PHARMACOLOGY
ADVERSE DRUG INTERACTIONS IN DENTAL PRACTICE:
INTERACTIONS ASSOCIATED WITH ANALGESICS
PART III IN A SERIES
DANIEL A. HAAS, D.D.S., PH.D.
A B S T R A C T
Background. Numerous reports of drug
interactions exist, yet not all are valid in or perti-
nent to dentistry. This article provides an overview
of drug interactions with analgesics and identifies
those that are clinically relevant.
Methods. The author reviewed reports of
drug interactions involving nonsteroidal anti-in-
flammatory drugs, or NSAIDs (including aspirin),
acetaminophen and opioids to determine the inter-
actions’ validity and clinical relevance. Consistent
with the practice followed in other articles in this
series, the author determined the significance of
the proposed interaction by gauging its reported
severity and the quality of the documentation.
Results and Conclusions.
NSAIDs should not be taken by patients taking
high-dose methotrexate, anticoagulants or alcohol.
They should be avoided in elderly or renally im-
paired patients taking digoxin, and avoided over
the long term in those taking other NSAIDs. It is
possible that NSAIDs should not be given to pa-
Pain management is inherent to dental prac-
tice. It has been estimated that dentists write ap-
proximately 16 million prescriptions for anal-
gesics each year in the United States alone.1 Such
widespread use of these drugs necessitates that
dentists have a clear understanding of their ap-
plied pharmacology, including an awareness of
clinically important drug interactions.
Analgesics commonly used to manage acute
postoperative pain in dentistry are listed in Table
1. Monographs on their use contain many precau-
tions regarding potential drug interactions. The
clinical relevance of these interactions is not al-
ways clear, in view of the fact that the use of
analgesics in dentistry can differ from their use in
medicine. As an example, nonsteroidal anti-in-
Copyright ©1998-2001 American Dental Association. All rights reserved.
tients taking lithium, but future studies should be
conducted to confirm this. Use of NSAIDs likely is
appropriate in the short term with patients taking
antihypertensives, unless they have severe conges-
tive heart disease. Aspirin should not be given to
patients taking oral hypoglycemics, valproic acid
or carbonic anhydrase inhibitors. Acetaminophen
may be given in the short term to any patient with
a healthy liver, but it should not be given to a pa-
tient who has stopped drinking alcohol after chron-
ic intake. Opioids should not be combined with al-
cohol, and meperidine must be avoided in the
patient who has taken monoamine oxidase in-
hibitors in the previous 14 days.
Clinical Implications. Drug inter-
actions with analgesics are often reported, but only
a small number have clinical relevance in den-
tistry. Awareness of the significance of these inter-
actions will allow dentists to prescribe analgesics
optimally and minimize the potential for adverse
reactions.
flammatory drugs, or NSAIDs, are commonly pre-
scribed over the long term for the management of
arthritis. Specific interactions may manifest
themselves after such prolonged use. In dentistry,
the major indication for NSAIDs is to manage
acute postoperative pain, which requires prescrip-
tions of only a few days’ duration.
If unfounded warnings against using certain
analgesics are followed, patients may be denied
appropriate pain relief. Therefore, it is important
to know which interactions are meaningful.
Consistent with the other articles in this series,
this article will assess the validity of reported
drug interactions involving analgesics used in
dentistry. The rating system is summarized in
Table 2. The conclusions regarding reports of
JADA, Vol. 130, March 1999 397
 PHARMACOLOGY

TABLE 1 were to be accepted at face

value, it would be reasonable to
ANALGESICS USED FOR POSTOPERATIVE DENTAL PAIN. avoid prescribing NSAIDs alto-
GENERIC NAME TRADE NAME* gether, as the likelihood of an
Nonsteroidal Anti-inflammatory Drugs
interaction appears great. Yet,
many of these recommendations
Aspirin (many available)
are based on case reports, small
Ibuprofen Advil, Motrin, Nuprin clinical trials, studies of specific
Flurbiprofen Ansaid
patient populations (such as el-
derly patients) or situations in
Diflunisal Dolobid
which NSAIDs are used chroni-
Naproxen Aleve, Anaprox, Naprosyn cally. It is beyond the scope of
Ketorolac Toradol
this article to provide a detailed
discussion of the validity of
Ketoprofen Orudis
each of these putative interac-
Meclofenamate Meclomen tions. The following will briefly
Etodolac Lodine
describe those of clinical rele-
vance in dentistry.
Other Nonopioid Drug
NSAIDs and antihyperten-
Acetaminophen Tylenol sives. There is sufficient evi-
Opioids† dence to support an interaction
Codeine (combined in numerous
between NSAIDs and three
formulations) major classes of antihyperten-
Oxycodone in Percocet, Percodan
sives:
dACEIs, such as captopril
Meperidine Demerol (Capoten, Bristol-Myers
Pentazocine Talwin-Nx Squibb), enalapril maleate
Hydrocodone in Lortab, Vicodin
(Vasotec, Merck), fosinopril
(Monopril, Bristol-Myers
Dihydrocodeine in Synalgos-DC Squibb) and lisinopril (Prinivil,
Propoxyphene Darvon Merck);
* The trade-named analgesics are listed only as examples. Manufacturers are as follows:
ddiuretics such as furosemide
Advil, Whitehall-Robins; Motrin, Upjohn; Nuprin, Bristol-Myers Squibb; Ansaid, Upjohn; (Lasix, Hoeschst-Roussel),
Dolobid, Merck; Aleve, Procter & Gamble; Anaprox, Roche; Naprosyn, Roche; Toradol,
Roche; Orudis, Wyeth-Ayerst; Meclomen, Parke-Davis; Lodine, Wyeth-Ayerst; Tylenol,
ethacrynic acid (Edecrin,
McNeil; Percocet, DuPont Pharma; Percodan, DuPont Pharma; Demerol, Sanofi Winthrop Merck), hydrochlorothiazide
Pharmaceuticals; Talwin-Nx, Sanofi Winthrop Pharmaceuticals; Lortab, Whitby; Vicodin,
Knoll; Synalgos-DC, Wyeth-Ayerst; Darvon, Lilly.
(HydroDIURIL, Merck) and
†
Opioids should be used in combination with acetaminophen or an NSAID. chlorothiazide (Diuril, Merck);
dß-blockers such as propran-
analgesic interactions are sum- A review of the literature re- olol (Inderal, Wyeth-Ayerst),
marized in Table 3. veals reports of many interac- nadolol (Corgard, Bristol
tions with NSAIDs. These in- Laboratories), metoprolol
NSAIDS
clude interactions with (Lopressor, Geigy) and atenolol
NSAIDs inhibit the synthesis of angiotensin-converting enzyme (Tenormin, Zeneca).
prostaglandins and thrombox- inhibitors, or ACEIs, diuretics, At least part of these drugs’
anes. Interactions may be ex- ß-adrenergic-blockers, lithium, actions depend on renal
pected to occur with other drugs anticoagulants, methotrexate, prostaglandin mechanisms,
whose action depends on physi- sulfonylureas, cyclosporine, which exert an antihyperten-
ological levels of these media- phenytoin, valproic acid, digox- sive effect of their own.2 Anti-
tors. In addition, all NSAIDs in, corticosteroids, aminoglyco- hypertensive drugs that do not
are highly protein-bound, which sides, carbonic anhydrase in- depend on renal prostaglandins
may predispose them to interac- hibitors, alcohol, probenecid, are not implicated. Therefore,
tions with drugs that share this zidovudine, acetaminophen and the calcium-channel blockers—
characteristic. other NSAIDs. If these reports such as nifedipine (Adalat,

398 JADA, Vol. 130, March 1999

Copyright ©1998-2001 American Dental Association. All rights reserved.

Miles Pharmaceutical), verap-
amil (Isoptin, Knoll) or dilti-
azem (Cardizem, Marion
Merrell Dow)—do not interact
with NSAIDs.
Prostaglandins modulate
vasodilatation, glomerular fil-
tration, renal tubular secretion
of sodium and water, and the
renin-angiotensin-aldosterone
system. NSAIDs may attenuate
ACEI action directly by inhibit-
ing renal prostaglandin synthe-
sis and indirectly by interfering
with ACEI-induced prosta-
glandin production. Prosta-
glandins are even more impor-
tant in mediating the actions of
ACEIs in hypertensive patients
who have low renin production.
Black and elderly people tend to
have low-renin hypertension.3
ß-blockers reduce blood pres-
sure by a number of mecha-
nisms, including an increase in
circulating prostaglandins.
Their effect may be inhibited by
an NSAID-induced blockade of
prostaglandin synthesis.
NSAIDs interfere with diuretics
by reducing their efficacy in se-
creting sodium and affecting
plasma renin activity.
The evidence for this NSAID-
antihypertensive interaction
comes from numerous case re-
ports and clinical trials, many
of which have yielded conflict-
ing conclusions. With such a
large number of clinical trials,
one might expect that meta-
analyses would yield valuable
information, as they group re-
sults from a number of studies
to provide a quantitative basis
for conclusions. Unfortunately,
the conclusions from two meta-
analyses investigating this in-
teraction are not entirely con-
sistent.4,5 One meta-analysis
showed that indomethacin and
naproxen were associated with
a statistically significant in-
Copyright ©1998-2001 American Dental Association. All rights reserved.
TABLE 2
THE DRUG INTERACTION SIGNIFICANCE RATING SCALE.*
SIGNIFICANCE SEVERITY RATING
RATING
1 Major
2 Moderate
3 Minor
4 Major or moderate
5 Minor
All
* This rating scale was described in depth in the first article in this series.10
crease in blood pressure, where-
as aspirin and ibuprofen had
negligible effects.4 The other
meta-analysis showed that only
piroxicam was implicated in a
statistically significant increase
in blood pressure and in-
domethacin and ibuprofen in
nonsignificant increases, where-
as naproxen, aspirin, flurbipro-
fen and sulindac had no effect.5
These conflicting results pre-
vent us from making definitive
recommendations.
To what degree do NSAIDs
inhibit antihypertensive action?
One meta-analysis has shown
that, on average, NSAIDs raise
mean blood pressure by 5 mil-
limeters of mercury.5 An eleva-
tion of this magnitude is signifi-
cant in that antihypertensive
drugs are deemed to be effica-
cious if they decrease mean ar-
terial pressure by at least 5 mm
Hg. This same decrease may
lower the relative risk of a
stroke in hypertensive patients
by 45 percent.6,7 Other studies
have shown that an increase of
5 mm Hg in diastolic pressure
over a number of years increas-
es the risk of stroke by 67 per-
cent and coronary artery dis-
ease by 15 percent.8 Clearly, the
amount of blood pressure in-
JADA, Vol. 130, March 1999
PHARMACOLOGY
DOCUMENTATION
RATING
Established, proba-
ble or suspected
Established, proba-
ble or suspected
Established, proba-
ble or suspected
Possible
Possible
Unlikely
crease induced by NSAIDs is
clinically relevant when
NSAIDs are used chronically.
Conversely, it has been stat-
ed that increases approximating
10 mm Hg may be of little im-
portance in the short term.2
Short-term prescribing is most
common in dentistry, and this
fact may allow dentists to con-
tinue using NSAIDs when indi-
cated to manage acute pain.
What period of NSAID ad-
ministration is required to af-
fect antihypertensives? The ma-
jority of the durations in 50
randomized controlled trials re-
ported in a meta-analysis5
ranged from two weeks to six
weeks. The minimum time for
an effect to appear has been re-
ported as one day for indo-
methacin, a drug used to treat
inflammatory conditions but not
often used in dentistry.5 An as-
sessment of the agents used in
dentistry shows that the mini-
mum time for an effect has been
eight days for ibuprofen, seven
days for flurbiprofen and seven
days for naproxen.5 These data
strongly support the recommen-
dation that patients who are
taking ACEIs, diuretics or ß-
blockers also may be prescribed
an NSAID, provided the dura-
399
 PHARMACOLOGY
TABLE 3
ADVERSE DRUG INTERACTIONS IN DENTISTRY INVOLVING ANALGESICS USED FOR A PERIOD
OF LESS THAN FIVE DAYS.
DRUG INTERACTION
Nonsteroidal Anti-inflammatory Drugs†
NSAIDs and certain antihyperten-
sives: angiotensin-converting en-
zyme inhibitors, diuretics, ß-block-
ers (interaction does not apply to
calcium-channel blockers)
NSAIDs and lithium
NSAIDs and anticoagulants
NSAIDs and methotrexate
NSAIDs and alcohol
NSAIDs and digoxin
NSAIDs and cyclosporine
NSAIDs and other NSAIDs,
acetaminophen
Aspirin and oral hypoglycemics
Aspirin and anticonvulsants
Aspirin and carbonic anhydrase
inhibitors
* This rating system was described previously.10 See Table 2.
†
The considerations listed for NSAIDs include aspirin.
tion is four days or less. 9
Using the system introduced
in the original article in this se-
ries,10 which is shown in Table
2, this interaction may be rated
400 JADA, Vol. 130, March 1999
Copyright ©1998-2001 American Dental Association. All rights reserved.
SIGNIFICANCE CLINICAL IMPLICATIONS
RATING*
4 An NSAID may be coprescribed if
required for four days or less.
Coadministration should be avoided
in patients with severe congestive
heart disease. Should be combined
cautiously in elderly or black pa-
tients.
2 Toxicity may result; however, evi-
dence is not clear at the present
time. Combination should be avoid-
ed, or NSAIDs should be prescribed
for very short term. Use with elderly
patients should be avoided.
2 Gastrointestinal bleeding may
result. Combination should be avoid-
ed. High-dosage aspirin (more than
3 grams per day) is rated 1.
1 Toxicity may result. Combination with
high-dosage methotrexate, as
used for cancer therapy, should be
avoided. Low-dosage methotrexate,
as used for arthritis, is of little concern.
2 Predisposes patient to gastrointesti-
nal bleeding. Combination should be
avoided.
2 Toxicity may result. Combination
should be avoided if patient is
elderly or has renal disease; less
concern is necessary if patient’s
renal function is normal.
4 Toxicity may result. Combination
should be avoided if possible.
5 Renal damage may result when
given long term. Combination
should be avoided if possible.
2 Hypoglycemic effect may be in-
creased. Combination should be
avoided.
4 Toxicity may result with valproic
acid. Combination should be avoided
if possible.
3 Toxicity may result. Combination
should be avoided if possible.
Continued on page 401
as follows. For long-term use, nificance rating of 2. When we
the severity of the interaction is assess short-term use of the
moderate and the documenta- NSAIDs common in dentistry
tion indicates that it is proba- (Table 1), the documentation in-
ble, which would lead to a sig- dicates that the interaction is,
 PHARMACOLOGY

TABLE 3 (CONTINUED)

ADVERSE DRUG INTERACTIONS IN DENTISTRY INVOLVING ANALGESICS USED FOR A PERIOD

OF LESS THAN FIVE DAYS.
DRUG INTERACTION SIGNIFICANCE CLINICAL IMPLICATIONS
RATING*

Acetaminophen

Acetaminophen and alcohol

Acute alcohol ingestion 5 Combination may be used if patient
has a healthy liver. In alcoholics, or
those with liver disease, lower maxi-
mum dosage (less than 4 g/day)
should be used.

Cessation of alcohol ingestion 1 Liver damage may result. Alcoholic

after chronic intake patients should not be told to stop
drinking if acetaminophen is used.

Opioids and alcohol 2 Additive sedation may result.

Combination should be avoided.

Meperidine and monoamine oxidase 1 Toxicity may result. Combination

inhibitors, or MAOIs should be avoided if patient has
taken MAOI in past 14 days.

* This rating system was described previously.10 See Table 2.

at most, possible. Therefore, in
our context, a significance rat-
ing of 4 is most appropriate.
In conclusion, it appears that
we still can safely prescribe
NSAIDs for a short duration to
patients taking antihyperten-
sives. Possible exceptions are
those who are most susceptible
to this interaction: patients who
are elderly, have severe conges-
tive heart failure or have low
concentrations of renin. In
these cases, use of acetamino-
phen is most appropriate.
NSAIDs and lithium.
Lithium therapy is the treat-
ment of choice for patients with
bipolar depression. This drug is
associated with a low therapeu-
tic index; that is, the effective
dose is close to the toxic dose.
Adverse effects of excessive
lithium concentrations include
polyuria, polydipsia, nausea,
vomiting, diarrhea, tremors and
sedation. Even higher concen-
trations can lead to convulsions,
coma and death.
It has been suggested that
NSAIDs increase the serum
concentration of lithium and
thereby predispose the patient
to toxicity. The mechanism is
not known with certainty, but
it may involve inhibition of
renal prostaglandins that leads
to increased lithium reabsorp-
tion, which is relevant because
lithium is excreted primarily by
the kidneys.11,12 This proposed
interaction is described in a
number of case reports and
small clinical trials. Indo-
methacin is reported to have
the greatest effect, whereas
sulindac and aspirin do not
alter lithium levels.13,14 Two
studies have implicated ibupro-
fen and naproxen, two drugs
used in dentistry.13,15 It must be
pointed out that these latter
studies were small, with sam-
ple sizes of nine and seven, re-
spectively. Furthermore, they
involved only older patients,
and there was great interindi-
vidual variability in the mea-
sured lithium levels. In addi-
tion, the researchers did not
assess patients until three days
after NSAID administration.
The accumulation of these fac-
tors should lead to a reconsid-
eration of the validity of this in-
teraction. It may occur in
specific population groups such
as elderly people, but the evi-
dence for a general recommen-
dation is weak. Given the lack
of rigorous evidence supporting
the interaction, it is tempting
to ignore it, except for lithium’s
low therapeutic index. If the in-
teraction truly exists, the con-
sequences are serious. Clearly,
more study is needed before
definitive recommendations can
be made.
This NSAID-lithium interac-
tion has the potential to be se-
vere in susceptible people tak-
ing lithium. It is not clear who
is predisposed to this interac-
tion, but elderly people likely
are involved. In terms of the
quality of the documentation, it
seems equivocal whether this
interaction should be rated as
suspected, because there are

JADA, Vol. 130, March 1999 401

Copyright ©1998-2001 American Dental Association. All rights reserved.
 PHARMACOLOGY
data suggesting an interaction,
or as possible, because the data
are limited. At this point, it
may be more prudent to assign
a rating of suspected, which
then leads to a significance rat-
ing of 2. We can hope that clini-
cal trials soon will be published
that will help resolve this issue.
In the interim, it may be best to
prescribe NSAIDs for only very
short durations, if at all, to pa-
tients taking lithium, especially
if they are elderly.
NSAIDs and anticoagu-
lants. Upper gastrointestinal
bleeding is the most common se-
rious adverse event associated
with NSAIDs.16 It may account
for more than 2,600 deaths in the
United States yearly in patients
who have rheumatoid arthritis.16
Fatal hemorrhage has been re-
ported with all NSAIDs.17
Therefore, when an NSAID is
combined with an anticoagulant
such as warfarin (Coumadin,
DuPont), there clearly is poten-
tial for excessive bleeding.
Anticoagulants are indicated
for the prophylaxis or treatment
of deep venous thrombosis or
pulmonary thromboembolism, or
for prophylaxis for thromboem-
bolism associated with atrial fib-
rillation, myocardial infarction
or prosthetic heart valves.
Predisposition to gastric bleed-
ing stems from the NSAIDs’
damaging effects on the gastric
mucosa as well as their inhibi-
tion of platelet function. Fur-
thermore, the hypoprothrom-
binemic effect of anticoagulants
is increased by specific NSAIDs,
most likely by increased serum
concentrations of warfarin sec-
ondary to displacement from
plasma proteins.18 Moreover,
higher dosages of aspirin (for ex-
ample, more than 3 grams per
day) can lead to reduced levels of
prothrombin, which will further
402 JADA, Vol. 130, March 1999
Copyright ©1998-2001 American Dental Association. All rights reserved.
exacerbate the bleeding prob-
lems already mentioned.
Aspirin has the greatest po-
tential for causing this interac-
tion. Other NSAIDs also may be
problematic, as shown in a ret-
rospective survey of NSAID use
in elderly people,19 who have an
increased risk of gastric bleed-
ing. In particular, high-dose as-
pirin, mefenamic acid and keto-
profen should be avoided in
patients receiving warfarin.20
The severity of this interac-
tion with NSAIDs in general is
at least moderate, and perhaps
major when considering aspirin.
The documentation indicates
that it is at least suspected, if
not probable. Therefore,
NSAIDs in general have a sig-
nificance rating of 2, and for
high-dosage aspirin (more than
3 g per day), a significance rat-
ing of 1 is appropriate. There-
fore, NSAIDs, and particularly
aspirin, should not be taken by
patients who are receiving anti-
coagulant therapy. Avoidance is
most important in elderly pa-
tients, as they are most suscep-
tible to this adverse event.
NSAIDs and methotrex-
ate. Methotrexate (Rheuma-
trex, Lederle) has a number of
indications, such as the treat-
ment of rheumatoid arthritis or
psoriasis, and there is little con-
cern regarding an interaction
with NSAIDs when used for
these purposes. However, it is
believed that NSAIDs reduce
the renal clearance of metho-
trexate, which can lead to toxic-
ity when the latter drug is used
in much higher dosages, as it is
for the treatment of cancer.21
Ketoprofen, flurbiprofen,
naproxen and ibuprofen all
have been implicated.22-24
The severity of this interac-
tion is major, as renal failure
and pancytopenia can result.
The documentation indicates
that the interaction is suspect-
ed, leading to a significance rat-
ing of 1. Therefore, NSAIDs
should be avoided in patients
who are taking high dosages of
methotrexate for the treatment
of cancer.
NSAIDs and ethanol. Both
ethanol and NSAIDs, particu-
larly aspirin, damage the gastric
mucosal barrier. Ethanol stimu-
lates gastric acid, potentiates
aspirin-induced gastrointestinal
blood loss and prolongs bleeding
time.25 It has been recommended
to separate the ingestion of as-
pirin and alcohol by at least 12
hours.26 The severity of this in-
teraction usually is moderate;
the documentation indicates
that it is probable, leading to a
significance rating of 2.
NSAIDs and digoxin.
Digoxin (Lanoxin, Burroughs
Wellcome), a drug used for the
treatment of congestive heart
disease, has a low therapeutic
index. It is cleared primarily by
the kidneys, and a potential in-
teraction may occur owing to
the NSAIDs’ ability to reduce
renal function. Plasma levels of
digoxin have been reported to
be elevated when NSAIDs were
taken concurrently.27 This ap-
pears to be clinically relevant
only in those predisposed to tox-
icity, namely elderly people or
those with renal disease, for
whom a significance rating of 2
is warranted. If renal function
is normal, there should be little
concern regarding concurrent
prescribing.
NSAIDs and cyclosporine.
Cyclosporine (Sandimmune,
Sandoz Pharmaceuticals) is used
to prevent rejection of trans-
planted organs. There are case
reports of patients experiencing
nephrotoxicity when NSAIDs
are used concurrently.18,28 The

potential consequences of
nephrotoxicity are serious, but
the documentation is not yet
strong; therefore, a significance
rating of 4 is appropriate.
Future studies will help clarify
the likelihood of this interaction.
NSAIDs and other NSAIDs
or acetaminophen. NSAIDs
are associated with nephrotoxic-
ity when used improperly, par-
ticularly when taken chronical-
ly or in combination with other
NSAIDs or acetaminophen.
According to Henrich and col-
leagues,29 the development of
analgesic nephropathy as a re-
sult of analgesic combinations is
clearly supported by one
prospective cohort study and
five case-control studies. The
National Kidney Foundation
produced a position paper in
1996 recommending that as-
pirin not be taken within 48
hours of another NSAID and
vice versa.29 Combining ac-
etaminophen and aspirin over
the long term must be avoided,
whereas a short duration of this
combination is less likely to
cause renal damage. For dura-
tions typically used in den-
tistry—for instance, less than
five days—a significance rating
of 5 would be warranted, as a
reaction is unlikely. Neverthe-
less, it would appear to be most
prudent to avoid these combina-
tions and select either acet-
aminophen, aspirin or another
NSAID, with or without an opi-
oid, for postoperative pain man-
agement.
Aspirin and sulfonylureas
(oral hypoglycemics).
Salicylates interact with the
sulfonylurea oral hypoglycemics
such as chlorpropamide
(Diabinese, Pfizer Labora-
tories), glyburide (DiaBeta,
Hoechst Marion Roussel), aceto-
hexamide (Dymelor, Lilly), to-
Copyright ©1998-2001 American Dental Association. All rights reserved.
lazamide (Tolinase, Upjohn)
and tolbutamide (Orinase,
Hoechst Marion Roussel). An
increase in hypoglycemic effect
occurs when aspirin is used
with one of these agents.21
Salicylates enhance insulin se-
cretion and reduce plasma glu-
cose levels. Displaced protein
binding of sulfonylureas may
also be a factor. The severity of
this interaction is moderate and
the documentation indicates
that it is probable, leading to a
significance rating of 2.
Therefore, aspirin should be
avoided in diabetic patients who
are taking the sulfonylurea oral
hypoglycemics.
Aspirin and anticonvul-
sants. The anticonvulsant val-
proic acid (Depakene, Abbott)
may be displaced from plasma
proteins, and its main metabolic
pathway may be inhibited by
high doses of aspirin. These in-
teractions result in an increase
in free serum valproate concen-
trations and subsequent toxici-
ty, which may manifest as neu-
rological, hematologic or
gastrointestinal signs and
symptoms. The evidence sup-
porting this interaction is based
primarily on case reports,30
which leads to a significance
rating of 4.
The anticonvulsant phenytoin
(Dilantin, Parke-Davis) also may
be affected, as it is displaced
from plasma protein by aspirin.
However, there is little evidence
to support any increased likeli-
hood of phenytoin toxicity.
Aspirin and carbonic an-
hydrase inhibitors. Carbonic
anhydrase inhibitors, such as
acetazolamide (Diamox,
Lederle), are used for a number
of indications, including man-
agement of altitude sickness,
glaucoma and epilepsy. Aspirin
displaces acetazolamide from
PHARMACOLOGY
plasma protein and inhibits
renal clearance. Excessive levels
of acetazolamide may result,
and patients have developed
lethargy, incontinence and con-
fusion.31 Elderly people and
those with renal failure are pre-
disposed to this interaction. A
significance rating of 3 is most
appropriate. Therefore, aspirin
should not be taken by patients
who are taking these drugs.
ACETAMINOPHEN
Acetaminophen (Tylenol,
McNeil), commonly used to
manage mild-to-moderate pain,
has the advantage of providing
analgesia without the side ef-
fects associated with NSAIDs or
opioids. Acetaminophen is safe
when taken in recommended
dosages for a short duration,
consistent with the manage-
ment of acute pain in dentistry.
Analgesic nephropathy may
occur with habitual use, which
is defined as daily use for more
than five years, especially if
combined with aspirin or anoth-
er NSAID.32 A number of drug
interactions are reported, but
the only one clinically relevant
to dentistry is the result of acet-
aminophen’s combination with
alcohol.
Acetaminophen and alco-
hol. This particular interaction
has gained notoriety from re-
ports in the lay press, including
a lawsuit award in favor of a
patient who required a liver
transplantation after taking
acetaminophen.33 This patient,
who regularly consumed two or
three glasses of wine with din-
ner, came down with the flu, ab-
stained from drinking wine and
then began taking acetami-
nophen in recommended doses.
Several days later, he developed
liver failure and eventually re-
quired a transplantation. In ad-
JADA, Vol. 130, March 1999 403
 PHARMACOLOGY
METABOLISM OF ACETAMINOPHEN
95%
Acetaminophen
CYP2E1
NAPQI
Figure 1. A simplified summary of the metabolism of acetamino-
phen. The hepatotoxic potential of N-acetyl-p-benzoquinone imine,
or NAPQI, is prevented by rapid breakdown by glutathione.
dition to this case report, there
are others describing severe
hepatotoxicity in alcoholic pa-
tients who took recommended
doses of acetaminophen.
Should such case reports lead
us to recommend against acet-
aminophen use if we suspect
that the patient will be taking a
drink of alcohol? The alterna-
tives are to use aspirin or an-
other NSAID, with or without
an opioid analgesic. The risks of
combining alcohol with aspirin
or other NSAIDs already have
been described. Alcohol and opi-
oids are both central nervous
system, or CNS, depressants
and should not be combined, be-
cause excessive sedation may
result. Therefore, alcohol has
significant interactions with
each of these drug groups. This
brings us back to considering
acetaminophen and alcohol.
To understand the interac-
tion between acetaminophen
and alcohol, one must consider
the normal metabolism of acet-
aminophen. As illustrated in
Figure 1, 95 percent of acet-
aminophen is conjugated in the
liver to inactive compounds.
Cytochrome P-450 enzymes,
primarily CYP2E1, convert the
remainder into the highly reac-
tive metabolite N-acetyl-p-ben-
404 JADA, Vol. 130, March 1999
Copyright ©1998-2001 American Dental Association. All rights reserved.
Inactive Metabolites
Glutathione
zoquinone imine, or NAPQI. In
the healthy patient, this com-
pound is quickly rendered
harmless by conjugation with
glutathione. Excessive amounts
of NAPQI can cause hepatic cell
death. Toxicity may occur if
hepatic glutathione is depleted,
there is an overdose of acet-
aminophen, or the patient is
taking other drugs that induce
the production of liver enzymes.
The interaction with alcohol
arises because the enzyme
CYP2E1 is also involved in the
metabolism of ethanol, as
shown in Figure 2A. Interest-
ingly, CYP2E1 not only trans-
forms ethanol, but also is in-
duced by it, leading to increased
concentrations of CYP2E1.
Given that ethanol is a sub-
strate for CYP2E1, its presence
will inhibit the action of
CYP2E1 on other substrates,
such as acetaminophen, as
shown in Figure 2B. Therefore,
ethanol both inhibits and in-
duces this enzyme.
This simultaneous induction
and inhibition leads to a com-
plex interaction with acet-
aminophen, which has been de-
scribed in detail by Slattery and
colleagues.34 Chronic ingestion
of alcohol causes a gradual in-
crease in the amount of
CYP2E1. Yet, as long as the pa-
tient continues to drink alcohol,
CYP2E1 will be occupied by
ethanol and not be fully avail-
able to metabolize other drugs.
If the patient takes acet-
aminophen concurrently, the
formation of the toxic NAPQI
actually is decreased. When al-
cohol ingestion stops, CYP2E1
is no longer occupied and now
turns to a substrate that it can
metabolize. If acetaminophen is
present, it will become the me-
tabolized substrate and the for-
mation of NAPQI will rise
markedly, as shown in Figure 3.
As long as ethanol is present,
the patient is protected from
the formation of this toxic
metabolite. Once ethanol is no
longer there, that protection is
lost.
This mechanism is supported
by animal studies showing that
there is decreased NAPQI forma-
tion, and hence decreased toxici-
ty, if acetaminophen administra-
tion is accompanied by ethanol,
either taken as a single dose or
used continuously over the long
term.35-37 In nonalcoholic humans,
ethanol inhibits NAPQI forma-
tion.38 If ethanol is given over the
long term to animals and then
stopped for a sufficient amount of
time, NAPQI formation increas-
es. Similarly, in human alcoholic
patients, if ethanol ingestion is
interrupted for at least 12 hours,
CYP2E1 activity, NAPQI forma-
tion and acetaminophen toxicity
are all increased.39
Therefore, ethanol may lead
to either an increase or a de-
crease in NAPQI formation, de-
pending on the extent of chronic
ethanol consumption and when
it was last taken before acet-
aminophen administration.34
Ethanol’s protective effect is
best illustrated by the report of
a patient who took 60 g of acet-
 PHARMACOLOGY

aminophen after 48 hours of

heavy drinking—and survived.40
METABOLISM OF ALCOHOL
This amount of acetaminophen
is equivalent to more than 180
regular-strength tablets; the CYP2E1
Ethanol Metabolites
recommended daily maximum
is 4 g—that is, 12 regular-
strength tablets. Figure 2. A. A simplified summary of part of the metabolism of alco-
Does this mean that acet-
aminophen is safe for alcoholic
patients? The answer is a quali- METABOLISM OF ACETAMINOPHEN IN A PATIENT
fied “no,” as there are other rel- INGESTING ALCOHOL
evant factors. Glutathione,
which normally detoxifies Inactive
Acetaminophen
Metabolites
NAPQI, is decreased in alco-
holic patients, and therefore its
beneficial effect is diminished.
Also, alcoholic people are more
likely to have a fatty liver and CYP2E1 Glutathione
suffer from malnutrition, which
NAPQI
may further predispose them to
toxicity. It is now believed that
CYP2E1
some of the case reports of se- Ethanol Metabolites
vere liver injury in alcoholic pa-
tients who took recommended Figure 2. B. A simplified summary of the metabolism of acetamino-
doses of acetaminophen may phen in a patient using alcohol. The CYP2E1 enzyme is occupied by
ethanol as long as a sufficient amount of ethanol is present. If occu-
have occurred in patients who pied, CYP2E1 is not available to metabolize acetaminophen into N-
abstain from alcohol for a peri- acetyl-p-benzoquinone imine, or NAPQI. Chronic ethanol ingestion
od of time, such as overnight, increases CYP2E1 content.

and then take acetaminophen.34

Sufficient time elapsed for
ethanol levels to drop, thereby
freeing the induced CYP2E1 to
metabolize acetaminophen. This
same mechanism likely can be
applied to the patient involved
in the lawsuit described above.
In conclusion, in the nonalco-
holic patient, administration of
ethanol and acetaminophen to-
gether results in less NAPQI
formation than administration
of acetaminophen alone. In alco-
holic people, or those who ingest
alcohol on a regular basis, sud-
den cessation of alcohol inges-
tion creates a major risk of
enhanced acetaminophen toxici-
ty. The alcoholic patient should
not abstain from alcohol if
acetaminophen is required.
Even so, the maximum daily
dose of acetaminophen in alco-
holic patients should be reduced
below the usual recommenda-
tion of 4 g.
Classification of this interac-
tion is complicated by the fact
that the outcome depends on
the timing of alcohol intake and
cessation. When alcohol and
acetaminophen are ingested to-
gether, a significance rating of 5
is warranted, as no adverse out-
come is expected. However, if
acetaminophen is given to a pa-
tient who chronically ingests al-
cohol and then stops for a peri-
od of at least 12 hours, the
severity increases to major be-
cause permanent liver damage
may result. Given that the doc-
umentation for this indicates
that it is a probable occurrence,
this combination would warrant
a significance rating of 1.
Acetaminophen and
other drugs. Other interac-
tions with acetaminophen,
such as those with sulfinpyra-
zone, phenytoin or zidovudine,
are not clinically relevant in
dentistry if acetaminophen is
used in therapeutic dosages for
the short term. Recently, ac-
etaminophen was reported to
be an unrecognized cause of
overanticoagulation in patients
taking the anticoagulant war-
farin.41 This dose-dependent in-
teraction was demonstrated in
a case-control study of patients
whose mean age was 70 years.
If the validity of this interac-
tion is confirmed by future
studies, patients taking war-
farin should be advised to
avoid not only NSAIDs, but
also acetaminophen.

JADA, Vol. 130, March 1999 405

Copyright ©1998-2001 American Dental Association. All rights reserved.
 PHARMACOLOGY

has taken an MAOI in the pre-

METABOLISM OF ACETAMINOPHEN AFTER CHRONIC vious 14 days.
ALCOHOL INGESTION, THEN ABRUPT ABSTINENCE
CONCLUSIONS

Drug interactions with anal-

Acetaminophen Inactive Metabolites
gesics are often reported, but
only a small number are clini-
cally relevant in dentistry.
CYP2E1 Glutathione
NSAIDs (including aspirin)
should not be taken by patients
NAPQI
who are taking high-dosage
methotrexate, anticoagulants
or alcohol. Their use should be
Liver Toxicity avoided by elderly or renally
impaired patients who are tak-
Figure 3. Metabolism of acetaminophen after chronic alcohol inges-
tion followed by abrupt abstinence. If alcohol is removed after
ing digoxin, and also over the
chronic ingestion, the excess CYP2E1 now metabolizes ac- long term by patients taking
etaminophen. The capacity of glutathione to inactivate the metabo- other NSAIDs or acet-
lite N-acetyl-p-benzoquinone imine, or NAPQI, is then exceeded.
aminophen. It is possible that
OPIOIDS antidepressants, which include NSAIDs should not be given to
phenelzine (Nardil, Parke- patients taking lithium, but fu-
Opioid analgesics are indicated Davis), tranylcypromine ture studies should be conduct-
for moderate-to-severe levels of (Parnate, SmithKline ed to confirm this. NSAIDs may
pain and should be used in com- Beecham), isocarboxazid be given in the short term to
bination with acetaminophen or (Marplan, Hoffman-LaRoche) patients taking antihyperten-
an NSAID. Although associated and the antiparkinsonian agent sives, unless they have severe
with numerous side effects, such selegiline (Eldepryl, Somerset/ congestive heart disease.
as respiratory depression, seda- Sandoz). The mechanism for Aspirin should not be given to
tion, constipation and nausea, this interaction is not clear, but patients taking oral hypo-
opioids have only a few interac- may be the result of an accumu- glycemics, the anticonvulsant
tions of concern for the dentist. lation of serotonin secondary to valproic acid or carbonic anhy-
Opioids and alcohol. the MAO inhibition.42 Patients drase inhibitors. Acetami-
Alcohol and opioid analgesics who have taken therapeutic nophen may be given in the
are both CNS depressants, and dosages of meperidine while short term to any patient with
their combination will produce taking an MAOI or within the a healthy liver, but it should
increased sedation. The severity previous 14 days have devel- not be given to a patient who
of this interaction may be con- oped excitatory signs: agitation, has stopped drinking alcohol
sidered moderate. The docu- hypertension, hyperpyrexia, after chronic intake. Opioids
mentation indicates that it is tachycardia and seizures. should not be combined with al-
probable, leading to a signifi- Alternatively, depressive reac- cohol, and meperidine must be
cance rating of 2. Patients tions—respiratory depression, avoided in a patient who has
should be advised to avoid in- hypotension, cyanosis and taken MAOIs in the previous
gesting alcohol if they are tak- coma—have occurred. There are 14 days. ■
ing an opioid. Ideally, opioids reports of fatalities as a result
Dr. Haas is an associate professor and head
should not be selected as anal- of this interaction.43,44 of anaesthesia, Faculty of Dentistry, and an
gesics for alcoholic patients. The severity of this interac- associate professor, Department of
Pharmacology, Faculty of Medicine,
Meperidine and tion is major, as fatality is a University of Toronto, 124 Edward St.,
monoamine oxidase in- possibility; the documentation Toronto, Ontario M5G 1G6, Canada. Address
reprint requests to Dr. Haas.
hibitors. Meperidine (Demerol, indicates that it is probable.
Sanofi Winthrop Pharmaceu- Therefore, this interaction war-
This article is based on material presented
ticals) has a significant interac- rants a significance rating of 1, March 4, 1998, in a symposium entitled
tion with the monoamine oxi- and meperidine should not be “Adverse Drug Interactions in Dentistry:
Separating the Myths From the Facts.” The
dase inhibitor, or MAOI, prescribed to any patient who

406 JADA, Vol. 130, March 1999

Copyright ©1998-2001 American Dental Association. All rights reserved.

symposium was presented at the 27th
General Session of the American Association
for Dental Research in Minneapolis. The sym-
posium was jointly sponsored by the
International Association for Dental
Research, the American Association of Dental
Schools and the American Dental Association
and was supported in part by a grant from
Warner Lambert Pharmaceuticals.
The drugs listed by brand name in this arti-
cle are given as examples only. Their listing
does not imply any endorsement.
1. Dionne RA, Gordon SM. Nonsteroidal
anti-inflammatory drugs for acute pain con-
trol. Dent Clin North Am 1994;38:645-67.
2. Houston MC. Nonsteroidal anti-inflam-
matory drugs and antihypertensives. Am J
Med 1991;90(suppl 5A):42-7S.
3. Dowd FJ, Jeffries WS, Matheny JL.
Antihypertensive drugs. In: Yagiela JA,
Neidle EA, Dowd FJ, eds. Pharmacology and
therapeutics for dentistry. 4th ed. St. Louis:
Mosby–Year Book; 1998:381.
4. Pope JE, Anderson JJ, Felson DT. A
meta-analysis of the effects of nonsteroidal
anti-inflammatory drugs on blood pressure.
Arch Intern Med 1993;153:477-84.
5. Johnson AG, Nguyen TV, Day RO. Do
nonsteroidal anti-inflammatory drugs affect
blood pressure? A meta-analysis. Ann Intern
Med 1994;121:289-300.
6. MacFarlane LL, Orak DJ, Simpson WM.
NSAIDs, antihypertensive agents and loss of
blood pressure control. Am Fam Physician
1995;51(4):849-56.
7. Medical Research Council Working Party.
MRC trial of treatment of mild hypotension:
principal results. BMJ 1985;291:97-104.
8. Collins R, Peto R, MacMahon S, et al.
Blood pressure, stroke, and coronary heart
disease, Part 2: short-term reductions in
blood pressure: overview of randomised drug
trials in their epidemiological context. Lancet
1990;335(8693):827-38.
9. Becker DE. Drug interactions in dental
practice: a summary of facts and controver-
sies. Compend Contin Educ Dent
1994;15:1228-42.
10. Moore PA, Gage TW, Hersh EV, Yagiela
JA, Haas DA. Adverse drug interactions in
dental practice: professional and educational
implications. JADA 1999;130:47-54.
11. Harvey NS, Merriman S. Review of clin-
ically important drug interactions with lithi-
um. Drug Safety 1994;10:455-63.
12. Finley PR, Warner MD, Peabody CA.
Clinical relevance of drug interactions with
lithium. Clin Pharmacokinet 1995;29:172-91.
13. Ragheb M, Powell AL. Lithium interac-
Copyright ©1998-2001 American Dental Association. All rights reserved.
tion with sulindac and naproxen. J Clin
Psychopharmacol 1986;6:150-4.
14. Ragheb M. Aspirin does not significantly
affect patients’ serum lithium levels. J Clin
Psychiatry 1987;48:425.
15. Ragheb M. Ibuprofen can increase
serum lithium level in lithium-treated pa-
tients. J Clin Psychiatry 1987;48:161-3.
16. Fries JF, Miller SR, Spitz PW, Williams
CA, Hubert HV, Bloch DA. Toward an epi-
demiology of gastropathy associated with non-
steroidal antiinflammatory drug use.
Gastroenterology 1989;96(2 pt 2 suppl):647-55.
17. Acetaminophen, NSAIDs and alcohol.
Med Lett Drugs Ther 1996;38(977):55-6.
18. Brouwers JR, deSmet PA.
Pharmacokinetic-pharmacodynamic drug in-
teractions with nonsteroidal antiinflammato-
ry drugs. Clin Pharmacokinet 1994;27:462-85.
19. Shorr RI, Ray WA, Daugherty JR,
Griffen MR. Concurrent use of nonsteroidal
antiinflammatory drugs and oral anticoagu-
lants places elderly persons at high risk for
hemorrhagic peptic ulcer disease. Arch Intern
Med 1993;153:1665-70.
20. Chan TY. Adverse interactions between
warfarin and nonsteroidal antiinflammatory
drugs: mechanisms, clinical significance, and
avoidance. Ann Pharmacother 1995;29:1274-
83.
21. Tatro DS. Drug interaction facts. St.
Louis: Facts and Comparisons; 1997:496,666.
22. Thyss A, Kubaz J, Milano G. Clinical
and pharmacokinetic evidence of life threat-
ening interaction between methotrexate and
ketoprofen. Lancet 1986;1:256-8.
23. Frenia ML, Long KS. Methotrexate and
nonsteroidal antiinflammatory drug interac-
tions. Ann Pharmacother 1992;26:234-7.
24. Tracy TS, Krohn K, Jones DR, Bradley
JD, Hall SD, Brater DC. The effects of a sali-
cylate, ibuprofen, and naproxen on the dispo-
sition of methotrexate in patients with
rheumatoid arthritis. Eur J Clin Pharmacol
1992;42:121-5.
25. Hansten PD, Horn JR, eds. Hansten and
Horn’s drug interactions analysis and man-
agement. Vancouver, Wash.: Applied
Therapeutics; 1997:65.
26. Tatro DS. Drug interaction facts. St.
Louis: Facts and Comparisons; 1996:618a.
27. Johnson AG, Seidemann P, Day RO.
NSAID-related adverse drug interactions
with clinical relevance. Int J Clin Pharmacol
Ther 1994;32:509-32.
28. Harris KP, Jenkins D, Walls J.
Nonsteroidal antiinflammatory drugs and cy-
closporine. Transplantation 1988;46:598-9.
29. Henrich WL, Agodoa LE, Barrett B, et
al. Analgesics and the kidney: summary and
PHARMACOLOGY
recommendations to the Scientific Advisory
Board of the National Kidney Foundation
From an Ad Hoc Committee of the National
Kidney Foundation. Am J Kidney Dis
1996;27:162-5.
30. Goulden KJ, Dooley JM, Camfield PR,
Fraser AD. Clinical valproate toxicity induced
by acetylsalicylic acid. Neurology 1987;
37:1392-4.
31. Tatro DS. Drug interaction facts. St.
Louis: Facts and Comparisons; 1994:616.
32. DeBroe ME, Elseviers MM. Analgesic
nephropathy. N Engl J Med 1998;338:446-52.
33. Greenwald J. Bitter ads to swallow.
Time 1996;April 1:48-9.
34. Slattery JT, Nelson SD, Thummel KE.
The complex interaction between ethanol and
acetaminophen. Clin Pharmacol Ther
1996;60:241-6.
35. Sato C, Nakano M, Lieber CS.
Prevention of acetaminophen-induced hepato-
toxicity by acute ethanol administration in
the rat: comparison with carbon tetrachlo-
ride-induced hepatotoxicity. J Pharmacol Exp
Ther 1981;218:805-10.
36. Banda PW, Quart BD. The effect of alco-
hol on the toxicity of acetaminophen in mice.
Res Commun Chem Pathol Pharmacol
1984;43:127-38.
37. Tredger JM, Smith HM, Read RB,
Portmann B, Williams R. Effects of ethanol
ingestion on the hepatotoxicity and
metabolism of paracetamol in mice.
Toxicology 1985;36:341-52.
38. Banda PW, Quart BD. The effect of mild
alcohol consumption on the metabolism of
acetaminophen in man. Res Commun Chem
Pathol Pharmacol 1982;38:57-70.
39. Girre C, Lucas D, Hispard E, Menez C,
Dally S, Menez JF. Assessment of cytochrome
P4502E1 induction in alcoholic patients by
chlorzoxazone pharmacokinetics. Biochem
Pharmacol 1994;47:1503-8.
40. Lyons L, Studdiford JS, Sommaripa AM.
Treatment of acetaminophen overdosage with
N-acetylcysteine. N Engl J Med 1997;296:174.
41. Hylek EM, Heiman H, Skates SJ,
Sheehan MA, Singer DE. Acetaminophen and
other risk factors for excessive warfarin anti-
coagulation. JAMA 1998;279:657-62.
42. Drug information for the health care
professional. Vol. I. 18th ed. Rockville, Md:
United States Pharmacopeial Convention;
1998:256.
43. Janowsky ED, Risch C, Janowsky DS.
Effects of anesthesia on patients taking psy-
chotropic drugs. J Clin Psychopharmacol
1981;1:14-20.
44. Asch DA, Parker RM. The Libby Zion
case. New Engl J Med 1988;318:771-5.
JADA, Vol. 130, March 1999 407