Expert Opinion on Drug Safety

ISSN: 1474-0338 (Print) 1744-764X (Online) Journal homepage: https://www.tandfonline.com/loi/ieds20

Gastrointestinal safety of paracetamol: is there

any cause for concern?

Bernard Bannwarth

To cite this article: Bernard Bannwarth (2004) Gastrointestinal safety of paracetamol: is there any
cause for concern?, Expert Opinion on Drug Safety, 3:4, 269-272, DOI: 10.1517/14740338.3.4.269

To link to this article: https://doi.org/10.1517/14740338.3.4.269

Published online: 23 Feb 2005.

Submit your article to this journal

Article views: 3796

View related articles

Citing articles: 10 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=ieds20

1. Introduction
2. Gastroduodenal ulcers and
ulcer complications
3. Upper gastrointestinal
symptoms
4. Expert opinion
For reprint orders, please
contact:
reprints@ashley-pub.com
Ashley Publications
www.ashley-pub.com
Editorial
Gastrointestinal safety of
paracetamol: is there any cause
for concern?
Bernard Bannwarth
Department of Rheumatology, Pellegrin Teaching Hospital & Division of Therapeutics, Victor Segalen
University, 33076 Bordeaux, France
Paracetamol is classically considered as a very safe analgesic/antipyretic com-
pound and, more specifically, as being virtually devoid of any gastrointestinal
(GI) ulcerogenic potential. Accordingly, it is widely stated that paracetamol is
particularly suitable for patients at high risk of developing GI ulcers or bleeds.
This view has been challenged by recent epidemiological studies using com-
puterised prescription data, which indicated that paracetamol exhibits dose-
dependent GI toxicity. However, the results of these studies are most likely
incorrect for reasons of inherent biases and confounding. Furthermore, their
findings conflict with those of clinical trials and case-control studies in which
information about drug exposure was obtained by direct questioning of both
cases and controls. Finally, paracetamol, especially at high doses, may induce
upper GI symptoms such as abdominal pain/discomfort, heartburn, nausea or
vomiting. Conversely, the risk for ulcers and ulcer complications due to
paracetamol is not supported by available data.
Keywords: dyspepsia, gastrointestinal (GI) toxicity, GI ulcers, paracetamol
Expert Opin. Drug Saf. (2004) 3(4):269-272
1. Introduction
Paracetamol is usually regarded as an effective and very safe analgesic/antipyretic
agent. Accordingly, a great variety of national and international organisations,
including the World Health Organisation, have recommended it as first-line treat-
ment for fever and pain in children [1]. Similarly, expert guidelines have recom-
mended paracetamol as the initial drug treatment for symptomatic osteoarthritis [2]
and mild-to-moderate pain of musculoskeletal origin in older persons [3]. A major
reason for these statements is the classical view that, unlike NSAIDs, paracetamol
does not cause significant gastrointestinal (GI) mucosal injury. As a result, paraceta-
mol is the preferred non-opioid analgesic for patients with GI risk factors in clinical
practice [4]. However, recent epidemiological studies suggested that the overall
occurrence of GI adverse events with paracetamol is comparable to that with con-
ventional NSAIDs. How strong is the evidence?
2. Gastroduodenal ulcers and ulcer complications
Endoscopic studies allow a direct approach to the assessment of drug-induced
GI mucosal damage. In this way, a randomised, double-blind, three-way crossover
study compared the gastroduodenal effects of a 7-day regimen of ketoprofen
75 mg/day, paracetamol 4 g/day and placebo in 24 healthy subjects [5]. No clinically
significant mucosal injury developed in the paracetamol and placebo groups whereas
ketoprofen produced gastroduodenal lesions, including two frank gastric ulcers in
50% of the subjects [5]. Furthermore, paracetamol did not increase whole gut blood
loss as assessed by the radio-labelled erythrocyte technique [6]. It may be argued that
these findings are poor predictors of the actual incidence of symptomatic ulcers and
2004 © Ashley Publications Ltd ISSN 1474-0338 269
Gastrointestinal safety of paracetamol: is there any cause for concern?
ulcer complications in paracetamol users. Fortunately, early
epidemiological studies confirmed that habitual intake of para-
cetamol was not associated with GI bleeding [6,7]. Since spon-
taneous reporting schemes are valuable in providing early
warnings or signals of possible adverse drug reactions, it should
be stressed that no case of peptic ulcer disease related to para-
cetamol was reported to the Spanish Drug Monitoring System
between November 1982 and December 1991 [8]. These data
are consistent with the weak effects of paracetamol on periph-
eral cyclooxygenase isoenzymes. Unlike NSAIDs, paracetamol
did not alter the ability of gastric mucosa to generate
cytoprotective prostaglandins in humans [9].
Consequently, the findings of Garcia Rodriguez et al. [10]
appeared surprising. These authors conducted a nested
case-control study using the United Kingdom General Prac-
tice Research database to investigate the association between
paracetamol or traditional NSAIDs and the risk of upper GI
bleed/perforation. Compared with controls, patients who
were prescribed paracetamol at doses < 2 g/day did not have
an increased risk of upper GI complications. Conversely,
doses of paracetamol > 2 g/day were found to confer a risk
for upper GI bleed/perforation as great as that with conven-
tional NSAIDs. The corresponding adjusted relative risks
were 3.6 (95% confidence interval [CI]: 2.6 – 5.1) and 4.1
(95% CI: 3.6 – 4.8) for paracetamol and NSAIDs, respec-
tively [10]. Additional features suggested that paracetamol
doses of > 2 g/day might be toxic to the upper GI tract. First,
prescription of paracetamol was still associated with similar
raised risk even though the analysis had been restricted to
individuals who had no recorded antecedents of upper GI
disorders, including dyspepsia [10]. Second, a joint effect of
paracetamol and NSAIDs was reported, the relative risk
being 13.2 (95% CI: 9.2 – 18.9) in those patients who were
prescribed both compounds together [9]. What can be
inferred from this study is that prescription of high-dose
paracetamol alone or in combination with NSAIDs was asso-
ciated with serious GI adverse events. Whether this indicates
a causal relationship between paracetamol use and severe GI
complications should, however, be discussed.
Garcia Rodriguez et al. [10] acknowledged that a preferential
prescription of paracetamol to patients at high risk of upper GI
complications might account for their results and that subse-
quent biases would not have been completely eliminated after
adjustment for GI susceptibility. In this respect, paracetamol is
prone to these forms of biases, commonly referred to as ‘chan-
nelling bias’ and ‘confounding by indication’. This is exempli-
fied by a population-based cohort study, which found a nearly
two-fold increase in overall mortality among adults prescribed
paracetamol [11]. The excess of death rates was observed regard-
less of cause of death, including cancers, cardiovascular diseases
or diabetes, and this was more pronounced during the first
year of follow-up [11]. Thus, the most plausible explanation for
these findings is a channelling – that is, patients with chronic
or serious illnesses were more likely to be prescribed paraceta-
mol for pain relief. The channelling led to confounding by
270 Expert Opin. Drug Saf. (2004) 3(4)
indication – that is, the underlying condition for which the
drug was prescribed, rather than the drug itself, resulted in
increased mortality risk [11].
A further limitation of the study by Garcia Rodriguez et al.
[10] includes the lack of data on purchases of over-the-counter
(OTC) paracetamol and NSAIDs. Overall, it lacked informa-
tion on true consumption of drugs, whether they had been
prescribed or not. Interestingly, a meta-analysis of individual
patient data from three case-control studies in which informa-
tion about drug exposure was obtained by direct questioning
of cases and controls found no evidence of any increased risk
of upper GI bleeding with any dose of paracetamol [12]. No
excess of GI bleeding was observed among paracetamol users
in other case-control studies that assessed drug exposure in the
same manner [13,14]. Moreover, combined use of paracetamol
and NSAIDs did not increase the risk of GI bleeding com-
pared with NSAIDs given alone [12].
Finally, available data do not support the view that para-
cetamol may cause GI ulcers and ulcer complications. Gener-
ally speaking, it seems unlikely that such serious effects would
have been overlooked because paracetamol has been used for
several decades by millions of patients, especially those at
higher risk for GI events.
3. Upper gastrointestinal symptoms
Patients receiving paracetamol occasionally complain of upper
GI symptoms, including epigastric pain/discomfort, heart-
burn, nausea or vomiting. Such symptoms, usually termed
dyspepsia, accounted for 11.3% of the 239 adverse reactions to
paracetamol reported to the Spanish Drug Monitoring System
[8]. Since dyspepsia is extremely common even in patients not
taking drugs recognised as being gastrotoxic, a comparison
with a control group is necessary to determine the increase in
risk associated with paracetamol intake.
Using the Government of Quebec’s health insurance data-
base, which contains demographic, medical and prescription
information on almost all elderly patients (> 65 years of age)
in Quebec, Rahme et al. [4] carried out a retrospective cohort
analysis that examined rates of adverse GI events in patients
who received a prescription for paracetamol or non-aspirin
NSAIDs between 1994 and 1996. Subjects were required to
have at least 1 full year of observation in the database prior to
enrollment, which was initiated by the first filled non-renewal
of study drugs. GI events included dyspepsia, ulcer and
‘GI hospitalisation’, that is, any hospitalisation with a diagno-
sis of gastroduodenal ulcer, perforation or bleeding or any
hospitalisation during which an endoscopy was performed
within the first 2 days [4]. Regarding patient characteristics,
there were striking differences between the paracetamol
cohort (n = 21,207) and the NSAID cohort (n = 26,978).
The key determinants of paracetamol utilisation compared
with NSAIDs included older age (> 75 years), prior history of
a GI event, especially prior ulcer and GI hospitalisation
(as defined above), concomitant use of anticoagulants, use of
 Bannwarth

low-dose aspirin in the previous 3 months and recent prior
hospitalisation whatever the cause [4]. In other words, the ini-
tial choice of paracetamol versus a NSAID was strongly influ-
enced by pre-existing GI risk factors. After adjustment for GI
risk factors, patients who were prescribed higher doses of
paracetamol (> 2.6 g/day) were more likely to experience a GI
event compared with those who were prescribed lower doses.
Moreover, the former experienced similar rates of GI events to
those patients who were prescribed high-dose NSAIDs (rela-
tive risk: 0.98; 95% CI: 0.85 – 1.13) [4]. It is important to
note that dyspepsia, not ulcers or GI hospitalisation, was
largely responsible for the observed increased GI toxicity
attributed to high-dose paracetamol [15]. Furthermore, this
study had major weaknesses. Again, a degree of scepticism is
needed about the reliability of statistical neutralisation of con-
founding variables inasmuch as a number of GI risk factors
were not captured by the database used [15,16]. Another
shortcoming of this study was the use of computerised
prescription data which likely differed from actual drug expo-
sure [17]. For instance, OTC use of NSAIDs cannot be ruled
out in patients who were prescribed paracetamol, and vice
versa. Both situations would generate an exaggerated relation-
ship between paracetamol prescription and GI events.
4. Expert opinion
The weight of the clinical evidence continues to support the
superior GI safety profile of paracetamol compared with
NSAIDs. Paracetamol, especially at high doses, may induce
upper GI symptoms such as abdominal pain/discomfort,
heartburn, nausea or vomiting. Conversely, the risk for
GI ulcers and ulcer complications due to paracetamol is not
supported by available biological or clinical data. Thus,
paracetamol should still be recommended as a first-line
mild analgesic, particularly in patients prone to
gastroduodenal ulcers.

Bibliography
Papers of special note have been highlighted as
either of interest (•) or of considerable interest (••)
to readers.
1. CRANSWICK N, COGHLAN D:
Paracetamol efficacy and safety in children: the
first 40 years. Am. J. Ther. (2000) 7:135-141.
2. JORDAN KM, ARDEN NK,
DOHERTY M et al.:
EULAR recommendations 2003:
an evidence based approach to the
management of knee osteoarthritis. Report
of a Task Force of the Standing Committee
for International Clinical Studies Including
Therapeutic Trials (ESCISIT). Ann. Rheum.
Dis. (2003) 62:1145-1155.
3. AMERICAN GERIATRICS SOCIETY
PANEL ON PERSISTENT PAIN IN
OLDER PERSONS: The management
of persistent pain in older persons.
J. Am. Geriatr. Soc. (2002) 50:S205-S224.
4. RAHME E, PETTITT D, LELORIER J:
Determinants and sequelae associated with
utilization of acetaminophen versus
traditional nonsteroidal anti-inflammatory
drugs in an elderly population. Arthritis
Rheum. (2002) 46:3046-3054.
• A study assessing the determinants
of paracetamol prescription in
clinical practice.
5. LANZA FL, CODISPOTI JR,
NELSON EB: An endoscopic
comparison of gastroduodenal injury with
over-the-counter doses of ketoprofen and
acetaminophen. Am. J. Gastroenterol.
(1998) 93:1051-1054.
6. IVEY KJ: Gastrointestinal effects of
antipyretic analgesics. Am. J. Med. (1983)
75(Suppl. 5A):53-64.
7. HOLVOET J, TERRIERE L,
VAN HEE W, VERBIST L, FIERENS E,
HAUTEKEETE ML: Relation of upper
gastrointestinal bleeding to non-steroidal
anti-inflammatory drugs and aspirin: a
case-control study. Gut (1991)
32:730-734.
8. CARVAJAL A, PRIETO JR,
ALVAREZ REQUEJO A,
MARTIN ARIAS LH:
Aspirin or acetaminophen? A comparison
from data collected by the Spanish Drug
Monitoring System. J. Clin. Epidemiol.
(1996) 49:255-261.
9. CRYER B, FELDMAN M:
Cyclooxygenase-1 and cyclooxygenase-2
selectivity of widely used nonsteroidal
anti-inflammatory drugs. Am. J. Med.
(1998) 104:413-423.
10. GARCIA RODRIGUEZ LA,
HERNANDEZ-DIAZ S: Relative risk of
upper gastrointestinal complications among
users of acetaminophen and nonsteroidal
anti-inflammatory drugs. Epidemiology
(2001) 12:570-576.
11. LIPWORTH L, SOREN F,
MELLEMKJAER L et al.:
A population-based cohort study of
mortality among adults prescribed
paracetamol in Denmark. J. Clin.
Epidemiol. (2003) 56:796-801.
•• An illustration of confounding
by indication.
12. LEWIS SC, LANGMAN MJ, LAPORTE JR,
MATTHEWS JN, RAWLINS MD,
WIHOLM BE: Dose-response relationships
between individual nonaspirin nonsteroidal
anti-inflammatory drugs (NANSAIDs)
and serious gastrointestinal bleeding:
a meta-analysis based on individual patient
data. Br. Clin. Pharmacol. (2002) 54:320-326.
•• A clear explanation on how information
about drug exposure influences the results
of case-control studies.
13. LANGMAN MJ, WEIL J,
WAINWRIGHT P et al.: Risks of bleeding
peptic ulcer associated with non-steroidal
anti-inflammatory drugs. Lancet (1994)
343:1075-1078.
14. LANAS A, SERRANO P, BAJADOR E,
FUENTES J, SAINZ R: Risk of upper
gastrointestinal bleeding associated with
non-aspirin cardiovascular drugs, analgesics
and nonsteroidal anti-inflammatory drugs.
Eur. J. Gastroenterol. Hepatol. (2003)
15:173-178.
15. ABRAMSON SB: Et tu, acetaminophen?
Arthritis Rheum. (2002) 46:2831-2835.
•• An evidence-based editorial on GI
safety of paracetamol.
16. CRYER BL: Acetaminophen not associated
with gastrointestinal toxicity: comment on
the article by Rahme et al. Arthritis Rheum.
(2003) 48:2074-2075.
17. BANNWARTH B: Concerns about
gastrointestinal safety of acetaminophen:
comment on the article by Rahme et al.
Arthritis Rheum. (2003) 46:2075.

Expert Opin. Drug Saf. (2004) 3(4) 271

Gastrointestinal safety of paracetamol: is there any cause for concern?

Affiliation
Bernard Bannwarth MD
Service de Rhumatologie, Hôpital Pellegrin,
CHU de Bordeaux, 33076 Bordeaux Cedex,
France
Tel: +33 (0) 5 56 79 55 56;
Fax: +33 (0) 5 56 93 04 07;
E-mail: bernard.bannwarth@u-bordeaux2.fr

272 Expert Opin. Drug Saf. (2004) 3(4)