Int. J. Oral Maxillofac. Surg.

2013; 42: 835842

http://dx.doi.org/10.1016/j.ijom.2013.02.017, available online at http://www.sciencedirect.com

Clinical Paper
Oral Surgery

Efficacy of anti-inflammatory P. Mehra1, U. Reebye2,

M. Nadershah1, D. Cottrell1
1
Department of Oral and Maxillofacial

drugs in third molar surgery: a Surgery, Boston University School of Dental

Medicine and Boston University Medical
Center, Boston, MA, USA; 2Private Practice of
Oral and Maxillofacial Surgery, Durham, NC,

randomized clinical trial USA

P. Mehra, U. Reebye, M. Nadershah, D. Cottrell: Efficacy of anti-inflammatory

drugs in third molar surgery: a randomized clinical trial. Int. J. Oral Maxillofac. Surg.
2013; 42: 835842. # 2013 International Association of Oral and Maxillofacial
Surgeons. Published by Elsevier Ltd. All rights reserved.

Abstract. This was a double-blind randomized clinical trial to assess the effect of
different pharmacological regimens on the level of prostaglandin E2 (PGE2) in
urine and saliva, and to correlate the findings to the clinical course after removal of
impacted lower third molars. Eighty patients were randomly divided into four
groups: group 1 received placebo; group 2 received preoperative ibuprofen, which
was continued for a week; group 3 received intraoperative dexamethasone; and
group 4 received preoperative ibuprofen, which was continued for a week, in
addition to intraoperative dexamethasone. Saliva and urine samples were taken at
scheduled intervals. Patients receiving ibuprofen fared significantly better in most
parameters. A single dose of dexamethasone alone had a potent but transient
beneficial effect when compared to the results with ibuprofen, which showed
significant improvement in both subjective and objective parameters. Use of a
single dose of intravenous steroids perioperatively helped reduce untoward
Key words: third molar; prostaglandins; PGE2;
sequelae, although to a lesser degree and for a shorter duration than continuous oral surgery; inflammation; pain.
ibuprofen. Combining ibuprofen with perioperative dexamethasone added some
benefit in some of the measured parameters, but without a statistically significant Accepted for publication 21 February 2013
advantage over using ibuprofen only. Available online 25 March 2013

Third molar surgery remains the mainstay
of most oral and maxillofacial surgery
practices. Pain, swelling, and transient
loss of normal jaw function are usually
associated with the removal of impacted
third molar teeth. Management of these
postoperative symptoms is frequently
based on pharmacological manipulation
of local and systemic mediators of pain
and inflammation.
Unfortunately, the parameters for
manipulation of these pathways with spe-
cificity in third molar surgery have not yet
been defined. Currently, the use of anti-
inflammatory drugs varies among practi-
tioners based on a variety of factors,
including personal experience and prefer-
ence, information derived from pharma-
ceutical salespeople, or anecdotal
information. If a specific marker of pain
and inflammation, such as prostaglandins,
could be quantitatively manipulated by
preoperative and postoperative drug ther-
apy, an improved clinical result with mini-
mization of the usual postoperative
symptomatology may result.
Prostaglandin E2 (PGE2) is a measur-
able marker of post-traumatic pain,1
fever,2,3 and inflammation,47 and has
been isolated in many body systems,
including saliva in the oral cavity.8
Decreasing local and systemic prostaglan-
din levels with the use of drugs is a well
known method for reducing clinical
inflammation.911 Some previous studies
have evaluated the effects of a non-ster-
oidal anti-inflammatory drug (NSAID) on
tissue prostaglandin levels in gingival cre-
vicular fluid.12,13 Roszkowski et al.

0901-5027/070835 + 08 $36.00/0 # 2013 International Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved.
836 Mehra et al.
reported that the administration of the
NSAID effectively decreased prostaglan-
din release at extraction socket sites.14
PGE2 is known to cause hyperalgesia in
psychophysical studies and also to evoke
sensitization in electrophysiological stu-
dies.1518 Thus, a comparative difference
in preoperative and postoperative PGE2
levels serves as a useful tool in gauging
quantitatively the efficacy of the admin-
istration of pharmacological agents, and a
corresponding reduction of pain and
inflammation.
The objectives of this prospective study
were to: (1) evaluate PGE2 concentrations
in urine and saliva after mandibular third
molar surgery; (2) correlate PGE2 concen-
trations with subjective and objective clin-
ical symptomatology after mandibular
third molar surgery; and (3) compare the
efficacy of three commonly used pharma-
cological regimens to a control group in
manipulating PGE2 levels in patients
undergoing impacted mandibular third
molar removal. To the best of our knowl-
edge, there is no published literature that
relates salivary and urinary PGE2 levels to
detailed objective and subjective evalua-
tion of clinical symptoms over a 7-day
postoperative period after mandibular
third molar surgery. Our hypothesis was
that there would be no statistically signif-
icant differences between the placebo
group and the other three groups relative
to the parameters examined.
Patients and methods
This prospective study included 80 ASA I
patients (American Society of Anesthe-
siology classification, normal healthy)
with bilateral full-bony mandibular third
molar impactions. Institutional review
board approval was obtained prior to com-
mencement of the study. The patient popu-
lation was randomly and consecutively
selected from an outpatient oral and max-
illofacial surgery clinic. Care was taken to
maintain that the trial was in compliance
with the current CONSORT (Consolidated
Standards Of Reporting Trials) statement
guidelines. Other criteria for inclusion into
the study included: (1) age between 18 and
30 years; (2) no systemic disease (ASA I
status); (3) taking no medications; (4) no
allergies to any of the study drugs; and (5)
absence of local or systemic infection. All
patients underwent an initial preoperative
screening consult with a single oral and
maxillofacial surgery resident and a
faculty member. All patients enrolled in
the study completed the study without any
postoperative complications.
Table 1. Preoperative (oral), intraoperative (intravenous), and postoperative (oral) medication
regimens.
Preoperative p.o.
Group drug Intraoperative i.v. drug Postoperative p.o. drug
Group 1 Placebo tablet 2 ml of 0.9% saline Placebo qid  7 days
Group 2 Ibuprofen 600 mg 2 ml of 0.9% saline Ibuprofen 600 mg qid  7 days
Group 3 Placebo tablet 8 mg of dexamethasone Placebo qid  7 days
Group 4 Ibuprofen 600 mg 8 mg of dexamethasone Ibuprofen 600 mg qid  7 days
i.v., intravenous; p.o., by mouth; qid, four times a day.
All 80 patients underwent surgical patients in sealed envelopes with only a
removal of only bilateral full-bony study assigned number identifying the
impacted mandibular third molars under specific patient. They were prepacked
intravenous ambulatory general anaesthe- and consecutively numbered for each par-
sia. All extractions were performed by one ticipant according to the randomization
surgeon and required full-thickness muco- schedule by the pharmacist. The intrave-
periosteal flaps and bone removal (per- nous medications (dexamethasone and
formed under irrigation) using an air placebo) were drawn up aseptically just
driven rotary instrument. A uniform local prior to the time of surgery as 2-ml
anaesthetic technique was used which volumes in identical, unmarked 3-ml ster-
included bilateral inferior alveolar, lin- ile plastic syringes by the pharmacist and
gual, and long buccal nerve blocks using given to the anaesthesiologist for use. The
2% lidocaine with 1:100,000 epinephrine. study was double-blinded, and the sur-
Intravenous anaesthesia was provided by geon, anaesthesiologist, and patients were
the same board-certified oral and maxillo- not aware of the study group to which each
facial surgery faculty member for all cases individual patient belonged.
and included a combination of midazolam, Preoperatively (0 h), and at each post-
fentanyl, and methohexital. operative follow-up visit (24, 48, 72, and
Patients were randomly divided into 168 h), all patients were evaluated by a
four groups using a random number gen- single clinician who was not involved in
erated by a computer (20 patients per any of the surgical procedures. Clinical
group). The allocation sequence was con- examination included: (1) subjective eva-
cealed from the surgeon and anaesthesiol- luation using visual analogue scale (VAS)
ogist. An independent pharmacist scores for pain, loss of jaw function, swel-
dispensed either active or placebo medi- ling, diet restriction, and general wellness;
cations according to a computer generated and (2) objective evaluation of maximum
randomization list. The four groups were inter-incisal opening (MIO) and lateral
as follows: group 1 received immediate excursions (LE), presence or absence of
preoperative placebo tablet, intraoperative any temporomandibular joint (TMJ)
normal saline (2 ml intravenous (i.v.)), symptoms, and muscle tenderness. A note
and a postoperative placebo tablet every was also made of the number of codeine
6 h for a week; group 2 received immedi- tablets consumed since last visit, and of
ate preoperative ibuprofen 600 mg, which the presence of dry socket or infection.
was continued every 6 h postoperatively Saliva and urine samples were taken
for a week, in addition to intraoperative from each patient preoperatively and post-
normal saline (2 ml i.v.); group 3 received operatively at scheduled intervals (24 h,
immediate preoperative placebo tablet, 48 h, 72 h, and 168 h). Prior to sample
intraoperative dexamethasone (8 mg collection, all patients were asked to wash
i.v.), and a placebo postoperative tablet their mouths thoroughly with tap water.
every 6 h for a week; group 4 received Saliva was collected through an Eppen-
immediate preoperative ibuprofen dorf blue tip into a sterile polystyrene tube.
600 mg, which was continued every 6 h No stimulation to enhance secretion was
postoperatively for a week, in addition to used. A clean catch urine specimen was
intraoperative dexamethasone (8 mg i.v.). also collected into a sterile plastic cup.
All groups received 30 mg codeine tablets Both saliva and urine samples were imme-
every 4 h as needed for postoperative pain diately stored at 80 8C.
management (Table 1). A kit (plate) specific for PGE2 (Assay
Thus, all patients received one preo- Designs Inc., Ann Arbor, MI, USA)
perative (oral), one intraoperative (intra- was used to calculate the PGE2 level in
venous), and postoperative (oral) study each specimen.19 Each specimen was
medications besides codeine tablets. The tested three times to reduce laboratory
oral medications (placebo and ibuprofen) error, and the average value was used
were identical in appearance and given to for statistical analysis. Enzyme-linked

immunosorbent assay (ELISA) testing
using the double-antibody sandwich
technique was used for calculating the
optical density (OD) of each well in the
sample plates. A prostaglandin synthe-
tase inhibitor (indomethacin at a concen-
tration of 10 mg/ml) was initially added
to both urine and saliva samples in each
plate. Next, the assay procedure was
completed as per the manufacturers
instructions. The OD of the PGE2 plates
was read at 405 nm by a spectrometer.
The binding of each pair of standard
wells in a control plate as a percentage
of the maximum binding wells was then
calculated (% bound wells in the plates).
Using the logitlog paper plot, the per-
cent bound wells versus concentration of
PGE2 for the standard controls were
plotted on a graph. The concentrations
of PGE2 in the sample wells were then
calculated.
Clinical and laboratory data were
entered into an Excel spread sheet and
run with SPSS software. A multivariate
analysis of variance test was used to deter-
mine the statistical significance of the
results between groups. A P-value of less
than 0.05 was considered to be statistically
significant.
Results
Patient demographic data are presented in
Table 2. There were no significant differ-
ences in the results between the two sexes
for any of the clinical parameters assessed
in this study.
Laboratory data
Levels of PGE2 in saliva and urine (Figs 1
and 2) showed a similar trend over 7 days
and were measured individually at each
point in time. An analysis of the control
group showed two interesting findings: (1)
after third molar surgery, salivary and
urinary PGE2 levels increased from post-
operative day 1 to 3, and (2) levels of
PGE2 at local sites (saliva) were higher as
compared to systemic levels (urine) except
Table 2. Demographics for all patient groups.
Group 1
Average age (years) 22.1
Age range (years) 1728
Number of males 11
Number of females 9 10
Distoangular impactions 9 9
Mesioangular impactions 17
Horizontal impactions 14
Procedure duration average (min) 47
Procedure duration range (min) 4255
PGE2 manipulation in third molar surgery
at 3 days postsurgery when the urinary
levels had a significant spike.
PGE2 concentrations were significantly
higher in the placebo (control) group than
in the other three groups on all postopera-
tive days. The group of patients receiving
dexamethasone and ibuprofen had the
lowest PGE2 concentration in both body
fluids within 24 h of surgery. However, the
PGE2 concentration in this group then
increased during the 24168 h postopera-
tive period to levels greater than the ibu-
profen group, but still lower than the
placebo group. Patients in the ibuprofen
group had a sustained depression of PGE2
levels in both urine and saliva throughout
the 7-day follow-up period.
Subjective clinical data
Pain (Fig. 3)
At postoperative day 1, patients in group 4
(dexamethasone and ibuprofen) had the
lowest pain scores of all four groups. This
was statistically significant when com-
pared to the placebo group, but not to
the other two groups. At postoperative
days 2 and 3, all three groups had reduced
pain scores at levels that were significantly
lower than the pain scores of the placebo
group (P = 0.038). However, ibuprofen
alone was more effective in reducing pain
during this period when compared to dex-
amethasone alone, with statistical signifi-
cance. The effect of the combination of
ibuprofen and dexamethasone was not
significantly different from the use of
ibuprofen alone. At postoperative day 7,
patients taking placebo still had signifi-
cantly worse scores relative to the other
three groups.
Swelling (Fig. 4)
Patients taking placebo had the maximum
swelling on all days of examination. The
difference was statistically significant
when compared to the other three groups
(P = 0.042). On postoperative day 7,
patients on placebo still felt that they
Group 2 Group 3 Group 4
23.3 21.3 20.9
1829 1727 1828
10 12 9 and right and left LE values at each of the
8 11
9 11
15 14 19
16 17 10
47 43 45
4064 3954 3858
837
suffered from residual swelling compared
to the other three groups, in which the
patients felt normal. Patients taking dex-
amethasone with ibuprofen (group 4) had
the least swelling (as felt by the patient)
throughout the postoperative course.
When we compared the results of group
2 (ibuprofen) and group 3 (dexametha-
sone), the average swelling for postopera-
tive day 1 was lower in group 3, but in the
following days it was lower in patients in
group 2. The difference between the ibu-
profen alone, dexamethasone alone, and
ibuprofen with dexamethasone was, how-
ever, not statistically significant.
Jaw function (Fig. 5)
With regard to jaw function, the patients
taking ibuprofen only and those who had
received both dexamethasone and ibupro-
fen fared significantly better than the pla-
cebo and dexamethasone only groups on
postoperative days 2 and 3 (P = 0.037). At
7 days postoperatively, results for all four
groups were similar.
Diet (Fig. 6)
Patients on placebo were only able to take
liquids for approximately 3 days post-
operatively. This was in striking contrast
to the other groups where patients could
take semi-solid foods from postoperative
day 1.52 onwards. This difference
between the placebo group and the other
groups for the first 3 postoperative days
was statistically significant (P = 0.029).
On postoperative day 7, patients in the
dexamethasone, ibuprofen, and dexa-
methasone with ibuprofen groups were
almost on a normal diet, while patients
in the placebo group still had some dietary
restrictions secondary to discomfort at the
surgical sites.
General health
No significant difference was observed
between the four groups with regard to
general health.
Objective clinical data
Jaw function (Fig. 7)
This was evaluated by measuring the MIO
postoperative visits. The patients in the
steroid and NSAID groups had trismus
approximately 2 mm lesser than placebo
patients on the first postoperative day. By
the third postoperative day, the average
inter-incisal opening measured was as
838 Mehra et al.

Salivary PGE2 levels

900

800

PGE2 level (pg/ml) 700

600

500
Placebo
Ibuprofen
400 Dexamethasone
Dexa+ Ibuprofen
300

200

100

0
0 24 48 72 96 120 144 168 192
Time (Hours)

Fig. 1. PGE2 concentrations in saliva over 7 days.

follows: 64% of preoperative opening for
the placebo group, 82% of preoperative
opening for the dexamethasone group,
85% of preoperative opening for the ibu-
profen alone group, and 82% of preopera-
tive opening for the dexamethasone and
ibuprofen group. All but the placebo group
had no limitation of mouth opening on
postoperative day 7; placebo patients had
some residual trismus (due to guarding) at
this stage.
Narcotic use (Fig. 8)
The average number of 30 mg codeine
tablets used by the patients over the 7-
day postoperative period was 21 in the
placebo group, 15 in the dexamethasone
group, 9.25 in the ibuprofen group, and 10
in the dexamethasone and ibuprofen
group. Comparison between the placebo
group and the other groups was statisti-
cally significant (P = 0.022).
Muscle tenderness, dry socket, TMJ
symptoms, and infection
There were no significant differences in
these parameters between the four groups.
Discussion
All patients enrolled in this study had
bilateral mandibular third molars
extracted under intravenous general
anaesthesia in a manner that is considered
standard for many oral surgery practices.
In an attempt to standardize the study
sample, all screening appointments and
follow-up visits were supervised by the
same board-certified oral surgeon and the
surgery performed by one resident. All the
extracted teeth required cutting of man-
dibular bone via a rotary instrument under
saline irrigation. Codeine was used as the
common postoperative pain medication in
all groups as it does not have any known
anti-inflammatory effects, and thus did not
interfere with the laboratory evaluation.
Levels of PGE2 in serum are extremely
difficult to measure accurately over a

Urine PGE2 levels

1000

900

800
PGE2 level (pg/ml)

700

600
Placebo
500 Ibuprofen
Dexamethasone
400 Dexa+ Ibuprofen

300

200

100

0
0 24 48 72 96 120 144 168 192
Time (Hours)

Fig. 2. PGE2 concentrations in urine over 7 days.

 PGE2 manipulation in third molar surgery 839

Subjective Patient Pain Data

70

60

50
Visiual Analog Scale (points)

40

30

20

10

0
0 24 48 72 96 120 144 168 192
Time (Hours)

Placebo Ibuprofen Dexamethasone Dexa + Ibuprofen

Fig. 3. Visual analogue scores for pain as perceived by the patients.

7-day postoperative period. As saliva and significant and favourable differences in administered). However, from postopera-
urine samples are well accepted and many parameters when compared with tive day 2 onwards, the PGE2 levels
recommended in the scientific community group 1 (placebo). This was valid even increased in group 3 (dexamethasone only)
for the measurement of PGE2 levels, they at postoperative day 7, which is the usual in contrast to group 2 (ibuprofen), where
were used in this study.13,14 follow-up visit time for many oral sur- there was a consistent decrease. This is
The study findings clearly indicate that geons. In summary, group 1 patients likely due to the anti-inflammatory effect
it may be beneficial to prescribe anti- reported both an increased frequency of the intravenous steroid administration
inflammatory drugs in patients undergoing and duration of adverse postoperative perioperatively, which was effective only
third molar surgery. Looking at the sub- sequelae at all postoperative visits. for a few hours.
jective data (graphs for pain, swelling, A close comparison of groups 2 and 3 Correlating the laboratory data with the
diet, and loss of jaw function), it is clear revealed some interesting findings. Gener- clinical data also showed some interesting
that patients in groups 2 (ibuprofen), 3 ally speaking, at postoperative day 1, the similarities. The clinical picture seemed to
(dexamethasone), and 4 (dexamethasone level of measured postoperative PGE2 was mimic the laboratory picture, thereby
and ibuprofen) reported statistically lowest in groups 3 and 4 (dexamethasone showing a definite link between systemic

Subjective Swelling
70

60
Visual Analog Scale (points)

50

40

30

20

10

0
0 24 48 72 96 120 144 168 192
Time (Hours)

Placebo Ibuprofen Dexamethasone Dexa + Ibuprofen

Fig. 4. Visual analogue scores for swelling as perceived by the patients.

840 Mehra et al.

Percentile decrease in Subjective Jaw Function

60

Visual Analog Scale (points)

50

40

30

20

10

0
0 24 48 72 96 120 144 168 192
Time (Hours)

Placebo Ibuprofen Dexamethasone Dexa + Ibuprofen

Fig. 5. Objective evaluation of jaw function (maximum inter-incisal opening) as measured by a clinician.

PGE2 levels (urine and saliva) and clinical
inflammation. The subjective and objec-
tive evaluation of patients showed that
patients in groups 24 did consistently
better in all measured parameters as com-
pared to group 1 patients. Patients receiv-
ing continuous NSAIDs with or without
dexamethasone seemed to be much more
comfortable subjectively, and did better in
most measured objective data in compar-
ison to the other two groups.
There is a definite link between salivary
and urinary prostaglandin concentrations
and adverse sequelae following minor
oral surgery. PGE2 is a measurable
marker of pain and inflammation, and
suppression of this fatty acid can be
effected via the administration of either
oral or parenteral medications. The results
of this study disproved our null hypoth-
esis. It clearly demonstrated that contin-
uous round-the-clock administration of a
NSAID had a better long-term effect in
depressing PGE2 (inflammation) than an
isolated intravenous dose of steroids or no
medications. Combining both drugs
resulted in a slight improvement in all
measures, but without statistical signifi-
cance when compared to patients receiving
ibuprofen only.
It is hoped that the results of this study
will help better define the role of com-
monly used anti-inflammatory drugs in
reducing PGE2 (a known and measurable
marker of pain and inflammation).
Although codeine does not directly
impact PGE2 levels, it may have affected
some of the outcome measures evaluated
in this study such as trismus, jaw function,
and diet. Future studies should be done to
test other combination drug regimens
(e.g. steroid dose packs, longer acting
steroids, other NSAIDs, and steroid and
NSAID combination therapy), and eval-
uate the role of other known inflammatory

Subjective Patient Dietary restrictions

100

90
Visual Analog Scale (points)

80

70

60

50

40

30

20

10

0
0 24 48 72 96 120 144 168 192
Time (Hours)
Placebo Ibuprofen Dexamethasone Dexa +Ibuprofen

Fig. 6. The ability of patients to take oral intake over 7 days.

 PGE2 manipulation in third molar surgery 841

Objective Patient Trismus

50
48
47.0
46

Max. Incisal Opening (mm)

44
42 42.0

40
38
Placebo
36 Ibuprofen
34 Dexamethasone
Dexa + Ibuprofen
32
30
30.1
29.5
28
26
26.5
24
22
0 24 48 72 96 120 144 168 192
Time (Hours)

Fig. 7. Objective trismus (maximal inter-incisal opening) as recorded by a clinician.

mediators like interleukin 1, leucotrienes, subjective and objective evaluation subjective patient comfort/discomfort
tumour necrosis factor, etc., in oral sur- postoperatively. over that prolonged period of time. There
gery. (4) PGE2 levels in the body can be seem to be distinct advantages of routinely
In conclusion, the results of this study manipulated with oral and intravenous using drugs to help reduce inflammation
show the following interesting findings: administration of anti-inflammatory pharmacologically, and thus promote
drugs in an attempt to improve patient patient comfort after office-based third
(1) Levels of prostaglandins like PGE2 comfort and satisfaction after third molar surgery. Prescribing preoperative
are measurable in body fluids like molar surgery. NSAIDs before and after third molar sur-
saliva and urine, both before and after gery seems to be beneficial. The use of a
third molar surgery. A correlation between systemic prosta- single dose of intravenous steroids perio-
(2) Rises and falls in prostaglandin glandin levels and clinical symptomatol- peratively may also help to suppress
(PGE2) over time after minor oral ogy following third molar surgery over a inflammation and reduce untoward seque-
surgery can be detected in both urine 7-day postoperative period has been lae, although to a lesser degree and for a
and saliva. demonstrated. Most oral and maxillofacial shorter duration than continuous round-
(3) Concentrations of PGE2 in urine and surgeons perform patient follow-up at 7 the-clock oral NSAIDs. Combining ibu-
saliva seem to correlate well with the days postsurgery and may not be aware of profen with perioperative intravenous
dexamethasone adds some benefit in some
of the measured parameters, but not with
statistical significance.
It is hoped that the results of this study
will help demonstrate an improved under-
standing of postoperative pain manage-
ment in patients undergoing third molar
removal, and possibly lead to more pre-
dictable postoperative pharmaceutical
therapy in the future.

Funding
Partially supported by a grant from the Oral
and Maxillofacial Surgery Foundation.

Competing interests
None.

Ethical approval
Approval obtained from the Boston Uni-
Fig. 8. The number of narcotic (30 mg codeine) tablets consumed by the patients over 7 days. versity Institutional Review Board.
842 Mehra et al.
Acknowledgement. The authors wish to
thank Dr Alexie Mizin for his help in
the completion of this project.
References
1. Minami T, Nakano H, Kobayashi T, Sugi-
moto Y, Ushikubi F, Ichikawa A, et al. Char-
acterization of EP receptor subtypes
responsible for prostaglandin E2-induced
pain responses by use of EP1 and EP3
receptor knockout mice. Br J Pharmacol
2001;133:43844.
2. Ushikubi F, Segi E, Sugimoto Y, Murata T,
Matsuoka T, Kobayashi T, et al. Impaired
febrile response in mice lacking the prosta-
glandin E receptor subtype EP3. Nature
1998;395:2814.
3. Yamagata K, Matsumura K, Inoue W, Shir-
aki T, Suzuki K, Yasuda S, et al. Coexpres-
sion of microsomal-type prostaglandin E
synthase with cyclooxygenase-2 in brain
endothelial cells of rats during endotoxin-
induced fever. J Neurosci 2001;21:266977.
4. Murakami M, Naraba H, Tanioka T, Semmyo
N, Nakatani Y, Kojima F, et al. Regulation of
prostaglandin E2 biosynthesis by inducible
membrane associated prostaglandin E2
synthase that acts in concert with cyclooxy-
genase-2. J Biol Chem 2000;275:3278392.
5. Mancini JA, Blood K, Guay J, Gordon R,
Claveau D, Chan CC, et al. Cloning, expres-
sion, and up-regulation of inducible rat
prostaglandin E synthase during lipopoly-
saccharide-induced pyresis and adjuvant-
induced arthritis. J Biol Chem 2001;276:
446975.
6. Stichtenoth DO, Thoren S, Bian H, Peters-
Golden M, Jakobsson PJ, Crofford LJ.
Microsomal prostaglandin E synthase is
regulated by proinflammatory cytokines
and glucocorticoids in primary rheumatoid
synovial cells. J Immunol 2001;167:46974.
7. Uematsu S, Matsumoto M, Takeda K, Akira
S. Lipopolysaccharide-dependent prosta-
glandin E(2) production is regulated by the
glutathione-dependent prostaglandin E(2)
synthase gene induced by the Toll-like recep-
tor 4/MyD88/NFIL6 pathway. J Immunol
2002;168:58116.
8. Ohidhi K, Ueno R, Nishino S, Sakai T,
Hayaishi O. Increased levels of salivary
prostaglandins in patients with major depres-
sion. Biol Psychiatry 1988;23:32634.
9. Marcucci M, Panelli G, Cambini S. Clinical
experience in the treatment of dental pain.
Clin J Pain 1991;7(Suppl.):726.
10. Turker MN, Turker RK. A study on the
peripheral mediators of dental pain. Experi-
entia 1974;15:9323.
11. Urquhart E. Analgesic agents and strategies
in the dental pain model. J Dent 1994;22:
33641.
12. OBrien TP, Roszkowski MT, Wolff LF,
Hinrichs JE, Hargreaves KM. Effect of a
non-steroidal anti-inflammatory drug on tis-
sue levels of immunoreactive prostaglandin
E2, immunoreactive leukotriene, and pain
after periodontal surgery. J Periodontol
1996;67:130716.
13. Alexander DC, Martin JC, King PJ, Powell
JR, Caves J, Cohen ME. Interleukin-1 beta,
prostaglandin E, and immunoglobulin G sub-
classes in gingival crevicular fluid in patients
undergoing periodontal therapy. J Period-
ontol 1996;67:75562.
14. Roszkowski MT, Swift JQ, Hargreaves KM.
Effect of NSAID administration on tissue
levels of immunoreactive prostaglandin
E2, leukotriene B4, and (S)-flurbiprofen
following extraction of impacted third
molars. Pain 1997;73:33945.
15. Sciberras DG, Goldenberg MM, Bolognese
JA, James I, Baber NS. Inflammatory
responses to intradermal injection of platelet
activating factor, histamine, and prostaglan-
din E2 in healthy volunteers: a double blind
investigation. Br J Clin Pharmacol 1987;24:
75361.
16. Wallengren J, Hakanson R. Effects of cap-
saicin, bradykinin and prostaglandin E2 in
the human skin. Br J Dermatol 1992;126:
1117.
17. Martin HA, Basbaum AI, Kwiat GC, Goetzl
EJ, Levine JD. Leukotriene and prostaglan-
din sensitization of cutaneous high-threshold
C- and A-delta mechanoreceptors in the
hairy skin of rat hind limbs. Neuroscience
1987;22:6519.
18. Pitchford S, Levine JD. Prostaglandins sen-
sitize nociceptors in cell culture. Neurosci
Lett 1991;132:1058.
19. Assay Designs Inc.. Laboratory instruction
manual for PGE2 kits. Ann Arbor, MI:
Assay Designs Inc.; 1996.
Address:
Pushkar Mehra
Department of Oral and Maxillofacial
Surgery
Boston University School of Dental Medicine
Boston University Medical Center
100 East Newton Street
G-407
Boston
MA 02118
USA
Tel: +1 617 638 4350; Fax: +1 617 638 4365
E-mails: pmehra@bu.edu,
pushkar.mehra@bmc.org