CLINICAL REVIEW

Frequency of nonsteroidal anti-inflammatory drug-associatedulcers

Hideyuki Hiraishi • Ryo Oki • Kohei Tsuchida • Naoto Yoshitake •

Keiichi Tominaga • Koji Kusano • Takashi Hashimoto • Mitsunori Maeda •

Takako Sasai • Tadahito Shimada

Received: 6 April 2012 / Accepted: 8 April 2012 / Published online: 24 April 2012

� Springer 2012

Abstract Nonsteroidal anti-inflammatory drugs (NSA-

IDs) are widely used for treatment of orthopedic diseases,

inflammatory diseases, etc., and low-dose aspirin is a

common antiplatelet therapy given mainly for secondary

prevention of atherothrombosis (e.g., myocardial infarction

and cerebral infarction). As to the history of NSAID-

induced gastric mucosal injury in Japan, the first case of an

aspirin-induced gastric ulcer was reported as early as 1934.

Based on a meta-analysis of risk factors for peptic ulcers,

Helicobacter pylori infection and NSAIDs are the main

etiologies of peptic ulcers. NSAIDs alone increase the odds

ratio for ulcer development to 19.4 and that for ulcer

bleeding to 4.85. In fact, the Japan Rheumatism Foundation

reported in 1991 that active gastric ulcers and active duo-

denal ulcers were detected in 15.5 and 1.9 % of 1008

patients, respectively, taking oral NSAIDs for 3 months or

longer. In Japan, which is becoming an increasingly aged

society, the numbers of patients taking NSAIDs and low-

dose aspirin are expected to increase dramatically in the

future. It is hoped that accumulation of evidence on gas-

trointestinal risk will allow many patients to rationally

avoid gastrointestinal complications while receiving the

benefits of NSAIDs and low-dose aspirin.

Keywords NSAIDs � Aspirin � Antiplatelet therapy �Risk factors � Upper gastrointestinal bleeding �Proton pump inhibitor

Introduction

Nonsteroidal anti-inflammatory drugs (NSAIDs) are

widely used for treatment of orthopedic diseases, febrile/

inflammatory diseases, collagen diseases, etc., and low-

dose aspirin is a common antiplatelet therapy mainly for

secondary prevention of atherothrombosis (e.g., myocar-

dial infarction and cerebral infarction) [1]. However, gas-

trointestinal (GI) lesions pose clinically important

problems in terms of adverse reactions. Mucosal injury

induced by NSAIDs is divided into acute diseases, such as

acute gastritis or acute gastric mucosal lesions, and chronic

diseases, such as peptic ulcers.

This clinical review discusses the history of peptic

ulcers associated with administration of NSAIDs, the fre-

quency of nonselective NSAID-induced ulcers, differences

in frequency between NSAID and low-dose aspirin or

cyclooxygenase (COX)-2 selective inhibitors, differences

in frequency due to dosage formulations, risk factors, and

stratification of risks for GI lesions and complications.

History of upper GI mucosal injury induced by NSAIDs

How long has GI mucosal injury induced by NSAIDs been

recognized? When we searched Medline for English arti-

cles on upper GI mucosal injury using the term ‘‘aspirin’’

as an example, the oldest article retrieved by the search was

published in 1955; tracing back through the references of

this article [2] led to the first case of mucosal injury due to

aspirin, which was reported in The Lancet in 1938 [3]. This

report seems to describe aspirin as a frequent cause of

gastric bleeding. In our search for Japanese articles, we

found a case report on aspirin-related gastric mucosal

injury published in 1934 [4]. In this case report,

H. Hiraishi (&) � R. Oki � K. Tsuchida � N. Yoshitake �K. Tominaga � K. Kusano � T. Hashimoto � M. Maeda �T. Sasai � T. Shimada

Department of Gastroenterology, Dokkyo Medical University,

Mibu, Tochigi 321-0293, Japan

e-mail: hiraishi@dokkyomed.ac.jp

123

Clin J Gastroenterol (2012) 5:171–176

DOI 10.1007/s12328-012-0300-y

epigastralgia and hematemesis/melena occurred after oral

administration of aspirin. Although whether bleeding was

caused by ulcers or hemorrhagic gastritis is unknown, this

case report appears to be the first in Japan. Thus, cases of

aspirin-related ulcers were reported as early as 1934 in

Japan and 1938 in the UK.

Odds ratios for ulcers and upper GI bleeding (UGIB)

As described above, it has long been recognized that

NSAIDs cause upper GI mucosal injury. A recent meta-

analysis of the etiology of and risk for peptic ulcers [5] also

indicated NSAIDs and Helicobacter pylori (H. pylori) to

be the main etiologies of peptic ulcers. If the odds ratio for

ulcer development in NSAIDs (-)/H. pylori (-) patients is

set at 1, the odds ratios are considered to increase to 19.4 in

NSAIDs (?) patients, 18.1 in H. pylori (?) patients, and

61.1 in NSAIDs (?)/H. pylori (?) patients. Moreover, the

odds ratios for ulcer bleeding are considered to increase to

4.85, 1.79, and 6.13, respectively (Table 1).

In Japan, a case–control study on the associations of

UGIB with NSAIDs and H. pylori infection was also

reported [6]. In patients with peptic ulcers or hemorrhagic

gastritis, risk with aspirin use is 5.5 times higher and risk

with oral administration of NSAIDs other than aspirin is

6.1 times higher (Fig. 1). Although the subjects included

patients with hemorrhagic gastritis, this study provides

important findings on the influence of aspirin/NSAIDs on

ulcer development and bleeding in Japanese.

Frequency of nonselective NSAID-induced ulcers

Results of the study by the Japan Rheumatism

Foundation

In the large-scale study reported by the Japan Rheumatism

Foundation in 1991 [7], upper GI endoscopy was per-

formed in 1008 patients with rheumatoid arthritis or

osteoarthritis who had been taking oral NSAIDs for

3 months or longer. Those ‘‘with upper GI mucosal

lesions’’ accounted for 62.2 %. Active gastric ulcers and

active duodenal ulcers were detected in 15.5 and 1.9 %,

respectively (Fig. 2). These rates are more than 10 times

higher than the detection rate of 1.42 % for gastric ulcers

and more than 3 times higher than the detection rate of

0.59 % for duodenal ulcers reported by the Japanese

Society of Gastroenterological Mass Survey in the same

year. Patients ‘‘without active lesion’’ accounted for

37.8 %. Of these, gastric ulcer scars and duodenal ulcer

scars were detected in 8.0 and 3.0 %, respectively. If

combined with those with ulcer scars, this study shows that

28.4 % of the patients with arthritis who were taking oral

NSAIDs for a long period had peptic ulcers or history of

such ulcers. This study was conducted more than 20 years

ago. Given the recently enhanced gastric acid secretion

capacity of Japanese subjects, and especially the low rates

of H. pylori infection in young people, it cannot be ruled

out that the incidence of ulcers may have changed. How-

ever, there is no other large-scale clinical study using upper

GI endoscopy like this in the world. The study may thus be

highly reliable.

Table 1 Odds ratios for development of peptic ulcers and ulcer

bleeding according to H. pylori infection and NSAID use (cited from

Ref. [5])

Ulcer development Ulcer bleeding

H. pylori(-)

H. pylori(?)

H. pylori(-)

H. pylori(?)

NSAIDs (-) 1.0 18.1 1.0 1.70

NSAIDs (?) 19.4 61.1 4.85 6.13

0

2

4

6

8

10Odds ratio

Overall Regular Occasional

5.5

7.7

1.9

Aspirin

0

2

4

6

8

10Odds ratio

Overall Regular Occasional

6.1

7.6

3.6

Non-aspirin NSAID

Fig. 1 Risk for upper gastrointestinal bleeding due to oral adminis-

tration of aspirin and nonaspirin NSAIDs (case–control study

conducted in Japan, cited from Ref. [6])

Active lesions62.2%

No active lesions37.8%

GU15.5% DU 1.9%

Gastritis38.5%

Others6.3%

Others 26.8%

DU scar 3.0%

GU scar 8.0%

Fig. 2 Frequency of upper gastrointestinal lesions in patients with

arthritis under long-term oral administration of NSAIDs (report by the

Japan Rheumatism Foundation, cited from Ref. [7])

172 Clin J Gastroenterol (2012) 5:171–176

123

Results from the Cochrane Database of Systematic

Reviews

In the Systematic Reviews of the Cochrane Library [8], we

searched for randomized controlled trials (RCT) of pro-

phylactic effects of drugs used for ulcers induced by

NSAID administration for 3 months or longer, that is,

misoprostol [prostaglandin (PG) preparation], H2 receptor

antagonists at the standard and double doses, and proton

pump inhibitors (PPI). Then, we attempted to calculate the

frequency of ulcers in the placebo (control) group, based on

meta-analysis of these RCTs (active drugs versus placebo).

The frequency of gastric ulcers in the control group was

277/1865 patients in RCT with PG, 52/495 in RCT with H2

receptor antagonists at the standard dose, 38/147 in RCT

with H2 receptor antagonists at double doses, and 124/465

in RCT with PPI, resulting in an overall frequency of

16.5 % (491/2972 patients). Moreover, based on the meta-

analysis, the frequency of duodenal ulcers in the placebo

(control) group was 86/1439 patients in RCT with PG,

27/487 in RCT with H2 receptor antagonists at the standard

dose, 20/147 in RCT with H2 receptor antagonists at double

doses, and 36/354 in RCT with PPI, resulting in an overall

frequency of 7.0 % (169/2427 patients). Compared with

the ratio between gastric and duodenal ulcers of 8.2 (15.5/

1.9 %) reported by the Japan Rheumatism Foundation, the

ratio in Europe and the USA is 2.4 (16.5/7.0 %), reflecting

a relatively higher proportion of duodenal ulcers.

Differences due to low-dose aspirin/COX-2 selective

NSAIDs

Development of ulcers due to low-dose aspirin and risk

for UGIB

We reviewed reports on the frequency of peptic ulcers

induced by administration of low-dose aspirin. The inci-

dence of ulcers was 10.7 % in patients who had been orally

taking low-dose aspirin for 1 month or longer to prevent

cardiovascular events [9]. In the later-described RCT on

the prophylactic effects of famotidine administration in

patients with moderate GI risk who were taking oral low-

dose aspirin, the incidences of gastric and duodenal ulcers

after 12 weeks of treatment in the placebo group were 15

and 8.5 %, respectively. In addition, the prophylactic effect

of famotidine was significant [10]. Similarly, in another

RCT with 26-week observation of patients aged 60 years

and older with moderate GI risk who were taking oral low-

dose aspirin (comparison between 2 doses of esomeprazole

and placebo), the incidences of ulcers were 7.4 % in the

placebo group, 1.1 % in the 20 mg esomeprazole group,

and 1.5 % in the 40 mg esomeprazole group, indicating

prophylactic effects of esomeprazole [11]. Furthermore, a

Japanese RCT with 1-year follow-up of high-risk patients

with verified history of ulcers who were taking oral low-

dose aspirin revealed PPI (lansoprazole 15 mg/day) to

prevent ulcer development, as compared with gefarnate

(100 mg/day) (3.7 versus 31.7 %) [12]. In this RCT, the

incidence (which can be regarded as recurrence because the

patients had apparent history) of ulcers in the gefarnate

group was high, at 31.7 %. Because there is no evidence

that gefarnate prevents recurrence of ulcers due to low-

dose aspirin, the incidence of ulcers in this group may be

assumed to be close to that in patients receiving placebo

administration.

Moreover, meta-analysis results of several studies on

doses of aspirin and risk for GI bleeding are available [13].

As shown in Fig. 3, the odds ratios varied greatly among

studies from 0.2 to 7. Based on the meta-analysis, the

frequency of GI bleeding is 2.47 % in the aspirin group

versus 1.42 % in the placebo group. The odds ratio is

calculated to be 1.68 [95 % confidence interval (CI)

1.51–1.88], indicating that the incidence is significantly

higher in the aspirin than in the placebo group. As regards

the correlation between doses of aspirin and odds ratios of

GI bleeding, the slope of the regression line is close to 0

(Fig. 3). In other words, aspirin increases the risk for UGIB

by approximately twofold regardless of whether low doses

exerting antiplatelet effects (70–330 mg/day) or anti-

inflammatory doses over 1000 mg/day are given. NSAIDs

increase the risk for ulcers depending on doses. However,

the risk for GI bleeding after aspirin administration does

not depend on dose, with low and high aspirin doses

yielding similar risks. In fact, PG levels in the gastric

mucosa significantly decrease even at a very low aspirin

dose of just 10 mg/day, which is supported by a study

showing endoscopy to cause gastric mucosal injury [14].

Aspirin Dose (mg/day)

75 300 1,000 1,500

10

5

2

10.2

Od

ds

rati

o f

or

GI b

leed

ing

Fig. 3 Metaregression of Peto odds ratio for gastrointestinal bleeding

against dose of aspirin (size of circle is proportional to size of trial)

(cited from Ref. [13])

Clin J Gastroenterol (2012) 5:171–176 173

123

Comparison between nonselective and COX-2 selective

NSAIDs

In regard to the risk for GI bleeding associated with non-

selective NSAIDs, coxib (COX-2 selective NSAID), and

meloxicam (which has COX-2 inhibitory action but low

COX-2 selectivity), there are results of a comparison of

risk for hospitalization due to UGIB in a cohort prescribed

NSAIDs from the General Practice Research Database in

the UK (nonselective NSAID group, 628000 patients/year;

meloxicam group, 7100 patients/year; coxib group,

1600 patients/year) [15]. After adjustment for factors such

as sex, age, history of GI bleeding, prescription of antiulcer

drugs or nitrate, and alcohol consumption, if risk for seri-

ous GI bleeding associated with nonselective NSAIDs is

set at 1, the risks with coxib and meloxicam are 0.36 (95 %

CI 0.14–0.97; P = 0.043) and 0.84 (95 % CI 0.60–1.17;

P = 0.303), respectively. Risk for GI bleeding is signifi-

cantly lower with administration of COX-2 selective

NSAIDs (Fig. 4). There is also a report comparing inci-

dence of upper GI events induced by nonselective NSAIDs

and celecoxib (COX-2 selective coxib) using endoscopy

[16]. After 13274 patients with osteoarthritis took NSAIDs

(either diclofenac or naproxen) or celecoxib for 12 weeks,

similar analgesic effects were observed in the two groups.

Meanwhile, the annual incidence of ulcer complications

(perforation, bleeding, and obstruction) and symptomatic

ulcers were, respectively, 0.8 and 1.2 % in the NSAID

group and 0.1 and 0.9 % in the celecoxib group. Incidence

of ulcer complications was significantly lower in the

celecoxib group (0.8 versus 0.1 %).

Differences due to dosage formulations

UGIB due to different dosage formulations of low-dose

aspirin

Several formulations of low-dose aspirin have been

developed, and Kelly et al. [17] reported on the results of a

study on incidence of aspirin ulcer induction by dosage

formulations in an effort to describe the current status. In

550 patients undergoing endoscopy for UGIB, risk for

ulcer development was examined according to dosage

formulations of aspirin taken during 1 week before

endoscopy. The risk was 2.6 with the plain tablet, 2.7 with

the enteric-coated tablet, and 3.1 with the antacid buffered

tablet. No significant difference was observed in relative

risk among these three formulations. There is also a report

on gastric mucosal injury due to 2-week administration of

enteric-coated aspirin tablets that were used to prevent

gastric mucosal injury directly caused by gastric acid [18].

The incidence of gastric mucosal bleeding or gastric ulcer

was reported to be significantly lower with the enteric-

coated tablet than with either the plain or the buffered

tablet, and the incidence was similar to with the placebo.

As described above, no consensus has been reached on

whether risks for ulcers or bleeding can be reduced by

changing the dosage formulations of aspirin, such as the

enteric-coated tablet. Thus, it may be clinically practical to

treat patients by assuming that adjustment of dosage for-

mulations does not ensure reduction of GI risk.

Prodrug

Direct gastric mucosal injury may also be caused by indi-

rect mechanisms of biliary excretion and duodenogastric

reflux of active NSAID metabolites; for example, sulindac,

which is orally administered as a non-mucosa-damaging

prodrug, is converted to its active form, sulindac sulfide,

via hepatic metabolism. Sulindac sulfide is excreted into

bile and may cause gastric mucosal injury via reflux from

the duodenum [19]. However, aspirin, which is not excre-

ted into bile, causes ulcers even with venous administration

to humans. As described above, ulcers and bleeding are

highly likely to occur even without direct contact between

the gastric mucosa and aspirin, such as with the enteric-

coated tablet. Thus, NSAIDs apparently exert a systemic

action via circulating blood.

Non specific NSAIDRelative risk = 1.0

0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Meloxicam

Relative risk = 0.84(95% CI: 0.60-1.17)

p=0.303

Non-specific NSAID

Relative risk = 0.36(95% CI: 0.14-0.97)

p=0.043

Coxib

Adjusted relativerisks

Fig. 4 Risk for gastrointestinal

bleeding with nonselective

NSAIDs, coxib, and meloxicam

(cited from Ref. [15])

174 Clin J Gastroenterol (2012) 5:171–176

123

Risk factors and stratification by risk factors

The main mechanisms of upper GI injury due to aspirin/

NSAIDs may be mediated by the local damaging action on

the mucosa and inhibition of COX, thereby decreasing

levels of PG, which has protective effects on GI mucosa

[20]. Risk factors for NSAID-induced ulcers reportedly

include advanced age, history of ulcers, concomitant glu-

cocorticoid use, oral administration of a high dose or

several types of NSAIDs, and concomitant systemic dis-

orders, and potential risk factors are H. pylori infection,

smoking, and alcohol consumption [19]. Recently, a high,

moderate, and low risk classification, based on the weight

and number of these risk factors, was proposed [21]

(Table 2). According to this classification, H. pylori

infection is an independent additive risk factor and is

considered to require separate handling from the other risk

factors.

Aspirin causes GI mucosal injury even at a very low

dose of only 10 mg/day [14], increasing the risk for peptic

ulcers and their potentially fatal complications (bleeding

and perforation). In general, administration of aspirin

approximately doubles GI risk in middle-aged users with

neither history of peptic ulcers nor concomitant drug use.

However, GI risk is dramatically increased in patients with

history of peptic ulcers, H. pylori infection, advanced age

(60 years or older), and concomitant use of other anti-

platelet or anticoagulant drugs such as NSAIDs, gluco-

corticoids, clopidogrel, etc. [22]. Based on these results,

Expert Consensus Documents [23] of 3 academic societies,

the American College of Cardiology Foundation, American

College of Gastroenterology, and American Heart Associ-

ation, define high risk as a case with any of the following

factors: (1) history of ulcer complications, (2) history of

nonbleeding ulcers, (3) GI bleeding, or (4) concomitant use

of 2 types of antiplatelet drugs or concomitant anticoagu-

lant therapy. Moderate risk is defined as a case with 2 or

more of the following factors: (1) advanced age (60 years

or older), (2) concomitant glucocorticoid use, and (3)

symptoms (dyspepsia or gastroesophageal reflux disease).

Implementation of aggressive prophylactic measures is

advocated for these cases. For patients with history of ulcer

complications or nonbleeding ulcers, tests for H. pylori

infection and eradication are recommended and, if positive,

appropriate treatment should be given. This is based on

RCT showing that recurrence of aspirin ulcers is lower in

patients undergoing eradication than in those without

H. pylori eradication [24]. Attention should, however, be

paid to the fact that this recommendation is not supported

by any meta-analyses of data [25].

Conclusions

Japan is facing an unprecedented aging of its society, and

medical care must be adjusted to changes in disease pat-

terns associated with this population aging phenomenon.

Although the annual consumption of nonaspirin NSAIDs

remains essentially constant in Japan, consumption of low-

dose aspirin as an antiplatelet drug has increased remark-

ably since 2000. The secondary prophylactic effects of

antiplatelet drugs on atherothrombotic events have been

demonstrated by many clinical studies, and use of anti-

platelet drugs is recommended by various guidelines.

However, the effects are limited by GI complications, such

as ulcers and bleeding. It should be considered that an

article published in 2011 indicates that acid suppressive

therapy to reduce risk for aspirin ulcers may reduce the

inhibitory effects of aspirin on cardiovascular events [26].

In regard to the merits of aspirin/NSAIDs and GI risk, as

well as the pros and cons of acid suppressive therapy to

prevent GI risk, immediate and careful studies are urgently

needed.

Japan clearly lags behind in studies on the current status

of and preventive strategies for GI injury due to NSAIDs

and low-dose aspirin, as compared with other countries.

However, we hope that accumulation of evidence will

allow more patients to rationally avoid GI complications

while still receiving the benefits of NSAIDs or low-dose

aspirin.

Conflict of interest The authors declare that they have no conflict

of interest.

References

1. Antiplatelet Trialists’ Collaboration. Collaborative overview of

randomized trials of antiplatelet therapy-I: prevention of death,

myocardial infarction, and stroke by prolonged antiplatelet ther-

apy in various categories of patients. BMJ. 1994;308:81–106.

2. Muir A, Cossar IA. Aspirin and ulcer. Br Med J. 1955;ii:7–12.

Table 2 Risk factors for development of NSAIDs ulcers and strati-

fication of risk (cited from Ref. [21])

High risk

1 History of a previously complicated ulcer

2 Multiple ([2) risk factors

Moderate risk (1–2 risk factors)

1 Age [65 years

2 High-dose NSAID therapy

3 Previous history of uncomplicated ulcer

4 Concurrent use of aspirin (including low dose), corticosteroids

or anticoagulants

Low risk

1 No risk factors

Clin J Gastroenterol (2012) 5:171–176 175

123

3. Douthwaite AH, Lond MD, Lintotto GAM. Gastroscopic obser-

vation of the effect of aspirin and certain other substances on the

stomach. Lancet. 1938;232:1222–4.

4. Morimoto M. A case of aspirin-induced gastric ulcer. Diagn

Treat. 1934;21:1044–5 (in Japanese).

5. Huang JQ, Sridhar S, Hunt RH. Role of Helicobacter pylori and

non-steroidal anti-inflammatory drugs in peptic ulcer disease: a

meta-analysis. Lancet. 2002;359:14–22.

6. Sakamoto C, Sugano K, Ota S, Sakaki N, Takahashi S, Yoshida

Y, et al. Case–control study on the association of upper gastro-

intestinal bleeding and nonsteroidal anti-inflammatory drugs in

Japan. Eur J Clin Pharmacol. 2006;62:765–72.

7. Shiokawa Y, Nobunaga M, Saito T, Asaki S, Ogawa N. Epide-

miology study on upper gastrointestinal lesions induced by non-

steroidal anti-inflammatory drugs. Ryumachi. 1991;31:96–111

(in Japanese).

8. Rostom A, Dube C, Wells G, Tugwell P, Welch V, Jolicoeur E,

et al. Prevention of NSAID-induced gastroduodenal ulcers.

Cochrane Database Syst Rev. 2002;4:CD002296 (Review content

assessed as up-to-date: 11 May 2009).

9. Yeomans ND, Lanas AI, Talley NJ, Thomson AB, Daneshjoo R,

Eriksson B, et al. Prevalence and incidence of gastroduodenal

ulcers during treatment with vascular protective doses of aspirin.

Aliment Pharmacol Ther. 2005;22:795–801.

10. Taha AS, McCloskey C, Prasad R, Bezlyak V. Famotidine for the

prevention of peptic ulcers and oesophagitis in patients taking

low-dose aspirin (FAMOUS): a phase III, randomised, double-

blind, placebo-controlled trial. Lancet. 2009;374:119–25.

11. Scheiman JM, Devereaux PJ, Herlitz J, Katelaris PH, Lanas A,

Veldhuyzen van Zanten S, et al. Prevention of peptic ulcers with

esomeprazole in patients at risk for ulcer development treated

with low-dose acetylsalicylic acid: a randomised, controlled trial

(OBERON). Heart. 2011;97:797–802.

12. Sugano K, Matsumoto Y, Itabashi T, Abe S, Sakaki N, Ashida K,

et al. Lansoprazole for secondary prevention of gastric or duo-

denal ulcers associated with long-term low-dose aspirin therapy:

results of a prospective, multicenter, double-blind, randomized,

double-dummy, active-controlled trial. J Gastroenterol. 2011;46:

724–35.

13. Derry S, Loke YK. Risk of gastrointestinal haemorrhage with

long term use of aspirin: meta-analysis. BMJ. 2000;321:1183–7.

14. Cryer B, Feldman M. Effects of very low dose daily, long-term

aspirin therapy on gastric, duodenal, and rectal prostaglandin

levels and on mucosal injury in healthy humans. Gastroenterol-

ogy. 1999;17:17–25.

15. MacDonald TM, Morant SV, Goldstein JL, Burke TA, Pettitt D.

Channelling bias and the incidence of gastrointestinal haemor-

rhage in users of meloxicam, coxibs, and older, non-specific non-

steroidal anti-inflammatory drugs. Gut. 2003;52:1265–70.

16. Singh G, Fort JG, Goldstein JL, Levy RA, Hanrahan PS, Bello

AE, et al. Celecoxib versus naproxen and diclofenac in osteoar-

thritis patients: SUCCESS-I Study. Am J Med. 2006;119:255–66.

17. Kelly JP, Kaufman DW, Jurgelon JM, Sheehan J, Koff RS,

Shapiro S. Risk of aspirin-associated major upper-gastrointestinal

bleeding with enteric-coated or buffered product. Lancet. 1996;

348:1413–6.

18. Petroski D. Endoscopic comparison of three aspirin preparations

and placebo. Clin Ther. 1993;15:314–20.

19. Wolfe MM, Lichtenstein DR, Singh G. Gastrointestinal toxicity

of nonsteroidal antiinflammatory drugs. N Engl J Med. 1999;340:

1888–99.

20. Study Group of the guidelines for treatment of gastric ulcer based

on evidence: NSAID-induced ulcer. Guidelines for treatment of

gastric ulcer based on evidence. 2nd ed. Tokyo: Jiho Publishing

Co.; 2007. p. 101–10 (in Japanese).

21. Lanza FL, Chan FK, Quigley EM, Practice Parameters Com-

mittee of the American College of Gastroenterology. Guidelines

for prevention of NSAID-related ulcer complications. Am J

Gastroenterol. 2009;104:728–38.

22. Lanas A, Scheiman J. Low-dose aspirin and upper gastrointesti-

nal damage: epidemiology, prevention and treatment. Curr Med

Res Opin. 2007;23:163–73.

23. Bhatt DL, Scheiman J, Abraham NS, Antman EM, Chan FK,

Furberg CD, et al. ACCF/ACG/AHA 2008 expert consensus

document on reducing the gastrointestinal risks of antiplatelet

therapy and NSAID use. Am J Gastroenterol. 2008;103:

2890–907.

24. Chan FK, Chung SC, Suen BY, Lee YT, Leung WK, Leung VK,

et al. Preventing recurrent upper gastrointestinal bleeding in

patients with Helicobacter pylori infection who are taking low-

dose aspirin or naproxen. N Engl J Med. 2001;344:967–73.

25. Fletcher EH, Johnston DE, Fisher CR, Koerner RJ, Newton JL,

Gray CS. Systematic review: Helicobacter pylori and the risk of

upper gastrointestinal bleeding risk in patients taking aspirin.

Aliment Pharmacol Ther. 2010;32:831–9.

26. Charlot M, Grove EL, Hansen PR, Olesen JB, Ahlehoff O, Sel-

mer C, et al. Proton pump inhibitor use and risk of adverse car-

diovascular events in aspirin treated patients with first time

myocardial infarction: nationwide propensity score matched

study. BMJ. 2011;342:d2690. doi:10.1136/bmj.d2690.

176 Clin J Gastroenterol (2012) 5:171–176

123