A Randomized Study of the Efcacy and

Safety of Intravenous Acetaminophen

Compared to Oral Acetaminophen
for the Treatment of Fever
W. Frank Peacock, MD, James B. Breitmeyer, MD, PhD, Christine Pan, MS, William B. Smith, MD,
and Mike A. Royal, MD, JD
Abstract
Objectives: The purpose of this study was to assess the safety and dynamics of the onset of antipyretic
efcacy of intravenous (IV) acetaminophen versus oral (PO) acetaminophen in the treatment of
endotoxin-induced fever.
Methods: This randomized, double-blind, double-dummy, single-dose study was conducted at a single
center in the United States in healthy volunteer adult males with an endotoxin-induced fever to assess
the antipyretic efcacy and safety of IV acetaminophen 1 g versus PO acetaminophen 1 g over 6 hours.
Subjects who achieved a sufcient fever response to a test dose of reference standard endotoxin were
randomly assigned to receive either IV acetaminophen and PO placebo (n = 54) or PO acetaminophen
and IV placebo (n = 51). The primary efcacy outcome was the weighted sum of temperature differences
from baseline at time T0 through T120 minutes. Safety evaluations included adverse event (AE), physical
exam, and laboratory assessments.
Results: Of 105 subjects receiving study medication, 24 vomited within 2 hours postdose (PO acetamino-
phen, n = 15; and IV acetaminophen, n = 9) and were excluded from the modied intent-to-treat popula-
tion that consisted of 36 and 45 subjects treated with PO and IV acetaminophen, respectively. While this
was done to not confer an advantage to the IV formulation, a sensitivity analysis including these subjects
did not change the overall efcacy results. Statistically signicant results favoring IV acetaminophen
were observed for the primary endpoint (weighted sum of temperature differences over 120 minutes,
p = 0.0039) and also at each time point from T30 to T90 minutes, although the maximum mean observed
temperature difference was only 0.3C. The study drugs were well tolerated. The AE frequency was
comparable between the IV and PO groups.
Conclusions: A single dose of IV acetaminophen is as safe and effective in reducing endotoxin-induced
fever as PO acetaminophen. IV acetaminophen may be useful where patients are unable to tolerate PO
intake or when an earlier onset of action is desirable.
ACADEMIC EMERGENCY MEDICINE 2011; 18:360366 2011 by the Society for Academic Emergency
Medicine
A
cetaminophen is a synthetic, nonopioid, cen-
trally acting analgesic and antipyretic agent.
1
It has a well-established efcacy prole, a
well-understood risk benet ratio, and a very low
potential for harmful drugdrug interactions.
2
In
recommended doses, acetaminophen is considered safe
ISSN 1069-6563 2011 by the Society for Academic Emergency Medicine
360 PII ISSN 1069-6563583 doi: 10.1111/j.1553-2712.2011.01043.x
From the Department of Emergency Medicine, The Cleveland Clinic (WFP), Cleveland, OH; Cadence Pharmaceuticals (JBB, CP,
MAR), San Diego, CA; and the New Orleans Center for Clinical Research and the Department of Medicine, University of
Tennessee Medical Center (WBS), Knoxville, TN.
Received July 28, 2010; revision received September 15, 2010; accepted September 18, 2010.
Presented at the 12th Annual Meeting of the Society of Hospital Medicine, May 2009, Chicago, IL.
This study was funded by Cadence Pharmaceuticals, Inc.
Drs. Breitmeyer and Royal and Ms. Pan are employees of Cadence Pharmaceuticals, Inc., and Dr. Smith was hired as an independent
contractor.
Supervising Editor: Mark Mycyk, MD.
Address for correspondence and reprints: W. Frank Peacock, MD; e-mail: Peacocw@ccf.org.
A related commentary appears on page 428.
for infants, children, and adults. Although the exact site
and mechanism of action of acetaminophen are not
clearly dened, its effectiveness as an antipyretic agent
has been attributed to its effect on the hypothalamic
heat-regulating center.
35
Worldwide, acetaminophen is currently the most
widely used analgesic and antipyretic. Oral (PO) acet-
aminophen was initially approved by the U.S. Food and
Drug Administration (FDA) in 1951 and was rst mar-
keted in the United States in 1953. Acetaminophen has
been well known as an effective analgesic and antipy-
retic. Intravenous (IV) acetaminophen is approved for
the short-term treatment of acute pain and fever in
approximately 80 countries outside of the United States
and was rst approved in Europe in 2001. OFIRMEV
(acetaminophen for injection; Cadence Pharmaceuticals,
Inc., San Diego, CA) is currently under review by the
FDA for approval in the United States. The purpose of
this study was to assess the safety and dynamics of the
onset of antipyretic efcacy of IV versus PO acetamino-
phen in the treatment of endotoxin-induced fever.
METHODS
Study Protocol
This was a randomized, double-blind, double-dummy,
active comparator-controlled, single-dose, parallel group
study (CPI-APA-303). The protocol and the informed
consent form were approved by an independent institu-
tional review board in compliance with current regula-
tions of the local regulatory authorities and the
International Conference on Harmonisation (ICH) Good
Clinical Practices (GCP) guidelines. This study was con-
ducted in compliance with the protocol, ICH and GCP
guidelines, as well as Title 21 of the U.S. Code of
Federal Regulations Part 50 (Protection of Human Sub-
jects), Part 56 (Institutional Review Boards), and Part
312 (Investigational New Drug Application). This study
was also registered at ClinicalTrials.gov (Identication
Number: NCT00564629).
Study Setting and Population
The study was conducted at a single center in the Uni-
ted States, the New Orleans Center for Clinical
Research, University of Tennessee Medical Center, from
August 2007 to October 2007. Healthy males between
the ages of 18 and 75 were recruited from the research
institutions current database of research volunteers
and through multimedia advertisements. The volunteers
were paid a stipend for participating. After signing
informed consent forms, volunteers were eligible for
screening. Subjects had to have a body mass index
between 19 and 45 lbs in.
2
and no physical, mental, or
medical conditions that the principle investigator (WS)
deemed could have confounded study participation.
Subjects had to abstain from smoking cigarettes or
using nicotine products from the time of admission to
the clinic until study completion. Subjects were
excluded if they had any of the following: been treated
with any medication having antipyretic effects within
2 days of clinic admission; known hypersensitivity
or contraindication to receiving endotoxin, acetamino-
phen, or the excipients of the IV acetaminophen
formulation or rescue medications (e.g., ibuprofen or
ketorolac); known or suspected recent history of
alcohol or drug dependence; history of nasal polyps,
angioedema, or signicant or actively treated broncho-
spastic disease; active infection or condition that might
cause abnormal alterations in body temperature; or
impaired liver function, active hepatic disease, or evi-
dence of clinically signicant liver disease. Females
were excluded from study participation due to the
known fetotoxic effects of endotoxin.
Study Protocol
Reference standard endotoxin was manufactured by the
Bureau of Biologics under Good Manufacturing Prac-
tices guidelines and obtained from the United States
National Institutes of Health. Reference standard endo-
toxin was derived from the original bulk material
extracted from Escherichia coli O:113 (data on le with
the FDA). Reference standard endotoxin at a concentra-
tion of 10,000 EU vial from Lot No. 67801 was used in
this study.
Oral study medication consisted of acetaminophen
caplets (500 mg Tylenol extra strength caplets;
Lot PAA260; McNeil Consumer Healthcare, a Division
of McNeil-PPC, Inc., Parsippany, NJ) and identical
appearing placebo caplets. IV study medication was
acetaminophen for injection (Lot 6H11817; Cadence
Pharmaceuticals, Inc.) and identical appearing IV
placebo.
After an overnight admission, each subject was
assessed to ensure that he was afebrile and free from
occult infection based on clinical assessment, blood,
and urine testing. Afebrile was dened as a mean base-
line PO temperature around 37C that did not vary
more than 0.4C from the lowest to highest on three
repeated assessments during a 30-minute period. After
a standardized breakfast, subjects swallowed a Jonah
Ingestible Core Temperature Capsule, part of the previ-
ously validated VitalSense Integrated Physiological
Monitoring System (Respironics, Bend, OR), to allow
for remote continuous monitoring of core tempera-
ture.
6
Subjects were administered a test dose of IV ref-
erence standard endotoxin (1 ng kg body weight) to
verify the absence of a medically signicant allergic or
exaggerated systemic response. Subjects were then
administered IV reference standard endotoxin 4 ng kg.
The threshold temperature for antipyretic administra-
tion was 38.6C. Once the threshold temperature was
exceeded and it appeared that a peak fever response
was present, subjects were randomized (1:1) to receive
either IV acetaminophen 1 g (100 mL; given as a 15-
minute IV infusion) or PO acetaminophen 1 g.
Prior to study database lock and unblinding, data
from another study comparing IV acetaminophen ver-
sus placebo using the same lot of endotoxin became
available (unpublished data), which indicated that mod-
eling based on historical data (e.g., Vargas et al., which
used an earlier lot of endotoxin
7
) overestimated the
extent of the expected temperature response. Based on
these new data, a new primary endpoint similar to the
one in the more recent endotoxin study was selected to
better demonstrate the onset of action comparison.
Therefore, the primary efcacy outcome measure was
ACAD EMERG MED
April 2011, Vol. 18, No. 4

www.aemj.org 361
set as the weighted sum of temperature differences
(WSTD) at each assessment time point from T0 to
T120 minutes, weighted by the time elapsed between
each two consecutive time points. Safety endpoints
included the incidence, severity, and causality of treat-
ment-emergent adverse events (AEs); changes in endo-
toxin-induced fever-associated AEs from T0 to T360;
and clinical laboratory test assessments.
Data Analysis
The target sample size of 80 patients with a 1:1 random-
ization was estimated using data from several sources.
These sources included an outpatient trial by Bachert
et al.
8
comparing PO acetaminophen with PO aspirin in
adults with spontaneous fever from upper respiratory
tract infections, a trial by Walson et al.
7
comparing IV
propacetamol (the prodrug to acetaminophen) versus
placebo in children using a spontaneous fever model
and a trial by Vargas et al.
9
in adults evaluating the ef-
cacy of PO ketorolac and PO acetaminophen versus pla-
cebo in endotoxin-induced fever. Assuming median
times to a 1.5C reduction in temperature of 1 and
3 hours for IV and PO acetaminophen, respectively,
resulted in a sample size estimate of 39 in each group
at 80% power to detect a difference between the
groups with a 0.05 signicance level based on a two-
sided t-test.
Eligible participants who were afebrile prior to, and
who developed a fever after endotoxin dosing, were
centrally randomized in a 1:1 ratio to the IV or PO acet-
aminophen groups using a double-dummy design. The
randomization scheme was programmatically generated
using a prespecied block size and seed value for a ran-
dom generator with equal treatment allocation.
Core temperatures were recorded at the following
time points after randomization: T0, T5, T10, T15, T20,
T25, T30, T40, T50, T60, T75, T90, T105, T120, T150,
T180, T210, T240, T270, T300, T330, and T360 minutes;
T0 was dened as the start time of IV medication
infusion swallowing time for PO medication. If applica-
ble, temperature was recorded just prior to administra-
tion of rescue medication or at the time of early
termination. Subjects who vomited within 2 hours of T0
were not included in the evaluable population for
efcacy analyses due to concern over complete PO
medication absorption as a result of the known
impaired gastric emptying observed after endotoxin
administration.
10
However, sensitivity analyses were
performed including these subjects in efcacy analyses.
Subjects had access to rescue medication at all times
for the treatment of severe fever-associated symptoms.
Efcacy analyses were carried out using the modied
intent-to-treat population, dened as those subjects
who received a full dose of both study medications and
did not experience vomiting within the rst 2 hours
after T0. Subjects who received any part of a dose of
the study medications were included in the safety popu-
lation. All analyses were performed at the 0.05 signi-
cance level. The primary efcacy endpoint, the WSTD
over the rst 120 minutes, was analyzed using an analy-
sis of covariance (ANCOVA) model with treatment
group as the xed effect and temperature score at T0
as the covariate. All assumptions for ANCOVA were
met. The formula used to derive the primary endpoint
was WSTD = sum of (tmdiff ((val+prevval) 2)) through
120 minutes, where tmdiff = time difference, val = tem-
perature value at current time point, and preval = tem-
perature value at previous time point.
Other efcacy endpoints included time to a tempera-
ture reduction analyzed using a log-rank test; time to the
specic event (e.g., time to specic temperature or
rescue) estimated based on the Kaplan-Meier method
(censored at 360 minutes if a subject did not achieve the
specied temperature reduction); global evaluation at
T360 minutes summarized for each group by frequency
and percentage for each categorical response and
analyzed using unstratied Cochran-Mantel-Haenzel
mean score test using integer scores; and continuous
variables such as change in temperature, maximum
temperature reduction, and percentage of subjects with
a temperature of <38.5C, analyzed using ANCOVA. All
statistical analyses were performed using SAS statistical
software (Version 8.2, SAS Institute, Cary, NC).
RESULTS
Subject Demographics and Clinical Characteristics
A total of 105 participants were enrolled, randomized,
and received study medication: 54 in the IV group and
51 in the PO acetaminophen group (Figure 1). A total of
81 subjects completed the study, 45 in the IV and 36 in
the PO group. There were nine and 15 withdraw-
als < 2 hours due to vomiting in the IV and PO groups,
respectively. This left 45 IV and 36 PO acetaminophen-
treated subjects in the modied intention-to-treat popu-
lation. Demographics and baseline characteristics were
similar between the two groups and were normally dis-
tributed. The mean (SD) age was 33.0 (10.5) years,
the mean weight was 192.1 ( 43.2) lbs, and the majority
of subjects were white (77.8%) or African American
(21.0%; Table 1). Table 1 also includes the core temper-
ature results at the various time points: prior to the ref-
erence standard endotoxin test dose, prior to full
reference standard endotoxin dose, before randomiza-
tion, and prior to study drug administration. Mean core
temperature for each group at baseline prior to the test
dose was 36.6C and prior to study medication was
38.8C. Note that the more than 2C rise in core tem-
perature occurred within 2 to 3 hours after endotoxin
administration.
Efcacy
Statistically signicant differences in the WSTD through
120 minutes (p < 0.004) were observed in favor of the
IV acetaminophen group when compared to those
receiving PO acetaminophen (Figure 2). After 2 hours,
there was no difference in the WSTD between the
treatment groups. Sensitivity analyses using the inten-
tion-to-treat population (including the nine and 15 sub-
jects who vomited during the rst 2 hours) did not
change the efcacy results.
Signicant changes in temperature were observed in
favor of IV acetaminophen over PO at each time point
from T30 (15 minutes after the end of the IV infusion)
through T90, inclusive. Figure 2 presents the mean tem-
peraturetime curves for each treatment group. Up to
362 Peacock et al.
IV ACETAMINOPHEN FOR THE TREATMENT OF FEVER
Figure 1. Consort diagram. Modied intent-to-treat: All randomized subjects who received at least one full dose of PO or IV study
medication without vomiting < 2 hours after study medication dose. IV = intravenous; PO = oral.
Table 1
Summary of Subject Demographics and Baseline Characteristics
Characteristic
Acetaminophen Treatment Group
Overall (N = 81) PO (n = 36) IV (n = 45)
Race, n [%]
White 29 (80.6) 34 (75.6) 63 (77.8)
African American 7 (19.4) 10 (22.2) 17 (21.0)
Asian 0 (0.0) 1 (2.2) 1 (1.2)
Age (yr) 32.1 (9.07) 33.8 (11.60) 33.0 (10.52)
Weight (lbs) 191.9 (48.96) 192.3 (38.65) 192.1 (43.25)
Heart rate (beats min) 95.6 (13.27) 95.9 (11.67) 95.8 (12.33)
Systolic blood pressure (mm Hg) 119.5 (16.28) 121.1 (16.27) 120.4 (16.19)
Diastolic blood pressure (mm Hg) 69.3 (9.89) 70.7 (9.81) 70.1 (9.81)
Temperature prior to RSE test dose (C) 36.6 (0.31) 36.6 (0.27) 36.6 (0.29)
RSE test dose (ng) 86.5 (21.40) 87.0 (17.32) 86.8 (19.12)
Temperature prior to RSE dose (C) 36.7 (0.28) 36.7 (0.21) 36.7 (0.24)
RSE dose (ng) 342.4 (90.51) 345.7 (74.90) 344.2 (81.68)
Temperature prior to randomization (C) 38.7 (0.08) 38.7 (0.06) 38.7 (0.07)
Temperature prior to study drug administration (C) 38.8 (0.14) 38.8 (0.13) 38.8 (0.13)
Values are mean (SD) except race.
RSE = reference standard endotoxin.
ACAD EMERG MED
April 2011, Vol. 18, No. 4

www.aemj.org 363
2.5 hours, the IV acetaminophen group had lower over-
all mean temperatures, with a maximum mean tempera-
ture 0.3C lower than the PO acetaminophen group
from 40 to 75 minutes. A statistically signicant differ-
ence in the maximum temperature reduction achieved
during the period T0 to T360 was observed for the sub-
jects who received PO acetaminophen when compared
with subjects who received IV acetaminophen due to
temperature differences in the last 2 hours.
No statistically signicant differences between the
two groups were observed for percentage of subjects
with a temperature under 38.5C at any time point dur-
ing T0 to T360 minutes or for global evaluations at T360.
No subjects requested or received rescue medication.
Safety
There were no clinically relevant differences observed
between the treatment groups in the frequency of over-
all treatment-emergent AEs (Table 2). Nausea and vom-
iting were not considered treatment-emergent AEs, as
both are typical after endotoxin administration, and the
nausea began prior to T0. There was no difference in
the incidence of vomiting between groups. The most
signicant AEs in the study (fever, constitutional symp-
toms, mildly elevated liver function tests [LFTs]) were
most likely due to endotoxin administration, which is
known to induce elevations LFTs as part of its inam-
matory response. Study treatments were well tolerated,
with all AEs reported as mild or moderate in severity.
No deaths, severe AEs, hepatic AEs, serious AEs, or
discontinuations due to AEs were reported during the
study. There were no subjects in the PO acetaminophen
group who experienced an AE considered to be related
to study medication by the investigator. Two subjects in
the IV acetaminophen group experienced one AE each
(diarrhea and nausea), considered at least possibly
related to study medication. No clinically relevant dif-
ferences were observed between the treatment groups
regarding vital signs or physical examinations, nor
were clinically relevant vital sign changes (except for
temperature) from baseline observed.
While there were no reported hepatic AEs, both
treatment groups displayed a similar percentage of ala-
nine aminotransferase (ALT) or aspartate aminotrans-
ferase (AST) elevations on quantitative assessments of
pre- and poststudy clinical laboratory values (PO acet-
aminophen, 14 36, 38.9%; IV acetaminophen, 17 45,
37.8%; p = NS). Such elevations are typical after endo-
toxin given the constellation of inammatory processes
it induces, including an increase of a number of acute-
phase reactants and elevations of LFTs. It is important
to note that none of these elevations were associated
with AEs.
The LFT results (ALT, AST, alkaline phosphatase,
total bilirubin, or gamma-glutamyl transpeptidase) for
the IV and PO acetaminophen groups were similar.
While the numbers of subjects in the PO acetamino-
phen group with observed ALT or AST elevations were
less than those seen in the IV group (4 vs. 10 for ALT
and 7 vs. 14 for AST) the extent of the elevations on
quantitative analysis was not clinically signicant. After
PO acetaminophen, the maximum individual values for
ALT and AST were increased 7.0 and 4.7 times the
upper limit of normal for these enzymes, respectively.
(a)
(b)
Figure 2. Distribution of mean temperature over time. p-values
were derived from an ANCOVA model with treatment group as
the xed effect and temperature score at T0 as the covariate.
The vertical dotted line represents the time period for the
weighted sum of temperature difference primary endpoint.
Vertical solid lines represent the 95% condence intervals.
ANCOVA = analysis of covariance.
Table 2
Summary of Overall TEAEs
IV
Acetaminophen
(n = 54)
PO
Acetaminophen
(n = 51)
Subjects with 1 TEAE 17 (31.5) 18 (35.3)
Subjects with a severe
TEAE (does not include
endotoxin-induced
fever-associated AEs)
0 0
Subjects with a related
TEAE (evaluated by the
investigator as certainly,
probably or possibly
related to study
medication)
2 (3.7) 0
Subjects with a severe
related TEAE
0 0
Subject who discontinued
the study due to a TEAE
0 0
Any hepatotoxic TEAEs 0 0
Any endotoxin-induced AE (AEs that started after endotoxin
administration, but before T0)
Severe 2 (3.7) 5 (9.8)
Moderate 26 (48.1) 24 (47.1)
Mild 26 (48.1) 22 (43.1)
Values are n (%).
AE = adverse event; TEAEs = treatment-emergent adverse
events.
364 Peacock et al.
IV ACETAMINOPHEN FOR THE TREATMENT OF FEVER
This compared to a maximum individual value for ALT
or AST in the IV acetaminophen group of 3.0 and 4.2
times the upper limit of normal for ALT and AST,
respectively. In cases where follow-up values were
obtained, the values appeared to resolve quickly to nor-
mal or to a new baseline just above the upper limit of
normal.
DISCUSSION
We found that IV acetaminophen produced a more
rapid speed of onset of temperature reduction versus
PO acetaminophen in an endotoxin-induced fever
model. Statistically signicant temperature differences
were observed in the rst 2 hours, favoring IV com-
pared to PO acetaminophen. Additionally, signicant
reductions in mean temperature were observed in favor
of the IV acetaminophen group compared to those sub-
jects who received PO acetaminophen at time points
from T30 (within 15 minutes of the end of the IV infu-
sion) to T90, inclusive. After 2 hours, the groups were
not signicantly different based on WSTD analyses. It
should not be surprising that the IV route achieves a
faster therapeutic effect, given its pharmacokinetic
advantages. Over the full 6-hour treatment period, the
results for the two groups were clinically similar.
There were no clinically meaningful differences
between IV and PO acetaminophen in terms of safety
parameters. No deaths, no severe, hepatic, or serious
AEs or discontinuations due to AEs were reported dur-
ing the study. The observed increase in LFTs may be
explained by the endotoxin-induced inammatory
response. The extent of the elevations on quantitative
analysis was not clinically signicant.
Having the option of an IV formulation of acetamino-
phen for the acute treatment of fever may be particu-
larly benecial for the pediatric population, where PO
delivery may be impractical. Rectal administration of
acetaminophen, while widely used in children, may
result in slow and erratic absorption
11,12
due to several
factors that may include variability in absorption of
acetaminophen from suppositories, erratic placement of
the suppository, lipophilicity of the vehicle in the rectal
formulation, and the pH within the rectum. The absorp-
tion of acetaminophen from suppositories is gradual
and highly variable, resulting in a T
max
about 4 hours
and C
max
that is subtherapeutic unless one administers
a loading dose.
1315
However, the higher rectal loading
doses may result in a higher risk of drug accumulation
and toxicity for those subjects who absorb well through
the rectal mucosa.
Because the active site for the antipyretic effect is the
central nervous system, the acetaminophen concentra-
tiontime curve in the cerebrospinal uid (CSF) corre-
lates better with the antipyretic prole than does the
plasma concentrationtime curve. The CSF curve is
similar to that observed in plasma, but it is delayed by
the time necessary to achieve CSF penetration.
16
In a
prior pharmacokinetic study comparing IV and PO
acetaminophen,
17
in repeated 1-g doses over 48 hours,
the mean C
max
was approximately 70% higher, and
median T
max
approximately 30 minutes sooner, with IV
acetaminophen than observed with the equivalent PO
dose. Kumpulainen and colleagues
16
noted that while
PO and rectal acetaminophen are both effective, IV
acetaminophen leads to earlier peak acetaminophen
plasma levels, earlier peak CSF levels, and faster onset
of efcacy. The faster antipyretic effect also may lead to
improved early management of systemic fever-associ-
ated symptoms, such as nausea and vomiting, which in
turn may help to ensure absorption of orally adminis-
tered medications. Further investigation is needed to
determine the clinical relevance of these ndings.
Our study used endotoxin to create a reproducible
fever model. Reference standard endotoxin (or lipopoly-
saccharide) is a major constituent of bacterial cell walls
and can produce an overwhelming spectrum of biologic
responses when injected or as part of Gram-negative
sepsis. The endotoxin-induced fever model has been
used in a number of published clinical trials designed to
demonstrate antipyretic effects of pharmaceutical
agents.
9,18
Most studies have evaluated the physiologic
effects of a 4 ng kg dose, the dose used in this study.
19
Endotoxin produces a rapid-onset acute inammatory
response typically associated with u-like symptoms
that include fever, chills, arthralgias, myalgias, malaise,
headache, nausea, vomiting, and hypotension or hyper-
dynamic cardiac responses. After endotoxin admini-
stration, symptoms peak within a few hours, begin to
decline after 8 hours, and resolve completely by
24 hours.
20
We chose an endotoxin-induced fever
model as it avoids many of the complexities and con-
founding variables associated with spontaneous fever
trial designs, such as variability of the infectious model
chosen, variability of fever responses within specic
types of infections, and variability in fever patterns and
responses.
LIMITATIONS
By using an endotoxin model to create a febrile state,
the febrile response can be standardized. While this
reduces variability, it decreases the generalizability of
our results to routine clinical practice. Additionally, due
to acute tachyphylaxis with repeated dosing, endotoxin
cannot be used to study repeated doses of antipyretics.
Likewise, the use of a healthy, relatively young male
population has similar effects on generalizability. Ulti-
mately, we cannot objectively comment on the differ-
ences between the IV and PO routes of acetaminophen
administration in populations outside of the one studied
here. However, our results are consistent and plausible
with general clinical experience.
Second, while we demonstrated statistically signi-
cant differences over the rst 2 hours in favor of the IV
route, the clinical benet derived from a 0.3C improve-
ment in fever compared to the PO route must be con-
sidered. Nevertheless, in patients who are unable to
take PO medications, or who have contraindications to
PO antipyretic administration, IV acetaminophen can be
an effective option.
CONCLUSIONS
A single dose of intravenous acetaminophen is safe
and effective in reducing endotoxin-induced fever.
ACAD EMERG MED
April 2011, Vol. 18, No. 4

www.aemj.org 365
Intravenous acetaminophen may be useful where
patients are unable to tolerate oral intake or when an
earlier onset of action is desirable.
The authors appreciate the dedication of the clinical research and
hospital staff involved in this clinical study.
References
1. Duggan ST, Scott LJ. Intravenous paracetamol
(acetaminophen). Drugs. 2009; 69:10113.
2. Prescott LF. Paracetamol: past, present, and future.
Am J Ther. 2000; 7:1437.
3. Li S, Dou W, Tang Y, Goorha S, Ballou LR, Blatteis
CM. Acetaminophen: antipyretic or hypothermic in
mice? In either case, PGHS-1b (COX-3) is irrelevant.
Prostaglandins Other Lipid Mediat. 2008; 85:8999.
4. Huang WT, Wang JJ, Lin MT. Antipyretic effect of
acetaminophen by inhibition of glutamate release
after staphylococcal enterotoxin A fever in rabbits.
Neurosci Lett. 2004; 355:336.
5. Huang WT, Wang JJ, Lin MT. Cyclooxygenase
inhibitors attenuate augmented glutamate release in
organum vasculosum laminae terminalis and fever
induced by staphylococcal enterotoxin A. J Pharma-
col Sci. 2004; 94:1926.
6. McKenzie JE, Osgood DW. Validation of a new tele-
metric core temperature monitor. J Thermal Biol.
2004; 29:60511.
7. Walson PD, Jones J, Chesney R, et al. Antipyretic
efcacy and tolerability of a single intravenous dose
of the acetaminophen prodrug propacetamol in
children: a randomized, double-blind, placebo-
controlled trial. Clin Ther. 2006; 28:7629.
8. Bachert C, Chuchalin AG, Eisebitt R, et al. Aspirin
compared with acetaminophen in the treatment of
fever and other symptoms of upper respiratory
tract infections in adults: a multicenter, randomized,
double-blind, double-dummy, placebo-controlled,
parallel-group, single-dose, 6-hour dose-ranging
study. Clin Ther. 2005; 27:9931003.
9. Vargas R, Maneatis T, Bynum L, Peterson C, McMa-
hon FG. Evaluation of the antipyretic effect of ket-
orolac, acetaminophen, and placebo in endotoxin-
induced fever. J Clin Pharmacol. 1994; 34:84853.
10. Hamano N, Inada T, Iwata R, Asai T, Shingu K. The
a
2
-adrenergic receptor antagonist yohimbine
improves endotoxin-induced inhibition of gastroin-
testinal motility in mice. Br J Anaesth. 2007; 98:
48490.
11. Coulthard KP, Nielson HW, Schroder M, et al. Rela-
tive bioavailability and plasma paracetamol proles
of Panadol suppositories in children. J Paediatr
Child Health. 1998; 43:42531.
12. Gaudreault P, Guay J, Nicol O, Dupuis C. Pharma-
cokinetics and clinical efcacy of intrarectal solution
of acetaminophen. Can J Anaesth. 1998; 35:14952.
13. Beck DH, Schenk MR, Doepfmer UR, Kox WJ. The
pharmacokinetics and analgesic efcacy of larger
dose rectal acetaminophen (40 mg kg) in adults: a
double-blinded, randomized study. Anesth Analg.
2000; 90:4316.
14. Montgomery CJ, McCormack JP, Reichert CC, et al.
Plasma concentrations after high-dose rectal acet-
aminophen in children. Can J Anaesth. 1995;
42:9826.
15. Birmingham PK, Tobin MJ, Henthorn TK, et al.
Twenty-four-hour pharmacokinetics of rectal acet-
aminophen in children: an old drug with new rec-
ommendations. Anesthesiology. 1997; 87:24452.
16. Kumpulainen E, Kokki H, Halonen T, Heikkinen M,
Savolainen J, Laisalmi M. Paracetamol (acetamino-
phen) penetrates readily into the cerebrospinal uid
of children after intravenous administration. Pediat-
rics. 2007; 119:76671.
17. Schutz RA, Fong L, Chang Y, Royal MA. Open-
label, 4-period, randomized crossover study to
determine the comparative pharmacokinetics of
oral and intravenous acetaminophen administration
in healthy male volunteers, poster presentation at
the 32nd Annual Regional Anesthesia Meeting and
Workshops (American Society of Regional Anesthe-
sia and Pain), April 1922, 2007, the Westin Bay-
shore Hotel, Vancouver, British Columbia, Canada.
18. Pernerstorfer T, Schmid R, Bieglmayer C, Eichler
HG, Kapiotis S, Jilma B. Acetaminophen has
greater antipyretic efcacy than aspirin in endo-
toxemia: a randomized, double-blind, placebo-
controlled trial. Clin Pharmacol Ther. 1999; 66:517.
19. Suffredini AF, Hochstein HD, McMahon FG. Dose-
related inammatory effects of intravenous endo-
toxin in humans: evaluation of a new clinical lot of
Escherichia coli O:113 endotoxin. J Infect Dis. 1999;
179:127882.
20. Suffredini AF, OGrady NP. Pathophysiologic
responses to endotoxin in humans. In: Brade H,
Opal SM, Vogel SN, Morrison DC (eds). Endotoxin
in Health and Disease. New York, NY: Marcel
Dekker, 1999, pp 81730.
366 Peacock et al.
IV ACETAMINOPHEN FOR THE TREATMENT OF FEVER