The Scientic World Journal

Volume 2012, Article ID 606404, 14 pages

doi:10.1100/2012/606404
The cienticWorldJOURNAL
Review Article
Do Nonsteroidal Anti-Inammatory Drugs Affect Bone Healing?
A Critical Analysis
Ippokratis Pountos,
1
Theodora Georgouli,
1
Giorgio M. Calori,
2
and Peter V. Giannoudis
1, 3
1
Academic Department of Trauma & Orthopaedics, School of Medicine, University of Leeds, Leeds LS1 3EX, UK
2
Academic Department of Trauma & Orthopaedics, School of Medicine, University of Milan, 20122 Milano, Italy
3
Academic Unit, Department of Trauma and Orthopaedics, Clarendon Wing, Leeds Teaching Hospitals NHS Trust,
Great George Street, Leeds LS1 3EX, UK
Correspondence should be addressed to Peter V. Giannoudis, pgiannoudi@aol.com
Received 18 September 2011; Accepted 18 October 2011
Academic Editors: A. Ndreu and A. Sihoe
Copyright 2012 Ippokratis Pountos et al. This is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly
cited.
Nonsteroidal anti-inammatory drugs (NSAIDs) play an essential part in our approach to control pain in the posttraumatic
setting. Over the last decades, several studies suggested that NSAIDs interfere with bone healing while others contradict these
ndings. Although their analgesic potency is well proven, clinicians remain puzzled over the potential safety issues. We have
systematically reviewed the available literature, analyzing and presenting the available in vitro animal and clinical studies on this
eld. Our comprehensive review reveals the great diversity of the presented data in all groups of studies. Animal and in vitro studies
present so conicting data that even studies with identical parameters have opposing results. Basic science research dening the
exact mechanism with which NSAIDs could interfere with bone cells and also the conduction of well-randomized prospective
clinical trials are warranted. In the absence of robust clinical or scientic evidence, clinicians should treat NSAIDs as a risk factor
for bone healing impairment, and their administration should be avoided in high-risk patients.
1. Introduction
1.1. Bone Healing. Bone healing is one of the most complex
cascades of events aiming to the repair of fractured bone
without the formation of scar tissue [1]. In this physiological
process, several cell types participate along with signal
pathways and alternations in the biochemical prole of the
local area. Bone healing can be either primary (direct) or
secondary (indirect), [1, 2] with the majority of fractures
heal indirectly, that is, a process subdivided in several stages
[1]. The indirect fracture healing begins immediately after
fracture occurrence with the disruption of local blood supply,
hypoxia, and the formation of a hematoma, (Figure 1) [1].
Cytokines and growth factors are released both locally and
systemically and induce a mitogenic and osteogenic eect
on the osteoprogenitor cells [35]. The formation of new
blood vessels, in association with further growth factor
and prostaglandin production, promotes dierentiation of
mesenchymal stem cells (MSCs) toward chondrogenic or
osteogenic lineages, forming initially woven bone and in
turn, the hard callus [3, 68]. Finally, this process is fol-
lowed by an extended period of remodeling characterized
by resorption and new bone formation, resulting in the
restoration of mechanical strength and stability [8].
1.2. Factors Aecting Bone Healing. The outcome of bone
healing can be aected by a diversity of local and systemic
factors with varying degrees of aection including fracture
gap and comminution, disturbances of blood ow, degree of
soft tissue damage [9, 10], insucient mechanical stability,
[1013], poor nutritional state, age, and smoking [14, 15].
Another important factor that can interfere with the bodys
ability to heal a fracture is the administration of several
pharmacological agents [15]. Steroids, chemotherapy drugs,
and some classes of antibiotics have been reported to exert a
negative eect on bone healing[15, 16]. In addition, NSAIDs
that are one of the most commonly prescribed drugs for pain
relief and inammation to date have also been found to delay
union and to inhibit fracture healing [15].
2 The Scientic World Journal
Fracture occurrence
Blood vessel disruption
Hematoma formation
Chemotactic invasion of
osteoprogenitor
Mitogenic and osteogenic
molecules released
Cellular proliferation begins
Formation of granulation
tissue
Neovascularization
Soft callus is formed
Dierentiation starts
Hard callus is formed
Remodeling
-
-
-
-
-
-
-
-
-
-
-
-
Figure 1: The fracture healing cascade.
1.3. NSAIDs Physiology. Nonsteroidal anti-Inammatory
drugs (NSAIDs) have their origin in the extracts of salicylate-
containing plants initially described in ancient Roman and
Greek literature, with the willow tree extract to be renowned
for their antipyretic, analgesic, and anti-inammatory prop-
erties [17]. Their mode of action remained unknown till the
1970s when Sir John Vane demonstrated the inhibition of the
enzymatic production of prostaglandins by NSAIDs [5].
During the biosynthesis of prostaglandins, cyclo-oxygen-
ase (COX or prostaglandin H synthease) catalyses the con-
version of arachidonic acid to the prostaglandin endoperox-
idases PGG
2
and then PGH
2
[18, 19]. PGH
2
is the precursor
for the biological active prostaglandins and thromboxanes.
PGH
2
is then isomerized into various prostanoids such us
thromboxane A
2
(TXA
2
), prostacyclin (PGI
2
), PGD
2
, PGE
2
,
and PGF
2a
[18, 19].
However, COX-1 is constitutively expressed in the most
of cells and is involved in physiological processes. In the
gastrointestinal tract (GI), prostacyclin and PGE
2
exert a
protective eect by reducing acid secretion, vasodilatation
of blood vessels of gastric mucosa, and stimulation of
production of mucus which acts as a barrier [20]. In the
kidneys, prostaglandins play a key role in regulating blood
ow and enhancing organ perfusion [20]. COX-1 expres-
sion is also found in faetal and amniotic cells, uterine epi-
thelium in early pregnancy, and central nervous system, and
it is thought to exert complex integrative functions [21].
COX-2, on the other hand, is considered to be induced
by inammation and by the presence of proinammatory
cytokines and mitogens [22]. It has been suggested that
the anti-inammatory action of NSAIDs is due to the
inhibition of COX-2, whereas COX-1 inhibition is associated
to unwanted eects related to interference of the regulatory
and protective mechanisms [22, 23]. Recent studies, however,
have indicated that COX-2 is also constitutively expressed in
the brain, and in particular, in the hippocampus and cortical
glutaminergic neurons as well as the kidneys, uterus, and
prostate [24, 25]. Similarly, COX-1, despite its constitutively
expression, has been shown to participate in inammation
(e.g., lipopolysaccharide-induced inammation) where it
might be inducible [26].
1.4. Prostaglandins during Fracture Healing. Prostaglandins
(PGs) are autocrine and paracrine lipid mediators produced
by several cell types capable of mediating either a stimulatory
or resoptive role depending on the physiological or patho-
logical conditions [27]. Administration of prostaglandins in
animal models has shown to increase cortical and trabecular
mass and cause hyperostosis in infants [28, 29]. Similarly,
local administration of PGs in rat long bones had stimulatory
properties suggesting direct eect on bone by inducing
osteogenesis [30]. At a cellular level, PGs have a direct eect
on osteoclasts leading to increased bone resorption by a
mitogenic eect and increasing their functional activity [31].
The Scientic World Journal 3
On the other hand, PGs can exert an anabolic eect on the
bone by increasing the multiplication and dierentiation of
osteoblasts [32]. One could claim that PGs safeguard the
balance between bone resorption and bone formation [33].
Following a fracture, local release of PGs occurs early
as a result of the acute aseptic inammatory response [34].
COX-2 plays a critical role in this phase and its induction
in osteoblasts is essential for bone healing [35]. In COX-2
null mice, fracture healing was found impaired characterized
by reduced bone formation and persistence of mesenchyme
and cartilage [36]. In the same study, COX-1 knockout
animal was found to have the same healing potential to that
of the normal wild type [36]. COX-2 activation therefore
is a local regulator of cellular response within bone and
responsible for the production of PGs [37]. It is not yet
clear what the exact mechanism of PGs on bone cells is;
however, it was found that PGE2 regulates BMP-2, BMP-
7, and RANKL expression [3840], and it can increase cell
numbers through suppression of apoptosis without direct
eect on proliferation [41, 42]. PGs exert this range of action
through a variety of receptors expressed. These receptors
belong to the G-protein-coupled receptor family and are the
EP1, EP2, EP3, and EP4 subtypes [40]. Although the role of
each receptor is not fully explored, studies suggest that the
PGE2 binding to EP4 can stimulate osteoclastogenesis and
osteoblastic dierentiation, and animal models lacking the
EP2 and EP4 receptors had defects in bone metabolism [43].
On the contrary, EP1 null mice found to have accelerated
fracture repair and MSCs isolated from their bone marrow
had higher osteoblast dierentiation capacity and accelerated
bone nodule formation and mineralization in vitro [44].
1.5. NSAIDs and Analgesia. In acute pain after fracture
or during the postoperative period after fracture xation,
NSAIDs play an important role due to their pronounced
analgesic potency, anti-inammatory eects, and lesser side
eects compared to opioids [7, 45]. However, studies com-
paring opiates and NSAIDs have shown that NSAIDs are at
least as eective as opiates with some studies suggesting that
NSAIDs can achieve greater reductions in pain scores [46
52]. For acute pain, it has been suggested that NSAIDs should
be used as the rst line of treatment in pain therapy and
recommend that opioids should be added only if pain is not
controlled adequately with NSAIDs alone [50]. Furthermore,
the use of NSAIDs instead of narcotic analgesics avoids
signicant side eects like respiratory depression, sedation,
and cognitive eects [53]. For the postoperative patients, this
can be translated with decreased hospital stay, allowing early
mobilization and weight bearing [15, 46, 51, 52].
While there are clear benets supporting the administra-
tion of NSAIDs as pain relief agents following fractures, their
wide-spread use has been challenged due to their reported
negative impact on the bone repair processes [47, 51, 52]. Do
NSAIDs inhibit the healing of fractures? Can they safely be
administered?If so, at what time point and for how long? In
order to provide replies to the above queries, we undertook a
comprehensive review of the literature.
Records identied through database searching
n = 4443
New words added
(ex. name of NSAIDs)
n = 291
Studies on English language identied
n = 264
Studies included in the study
n = 90
Figure 2: Flow chart diagram of included studies.
2. Materials and Methods
We searched available literature through PubMed, OVID,
and EMbase with general keywords including mesenchymal
stem cells (MSCs), Bone healing, and Bone marrow-
derived stem cells both isolated or in combination with
specic words including generic words like NSAIDs or spe-
cic names of NSAIDs from January 1980 to January 2011.
For paper selection, the initial inclusion criteria were studies
publishing results on the eect of NSAIDs on bone healing
in vivo both in humans and animal models, and also in-vitro
studies on the eect of these agents on osteoprogenitor cells.
Exclusion criteria included publications in languages other
than English or studies with unclear methodology [54, 55].
The papers describing the eect of NSAIDs on bone healing
were reviewed and are presented below.
3. Results
Out of 4443 papers that were initially isolated, 90 meet the
inclusion criteria (Figure 2) [56145]. Studies selected were
grouped as experimental (in vivo or in vitro) or clinical as
described below.
3.1. In-Vitro Models. We identied 18 in-vitro studies ana-
lyzing the eect of NSAIDs on osteoblasts and MSCs
ability to proliferate and dierentiate toward osteogenic
lineages, (Table 1) [5673]. An early study, byT ornkvist et
al., using mesenchymal limb-bud cells reported no eect
on osteogenesis and chondrogenesis by indomethacin [56].
However, the latter studies presented a diversity of results.
The proliferation potential of osteogenic cells was found
inhibited, and the higher the concentration, the more
potent the antiproliferative eect was [57, 58, 60, 68, 71].
Interestingly, replenishment of PGE-1, PGE-2 and PGF2a did
not reverse this negative eect [64, 68]. Other studies showed
no eect on low concentration and reported a negative eect
at higher ones [63, 67, 69, 72, 73].
4 The Scientic World Journal
Table 1: In-Vitro Studies BM: Bone marrow, TB: Trabecular Bone.
Year/Study Model used Drug Outcome
T ornkvist et al.,
1984 [56]
Chicken mesenchymal
limb-bud cells
Indomethacin (25100 M) (i) No eect on osteogenesis and chondrogenesis
Ho et al., 1999 [57]
Osteoblasts derived from
fetal rat calvaria
Ketorolac (0.11000 M),
Indomethacin (0.01100 M)
(i) All concentration of Ketorolac inhibited
proliferation at 24 hours
(ii) 0.1 M of indomethacin or higher inhibited
proliferation
(iii) A dose dependant increase of ALP was found
for concentration between 0.1100 M of Ketorolac
(iv) Both NSAIDs stimulated collagen type I
synthesis
Evans and Butcher,
2004 [58]
Human trabecular bone
osteoblasts
Indomethacin (0.0030.3 M/L)
(i) Inhibition of proliferation and increase in
collagen synthesis and ALP in a dose dependant
manner
Wang et al., 2004
[59]
MG63 human
osteoblasts
Celecoxib (1120 M)
(i) Dose dependant decrease of cellular
proliferation and stimulation of Ca
++
production
Chang et al., 2005
[60]
Osteoblasts derived from
fetal rat calvaria
Diclofenac, piroxicam,
indomethacin Ketorolac
(0.0010.1 M)
(i) All NSAIDs resulted in cell cycle arrest and cell
death
(ii) Piroxicam had the least eect to produce
osteoblastic dysfunction
Wang et al., 2006
[61]
BM-derived Rat MSCs Aspirin 1, 5, 10 mmol/L (i) Inhibition of MSCs proliferation
Wiontzek et al.,
2006 [62]
MG63 human
osteoblasts
Celecoxib (10 M)
(i) No eect on Ca
++
production, COX-2
expression, ALP and osteocalcin
Wolfesberger et al.,
2006 [63]
Canine Osteosarcoma
cell line
Meloxicam (1200 g/mL)
(i) Marked untiproliferative eect for
concentrations over 100 while lower concentrations
resulted in an increase of cell numbers
Chang et al., 2007
[64]
Human MSCs and
D1-cells (Mice)
Indomethacin (10, 100 M),
Celecoxib (1, 10 M)
(i) Inhibition of proliferation for both NSAIDs but
no signicant cytotoxic eect
(ii) Replenishment of PGE-1, PGE-2 and PGF2a
did not reverse this negative eect
Kellinsalmi et al.,
2007 [65]
Human MSCs
Indomethacin (1, 10, 100 M),
Parecoxib (1, 10, 100 M),
NS398 (0.03, 0.3, 3 M)
(i) All studied NSAIDs inhibited osteoblastic and
osteoclastic dierentiation
(ii) Signicant increase of adipocytes suggesting
diversion to adipogenesis instead of osteogenesis
Arpornmaeklong et
al., 2008 [66]
Mouse calvaria cell line
MC3T3-E1
Indomethacin (0.1 M),
Celecoxib (1.5, 3, 9 M)
(i) Inhibition of growth with both NSAIDs
(ii) Indomethacin had a higher inhibitory eect
than Celecoxib
Abukawa et al., 2009
[67]
Porcine BM progenitor
cells
Ibuprofen (0.1, 1, 3 mmol/L)
(i) 0.1 mmol/L had no eect on proliferation, ALP,
bone matrix mineralization while inhibition found
for the higher studied concentrations
Chang et al., 2009
[68]
Human osteoblasts
Indomethacin (0.11 M),
Ketorolac (0.11 M),
Piroxicam (0.11 M),
Diclofenac (0.11 M),
Celecoxib (110 M)
(i) Inhibition of proliferation occurred with all
studied NSAIDs
(ii) Replenishment of PGE-1, PGE-2 and PGF2a
did not reverse this negative eect
Kolar et al., 2009
[69]
MG63 human
osteoblasts
Celecoxib (2, 10, 50 M)
(i) Marginal eect with the concentrations of 2 and
10 M but 50 M reduced cell viability and OPG
secretion and stimulated oxygen consumption and
GLUT-1 expression
Yoon et al., 2010
[70]
Human BM MSCs
Celecoxib (10, 20, 40 M ),
Naproxen (100, 200, 300 M)
(i) No eect on ALP and Calcium content in
absence of Interleukin 1 while in its presence ALP
and Calcium was reduced only with the highest
studied concentration
The Scientic World Journal 5
Table 1: Continued.
Year/Study Model used Drug Outcome
Guez et al., 2011
[71]
Human MG-63
Osteosarcoma Cell
Indomethacin (110 M)
Nimesulide (110 M)
Diclofenac (110 M)
(i) All NSAIDs had an inhibiting eect on
osteoblastic proliferation and signicant eects on
the antigenic prole
(ii) No treatment altered osteocalcin synthesis
M uller et al., 2011
[72]
Equine BM MSCs
Flunixin (101000 M),
phenylbutazone (101000 M),
Meloxicam (0.01200 M),
Celecoxib (0.01200 M)
(i) Low NSAIDs concentrations had positive eect
on proliferation while the higher ones inhibited
proliferation
(ii) Adipogenic and chondrogenic dierentiation
was found unaltered however osteogenesis was
signicantly disrupted
Pountos et al., 2011
[73]
BM and TB derived
MSCs
Diclofenac, Ketorolac, Parecoxib,
Ketoprofen, Piroxicam,
Meloxicam and Lornoxicam
(all 0.001 to 100 g/mL)
(i) No eect on MSCs proliferation when cellular
medium was supplemented with expected plasma
concentrations
(ii) Negative eect encountered when high
concentrations used (over 100 g/mL)
(iii) NSAIDs in plasma concentrations had no
eect on osteogenesis
(iv) Chondrogenesis was found inhibited by
NSAIDs
The osteogenic potential of the studied cells, measured
by the levels of ALP activity and calcium production,
was either found increased or unaected in the majority
of the studies [5659, 62, 73]. NSAIDs also reported to
stimulate collagen synthesis [57, 58]. On the contrary, other
researchers failed to reproduce this result showing disruption
of osteogenesis [65, 72]. Kellinsalmi et al. reported that
indomethacin, parecoxib, and NS398 inhibited osteoblastic
dierentiation of human MSCs and found a signicant
increase of adipocytes suggesting diversion to adipogenesis
instead of osteogenesis [65].
In terms of chondrogenesis, limited studies exist which
either present no eect [72] or a negative eect in expected
concentrations after administration [73]. In an attempt to
explain this wide diversity of results, it was apparent that
cells were isolated from a variety of species and sites [56
58, 66, 68, 69, 72, 73]. However, we could not nd any
association between them, and no association was apparent
between dierent NSAIDs or even selectivity toward the
COX-1 or COX-2 enzyme to explain these results.
3.2. Animal Models. A large volume of work has been under-
taken over the last 4 decades using experimental fracture
animal models. The majority of these studies were centred
over rodents or rabbits, and as with the in-vitro studies, great
diversity and controversial results have been presented in
the 54 studies identied (Table 2) [74127]. A proportion of
these studies suggest that NSAIDs adversely aect the bone
physiology by delaying bone healing and callus formation,
impairing bending stiness and the bones mechanical
properties leading to an increased rate of nonunions [74
105]. Some authors have even compared NSAIDs eect
on fracture healing with that of other pharmacological
agents like steroids [89, 91]. Hgevold et al. presented
that short-term administration of indomethacin inhibits
fracture healing while this was not the case with short-term
administration of methylprednisolone [89].
On the other hand, several studies failed to reproduce
this eect suggesting that NSAIDs have no eect of fracture
healing [110125]. The results were so controversial that
dierent researchers with identical animal fracture models,
same drugs, and same doses presented opposite outcomes
[122, 127]. Analysing these studies further, we could not
identify any association among the class or potency of the
studied NSAIDs to inhibit the COX-1 or COX-2 enzyme,
the dose, or the timing. In terms of timing, although some
authors suggest that short-term administration after fracture
could be safe, others contradict this nding suggesting that
NSAID administration is safe only if it is initiated a few
weeks after fracture [78, 85, 98, 101, 105, 106, 108, 126].
A link, however, can exist between the size of the animals.
The vast majority of the animal models involved rodents or
rabbits. There are two studies that involved dogs and goats
whose results showed no inhibition of bone healing or bone
ingrowth suggesting that the type or size of the animal model
used might be an explanation for the dierences seen in the
results presented [113, 117].
3.3. Clinical Studies. There are only a few retrospective hu-
man studies and even fewer prospective randomized trials
studying the eect of NSAIDs after fracture or spinal fusion,
(Tables 3 and 4) [128145]. In a double-blinded randomized
control trial, Adolphson et al. found that piroxicam had no
eect on the healing potential of 42 postmenopausal women
with displaced Colles fractures [137]. Similar ndings were
reported by Davis and Ackroyd who studied the eect of
uriprophen on Colles fracture healing potential [136]. In
cementless hip arthroplasty, indomethacin was found to have
no eect on the prosthetic loosening. No eect was also
reported in a randomized, controlled, and blinded study
6 The Scientic World Journal
Table 2: Animal studies: agents and model used in relation to the
presented eect.
Impaired bone healing
(1) Aspirin [84]
(2) Celecoxib [101, 102]
(3) Diclofenac [9799]
(4) Etodolac [104, 105]
(5) Ibuprofen [88, 9496, 127]
(6) Indomethacin [7891, 119, 127]
(7) Ketoprofen [77]
(8) Ketorolac [7476, 107]
(9) Meloxicam [85, 103]
(10) Naproxen [92, 93]
(11) Parecoxib [74]
(12) Rofecoxib [92, 94, 95, 108, 109]
(13) Tenoxicam [99, 100]
(14) Valdecoxib [107]
No eect
(1) Celecoxib [80, 111, 119, 120]
(2) Diclofenac [123]
(3) Etoricoxib [110]
(4) Ibuprofen [111, 121, 122]
(5) Indomethacin [81, 111, 114118]
(6) Ketoprofen [112, 113]
(7) Ketorolac [110, 111]
(8) Meloxicam [113]
(9) Nimesulid [124]
(10) Rofecoxib [111, 123, 125]
Short term has no eect
(1) Diclofenac [98, 106]
(2) Ketoprofen [126]
(3) Ketorolac [78]
(4) Parecoxib [106]
(5) Rofecoxib [85, 108]
(6) Valdecoxib [106]
Model used
(i) Rats
[74, 77, 78, 80, 81, 83, 84, 8991, 93, 94, 97, 98, 100
107, 110, 114, 116, 118, 121124, 126]
(ii) Mouse [109, 111, 120]
(iii) Rabbit
[75, 76, 79, 82, 8588, 92, 95, 96, 99, 108, 112, 115, 119, 125]
(iv) Dog [117]
(v) Goats [113]
by Sculean et al. who studied the eect of rofecoxib on
the healing of intrabony periodontal defects [143]. On the
contrary, four retrospective studies suggested that patients
using NSAIDs after fracture had a higher incidence of
nonunion compared to those that did not [140142, 144].
Bhattacharyya et al., have suggested that patients receiving
NSAIDs within 90 days after fracture had a 3.7-fold risk
for nonunion, while the risk for opioids users was 1.6 folds
[144].
Detrimental eects in spinal fusion are presented by
some authors, while others concluded that NSAIDs do
not aect union. Park et al. found that the incidence of
incomplete union or nonunion was much higher in patients
taking ketorolac and the relative risk was approximately 6
times higher compared to that of the control group [131]. A
more recent study by Lumawig et al. indicated that diclofenac
sodium showed a dose-dependent inhibitory eect toward
spinal fusion especially when used during the immediate
postoperative period [134]. In addition, it was pointed out
that patients who continued to take NSAIDs for more than
3 months postoperatively showed signicantly lower fusion
and success rates [128]. On the contrary, other studies failed
to support these ndings suggesting that NSAIDs do not
aect union after spinal fusion [130, 133, 135].
4. Discussion
For many years, NSAIDs have played an essential part in
our approach to control pain in the posttraumatic setting.
However, several authors highlighted that NSAIDs could
inhibit the bone healing process. The available data from
animal studies have all evaluated the properties of newly
formed bone in animals that NSAIDs were administered
in dierent doses and durations. One could expect some
uniformity in these results but on the contrary great diversity
and conicting results exist. It is of question how so many
studies have failed to provide a clear message with regards
to the eect and mode of action of NSAIDs on bone healing
in animals. The dierences reported were not only between
species, dose, and duration of administration but also
between identical parameters. For instance, 30 mg/kg/day
of ibuprofen given orally in rats had no eect on femoral
fracture by Huo et al. [122], but retardation of fracture
healing with signicant dierences of mechanical and histo-
logical properties was reported by Altman et al. [127]. Many
researchers have also chosen long-term administration of
NSAIDs or high doses that is against the intended human
therapies. For example, Leonelli et al. compared the eect
of 30 mg/kg/day of Ibuprofen and 8 mg/kg/day of rofecoxib
on the union potential of a closed femoral fracture in a rat
model [94]. Their results showed nonunions in 64.7% of
rofecoxib-treated rats and 17.6% of ibuprofen-treated rats
but the dose of rofecoxib used was more than 10 times
the dose given in humans for acute pain. It is also unclear
whether this diversity of results is due to inter- or intraspecies
dierences, compensatory local and systemic factors, or even
dierent pharmacokinetics of the drugs in the laboratory
models compared to humans. It is possible that secondary
factors inuence the nal result like the level of analgesia
achieved by a specic dose and class of NSAID aecting the
weight bearing status for some animals and thus to evolution
of healing.
The extent of trauma and the comminution of the
fractures produced in these experimental models, we believe,
is another major factor that could explain the dierences
The Scientic World Journal 7
Table 3: The eect of NSAIDs on spinal fusion in humans.
Study/Year Design NSAID used Conclusions and recommendations
Deguchi et al., 1998
[128]
Retrospective review of 73 patients
undergoing primary or revision one or
two level lumbar fusion
Not specied
(i) Patients who continued to take NSAIDs for
more than 3 months postoperatively showed
signicantly lower fusion and success rates
Glassman et al., 1998
[129]
Retrospective review of 288 patients
undergoing posterior L4 to sacral
fusion
Ketorolac
(i) High rate of nonunion in spinal fusion
(ii) Avoid NSAIDs in early postoperative period is
recommended
Vitale et al., 2003
[130]
Retrospective review of 208 children
undergoing scoliosis correction
Ketorolac
(i) No signicantly increase in complications,
including transfusion and reoperation
Park et al., 2005 [131]
Retrospective review of 88 consecutive
patients undergoing posterolateral
lumbar fusion
Ketorolac
(i) The incidence of incomplete union or
nonunion was much higher in the ketorolac
group, and the relative risk was approximately 6
times higher than control group
Pradhan et al., 2008
[132]
Retrospective review of 405
consecutive patients undergoing one,
two or three level posterolateral
lumbar fusion
Ketorolac
(i) The use of ketorolac limited to 48 hours after
surgery for adjunctive analgesia, has no
signicant eect on ultimate fusion rates.
Sucato et al., 2008
[133]
Retrospective review of 319 patients
undergoing scoliosis correction
Ketorolac
(i) Ketorolac does not increase the incidence of
developing a pseudoarthrosis when used as an
adjunct for postoperative analgesia
Lumawig et al., 2009
[134]
Retrospective review of 273 patients
undergoing one or two level posterior
lumbar fusion
Diclofenac
(i) Diclofenac sodium showed a dose-dependent
inhibitory eect toward spinal fusion especially
when used during the immediate postoperative
period
Horn et al., 2010
[135]
Retrospective review of 46 pediatric
patients who undergone spinal fusions
for scoliosis
Ketorolac
(i) No clinical or radiographic evidence of curve
progression, nonunion, or instrumentation
failure
between researchers. Dierences in the fracture comminu-
tion, force used, soft tissue damage, and fracture stability
achieved can all inuence the nal result. This argument can
be further strengthened by the observations of Engebretsen
et al., who reported that indomethacin exerted a negative
eect in unstable fracture condition while in stable ones
had no signicant eect compared to controls [78]. This
could also explain the high complication rates presented
even for the control animal groups. Long et al., for instance,
studied the eect of celecoxib and indomethacin on a rabbit
model of spinal fusion [119]. Their results showed fusion
rates of 64% in the control group, 45% for the celecoxib
group, and 18% for the indomethacin group. Although
a signicant negative eect is presented with these two
NSAIDs, a nonunion rate of 36% is high to be used as
baseline. It is possible, for example, that NSAIDs interfere
only in endochondral ossication, therefore, if the fracture
is comminuted and highly unstable, NSAIDs might have a
detrimental eect on the consolidation process, while in a
more stable fracture, NSAIDs could be totally ineective.
One positive association could be the fact that the
majority of the animal studies involved small animals, that
is, rodents and rabbits, while the available studies involving
goats and dogs show no eect of bone ingrowth or healing
[113, 117]. This could be an important nding as marked
interspecies dierences with regard to bone composition,
bone density, bone mechanical competence and bone cells
exist and are more pronounced in small rodents while dogs
approximate human bone properties the best [146]. On the
other hand, NSAIDs pharmacokinetics could be signicantly
dierent between species and humans, and in fact studies
analyzing NSAIDs kinetic parameters showsignicant dier-
ences [147, 148]. Despite the above-mentioned limitations of
animal models, further studies will be needed to strengthen
these assumptions.
The in-vitro studies follow exactly the same pathway,
being inconclusive and dicult to interpret. Some studies
present a strong eect of NSAIDs on the potential of
osteoprogenitor cells to proliferate and dierentiate toward
an osteogenic lineage while others refute it (Table 1). A
determining factor could be the use of cells from a variety
of species which include cells from rodents, chicken, horses,
dogs, and pigs as well as human cells [56, 73]. This diversity
of cell sources could be crucial as cells from animals could
react in a dierent fashion to the human cells and in fact
according to our experience, animal MSCs exert dierent
proliferation and dierentiation rates compared to human
cells. In terms of the human cells, some authors have used
osteosarcoma cells which are pathologic cell type character-
ized by aggressive numerous atypical mitoses, and their prop-
erties are far dierent to that of MSCs [59, 62, 69, 71]. Dier-
ences could even exist between the same osteoprogenitor cells
isolated from humans who suer from a fracture compared
to those who do not, as systemic signals are triggered forcing
the cells to proliferate or dierentiate toward a specic way
as a result of the trauma stimulus sustained [149, 150].
8 The Scientic World Journal
Table 4: Studies analyzing the eect of NSAIDs on bone healing in humans.
Study/Year Design NSAID used Conclusions and recommendations
Davis and Ackroyd,
1988 [136]
Prospective double-blinded
study of 100 patients with Colles
fracture
Fluriprophen
(50 mg TDS)
(i) No eect on Colles fracture.
Adolphson et al., 1993
[137]
Randomized double-blinded
study on 42 postmenopausal
women with colles fracture
Piroxicam
(i) No decrease of the rate of fracture healing
(ii) Patients receiving piroxicam had
signicantly less pain
(iii) No dierence in the rate of functional
recovery
Butcher and Marsh,
1996 [138]
Retrospective review of 94
patients with tibial fracture
Not specied
(i) Increase in the length of time to union by of
7.6 weeks (P = 0.0003) (16.7 weeks versus 24.3
weeks).
Wurnig et al., 1999
[139]
80 prospective patients receiving
indomethacin prophylaxis for
THR compared with 82 patients
without.
Indomethacin
(Oral 50 mg BD)
(i) No eect on prosthetic loosening after
cementless hip arthroplasty
Giannoudis et al., 2000
[140]
Retrospective review of 377
patients treated with IM nail
Ibuprophen and
Diclofenac
(i) Increased risk for nonunion in patients
receiving NSAIDs
Bhandari et al., 2003
[141]
Retrospective review of 192 tibial
shaft fractures
Not specied
(i) Relative risk of 2.02 (P = 0.035) for patient
who take NSAIDs
Burd et al., 2003 [142]
Retrospective review of 282 with
acetabular fractures
Indomethacin
(i) Patients receiving indomethacin had
increased risk for developing non-union
Sculean et al., 2003
[143]
Randomized blindied study on
20 patients with deep intrabony
defect
Rofecoxib
(25 mg/day for 14 days)
(i) No eect on the healing of intrabony
periodontal defects
Bhattacharyya et al.,
2005 [144]
Retrospective review of 9995
humeral shaft fractures treated
nonoperatively
Not specied
(i) Exposure to nonselective NSAIDs in the
period 6190 days after a humeral shaft fracture
was associated with nonunion
Meunier et al., 2009
[145]
Randomized study involving 50
patients undergoing total knee
replacement
Celecoxib
(200 mg BD)
(i) No dierences in prosthesis migration, pain
scores, range of motion, and subjective outcome
were found after 2 years
It merits saying that NSAIDs can aect the osteopro-
genitor cells by a pathway far dierent to that of the inhi-
bition of COX-1 and COX-2 enzymes, therefore, dierences
in tissue culture parameters could inuence the nal result.
This so-called non-Cox eect is unfortunately poorly
understood. According to this theory, the NSAIDs properties
related the protective eect against the tumors, cancer
inhibition, and the prevention of metastasis as well as the
prevention of other pathologies like Alzheimers disease or
cataract cannot be explained solely by the inhibition of
prostaglandins, therefore, an alternative unrelated to COX
enzymes inhibition pathway should exist to explain these
results [17]. This theory can be strengthened by the studies of
Chang et al., who presented that the replenishment of PGE-
1, PGE-2, and PGF2a did not reverse this negative eect on
bone cells produced by the studied NSAIDs [64, 68].
This comprehensive review includes the data of 18,
mainly retrospective, clinical studies trying to enlighten this
area of high interest [128145]. Two early studies involved
patients suering from Colles fracture reported no eect of
NSAIDs on union rates [136, 137]. It is of note, however,
that nonunion after Colles fracture is rare and only a few
case reports exist [148]. There are a few case studies that
tried to dene an association between NSAIDs exposure and
nonunion [140142, 144]. Most of them report an increased
incidence of nonunion among the patients that received
NSAIDs. Although this can be true, many covariates that
could inuence this result like smoking, extent of trauma,
comminution, patient demographics, and diabetes were not
isolated. Unavailable bias of variable forms could exist in
a large number of these studies. In a retrospective review
of approximately 10,000 humeral shaft fractures treated
nonoperatively, the authors reported increased incidence of
nonunion among patients receiving NSAIDs [144]. However,
opioids user had an increased risk as well. To our knowledge,
we are not aware of any studies highlighting a similar eect.
So, does this mean that NSAIDs increase the risk for a
nonunion or that patients with unhealed fractures require
signicant amounts of analgesia for prolonged period of
time? Another study showed that NSAIDs users had a
relative risk of 2.02 (P = 0.035) for reoperation compared
to nonusers [141]. Similar to the previous example, patients
with complications requiring a second operation would have
increased analgesic needs. In the same study, the risk for
patients receiving antibiotics was 3.01 (P = 0.002). Although
some antibiotics can interfere with the bone healing process
[16], it looks more plausible that the extent of injury and/or a
contaminated open fracture is the cause for this observation.
The Scientic World Journal 9
These two examples highlight the diculties faced by the
researchers on these observational studies and the potential
presence of bias.
In spinal fusion, the results presented are more uniform.
Some studies suggest that NSAIDs have no eect on union
rates, [130, 132135] while some others showed a dose-
dependant inhibitory eect underlining that patients who
continued to take NSAIDs for more than 3 months post-
operatively showed signicantly lower success rates [128].
Potential presence of bias can exist in these studies as well
and it is unclear whether COX-2 selective NSAIDs do have
an eect as the vast majority of these studies utilize ketorolac
which is a high COX-1 inhibitor. It is worth mentioning
that there are signicant dierences in the nonunion rates
between long bones and spine. The nonunion in spinal
fusion can be as high as 15% [151]. Dierences in structure,
mechanical loading, and function of vertebrae also exist. This
can result in dierent degrees of aection, if any, by NSAIDs.
The eect of NSAIDs on heterotopic ossication (HO)
has being used as an argument against NSAIDs administra-
tion during fracture healing. HO is dened as the process
by which marrow-containing boneis formed in soft tissues
outside the skeleton [152, 153]. NSAIDs have proven to have
a strong eect on this process although the pathophysiology
is not fully understood [15]. A Cochraine meta-analysis of
17 trials involving more than 1900 patients having hip joint
replacemnet suggested that NSAID administration reduces
HO by 59% [152]. Although this is true, HO should not be
confused with bone healing as it is a pathologic condition in
which a fully dierentiated and comitted cell turns into bone.
Contributing factors for the development of HO have been
proposed to be the locally released BMPs, inammation and
PGE-2 production, hypercalcemia, hypoxia, abnormal nerve
activities, immobilization, and disequilibrium of hormones
[153, 154]. HO does not follow the fracture healing cascade
and signicant dierences of the local microenviroment
characteristics exist as, for example, the mechanical loads
that are applied on these tissues are minimal compared to
those between the fragments of a broken bone. It is also
possible that the fully committed cells switch from one type
to another similar to what occurs in the transdierentiation
of MSCs [155].
NSAIDs, due to their ability to inhibit the production
of prostaglandins, alleviate the intrinsic local inamma-
tory response, desensitizing the peripheral pain receptors.
Although they are potent analgesics, some studies showed
that they can inhibit bone healing, while some others dis-
agreed with these ndings. This has triggered a wide range
of responses from the medical community, ranging from
recommendation of cautious use to statements like when
fracture healing or spine fusion is desired, nonsteroidal anti-
inammatory drugs should be avoided [156]. The out-
come is a widespread confusion with some centres having
ignored these recommendations while others rely on narcotic
analgesia for pain control even for nondisplaced fractures.
In the absence of robust scientic evidence concerning
the use of NSAIDs after fracture, a denite statement
regarding their use cannot be made. However, based on the
available literature, some simple carefully derived recom-
mendations can be issued. According to the authors opinion,
there is rather weak evidence to absolute contraindicate the
use of NSAIDs in patients suering from a fracture. NSAIDs
administration should be considered as a risk factor for
delayed fracture healing, at extreme as equal to smoking,
corticosteroids, or diabetes. Clinicians could consider them
as low-risk patients and for ashort period, probably not
exceeding a week after fracture. The need of prospective
well-controlled clinical trials is warranted. Patient selection
and recruitment, the randomization of patients, and ways to
overcome ethical issues are all crucial. In addition, dening
the pathway by which NSAIDs could aect bone cells would
be of paramount importance.
5. Conclusion
There is no robust clinical and/or scientic evidence to dis-
card the use of NSAIDs in patients suering from a fracture,
but equal lack of evidence does not constitute proof of the
absence of an eect. The majority of the available evidence is
based on animal ndings and these results should be inter-
preted with caution due to the dierences in physiological
mechanisms between humans and animals. The need of basic
science research dening the exact mechanism that NSAIDs
could interfere with bone cells and the conduction of well-
randomized prospective clinical trial are warranted. Till then,
clinician should treat NSAIDs as a risk factor for bone heal-
ing impairment and should be avoided in high-risk patients.
Conict of Interests
The authors declare that there is no conict of interest.
Ackowledgment
WORK ATTRIBUTED TO: Academic Unit of Trauma and
Orthopaedic Surgery, Clarendon Wing, Leeds General Inr-
mary, Great George Street, Leeds, LS1 3EX, UK.
References
[1] B. McKibbin, The biology of fracture healing in long bones,
Journal of Bone and Joint Surgery B, vol. 60, no. 2, pp. 150
162, 1978.
[2] B. A. Rahn, Direct and indirect bone healing after operative
fracture treatment, Otolaryngologic Clinics of North America,
vol. 20, no. 3, pp. 425440, 1987.
[3] R. Dimitriou, E. Tsiridis, and P. V. Giannoudis, Current
concepts of molecular aspects of bone healing, Injury, vol.
36, no. 12, pp. 13921404, 2005.
[4] T. A. Einhorn, The cell and molecular biology of fracture
healing, Clinical Orthopaedics and Related Research, no. 355,
pp. S7S21, 1998.
[5] J. E. Puzas, R. J. OKeefe, E. M. Schwarz, and X. Zhang,
Pharmacologic modulators of fracture healing: the role
of cyclooxygenase inhibition, Journal of Musculoskeletal
Neuronal Interactions, vol. 3, no. 4, pp. 308312, 2003.
[6] Z. A. Radi and N. K. Khan, Eects of cyclooxygenase inhi-
bition on bone, tendon, and ligament healing, Inammation
Research, vol. 54, no. 9, pp. 358366, 2005.
10 The Scientic World Journal
[7] N. M. Gajraj, The eect of cyclooxygenase-2 inhibitors on
bone healing, Regional Anesthesia and Pain Medicine, vol. 28,
no. 5, pp. 456465, 2003.
[8] R. L. Cruess and J. Dumont, Fracture healing, Canadian
Journal of Surgery, vol. 18, no. 5, pp. 403413, 1975.
[9] P. Augat, K. Margevicius, J. Simon, S. Wolf, G. Suger, and
L. Claes, Local tissue properties in bone healing: inuence
of size and stability of the osteotomy gap, Journal of
Orthopaedic Research, vol. 16, no. 4, pp. 475481, 1998.
[10] J. V. Vigorita, Orthopaedic Pathology, Lippincott & Wilkins,
1st edition, 1999.
[11] F. B. Castell a, F. B. Garcia, E. M. Berry, E. B. Perello, E.
Sanchez-Alepuz, and R. Gabarda, Nonunion of the humeral
shaft: long lateral buttery fracturea nonunion predictive
pattern? Clinical Orthopaedics and Related Research, no. 424,
pp. 227230, 2004.
[12] L. E. Claes, H. J. Wilke, P. Augat, S. Rubenacker, and
K. J. Margevicius, Eect of dynamization on gap healing
of diaphyseal fractures under external xation, Clinical
Biomechanics, vol. 10, no. 5, pp. 227234, 1995.
[13] T. Yamaji, K. Ando, S. Wolf, P. Augat, and L. Claes, The
eect of micromovement on callus formation, Journal of
Orthopaedic Science, vol. 6, no. 6, pp. 571575, 2001.
[14] S. M. Day and D. H. DeHeer, Reversal of the detrimental
eects of chronic protein malnutrition on long bone fracture
healing, Journal of Orthopaedic Trauma, vol. 15, no. 1, pp.
4753, 2001.
[15] I. Pountos, T. Georgouli, T. J. Blokhuis, H. C. Pape, and P.
V. Giannoudis, Pharmacological agents and impairment of
fracture healing: what is the evidence? Injury, vol. 39, no. 4,
pp. 384394, 2008.
[16] I. Pountos, T. Georgouli, H. Bird, G. Kontakis, and P.
V. Giannoudis, The eect of antibiotics onbone healing:
current evidence, Expert Opinion on Drug Safety, vol. 10, no.
6, pp. 935945, 2011.
[17] I. Pountos, T. Georgouli, H. Bird, and P. V. Giannoudis,
Nonsteroidal anti-inammatory drugs: prostaglandins,
indications, and side eects, International Journal of Inter-
feron, Cytokine and Mediator Research, vol. 3, no. 1, pp. 19
27, 2011.
[18] B. F. Mcadam, F. Catella-Lawson, I. A. Mardini, S. Kapoor,
J. A. Lawson, and G. A. Fitzgerald, Systemic biosynthesis
of prostacyclin by cyclooxygenase (COX)-2: the human
pharmacology of a selective inhibitor of COX-2, Proceedings
of the National Academy of Sciences of the United States of
America, vol. 96, no. 1, pp. 272277, 1999.
[19] T. D. Warner and J. A. Mitchell, Cyclooxygenases: new
forms, new inhibitors, and lessons from the clinic, FASEB
Journal, vol. 18, no. 7, pp. 790804, 2004.
[20] A. Whelton, Nephrotoxicity of nonsteroidal anti-
inammatory drugs: physiologic foundations and clinical
implications, American Journal of Medicine, vol. 106,
supplement 2, no. 5, pp. 13S24S, 1999.
[21] M. S. Trautman, S. S. Edwin, D. Collmer, D. J. Dudley, D.
Simmons, and M. D. Mitchell, Prostaglandin H synthase-
2 in human gestational tissues: regulation in amnion,
Placenta, vol. 17, no. 4, pp. 239245, 1996.
[22] J. R. Vane and R. M. Botting, Mechanism of action of
nonsteroidal anti-inammatory drugs, American Journal of
Medicine, vol. 104, supplement 1, no. 3, pp. 2S8S, 1998.
[23] J. van Ryn and M. Pairet, Clinical experience with
cyclooxygenase-2 inhibitors, Inammation Research, vol. 48,
no. 5, pp. 247254, 1999.
[24] G. A. Green, Understanding NSAIDs: from aspirin to COX-
2, Clinical Cornerstone, vol. 3, no. 5, pp. 5058, 2001.
[25] T. D. Warner, F. Giuliano, I. Vojnovic, A. Bukasa, J. A.
Mitchell, and J. R. Vane, Nonsteroid drug selectives for
cyclo-oxygenase-1 rather than cyclo-oxygenase-2 are asso-
ciated with human gastrointestinal toxicity: a full in vitro
analysis, Proceedings of the National Academy of Sciences of
the United States of America, vol. 96, no. 13, pp. 75637568,
1999.
[26] P. Brooks, P. Emery, J. F. Evans et al., Interpreting the clinical
signicance of the dierential inhibition of cyclooxygenase-1
and cyclooxygenase-2, Rheumatology, vol. 38, no. 8, pp. 779
788, 1999.
[27] R. N. Fracon, J. M. Te olo, R. B. Satin, and T. Lamano,
Prostaglandins and bone: potential risks and benets related
to the use of nonsteroidal anti-inammatory drugs in clinical
dentistry, Journal of Oral Science, vol. 50, no. 3, pp. 247252,
2008.
[28] R. E. Ringel, J. I. Brenner, P. J. Haney, J. E. Burns, A.
L. Moulton, and M. A. Berman, Prostaglandin-induced
periostitis: a complication of long-term PGE1 infusion in an
infant with congenital heart disease, Radiology, vol. 142, no.
3, pp. 657658, 1982.
[29] I. Suponitzky and M. Weinreb, Dierential eects of
systemic prostaglandin E2 on bone mass in rat long bones
and calvariae, Journal of Endocrinology, vol. 156, no. 1, pp.
5157, 1998.
[30] W. S. Jee, K. Ueno, Y. P. Deng, and D. M. Woodbury, The
eects of prostaglandin E2 in growing rats: increased meta-
physeal hard tissue and cortico-endosteal bone formation,
Calcied Tissue International, vol. 37, no. 2, pp. 148157,
1985.
[31] C. H. Lin, W. S. Jee, Y. F. Ma, and R. B. Setterberg, Early
eects of prostaglandin E2 on bone formation and resorption
in dierent bone sites of rats, Bone, vol. 17, supplement 4,
pp. 255S259S, 1995.
[32] J. R. Nefussi and R. Baron, PGE2 stimulates both resorption
and formation of bone in vitro: dierential responses of
the periosteum and the endosteum in fetal rat long bone
cultures, Anatomical Record, vol. 211, no. 1, pp. 916, 1985.
[33] H. Kawaguchi, C. C. Pilbeam, J. R. Harrison, and L. G.
Raisz, The role of prostaglandins in the regulation of bone
metabolism, Clinical Orthopaedics and Related Research, no.
313, pp. 3646, 1995.
[34] S. Dekel, G. Lenthall, and M. J. Francis, Release of
prostaglandins from bone and muscle after tibial fracture.
An experimental study in rabbits, Journal of Bone and Joint
Surgery B, vol. 63, no. 2, pp. 185189, 1981.
[35] C. C. Pilbeam, P. M. Fall, C. B. Alander, and L. G. Raisz,
Dierential eects of nonsteroidal anti-inammatory drugs
on constitutive and inducible prostaglandin G/H synthase in
cultured bone cells, Journal of Bone and Mineral Research,
vol. 12, no. 8, pp. 11981203, 1997.
[36] X. Zhang, E. M. Schwarz, D. A. Young, J. E. Puzas, R.
N. Rosier, and R. J. OKeefe, Cyclooxygenase-2 regulates
mesenchymal cell dierentiation into the osteoblast lineage
and is critically involved in bone repair, Journal of Clinical
Investigation, vol. 109, no. 11, pp. 14051415, 2002.
[37] L. C. Gerstenfeld and T. A. Einhorn, COX inhibitors and
their eects on bone healing, Expert Opinion on Drug Safety,
vol. 3, no. 2, pp. 131136, 2004.
[38] T. Arikawa, K. Omura, and I. Morita, Regulation of
bone morphogenetic protein-2 expression by endogenous
The Scientic World Journal 11
prostaglandin E2 in human mesenchymal stem cells, Journal
of Cellular Physiology, vol. 200, no. 3, pp. 400406, 2004.
[39] S. Jurado, N. Garcia-Giralt, A. Dez-P erez et al., Eect
of IL-1, PGE2, and TGF-1 on the expression of OPG
and RANKL in normal and osteoporotic primary human
osteoblasts, Journal of Cellular Biochemistry, vol. 110, no. 2,
pp. 304310, 2010.
[40] V. M. Paralkar, W. A. Grasser, A. L. Mansolf et al., Regulation
of BMP-7 expression by retinoic acid and prostaglandin
E(2), Journal of Cellular Physiology, vol. 190, no. 2, pp. 207
217, 2002.
[41] M. Coetzee, M. Haag, A. M. Joubert, and M. C. Kruger,
Eects of arachidonic acid, docosahexaenoic acid and
prostaglandin E(2) on cell proliferation and morphology of
MG-63 and MC3T3-E1 osteoblast-like cells, Prostaglandins
Leukotrienes and Essential Fatty Acids, vol. 76, no. 1, pp. 35
45, 2007.
[42] M. Machwate, S. B. Rodan, G. A. Rodan, and S. I. Harada,
Sphingosine kinase mediates cyclic AMP suppression of
apoptosis in rat periosteal cells, Molecular Pharmacology,
vol. 54, no. 1, pp. 7077, 1998.
[43] M. Machwate, S. Harada, C. T. Leu et al., Prostaglandin
receptor EP(4) mediates the bone anabolic eects of PGE(2),
Molecular Pharmacology, vol. 60, no. 1, pp. 3641, 2001.
[44] M. Zhang, H. C. Ho, T. J. Sheu et al., EP1(-/-) mice have
enhanced osteoblast dierentiation and accelerated fracture
repair, Journal of Bone and Mineral Research, vol. 26, no. 4,
pp. 792802, 2011.
[45] A. Beck, K. Salem, G. Krischak, L. Kinzl, M. Bischo,
and A. Schmelz, Nonsteroidal anti-inammatory drugs
(NSAIDs) in the perioperative phase in traumatology and
orthopedics. Eects on bone healing, Operative Orthopadie
und Traumatologie, vol. 17, no. 6, pp. 569578, 2005.
[46] A. Holdgate and T. Pollock, Systematic review of the
relative ecacy of non-steroidal anti-inammatory drugs
and opioids in the treatment of acute renal colic, British
Medical Journal, vol. 328, no. 7453, pp. 14011404, 2004.
[47] D. G. Monz on, J. Vazquez, J. R. Jauregui, and K. V.
Iserson, Pain treatment in post-traumatic hip fracture in the
elderly: regional block vs. systemic non-steroidal analgesics,
International Journal of Emergency Medicine, vol. 3, no. 4, pp.
321325, 2010.
[48] G. Stacher, H. Steinringer, and S. Schneider, Experimental
pain induced by electrical and thermal stimulation of the
skin in healthy man: sensitivity to 75 and 150 mg diclofenac
sodium in comparison with 60 mg codeine and placebo,
British Journal of Clinical Pharmacology, vol. 21, no. 1, pp.
3543, 1986.
[49] T. Kantor, M. B. Cavaliere, M. Hopper, and S. Roepke, A
double-blind parallel comparison of ketoprofen, codeine,
and placebo in patients with moderate to severe postpartum
pain, Journal of Clinical Pharmacology, vol. 24, no. 5-6, pp.
228234, 1984.
[50] M. H. Ebell, NSAIDs vs. opiates for pain in acute renal
colic, American Family Physician, vol. 70, no. 9, p. 1682,
2004.
[51] F. Camu, T. Beecher, D. P. Recker, and K. M. Verburg, Valde-
coxib, a COX-2-specic inhibitor, is an ecacious, opioid-
sparing analgesic in patients undergoing hip arthroplasty,
American Journal of Therapeutics, vol. 9, no. 1, pp. 4351,
2002.
[52] D. M. Turner, J. S. Warson, T. C. Wirt, R. D. Scalley, R. S.
Cochran, and K. J. Miller, The use of ketorolac in lumbar
spine surgery: a cost-benet analysis, Journal of Spinal
Disorders, vol. 8, no. 3, pp. 206212, 1995.
[53] W. M. ONeill, G. W. Hanks, P. Simpson, M. T. Fallon, E.
Jenkins, and K. Wesnes, The cognitive and psychomotor
eects of morphine in healthy subjects: a randomized
controlled trial of repeated (four) oral doses of dextro-
propoxyphene, morphine, lorazepamand placebo, Pain, vol.
85, no. 1-2, pp. 209215, 2000.
[54] E. Marret, N. Elia, J. B. Dahl et al., Susceptibility to
fraud in systematic reviews: lessons from the reuben case,
Anesthesiology, vol. 111, no. 6, pp. 12791289, 2009.
[55] H. L. Rittner, P. Kranke, M. Sch afer, N. Roewer, and A. Brack,
What can we learn from the Scott Reuben case? scientic
misconduct in anaesthesiology, Anaesthesist, vol. 58, no. 12,
pp. 11991209, 2009.
[56] H. T ornkvist, B. R. Danielsson, L. Dencker, and O. S. Nilsson,
Lack of eect of indomethacin on mesenchymal limb-bud
cells in vitro, Acta Orthopaedica Scandinavica, vol. 55, no. 3,
pp. 378380, 1984.
[57] M. L. Ho, J. K. Chang, L. Y. Chuang, H. K. Hsu, and G.
J. Wang, Eects of nonsteroidal anti-inammatory drugs
and prostaglandins on osteoblastic functions, Biochemical
Pharmacology, vol. 58, no. 6, pp. 983990, 1999.
[58] C. E. Evans and C. Butcher, The inuence on human
osteoblasts in vitro on non-steroidal anti-inammatory
drugs which act on dierent cyclooxygenase enzymes,
Journal of Bone and Joint Surgery B, vol. 86, no. 3, pp. 444
449, 2004.
[59] J. L. Wang, K. L. Lin, J. S. Chen et al., Eect of celecoxib on
Ca
2+
movement and cell proliferation in human osteoblasts,
Biochemical Pharmacology, vol. 67, no. 6, pp. 11231130,
2004.
[60] J. K. Chang, G. J. Wang, S. T. Tsai, and M. L. Ho,
Nonsteroidal anti-inammatory drug eects on osteoblastic
cell cycle, cytotoxicity, and cell death, Connective Tissue
Research, vol. 46, no. 4-5, pp. 200210, 2005.
[61] Y. Wang, X. Chen, W. Zhu, H. Zhang, S. Hu, and X. Cong,
Growth inhibition of mesenchymal stem cells by aspirin:
involvement of the wnt/-catenin signal pathway, Clinical
and Experimental Pharmacology and Physiology, vol. 33, no.
8, pp. 696701, 2006.
[62] M. Wiontzek, G. Matziolis, S. Schuchmann et al., Eects of
dexamethasone and celecoxib on calcium homeostasis and
expression of cyclooxygenase-2 mRNA in MG-63 human
osteosarcoma cells, Clinical and Experimental Rheumatology,
vol. 24, no. 4, pp. 366372, 2006.
[63] B. Wolfesberger, C. Hoelzl, I. Walter et al., In vitro eects
of meloxicam with or without doxorubicin on canine
osteosarcoma cells, Journal of Veterinary Pharmacology and
Therapeutics, vol. 29, no. 1, pp. 1523, 2006.
[64] J. K. Chang, C. J. Li, S. C. Wu et al., Eects of anti-
inammatory drugs on proliferation, cytotoxicity and osteo-
genesis in bone marrow mesenchymal stem cells, Biochemi-
cal Pharmacology, vol. 74, no. 9, pp. 13711382, 2007.
[65] M. Kellinsalmi, V. Parikka, J. Risteli et al., Inhibition of
cyclooxygenase-2 down-regulates osteoclast and osteoblast
dierentiation and favours adipocyte formation in vitro,
European Journal of Pharmacology, vol. 572, no. 2-3, pp. 102
110, 2007.
[66] P. Arpornmaeklong, B. Akarawatcharangura, P. Pripat-
nanont, S. H. Lee, and C. Ohkubo, Factors inuencing
eects of specic COX-2 inhibitor NSAIDs on growth and
dierentiation of mouse osteoblasts on titanium surfaces,
12 The Scientic World Journal
International Journal of Oral and Maxillofacial Implants, vol.
23, no. 6, pp. 10711081, 2008.
[67] H. Abukawa, M. Phelps, P. Jackson et al., Eect of ibupro-
fen on osteoblast dierentiation of porcine bone marrow-
derived progenitor cells, Journal of Oral and Maxillofacial
Surgery, vol. 67, no. 11, pp. 24122417, 2009.
[68] J. K. Chang, C. J. Li, H. J. Liao, C. K. Wang, G. J. Wang, and
M. L. Ho, Anti-inammatory drugs suppress proliferation
and induce apoptosis through altering expressions of cell
cycle regulators and pro-apoptotic factors in cultured human
osteoblasts, Toxicology, vol. 258, no. 2-3, pp. 148156, 2009.
[69] P. Kolar, S. Lach, T. Gaber et al., Eects of celecoxib on the
expression of osteoprotegerin, energy metabolism and cell
viability in cultured human osteoblastic cells, Clinical and
Experimental Rheumatology, vol. 27, no. 1, pp. 99107, 2009.
[70] D. S. Yoon, J. H. Yoo, Y. H. Kim, S. Paik, C. D. Han,
and J. W. Lee, The eects of COX-2 inhibitor during
osteogenic dierentiation of bone marrow-derived human
mesenchymal stem cells, Stem Cells and Development, vol.
19, no. 10, pp. 15231533, 2010.
[71] L. D. A. R. Guez, O. G. A. M. Nez, M. Arroyo-Morales,
L. R. Guez-P erez, B. Rubio-Ruiz, and C. Ruiz, Eects of
indomethacin, nimesulide,and diclofenac on human MG-63
osteosarcoma cell line, Biological Research For Nursing. In
press.
[72] M. M uller, O. Raabe, K. Addicks, S. Wenisch, and S.
Arnhold, Eects of non-steroidal anti-inammatory drugs
on proliferation, dierentiation and migration in equine
mesenchymal stem cells, Cell Biology International, vol. 35,
no. 3, pp. 235248, 2011.
[73] I. Pountos, P. V. Giannoudis, E. Jones et al., NSAIDS
inhibit in vitro MSC chondrogenesis but not osteogenesis:
implications for mechanism of bone formation inhibition in
man, Journal of Cellular and Molecular Medicine, vol. 15, no.
3, pp. 525534, 2011.
[74] L. C. Gerstenfeld, M. Thiede, K. Siebert et al., Dier-
ential inhibition of fracture healing by non-selective and
cyclooxygenase-2 selective non-steroidal anti-inammatory
drugs, Journal of Orthopaedic Research, vol. 21, no. 4, pp.
670675, 2003.
[75] G. J. Martin Jr., S. D. Boden, and L. Titus, Recom-
binant human bone morphogenetic protein-2 overcomes
the inhibitory eect of ketorolac, a nonsteroidal anti-
inammatory drug (NSAID), on posterolateral lumbar inter-
transverse process spine fusion, Spine, vol. 24, no. 21, pp.
21882193, 1999.
[76] M. L. Ho, J. K. Chang, and G. J. Wang, Eects of ketorolac on
bone repair: a radiographic study in modeled demineralized
bone matrix grafted rabbits, Pharmacology, vol. 57, no. 3, pp.
148159, 1998.
[77] M. V. Martins, M. A. da Silva, E. Medici Filho, L. C. de
Moraes, J. C. D. M. Castilho, and R. F. da Rocha, Evaluation
of digital optical density of bone repair in rats medicated with
ketoprofen, Brazilian Dental Journal, vol. 16, no. 3, pp. 207
212, 2005.
[78] O. Reikeraas and L. Engebretsen, Eects of ketoralac
tromethamine and indomethacin on primary and secondary
bone healing. An experimental study in rats, Archives of
Orthopaedic and Trauma Surgery, vol. 118, no. 1-2, pp. 50
52, 1998.
[79] K. D. Riew, J. Long, J. Rhee et al., Time-dependent
inhibitory eects of indomethacin on spinal fusion, Journal
of Bone and Joint Surgery A, vol. 85, no. 4, pp. 632634, 2003.
[80] K. M. Brown, M. M. Saunders, T. Kirsch, H. J. Donahue, and
J. S. Reid, Eect of COX-2-specic inhibition on fracture-
healing in the rat femur, Journal of Bone and Joint Surgery A,
vol. 86, no. 1, pp. 116123, 2004.
[81] P. E. Persson, G. Sisask, and O. Nilsson, Indomethacin
inhibits bone formation in inductive allografts but not in
autografts: studies in rat, Acta Orthopaedica, vol. 76, no. 4,
pp. 465469, 2005.
[82] E. Sudmann and G. Bang, Indomethacin-induced inhibi-
tion of haversian remodelling in rabbits, Acta Orthopaedica
Scandinavica, vol. 50, no. 6, pp. 621627, 1979.
[83] J. R. Dimar II, W. A. Ante, Y. P. Zhang, and S. D. Glassman,
The eects of nonsteroidal anti-inammatory drugs on
posterior spinal fusions in the rat, Spine, vol. 21, no. 16, pp.
18701876, 1996.
[84] H. L. Allen, A. Wase, and W. T. Bear, Indomethacin and
aspirin: eect of nonsteroidal anti-inammatory agents on
the rate of fracture repair in the rat, Acta Orthopaedica
Scandinavica, vol. 51, no. 4, pp. 595600, 1980.
[85] T. Karachalios, L. Boursinos, L. Poultsides, L. Khaldi, and K.
N. Malizos, The eects of the short-term administration of
low therapeutic doses of anti-COX-2 agents on the healing of
fractures: an experimental study in rabbits, Journal of Bone
and Joint Surgery B, vol. 89, no. 9, pp. 12531260, 2007.
[86] J. Keller, C. B unger, T. T. Andreassen, B. Bak, and U.
Lucht, Bone repair inhibited by indomethacin. Eects on
bone metabolism and strength of rabbit osteotomies, Acta
Orthopaedica Scandinavica, vol. 58, no. 4, pp. 379383, 1987.
[87] J. Keller, I. Bayer-Kristensen, B. Bak et al., Indomethacin and
bone remodeling. Eect on cortical bone after osteotomy in
rabbits, Acta Orthopaedica Scandinavica, vol. 60, no. 1, pp.
119121, 1989.
[88] H. T ornkvist, T. S. Lindholm, P. Netz, L. Str omberg, and T. C.
Lindholm, Eect of ibuprofen and indomethacin on bone
metabolism reected in bone strength, Clinical Orthopaedics
and Related Research, vol. 187, pp. 255259, 1984.
[89] H. E. Hgevold, B. Grgaard, and O. Reiker as, Eects of
short-term treatment with corticosteroids and indomethacin
on bone healing. A mechanical study of osteotomies in rats,
Acta Orthopaedica Scandinavica, vol. 63, no. 6, pp. 607611,
1992.
[90] L. B. Engesaeter, B. Sudmann, and E. Sudmann, Frac-
ture healing in rats inhibited by locally administered
indomethacin, Acta Orthopaedica Scandinavica, vol. 63, no.
3, pp. 330333, 1992.
[91] S. Sato, T. Kim, T. Arai, S. Maruyama, M. Tajima, and N.
Utsumi, Comparison between the eects of dexamethasone
and indomethacin on bone wound healing, Japanese Journal
of Pharmacology, vol. 42, no. 1, pp. 7178, 1986.
[92] S. Goodman, T. Ma, M. Trindade et al., COX-2 selec-
tive NSAID decreases bone ingrowth in vivo, Journal of
Orthopaedic Research, vol. 20, no. 6, pp. 11641169, 2002.
[93] M. A. Kaygusuz, C. C. Turan, N. E. Aydin et al., The eects
of G-CSF and naproxen sodium on the serum TGF-1 level
and fracture healing in rat tibias, Life Sciences, vol. 80, no. 1,
pp. 6773, 2006.
[94] S. M. Leonelli, B. A. Goldberg, J. Safanda, M. R. Bagwe, S.
Sethuratnam, and S. J. King, Eects of a cyclooxygenase-2
inhibitor (rofecoxib) on bone healing, American Journal of
Orthopedics, vol. 35, no. 2, pp. 7984, 2006.
[95] J. P. OConnor, J. T. Capo, V. Tan et al., A comparison of the
eects of ibuprofen and rofecoxib on rabbit bula osteotomy
healing, Acta Orthopaedica, vol. 80, no. 5, pp. 597605, 2009.
The Scientic World Journal 13
[96] G. Obeid, X. Zhang, X. Wang, and M. Jecoat, Eect
of ibuprofen on the healing and remodeling of bone and
articular cartilage in the rabbit temporomandibular joint,
Journal of Oral and Maxillofacial Surgery, vol. 50, no. 8, pp.
843859, 1992.
[97] A. Beck, G. Krischak, T. Sorg et al., Inuence of diclofenac
(group of nonsteroidal anti-inammatory drugs) on fracture
healing, Archives of Orthopaedic and Trauma Surgery, vol.
123, no. 7, pp. 327332, 2003.
[98] G. D. Krischak, P. Augat, T. Sorg et al., Eects of diclofenac
on periosteal callus maturation in osteotomy healing in an
animal model, Archives of Orthopaedic and Trauma Surgery,
vol. 127, no. 1, pp. 39, 2007.
[99] C. Sen, M. Erdem, T. Gunes, D. Koseoglu, and N. O.
Filiz, Eects of diclofenac and tenoxicam on distraction
osteogenesis, Archives of Orthopaedic and Trauma Surgery,
vol. 127, no. 3, pp. 153159, 2007.
[100] V. Giordano, M. Giordano, I. G. Knackfuss, M. I. Apfel, and
R. D. Gomes, Eect of tenoxicam on fracture healing in rat
tibiae, Injury, vol. 34, no. 2, pp. 8594, 2003.
[101] A. M. Simon and J. P. OConnor, Dose and time-dependent
eects of cyclooxygenase-2 inhibition on fracture-healing,
Journal of Bone and Joint Surgery A, vol. 89, no. 3, pp. 500
511, 2007.
[102] M. Bergenstock, W. Min, A. M. Simon, C. Sabatino, and
J. P. OConnor, A comparison between the eects of
acetaminophen and celecoxib on bone fracture healing in
rats, Journal of Orthopaedic Trauma, vol. 19, no. 10, pp. 717
723, 2005.
[103] F. V. Ribeiro, J. B. C esar-Neto, F. H. Nociti Jr et al., Selective
cyclooxygenase-2 inhibitor may impair bone healing around
titanium implants in rats, Journal of Periodontology, vol. 77,
no. 10, pp. 17311735, 2006.
[104] K. Endo, K. Sairyo, S. Komatsubara et al., Cyclooxygenase-2
inhibitor inhibits the fracture healing, Journal of Physiolog-
ical Anthropology and Applied Human Science, vol. 21, no. 5,
pp. 235238, 2002.
[105] K. Endo, K. Sairyo, S. Komatsubara et al., Cyclooxygenase-2
inhibitor delays fracture healing in rats, Acta Orthopaedica,
vol. 76, no. 4, pp. 470474, 2005.
[106] S. E. Utv ag, O. M. Fuskev ag, H. Shegar, and O. Reiker as,
Short-term treatment with COX-2 inhibitors does not
impair fracture healing, Journal of Investigative Surgery, vol.
23, no. 5, pp. 257261, 2010.
[107] L. C. Gerstenfeld, M. Al-Ghawas, Y. M. Alkhiary et al.,
Selective and nonselective cyclooxygenase-2 inhibitors and
experimental frature-healing: reversibility of eects after
short-term treatment, Journal of Bone and Joint Surgery A,
vol. 89, no. 1, pp. 114125, 2007.
[108] S. B. Goodman, T. Ma, L. Mitsunaga, K. Miyanishi, M. C.
Genovese, and R. L. Smith, Temporal eects of a COX-2-
selective NSAID on bone ingrowth, Journal of Biomedical
Materials Research A, vol. 72, no. 3, pp. 279287, 2005.
[109] M. Murnaghan, G. Li, and D. R. Marsh, Nonsteroidal anti-
inammatory drug-induced fracture nonunion: an inhibi-
tion of angiogenesis? Journal of Bone and Joint Surgery A,
vol. 88, supplement 3, pp. S140S147, 2006.
[110] R. N. Fracon, J. M. Te olo, L. C. Moris, and T. Lamano,
Treatment with paracetamol, ketorolac or etoricoxib did not
hinder alveolar bone healing: a histometric study in rats,
Journal of Applied Oral Science, vol. 18, no. 6, pp. 630634,
2010.
[111] B. H. Mullis, S. T. Copland, P. S. Weinhold, T. Miclau, G.
E. Lester, and G. D. Bos, Eect of COX-2 inhibitors and
non-steroidal anti-inammatory drugs on a mouse fracture
model, Injury, vol. 37, no. 9, pp. 827837, 2006.
[112] J. Urrutia, R. Mardones, and F. Quezada, The eect of
ketoprophen on lumbar spinal fusion healing in a rabbit
model. Laboratory investigation, Journal of Neurosurgery:
Spine, vol. 7, no. 6, pp. 631636, 2007.
[113] H. J. van der Heide, G. Hannink, P. Buma, and B. W. Schreurs,
No eect of ketoprofen and meloxicam on bone graft
ingrowth: a bone chamber study in goats, Acta Orthopaedica,
vol. 79, no. 4, pp. 548554, 2008.
[114] M. W. Elves, I. Bayley, and P. J. Roylance, The eect of
Indomethacin upon experimental fractures in the rat, Acta
Orthopaedica Scandinavica, vol. 53, no. 1, pp. 3541, 1982.
[115] J. Keller, P. Kjaersgaard-Andersen, I. Bayer-Kristensen, and
F. Melsen, Indomethacin and bone trauma. Eects on
remodelling of rabbit bone, Acta Orthopaedica Scandinavica,
vol. 61, no. 1, pp. 6669, 1990.
[116] I. Boiskin, S. Epstein, F. Ismail, M. D. Fallon, and W. Levy,
Long term administration of prostaglandin inhibitors in
vivo fail to inuence cartilage and bone mineral metabolism
in the rat, Bone and Mineral, vol. 4, no. 1, pp. 2736, 1988.
[117] S. W. Mbugua, L. A. Skoglund, and P. Lkken, Eects
of phenylbutazone and indomethacin on the post-operative
course following experimental orthopaedic surgery in dogs,
Acta veterinaria Scandinavica, vol. 30, no. 1, pp. 2735, 1989.
[118] E. Sudmann, T. Tveita, and J. Hald Jr. Jr, Lack of eect of
indomethacin on ordered growth of the femur in rats, Acta
Orthopaedica Scandinavica, vol. 53, no. 1, pp. 4349, 1982.
[119] J. Long, S. Lewis, T. Kuklo, Y. Zhu, and K. D. Riew, The eect
of cyclooxygenase-2 inhibitors on spinal fusion, Journal of
Bone and Joint Surgery, vol. 84, no. 10, pp. 17631768, 2002.
[120] B. Mullis, S. Copeland, P. Weinhold et al., Eect of COX-
2 inhibitors and NSAIDs on fracture healing in a mouse
model, Transactions of the Annual MeetingOrthopaedic
Research Society, vol. 27, p. 712, 2002.
[121] H. T ornkvist and T. S. Lindholm, Eect of ibuprofen on
mass and composition of fracture callus and bone. An
experimental study on adult rat, Scandinavian Journal of
Rheumatology, vol. 9, no. 3, pp. 167171, 1980.
[122] M. H. Huo, N. W. Troiano, R. R. Pelker, C. M. Gundberg, and
G. E. Friedlaender, The inuence of Ibuprofen on fracture
repair: biomechanical, biochemical, histologic, and histo-
morphometric parameters in rats, Journal of Orthopaedic
Research, vol. 9, no. 3, pp. 383390, 1991.
[123] B. C. Tiseo, G. N. Namur, E. J. de Paula, R. M. Junior,
and C. R. de Oliveira, Experimental study of the action of
COX-2 selective nonsteroidal anti-inammatory drugs and
traditional anti-inammatory drugs in bone regeneration,
Clinics, vol. 61, no. 3, pp. 223230, 2006.
[124] J. M. Te olo, G. S. Giovanini, R. N. Fracon, and T. Lamano,
Histometric study of alveolar bone healing in rats treated
with the nonsteroidal anti-inammatory drug nimesulide,
Implant Dentistry, vol. 20, no. 2, pp. e7e13, 2011.
[125] D. J. Hak, K. S. Schulz, B. Khoie, and S. J. Hazelwood, The
eect of Cox-2 specic inhibition on direct fracture healing
in the rabbit tibia, Journal of Orthopaedic Science, vol. 16, no.
1, pp. 9398, 2011.
[126] H. Nyangoga, E. Aguado, E. Goyenvalle, M. F. Basl e, and
D. Chappard, A non-steroidal anti-inammatory drug
(ketoprofen) does not delay -TCP bone graft healing, Acta
Biomaterialia, vol. 6, no. 8, pp. 33103317, 2010.
[127] R. D. Altman, L. L. Latta, R. Keer, K. Renfree, F. J. Hornicek,
and K. Banovac, Eect of nonsteroidal antiinammatory
14 The Scientic World Journal
drugs on fracture healing: a laboratory study in rats, Journal
of Orthopaedic Trauma, vol. 9, no. 5, pp. 392400, 1995.
[128] M. Deguchi, A. J. Rapo, and T. A. Zdeblick, Posterolateral
fusion for isthmic spondylolisthesis in adults: analysis of
fusion rate and clinical results, Journal of Spinal Disorders,
vol. 11, no. 6, pp. 459464, 1998.
[129] S. D. Glassman, S. M. Rose, J. R. Dimar, R. M. Puno, M.
J. Campbell, and J. R. Johnson, The eect of postopera-
tive nonsteroidal anti-inammatory drug administration on
spinal fusion, Spine, vol. 23, no. 7, pp. 834838, 1998.
[130] M. G. Vitale, J. C. Choe, M. W. Hwang et al., Use of ketorolac
tromethamine in children undergoing scoliosis surgery: an
analysis of complications, Spine Journal, vol. 3, no. 1, pp. 55
62, 2003.
[131] S. Y. Park, S. H. Moon, M. S. Park, K. S. Oh, and H. M.
Lee, The eects of ketorolac injected via patient controlled
analgesia postoperatively on spinal fusion, Yonsei Medical
Journal, vol. 46, no. 2, pp. 245251, 2005.
[132] B. B. Pradhan, R. L. Tatsumi, J. Gallina, C. A. Kuhns, J.
C. Wang, and E. G. Dawson, Ketorolac and spinal fusion:
does the perioperative use of ketorolac really inhibit spinal
fusion? Spine, vol. 33, no. 19, pp. 20792082, 2008.
[133] D. J. Sucato, J. F. Lovejoy, S. Agrawal, E. Elerson, T. Nelson,
and A. McClung, Postoperative ketorolac does not predis-
pose to pseudoarthrosis following posterior spinal fusion and
instrumentation for adolescent idiopathic scoliosis, Spine,
vol. 33, no. 10, pp. 11191124, 2008.
[134] J. M. Lumawig, A. Yamazaki, and K. Watanabe, Dose-
dependent inhibition of diclofenac sodium on posterior
lumbar interbody fusion rates, Spine Journal, vol. 9, no. 5,
pp. 343349, 2009.
[135] P. L. Horn, S. Wrona, A. C. Beebe, and J. E. Klamar, A
retrospective quality improvement study of ketorolac use
following spinal fusion in pediatric patients, Orthopaedic
Nursing, vol. 29, no. 5, pp. 342343, 2010.
[136] T. R. Davis and C. E. Ackroyd, Non-steroidal anti-
inammatory agents in the management of Colles fractures,
British Journal of Clinical Practice, vol. 42, no. 5, pp. 184189,
1988.
[137] P. Adolphson, H. Abbaszadegan, U. Jonsson, N. Dalen,
H. E. Sjoberg, and S. Kalen, No eects of piroxicam on
osteopenia and recovery after Colles fracture. A randomized,
double-blind, placebo-controlled, prospective trial, Archives
of Orthopaedic and Trauma Surgery, vol. 112, no. 3, pp. 127
130, 1993.
[138] C. K. Butcher and D. R. Marsh, Non-steroidal anti-
inammatory drugs delay tibial fracture union, Injury, vol.
27, no. 5, p. 375, 1996.
[139] C. Wurnig, E. Schwameis, P. Bitzan, and F. Kainberger,
Six-year results of a cementless stem with prophylaxis
against heterotopic bone, Clinical Orthopaedics and Related
Research, no. 361, pp. 150158, 1999.
[140] P. V. Giannoudis, D. A. MacDonald, S. J. Matthews, R.
M. Smith, A. J. Furlong, and P. de Boer, Nonunion of
the femoral diaphysis. The inuence of reaming and non-
steroidal anti-inammatory drugs, Journal of Bone and Joint
Surgery, vol. 82, no. 5, pp. 655658, 2000.
[141] M. Bhandari, P. Tornetta III, S. Sprague et al., Predictors of
reoperation following operative management of fractures of
the tibial shaft, Journal of Orthopaedic Trauma, vol. 17, no.
5, pp. 353361, 2003.
[142] T. A. Burd, M. S. Hughes, and J. O. Anglen, Heterotopic
ossication prophylaxis with indomethacin increases the risk
of long-bone nonunion, Journal of Bone and Joint Surgery,
vol. 85, no. 5, pp. 700705, 2003.
[143] A. Sculean, M. Berakdar, N. Donos, T. M. Auschill, and
N. B. Arweiler, The eect of postsurgical administration
of a selective cyclo-oxygenase-2 inhibitor on the healing of
intrabony defects following treatment with enamel matrix
proteins, Clinical Oral Investigations, vol. 7, no. 2, pp. 108
112, 2003.
[144] T. Bhattacharyya, R. Levin, M. S. Vrahas, and D. H. Solomon,
Nonsteroidal antiinammatory drugs and nonunion of
humeral shaft fractures, Arthritis Care and Research, vol. 53,
no. 3, pp. 364367, 2005.
[145] A. Meunier, P. Aspenberg, and L. Good, Celecoxib does not
appear to aect prosthesis xation in total knee replacement:
a randomized study using radiostereometry in 50 patients,
Acta Orthopaedica, vol. 80, no. 1, pp. 4650, 2009.
[146] J. Aerssens, S. Boonen, G. Lowet, and J. Dequeker, Inter-
species dierences in bone composition, density, and quality:
potential implications for in vivo bone research, Endocrinol-
ogy, vol. 139, no. 2, pp. 663670, 1998.
[147] S. Miyatake, H. Ichiyama, E. Kondo, and K. Yasuda, Ran-
domized clinical comparisons of diclofenac concentration in
the soft tissues and blood plasma between topical and oral
applications, British Journal of Clinical Pharmacology, vol.
67, no. 1, pp. 125129, 2009.
[148] A. Rahal, A. Kumar, A. H. Ahmad, and J. K. Malik,
Pharmacokinetics of diclofenac and its interaction with
enrooxacin in sheep, Research in Veterinary Science, vol. 84,
no. 3, pp. 452456, 2008.
[149] I. Pountos, T. Georgouli, and P. V. Giannoudis, The eect of
autologous serumobtained after fracture on the proliferation
and osteogenic dierentiation of mesenchymal stem cells,
Cellular and Molecular Biology, vol. 54, no. 1, pp. 3339, 2008.
[150] I. Pountos, T. Georgouli, S. Perry, J. Morley, and P.
Giannoudis, Systemic signals after fracture: their eect in
osteogenesis, Injury Extra, vol. 38, pp. 161162, 2007.
[151] P. M. Arnold and J. A. Klemp, Assessment of malunion in
spinal fusion, Neurosurgery Quarterly, vol. 15, no. 4, pp. 239
247, 2005.
[152] M. Fransen and B. Neal, Non-steroidal anti-inammatory
drugs for preventing heterotopic bone formation after hip
arthroplasty, Cochrane Database of Systematic Reviews, no.
3, Article ID CD001160, 2004.
[153] E. F. McCarthy and M. Sundaram, Heterotopic ossication:
a review, Skeletal Radiology, vol. 34, no. 10, pp. 609619,
2005.
[154] L. V. Bossche and G. Vanderstraeten, Heterotopic ossica-
tion: a review, Journal of Rehabilitation Medicine, vol. 37, no.
3, pp. 129136, 2005.
[155] L. Song and R. S. Tuan, Transdierentiation potential of
human mesenchymal stem cells derived from bone marrow,
FASEB Journal, vol. 18, no. 9, pp. 980982, 2004.
[156] L. E. Dahners and B. H. Mullis, Eects of nonsteroidal
anti-inammatory drugs on bone formation and soft-tissue
healing, The Journal of the American Academy of Orthopaedic
Surgeons, vol. 12, no. 3, pp. 139143, 2004.