, and cytoprotection. Gastroenterol Clin North Am 1996;25:279–98.
20. Jones MK, Wang H, Peskar BM, et al. Inhibition of angiogenesis by nonsteroidal anti-inflammatory drugs: insight
into mechanism and implications for cancer growth and
ulcer healing. Nat Med 1999;5:1418–23.
21. Lichtenberger LM, Wang ZM, Romero JJ, et al. Nonsteroidal anti-inflammatory drugs (NSAIDs) associate
with zwitterionic phospholipids: insight into the mechanism and reversal of NSAID-induced gastrointestinal
injury. Nat Med 1995;1:154–8.
22. Wallace JL, Reuter B, Cicala C, et al. Novel nonsteroidal
anti-inflammatory drug derivatives with markedly reduced ulcerogenic properties in the rat. Gastroenterology 1994;107:173–9.
23. Wallace JL, McKnight W, Reuter BK, Vergnolle N. NSAIDinduced gastric damage in rats: requirement for inhibition
of both cyclooxygenase 1 and 2. Gastroenterology 2000;
119:706–14.
24. Hawkey CJ, Karrasch JA, Szczepanski L, et al. Omeprazole compared with misoprostol for ulcers associated
with nonsteroidal antiinflammatory drugs. Omeprazole
versus Misoprostol for NSAID-induced Ulcer Management (OMNIUM) Study Group. N Engl J Med 1998;
338:727–34.
25. Chan FK. COX-2 inhibition, H. pylori infection and the
risk of gastrointestinal complications. Curr Pharm Des
2003;9:2213–9.
26. Chan FK, Sung JJ, Chung SC, et al. Randomized trial of
eradication of Helicobacter pylori before non-steroidal antiinflammatory drug therapy to prevent peptic ulcers.
Lancet 1997;350:975–9.
27. Hawkey CJ, Tulassay Z, Szczepanski L, et al. Randomized
controlled trial of Helicobacter pylori eradication in patients on non-steroidal anti-inflammatory drugs: HELP
NSAIDs study. Helicobacter Eradication for Lesion Prevention [published erratum appears in Lancet 1998;352:
1634]. Lancet 1998;352:1016–21.
28. Chan FK, Chung SC, Suen BY, et al. Preventing recurrent
upper gastrointestinal bleeding in patients with
Gastroenterology Volume 10, Part 2 11
NSAID-Induced Gastrointestinal Damage
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
Helicobacter pylori infection who are taking low-dose
aspirin or naproxen. N Engl J Med 2001;344:967–73.
Lai KC, Lam SK, Chu KM, et al. Lansoprazole for the prevention of recurrences of ulcer complications from longterm low-dose aspirin use. N Engl J Med 2002;346:2033–8.
Scheiman JM, Bandekar RR, Chernew ME, Fendrick
AM. Helicobacter pylori screening for individuals requiring
chronic NSAID therapy: a decision analysis. Aliment
Pharmacol Ther 2001;15:63–71.
Lancaster-Smith MJ, Jaderberg ME, Jackson DA. Ranitidine in the treatment of non-steroidal anti-inflammatory
drug associated gastric and duodenal ulcers. Gut 1991;32:
252–5.
Yeomans ND, Tulassay Z, Juhasz L, et al. A comparison of
omeprazole with ranitidine for ulcers associated with nonsteroidal antiinflammatory drugs. Acid Suppression Trial:
Ranitidine versus Omeprazole for NSAID-associated
Ulcer Treatment (ASTRONAUT) Study Group. N Engl J
Med 1998;338:719–26.
Agrawal NM, Campbell DR, Safdi MA, et al. Superiority
of lansoprazole vs ranitidine in healing nonsteroidal antiinflammatory drug-associated gastric ulcers: results of a
double-blind, randomized, multicenter study. NSAIDAssociated Gastric Ulcer Study Group. Arch Intern Med
2000;160:1455–61.
Graham DY, White RH, Moreland LW, et al. Duodenal
and gastric ulcer prevention with misoprostol in arthritis
patients taking NSAIDs. Misoprostol Study Group. Ann
Intern Med 1993;119:257–62.
Scheiman JM. Meta analysis: Misoprostol reduced
NSAID-induced gastrointestinal injury. ACP J Club 1997;
126:36. [Comment on Koch M, Dezi A, Ferrario F,
Capurso I. Prevention of nonsteroidal anti-inflammatory
drug-induced gastrointestinal mucosal injury. A metaanalysis of randomized controlled clinical trials. Arch
Intern Med 1996;156:2321.]
Simon LS, Hatoum HT, Bittman RM, et al. Risk factors
for serious nonsteroidal-induced gastrointestinal complications: regression analysis of the MUCOSA Trial. Fam
Med 1996;28:204–10.
Cullen D, Bardhan KD, Eisner M, et al. Primary gastroduodenal prophylaxis with omeprazole for non-steroidal
anti-inflammatory drug users. Aliment Pharmacol Ther
1998;12:135–40.
Graham DY, Agrawal NM, Campbell DR, et al. Ulcer prevention in long-term users of nonsteroidal anti-inflammatory drugs: results of a double-blind, randomized, multicenter, active- and placebo-controlled study of misoprostol
vs lansoprazole. NSAID-Associated Gastric Ulcer Prevention Study Group. Arch Intern Med 2002;162:169–75.
Scheiman JM, Yeomans N, Hawkey CJ, et al. Esomeprazole
reduces gastric and duodenal ulcer development among
high risk NSAID users. Am J Gastroenterol 2003;98:S52.
Taha AS, Hudson N, Hawkey CJ, et al. Famotidine for the
prevention of gastric and duodenal ulcers caused by nonsteroidal anti-inflammatory drugs. N Engl J Med 1996;
334:1435–9.
41. Deeks JJ, Smith LA, Bradley MD. Efficacy, tolerability, and
upper gastrointestinal safety of celecoxib for treatment of
osteoarthritis and rheumatoid arthritis: systematic review
of randomised controlled trials. BMJ 2002;325:619–26.
42. Laine L, Harper S, Simon T, et al. A randomized trial
comparing the effect of rofecoxib, a cyclooxygenase
2-specific inhibitor, with that of ibuprofen on the gastroduodenal mucosa of patients with osteoarthritis. Rofecoxib Osteoarthritis Endoscopy Study Group. Gastroenterology 1999;117:776–83.
43. Vioxx [package insert]. White House (NJ): Merck & Co.;
2003. Available at www.vioxx.com/vioxx/shared/
documents/english/pi.pdf. Accessed 26 Mar 2004.
44. Scheiman JM. Outcomes studies of the gastrointestinal
safety of cyclooxygenase-2 inhibitors. Cleve Clin J Med
2002;69 Suppl 1:SI40–6.
45. Chan FK, Hung LC, Suen BY, et al. Celecoxib versus
diclofenac and omeprazole in reducing the risk of recurrent ulcer bleeding in patients with arthritis. N Engl J
Med 2002;347:2104–10.
46. Watson DJ, Harper SE, Zhou PL, et al. Gastrointestinal
tolerability of the selective cyclooxygenase-2 (COX-2) inhibitor rofecoxib compared with nonselective COX-1
and COX-2 inhibitors in osteoarthritis. Arch Intern Med
2000;160: 2998–3003.
47. Lisse JR, Perlman M, Johansson G, et al. Gastrointestinal
tolerability and effectiveness of rofecoxib versus naproxen in the treatment of osteoarthritis: a randomized, controlled trial. ADVANTAGE Study Group. Ann Intern
Med 2003;139:539–46.
48. Fendrick AM, Bandekar RR, Chernew ME, Scheiman
JM. Role of initial NSAID choice and patient risk factors
in the prevention of NSAID gastropathy: a decision analysis. Arthritis Rheum 2002:47:36–43.
49. Allison MC, Howatson AG, Torrance CJ, et al. Gastrointestinal damage associated with the use of nonsteroidal
antiinflammatory drugs. N Engl J Med 1992;327:749–54.
50. Goldstein J, Eisen G, Lewis B, Gralnek I, et al. Abnormal
small bowel findings are common in healthy subjects
screened for a multi-center, double blind, randomized,
placebo controlled trial using capsule endoscopy. Gastroenterology 2003;124:A37.
51. Goldstein J, Eisen G, Lewis B, Gralnek I, et al. Celecoxib
is associated with fewer small bowel lesions than naproxen + omeprazole in healthy subjects as determined by
capsule endoscopy. Am J Gastroenterol 2002:98:S297.
52. Laine L, Connors LG, Reicen A, et al, Serious lower gastrointestinal clinical events with nonselective NSAID or
coxib use. Gastroenterology 2003;124:288–92.
53. Sigthorsson G, Crane R, Simon T, et al. COX-2 inhibition
with rofecoxib does not increase intestinal permeability
in healthy subjects: a double blind crossover study comparing rofecoxib with placebo and indomethacin. Gut
2000;47:527–32.
Copyright 2004 by Turner White Communications Inc., Wayne, PA. All rights reserved.
12 Hospital Physician Board Review Manual