You are on page 1of 6

Pharmacological Research 52 (2005) 485–490

Clinical evaluation of piroxicam-FDDF and azithromycin
in the prevention of complications associated with
impacted lower third molar extraction
F. Graziani
a
, L. Corsi
a
, M. Fornai
b
, L. Antonioli
b
, M. Tonelli
a
, S. Cei
a
,
R. Colucci
b
, C. Blandizzi
b
, M. Gabriele
a
, M. Del Tacca
b,∗
a
Section of Oral Surgery, Department of Neurosciences, University of Pisa, Pisa, Italy
b
Interdepartmental Centre for Research in Clinical Pharmacology and Experimental Therapeutics, University of Pisa, Via Roma 55, 56126 Pisa, Italy
Accepted 29 July 2005
Abstract
Combined treatments with non-steroidal anti-inflammatory drugs and antibiotics may offer significant benefits in the prevention of pain and
infections associated with oral surgery. In this study, piroxicam and azithromycin were administered to patients undergoing dental extraction
to examine the efficacy of piroxicam in the prevention of post-operative pain and inflammatory complications, either in the absence or in the
presence of a concomitant antibiotic treatment.
Thirty patients were randomly assigned to three groups and treated for 3 days, before impacted lower third molar removal, as follows:
(1) sublingual piroxicam-FDDF (fast dissolving dosage formulation) 20 mg/day; (2) oral azithromycin 500 mg/day; (3) piroxicam-FDDF
20 mg/day plus azithromycin 500 mg/day. Oral acetaminophen (500 mg tablets) was allowed as rescue analgesic medication. Pain intensity
was evaluated on a 100-mm visual-analogue scale after dental extraction (day 1), and at days 2, 3, 7 after surgery. Edema and trismus were
estimated at days 2 and 7.
At days 1 and 2, pain intensity was significantly lower in patients treated with piroxicam-FDDF, either alone (p <0.05) or in combination
with azithromycin (p <0.05), than in patients administered with azithromycin alone. A higher acetaminophen consumption was also recorded
in the latter group (p <0.01). Pain intensity values did not differ among treatment groups at days 3 and 7. At day 2, the facial edema was
significantly less intense in patients exposed to piroxicam-FDDF alone, as compared to patients treated with azithromycin, either alone
(p <0.05) or in combination with piroxicam-FDDF (p <0.05). No significant differences were detected when comparing groups for trismus
at days 2 and 7.
The present results indicate that, when given alone in the pre-operative period, piroxicam-FDDF effectively counteracts post-surgical pain
and inflammatory reactions in oral tissues. Upon combined treatment with piroxicam-FDDF and azithromycin, the macrolide antibiotic may
reduce the influence of piroxicam on post-operative inflammation, without affecting its beneficial effect on surgical pain.
© 2005 Elsevier Ltd. All rights reserved.
Keywords: Piroxicam; Azithromycin; Post-operative pain; Post-operative inflammation
1. Introduction
Treatments with non-steroidal anti-inflammatory drugs
(NSAIDs) and antibacterial agents may play important roles
in the prophylactic therapy preceding or following oral
surgery [1,2]. NSAIDs, when administered pre-operatively,

Corresponding author. Tel.: +39 050 830148; fax: +39 050 562020.
E-mail address: m.deltacca@med.unipi.it (M. Del Tacca).
can be absorbed and distributed to oral tissues before the
initiation of surgical trauma, thus ensuring a blockade of
arachidonic acid pathway, with a subsequent reduction in the
occurrence of post-operative swelling, discomfort and pain
[2]. Antibiotics may counteract the onset of infectious foci at
the surgical site as well as their systemic extension [1].
Piroxicamis a NSAIDendowed with oxicamstructure and
prolonged plasma half-life following oral administration [3].
Piroxicam can significantly inhibit the development of gin-
1043-6618/$ – see front matter © 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.phrs.2005.07.009
486 F. Graziani et al. / Pharmacological Research 52 (2005) 485–490
gival inflammation in dogs [4]. It appears to be an effective
analgesic also in post-surgical oral pain. However, due to its
slow onset of action, a pre-operative administration has been
proposed [5]. Subsequently, a fast dissolving dosage formula-
tion of piroxicam (piroxicam-FDDF), suitable for sublingual
administration, has been developed to attain a faster absorp-
tion and achieve a more prompt effectiveness in the treatment
of post-operative oral pain [6].
In the last decade, the use of macrolide antibiotics in
dentistry has been encouraged, due to therapeutic advan-
tages of novel derivatives, such as a broader antibacterial
spectrum, improved tissue distribution, and low incidence of
adverse effects [7]. Among these, azithromycin represents
an interesting option [8], since an extended antimicrobial
spectrum[9], a favourable disposition into normal and patho-
logical periodontal tissues [10,11], and a good effectiveness
in the treatment of odontogenic infections [12,13] have been
demonstrated for this macrolide antibiotic.
Despite the above arguments, there is currently a scarcity
of data concerning the combined use of NSAIDs and antibi-
otics in the perioperative management of patients subjected
to oral surgery. Therefore, the present study was designed
to evaluate the clinical efficacy of piroxicam-FDDF, either
alone or in combination with azithromycin, in the prophy-
lactic treatment of pain and edema associated with impacted
third molar extraction.
2. Material and methods
2.1. Patients
Patients enrolled in this investigation were scheduled for
lower wisdom tooth surgical removal at the Oral Surgery
Unit of the University of Pisa. They had no history of drug
allergy, and none of them had taken drugs within 2 weeks
before the pharmacological treatment was started. Patients
were excluded if they had infectious diseases, history of
asthma, peptic ulceration, or other disorders of the upper gas-
trointestinal tract. Patients who had any concomitant major
medical problemand/or ongoing pharmacological treatments
were also excluded, as were women who were pregnant or
lactating, or not using contraceptives. Physical examination,
electrocardiogram, blood chemistry, hematology and urine
analysis were performed in all patients before enrollment and
after completion of the study. Informed consent was obtained
from each patient before entering the trial, and the investiga-
tion was approved by the local University Hospital Ethics
Committee.
2.2. Study design
Thirty patients (15 males and 15 females; age range 18–39
years; mean age 25.3 ±1.03 years), who met the inclusion
criteria, were admitted to the study. Patients were randomized
with a computer-generated randomization list and assigned
to three groups to receive one of the following pharma-
cological treatments: (A) piroxicam-FDDF 20 mg (Feldene
Fast
®
, Pfizer Italiana, Rome, Italy) administered once daily
by sublingual route; (B) azithromycin 500 mg (Zitromax
®
,
Pfizer Italiana, Rome, Italy) administered once daily by
oral route; (C) piroxicam-FDDF 20 mg plus azithromycin
500 mg, administered in accordance with the same sched-
ule and formulations reported for patients of Group A and
Group B, respectively. In group C, patients were requested to
allow1 h interval between the administration of azithromycin
and piroxicam. Patients of each group were treated with their
respective test drugs for three consecutive days and were then
subjected to dental extraction on the fourth day. The ratio-
nale of this prophylactic treatment schedule was drawn on
previous pharmacokinetic studies indicating a favourable dis-
position of both drugs into oral tissues following their short
term administration in the pre-operative period [10,11,18].
Surgery was performed in a standardized fashion by the same
surgeon and a standardized local anesthetic technique was
used.
2.3. Evaluation of pain, edema, trismus and infectious
complications
Pain assessment was carried out using a 100-mm visual
analogue scale (VAS) at the end of anesthesia (day 1) and
subsequently at days 2, 3 and 7 after surgery. At day 1, 2
and 7, VAS was evaluated by face-to-face interview. At day
3, patients reported their VAS score on a log diary. Anchor
points were: 0 =no pain; 100 =worst pain possible. More-
over, patients were provided with acetaminophen 500 mg
tablets and were instructed to take the drug only when needed
and to record the number of consumed tablets on a log diary.
Edema was evaluated at days 2 and 7 after surgery by a
linear method [14]. Briefly, edema was estimated as the sum
of the distance fromthe corner of the mouth to the attachment
of the earlobe, following the bulge of the cheek, plus the
distance from the outer cantus of the eye to the angle of the
mandible. Trismus was recorded as the difference of inter-
incisal distance at maximum of mouth opening before and
after the operation. Both edema and trismus were expressed
as percentage changes from pre-operative measurements.
The presence of infection and alveolitis was assessed at
days 2 and 7 after surgery. Infection was recognized by the
presence of pus discharge and abscess in the alveolar area.
Alveolitis was indicated by the onset of neuralgic pain and the
formation of a necrotic socket without clot. Pain, edema and
trismus as well as the presence of infectious complications
were evaluated by an examiner who was blind to treatments.
The occurrence of infectious complications was quantitated
but not subjected to comparative statistical analysis.
2.4. Statistics
Results are given as mean ±S.E. of the mean (S.E.M.).
Edema andtrismus are expressedas percentages, but statistics
F. Graziani et al. / Pharmacological Research 52 (2005) 485–490 487
was performed on row data. The significance of differences
was evaluated by one way ANOVA test. In order to identify
differences between treatment groups, a post-hoc analysis
was carried out using the Bonferroni test. Statistical differ-
ences between groups were accepted for p-values lower than
0.05. The statistical analysis was carried out by a statistician
who was blind to treatments.
3. Results
Group A consisted of 10 patients (5 males and 5 females;
age range 18 to 35 years, mean age 24.2 ±2.1 years); group
Bof 9 patient (4 males and 5 females; age range 20–39 years;
mean age 27.8 ±2.2 years) and group C of 11 (6 males and 5
females; age range 18–31 years; mean age 24.1 ±1.2 years).
Among patients of group A, treated with piroxicam-FDDF
alone, one case of post-surgical infection and one case of
alveolitis did occur at day 2. These two patients were with-
drawn from subsequent comparative analysis of data, as they
required adjunctive antibiotic treatments which might have
interfered with the post-surgical evaluation of inflammation
and pain. Conversely, patients in groups Band C, treated with
azithromycin alone or azithromycin plus piroxicam-FDDF,
respectively, did not show any case of infectious complica-
Fig. 1. Pain intensity (panel A) throughout the study and acetaminophen
intake within the first 3 days after surgery (panel B) in groups A, B and
C. Each point or column indicates the mean value ±S.E.M. (vertical bars)
obtained from 8–11 patients.
*
p <0.05: significant difference versus values
from patients treated with piroxicam-FDDF alone.
tion. Overall, subsequent analysis of pain intensity, edema
and trismus was carried out on a total of 28 patients, with
groups A, B and C, consisting of 8, 9 and 11 patients, respec-
tively.
3.1. Pain intensity
The values obtained for VAS and acetaminophen intake
in the different treatment groups are shown in Fig. 1. At
days 1 and 2, VAS values obtained from patients treated with
piroxicam-FDDF, either alone (p <0.05) or in combination
with azithromycin (p <0.05), were significantly lower than
values recorded from patients who received azithromycin
alone. (ANOVA p-values: 0.020 and 0.012 at day 1 and 2,
respectively). Furthermore, patients of the former groups
showed a complete pain resolution by day 2, whereas in
patients treated with azithromycin alone post-operative pain
disappeared by day 3. No significant differences among treat-
ment groups were detected when pain intensity was estimated
at days 3 and 7 (Fig. 1A). Consistently with these results,
the daily average consumption of acetaminophen was sig-
nificantly higher in patients treated with azithromycin alone,
Fig. 2. Edema (panel A) and trismus (panel B) in groups A, B and C at 2
and 7 days after surgery. Each column indicates the mean value ±S.E.M.
(vertical bars) obtained from 8–11 patients.
*
p <0.05: significant difference
versus values from patients treated with piroxicam-FDDF alone.
488 F. Graziani et al. / Pharmacological Research 52 (2005) 485–490
in comparison with patients of groups A (p <0.01) and C
(p <0.01): ANOVA p-value: 0.0007 (Fig. 1B). All patients in
group A required acetaminophen within the first three days
after surgery, whereas in groups Band Conly 3 and 4 patients
needed the intake of acetaminophen tablets, respectively.
3.2. Edema and trismus
Edema and trismus score values are reported in Fig. 2. At
day 2, patients treated with piroxicam-FDDF alone showed
a significantly lower degree of facial edema than that esti-
mated for patients subjected to treatment with azithromycin,
either alone (p <0.05) or in combination with piroxicam-
FDDF (p <0.05): ANOVA p-value: 0.011. By contrast, no
significant differences among all treatment groups could be
observed when edema was evaluated at day 7 (Fig. 2A).
As far as trismus is concerned, no significant differences
could be appreciated upon evaluation of this parameter at
day 2 or 7 in patients administered with piroxicam-FDDF or
azithromycin, either alone or in combination (Fig. 2B).
4. Discussion
The main purpose of this study was to evaluate the efficacy
of piroxicam-FDDF in the control of pain and other inflam-
matory sequelae (edema and trismus) associated with dental
extraction. The surgical removal of impacted mandibular
third molar is used extensively as a model for the evalu-
ation of analgesic drugs, due to the high predictability of
the development of post-surgical inflammation and pain [15].
Accordingly, the estimation of different clinical parameters in
this model allowed to observe that, following repeated admin-
istration during the pre-operative period, piroxicam-FDDF is
able to modulate the intensity of both pain and inflammation
elicited by oral surgical procedures. Indeed, when compared
with the group of subjects receiving the antibiotic alone, pain
intensity in patients exposed to piroxicam-FDDF alone was
significantly lower at day 1, and underwent a complete relief
at day 2 after surgery. Consistently with this observation,
patients of piroxicam-FDDF groups reported a lower level
of mean acetaminophen consumption, allowed as a rescue
analgesic medication, and developed a less intense degree of
post-surgical edema.
Previous clinical experiences indicate that piroxicam can
be effective in the management of surgical oral pain when
administered, either as a regular or FDDF formulation, dur-
ing the post-operative period [16,17]. For instance, Selcuk
et al. [6] observed a superiority of piroxicam-FDDF (40 mg
as a single dose) over naproxen (550 mg as a single dose) in
the treatment of pain following oral surgery for removal of
impacted third molars. It has been reported also that pirox-
icam can induce a significant post-operative analgesia for
about 24 h following its administration by oral route, as a
single 40 mg dose, 2.5 h before oral surgery [5]. The present
results extend currently available knowledge on piroxicam-
induced analgesia, as they indicate that a 3-day course of
treatment with piroxicam-FDDF before dental extraction can
ensure a rapid blunting of both post-surgical inflammatory
reaction and onset of post-operative pain at the end of anes-
thesia, and show that such clinical benefits may persist for
several days after oral surgery, allowing a significant reduc-
tion of additional rescue analgesic medications. In addition,
the patterns of prophylactic analgesic and anti-inflammatory
responses, observed in the present study with the preopera-
tive administration of piroxicam-FDDF, are fully consistent
with data on the disposition of piroxicam into oral tissues,
as provided by a previous pharmacokinetic investigation
[18]. In that study, samples of blood, saliva and periodon-
tal tissues were collected at different times of post-surgical
period from dental patients treated with 20 mg/day of sub-
lingual piroxicam-FDDF during three consecutive days pre-
ceding oral surgery, and consistent piroxicam levels could be
detected in both plasma and gingival specimens up to 5 days
fromdental extraction, indicating that short termpreoperative
treatment with piroxicam-FDDF can ensure a good systemic
distribution as well as suitable and long-lasting therapeutic
concentrations of piroxicam in soft periodontal tissues [18].
Of note, acetaminophen was used as rescue analgesic med-
ication, due to its lack of anti-inflammatory effects and, as
expected, all patients undergoing only antibiotic prophylaxis
needed the intake of one or two acetaminophen tablets. This
circumstance may well explain the lowlevel of pain recorded
by VAS analysis at day 1 in the azithromycin group. On
the other hand, low pain intensity in the other two groups
was likely to result from piroxicam-FDDF administration, as
indicated also by the observation that few of those patients
required acetaminophen as a rescue analgesic.
The incidence of alveolitis is 1–3% following dental
extraction, or even higher after the extraction of impacted
third molar, and previous studies have shown a reduction in
the incidence of alveolitis following either topical or sys-
temic treatments with antibiotics [19]. Moreover, antibiotic
prophylaxis may help in reducing post-surgical infection
[20]. In line with these expectations, two cases of infec-
tious adverse events were observed among patients receiv-
ing piroxicam-FDDF alone, while no evidence of infection
could be demonstrated during the post-operative follow-up of
patients treated with azithromycin, either alone or in combi-
nation with piroxicam-FDDF. The lack of post-surgical infec-
tious complications in patients subjected to a short course of
azithromycin administration during the pre-operative period
is consistent with the peculiar pharmacokinetic and pharma-
codynamic properties demonstrated for this macrolide antibi-
otic in previous investigations. Indeed, when administered to
odontoiatric patients for three consecutive days at the oral
dose of 500 mg/day, azithromycin penetrates extensively into
periodontal tissues, where it remains accumulated for several
days at levels higher than minimuminhibitory concentrations
required to counteract the growth of Grampositive and Gram
negative bacteria most frequently implicated in the pathogen-
esis of acute odontogenic infections and chronic periodontal
F. Graziani et al. / Pharmacological Research 52 (2005) 485–490 489
disease [21,11]. Furthermore, azithromycin exerts signifi-
cant bactericidal actions on microbial flora isolated from
oral infectious foci, and it was proven to be clinically effec-
tive when administered to patients with dental infections
[13]. Taken together, these findings encourage further clini-
cal trials, to be performed in adequate patient numbers, and
designed specifically to evaluate the ability of azithromycin
to prevent post-operative infectious complications following
oral surgical procedures.
An interesting observation made in the present study
was the recording of discrepant profiles of analgesic and
anti-inflammatory actions exerted by piroxicam-FDDF in
patients to whom this oxicam derivative was administered
in combination with azithromycin. In particular, similar pat-
terns of analgesic response, in terms of both VAS values
and acetaminophen intake, were obtained when comparing
piroxicam-FDDF-treated patients with those exposed to the
NSAIDin combination with azithromycin, whereas this latter
groupdisplayedhigher degrees of post-surgical inflammatory
signs at day 2, as estimated by the intensity of post-operative
edema. Therefore, while the concomitant administration of
azithromycin allowed a full development of the analgesic
effect of piroxicam-FDDF, it appears that the macrolide
antibiotic didinterfere withthe inhibitoryeffect of piroxicam-
FDDF on post-surgical edema, leading to a delay in the onset
of its local anti-inflammatory action. To explain the above
discrepancies, both pharmacokinetic and pharmacodynamic
mechanisms can be advocated. A previous pharmacokinetic
investigation in odontoiatric patients demonstrated lower lev-
els of piroxicam in gingival and alveolar bone of patients
treated with piroxicam-FDDF plus azithromycin, as com-
pared to piroxicam concentrations detected in periodontal
tissues of patients exposed to piroxicam-FDDF alone [18].
In the same study, similar piroxicam levels could be detected
in plasma irrespectively of azithromycin co-administration,
and therefore it was suggested that azithromycin might exert
a negative influence on piroxicam distribution into periodon-
tal tissues, while not affecting the systemic pharmacokinetic
profile of this oxicam NSAID at the systemic level [18].
In addition, when considering the pharmacodynamic bases
underlying the pain modulating effects of piroxicam, recent
evidence indicates that an interaction with cyclooxygenase
enzymes in the central nervous system, rather than a down-
regulation of prostanoid production directly in the inflamma-
tory site, is more likely to account for the analgesic actions of
NSAIDs [22,23]. Overall, it is conceivable that azithromycin
may partly reduce or slow the local effects of piroxicam
on inflamed post-surgical oral tissues, without any signifi-
cant interference with the systemic analgesic action of this
NSAID.
In conclusion, the present results provide direct clinical
evidence that a short term administration in the pre-operative
period enables piroxicam-FDDF to significantly reduce the
occurrence of post-surgical pain, and azithromycin to pro-
tect oral tissues against post-operative infectious complica-
tions. Moreover, it was observed that, upon combined treat-
ment with piroxicam-FDDF and azithromycin, the macrolide
antibiotic may interfere with the therapeutic action of pirox-
icam on the recovery of oral tissues from surgical proce-
dures, without affecting its beneficial effect on surgical pain
throughout the whole post-operative period.
References
[1] Rikhotso E, Ferretti C. Prophylactic use in oral surgery—a review
of current concepts. S Afr Dent J 2002;57:408–13.
[2] Savage MG, Henry MA. Preoperative nonsteroidal anti-inflammatory
agents: review of the literature. Oral Surg Oral Med Oral Pathol Oral
Radiol Endod 2004;98:146–52.
[3] Olkkola KT, Brunetto AV, Mattila MJ. Pharmacokinetics of oxi-
cam nonsteroidal anti-inflammatory agents. Clin Pharmacokinet
1994;26:107–20.
[4] Howell TH, Fiorellini J, Weber HP, Williams RC. Effect of the
NSAID piroxicam, topically administered, on the development of
gingivitis in beagle dogs. J Periodontal Res 1991;26:180–3.
[5] Hutchison GL, Crofts SL, Gray IG. Preoperative piroxicam for post-
operative analgesia in dental surgery. Br J Anaesth 1990;65:500–3.
[6] Selcuk E, Gomel M, Apaydin S, Kose T, Tugiular I. The postopera-
tive analgesic efficacy and safety of piroxicam(FDDF) and naproxen
sodium. Int J Clin Pharmacol Res 1998;18:21–9.
[7] Moore PA. Dental therapeutic indications for the newer long-acting
macrolide antibiotics. J Am Dent Assoc 1999;130:1341–3.
[8] Addy LD, Martin MV. Azithromycin and dentistry—a useful agent.
Br Dent J 2004;197:141–3.
[9] Williams JD, Maskell JP, Shain H, Chrysos G, Sefton AM,
Fraser HY, et al. Comparative in vitro activity of azithromycin,
macrolides (erythromycin, clarithromycin and spiramycin) and strep-
togramin RP 59500 against oral organism. J Antimicrob Chemother
1992;30:27–37.
[10] Malizia T, Tejada MR, Ghelardi E, Senesi S, Gabriele M, Giuca MR,
et al. Periodontal tissue disposition of azithromycin. J Periodontol
1997;68:1206–9.
[11] Blandizzi C, Malizia T, Lupetti A, Pesce D, Gabriele M, Giuca
MR, et al. Periodontal tissue disposition of azithromycin in patients
affected by chronic inflammatory periodontal diseases. J Periodontol
1999;70:960–6.
[12] Lo Bue AM, Sammartino R, Chisari G, Gismondo MR, Nicoletti G.
Efficacy of azithromycin compared with spiramycin in the treatment
of odontogenic infections. J Antimicrob Chemother 1993;31(Suppl.
E):119–27.
[13] Smith SR, Foyle DM, Daniels J, Joyston-Bechal S, Smales
FC, Sefton A, et al. A double-blind placebo-controlled trial of
azithromycin as an adjunct to non-surgical treatment of periodon-
titis in adults: clinical results. J Clin Periodontol 2002;29:54–61.
[14] Rakprasitkul S. Mandibular third molar surgery with a primary clo-
sure and tube drain. Int J Oral Maxillofacial Surg 1997;26:187–90.
[15] Barden, Edwards JE, McQuay HJ, Wiffen PJ, Moore RA. Relative
efficacy of oral analgesics after third molar extraction. Br Dent J
2004;197:407–11.
[16] Desjardins PJ. Analgesic efficacy of piroxicam in post-operative den-
tal pain. Am J Med 1988;20(Suppl. 5):35–41.
[17] Wakeling HG, Creagh Barry P, Butler PJ. Post-operative analgesia in
dental day case surgery, a comparison between Feldene Melt2 (pirox-
icam and diclofenac suppositories). Anaesthesia 1996;51:784–6.
[18] Malizia T, Batoni G, Ghelardi E, Baschiera F, Graziani F, Blan-
dizzi C, et al. Interaction between piroxicam and azithromycin dur-
ing their distribution to human periodontal tissues. J Periodontol
2001;72:1151–6.
[19] Larsen PE. Alveolar osteitis after surgical removal of impacted
mandibular third molars. Oral Surg Oral Med Oral Pathol
1992;73:393–7.
490 F. Graziani et al. / Pharmacological Research 52 (2005) 485–490
[20] Peterson LJ. Antibiotic prophylaxis against wound infection in
oral and maxillo-facial surgery. J Oral Maxillofac Surg 1990;48:
617–20.
[21] Sefton AM, Maskell JP, Beighton D, Whiley A, Shain H, Foyle D,
et al. Azithromycin in the treatment of periodontal disease. Effect
on microbial flora. J Clin Periodontol 1996;23:998–1003.
[22] Fabbri A, Cruccu G, Sperti P, Ridolfi M, Ciampani T, Leardi MG, et
al. Piroxicam-induced analgesia: evidence for a central component
which is not oppioid mediated. Experientia 1992;48:1139–42.
[23] Samad TA, Sapirstein A, Woolf CJ. Prostanoids and pain: unravel-
ing mechanisms and revealing therapeutic targets. Trends Mol Med
2002;8:390–6.