*
^
One
hundred and four dogs diagnosed with osteoarthritis (OA) were
enrolled in a multi-centre, double-blind, comparator
controlled, randomised clinical study in Germany to establish
the effectiveness of the disease modifying anti-osteoarthritic
drug, sodium pentosan polysulfate (NaPPS - marketed as
CARTROPHEN VET®)
compared to the non steroidal anti-inflammatory control drug
carprofen (Rimadyl®,
Zenecarp) in treating the clinical signs of OA. Efficacy
was assessed by a reduction in the clinical signs of OA namely
lameness and pain on manipulation by the veterinarian during
treatment (Weeks 1, 2, 3 and 4) and four weeks after treatment
had ceased (Week 8). Four weekly subcutaneous injections
of NaPPS at 3mg/kg by the subcutaneous route and four weekly
daily oral administration of carprofen at 4mg/kg were found to
be effective in treating OA with significant improvements in
all primary outcome parameters i.e. lameness, pain and
orthopaedic socre (p<0.05). NaPPS maintained its
effectiveness longer than carprofen with significant
improvement following NaPPS treatment compared to carprofen
treatment in orthopaedic score at Week 8 (p=0.013).
Osteoarthritis
(OA) is a disorder which may affect all articulations but is
most prominent in the spine and the peripheral weight-bearing
joints [Felson, 1988; Altman, 1991; March and Brooks, 1996].
OA is characterised pathologically by focal
fibrillation and erosion of articular cartilage, subchondral
bone changes including sclerosis, osteolysis, osteophytosis
and synovial inflammation [Gardner, 1983; Mankin, Brandt and
Shulman, 1986; Ghosh, 1991; Hamerman, 1993] (see Figure 1).
While the number of studies on the natural history of
OA in the canine is limited [Van Pelt, 1965; McDevitt
et al., 1974; McDevitt and Muir, 1976; Alexander, 1979;
Lust and Summers, 1981; Pederson, Pool and Morgan, 1983; Kealy
et al., 1997] there is a plethora of reports in the
literature of canine arthropathies induced experimentally.
The most widely used models have been the induction of
traumatic OA in canine joints by menisectomy [Cox
et al., 1975; Ghosh
et al., 1983a,b, 1984; Hannan
et al., 1987; Smith and Ghosh, 2001] or transection of the
anterior cruciate ligament (ACL)
[Sandy et al., 1984; Altman et al., 1984;Pelletier
et al., 1985; Dunham
et al., 1985; Fife, 1986; Johnson and Poole, 1990; Myers et al., 1990; Brandt
et al., 1991a,b; Carney
et al., 1985, 1992; Ratcliffe
et al., 1993; Guilak
et al., 1994; Adams, 1994; Venn et al. 1993,
1995; Dourado et al.,
1996].
The clinical outcome of experimental transection of the
ACL mimic the naturally occurring situation [Elkins et
al., 1991; Vasseur and
Figure 1: In
osteoarthritic joints pathological changes occur in articular
cartilage
Medical
treatments for OA have, up until recently, targeted the
clinical signs of the disease, rather than the underlying
pathologies responsible. Analgesics
and steroidal and non-steroidal anti-inflammatory drugs (NSAIDs)
are, and still remain, the mainstay of treatment [Brandt and
Slowman-Kovac, 1986; Gabriel and Wagner, 1997; Johnston and
Budsberg, 1997; Fox and Johnston, 1997].
However, the deleterious side effects provoked in dogs
and humans with the use of many of these agents (e.g.
on the gastrointestinal tract, kidneys and articular
cartilage) [McKenzie et
al., 1976; Palmoski
and Brandt, 1980; Innes, 1995; Lichenstein
et al., 1995; Manoukian
et al., 1996; Isaacs, 1996] has led to a steady decline in
their usage in recent years.
A variety of new compounds now marketed have been
reported to be selective inhibitors of COX-2 at low plasma
concentrations [Vane and Botting, 1996; Engelhardt, 1996;
Noble and Balfour, 1996; Engelhardt et
al., 1995; Hulse 1998; McLaughlin, 2000].
While these new NSAIDs are reported to have diminished
adverse side effects on the gastrointestinal tract they are
still associated with other toxicities, which are now becoming
more apparent as their clinical use increases.
This is particularly evident for the kidney where COX-2
enzymes are expressed and have important physiological
functions [Perazella and Tray, 2001].
In addition in the dog carprofen (RimadylÒ)
is associated with liver toxicity [MacPhail et
al., 1998] in certain breeds.
Moreover, there is no evidence that steroidal or NSAIDs
provide any beneficial effects on the underlying
haematological abnormalities which exist in OA joints, which
can contribute not only to the clinical signs of the disorder
but also its progression.
Indeed chronic use of corticosteroids is known to
exacerbate intravascular coagulation and osteonecrosis [Jones,
1993] and from this standpoint alone may contribute to disease
progression.
One class of drugs, the pentosan polysulfates (PPSs) which have been actively researched for more than 40 years, have now been developed for the treatment of OA [Ghosh, 1999]. The disease modifying properties of PPS are outlined in Figure 2. Therapeutic intervention with sodium pentosan polysulfate (NaPPS) in the ACL deficient dog model of OA was shown to maintain cartilage structure and biochemistry [Rogachefsky et al., 1993]. It was hypothesised that this effect was due to NaPPS's ability to block proteinase activity and that this allowed the observed growth factor induced effects. In a similar study in the dog knee model of disuse atrophy, Grumbles et al. (1995) speculated about the role of concurrent treatment with PPS and insulin-like growth factor - 1 (IGF-1) as a prophylactic therapy for retardation of protease- tissue inhibitor of metalloproteases. It has also been shown that NaPPS inhibits human lysosomal elastase, a serine proteinase [Baici et al., 1981].
Figure 2: Disease
modifying properties of PPS (Ý
sites of PPS action)
Pain
in OA is the most important clinical sign in humans [Moskowitz,
1984] and domestic animals [Caron, 1996; Innes, 1995;
Johnston, 1997; Hulse, 1998; McLaughlin, 2000].
Pain is the principle cause of reduced performance and
its pathogenesis is usually multifactorial, however, many
aspects of the reaction of the nervous system to noxious
stimuli (nociception) and the sensory and emotional experience
associated with a noxious stimulus (pain) remain unclear
[Caron, 1996; Johnston 1997].
The sites where PPS acts on pain in OA are summarised
in Figure 3.
Figure 3: Biochemical
origins of pain in OA and sites (Ý)
where PPS intervenes
The
peripheral neuroanatomy of joints is reported to be similar in
many species [Caron, 1996] and the popular classification of
articular nerve endings in mammalian appendicular joints as
four receptor types - types 1, 2, 3 and 4 is universally
accepted. Type 1
mechanoreceptors are located in the superficial areas of the
joint capsule and are low-threshold receptors.
That is, they are stimulated by relatively mild
mechanical stimuli and remain active while a mechanical
stimuli persists. Type
2 receptors are found more deeply in the joint capsule and are
low-threshold, rapidly adapting mechanoreceptors.
They are inactive when joints are immobile and become
activated when joints undergo movement or experience tension.
Type 3 receptors are large and are restricted to intra-articular
and periarticular ligaments near their insertions.
They are high-threshold slowly adapting
mechanoreceptors that are inactive in stationary joints and
with active and passive movement over a limited range and
become activated only when joint excursions occur near
physiological limits or when ligaments containing them undergo
powerful traction forces.
Type 3 receptors are also capable of nociception and
modifying type 1 and 2 receptor-mediated reflexes.
Type 4 'receptors' are free nerve endings rather than
specific end organs like receptors 1 to 3 of which there are
two types - type 4a and 4b. Type
4 endings are high threshold, slowly adapting nociceptors and
their activation signals impending or actual tissue damage.
Type 4 endings are polymodal and respond to mechanical,
heat and chemical stimuli such as lactic acid, kinins,
serotonin, histamine and prostaglandin E2.
The
potent anti-inflammatory and anti-complement activities of PPS
have been consistently demonstrated in different models of
severe inflammation. Kalbhen
and co-workers [Kalbhen 1971, 1972, 1973; Kalbhen et
al., 1978] using oedemas induced in rat paws by injection
of dextran, formaldehyde, trypsin, hyaluronidase, carrageenan
or kaolin. For all
of these experimentally induced oedemas, a dose-response
correlation was observed using NaPPS within the concentration
range of 25 - 100 mg/kg when the drug was
administered subcutaneously.
It was concluded by Kalbhen (1973, 1978) that unlike
sodium salicylate, phenylbutazone or indomethacin, NaPPS was
effective against all the types of inflammogens his group had
examined. The
mechanism of action in this regard was attributed largely to
stabilisation of the peripheral vascular system and
improvement of the microcirculation in the inflamed tissues.
According
to Walb, Loos and Hadding (1971), NaPPS possesses marked
anti-complementary activities.
Using sensitised erythrocytes
in vitro, it was found that NaPPS was ten times more
potent than heparin in preventing the lysis by a complement
preparation when used over the concentration range 5.0 -
8.3 µg/mL. In
vitro NaPPS also inhibited the complement C1-esterase, the
ED50 lying in the range of 7 - 8 µg/mL.
In a series of consecutive papers, Berthoux and
co-workers (1977a,b) confirmed the anti-complementary activity
of NaPPS both in vitro
and in vivo..
From these studies it was concluded that the in
vivo anti-complement activity of NaPPS were sufficiently
strong to suggest that a decrease liberation of humoral
mediators of inflammation would occur during its clinical
usage.
The
anti-inflammatory action of PPS is mediated through several
pathways the most important being summarised in Figure 4.
Figure
4: Anti-inflammatory
action of PPS
The
study was a multi-centre, double-blind, comparator controlled,
randomised study. The
objective of the study was to establish the efficacy and
safety of 3mg/kg NaPPS marketed at CARTROPHEN
VETâ
for the treatment of OA. Animals
(dogs) were drawn from the existing client base of ten private
practices in
Diagnosis of OA was based on radiographic examination and the presence of the clinical signs of OA according to conventional criteria, namely lameness, pain on palpation, stiffness and a decrease in activity. Cases had a specific diagnosis based on history, full physical examination and radiographic examination (2 views). Radiographs had to demonstrate some evidence of OA. Changes consistent with a diagnosis of OA include osteophyte formation, synovial effusion, subchondral sclerosis, decreased joint space and soft tissue swelling.
Dogs
were assessed by the veterinarian at weekly intervals during
treatment (Weeks 1 to 4) and four weeks after treatment had
ceased (Week 8). The
same veterinarian for each individual dog undertook the
physical examination at each visit.
The
primary outcomes measured were improvement at Week 4 and Week
8 from baseline for lameness (scored on a five point scale
from 0=no lameness to 4=won't use), pain on manipulation
(scored on a five point Likert scale from 1=no pain to
5=screams with pain) and orthopaedic score (sum of pain and
lameness scores) and the investigator's overall impression at
Week 8. In
addition the secondary outcomes of improvement at Week 4 and
Week 8 from baseline for gait ataxia, gait weakness, gait
stiffness, coat condition, condition of dog, body weight and
"get-up-and-go" were measured.
Safety was measured by recording the frequency and
severity of any suspect adverse reactions.
Data
analysis was designed to establish if carprofen and NaPPS were
effective treatments for OA and to determine if there were any
differences between the efficacy of the treatments.
The
Stratified Wilcoxon rank-sum test [Bajorski and Petkau, 1999]
was used to determine differences between treatments in
improving lameness, pain, orthopaedic score, gait ataxia, gait
weakness, gait stiffness and get up and go.
This method was the definitive test for drawing
inferences and conclusions in the study.
The strata used were the baseline values and the
statistics over all the strata for each week were pooled.
Analysis of Veterinary Impression was undertaken using the
continuation-ratio model.
The
Continuation-odds ratio approach was used to determine
differences between treatments in improving lameness, pain,
orthopaedic score, veterinary impression, gait ataxia, gait
weakness, gait stiffness, get up and go and second joint
parameters (lameness, pain and orthopaedic score).
Analysis performed by this method was secondary to the
Stratified Wilcoxon rank sum test.
The
activity of each treatment was determined using the Wilcoxon
signed rank test for both primary and secondary outcomes.
Fishers
exact test was used to determine if there was a statistical
difference between treatments in the number of adverse events
reported.
The
level of significance for all statistical methods was
p<0.05.
In
addition the effectiveness of both carprofen and CARTROPHEN
VETâ
treatments in reducing
severe lameness, pain and orthopaedic score was assessed by
graphing the mean change from baseline (Week 1) in these
parameters. Severe
lameness corresponded to the clinical signs “won’t
use” or “just puts affected limb on ground”, severe pain
was defined as the dog “vocalising” or “wincing and
withdrawing” upon manipulation of the affected limb and
severe orthopaedic score was considered to be a score of 6, 7,
8 or 9.
One
hundred and four dogs were enrolled in the study (51 control
and 53 CARTROPHEN VETâ).
A total of 33 different breeds participated in the
study. The
greatest number of animals treated fell in the age range 6.1
to 9.0 years in the carprofen treatment group (16) and in the
3.1 to 6.0 year old age group for treated
animals (18). The
average age was 8.2 years in the carprofen treatment group
compared to 6.9 years in the NaPPS
treatment group. Twenty two males and 29 females
received carprofen treatment compared to 31 males and 22
females receiving NaPPS treatment.
Analysis of the primary outcomes of lameness, pain and orthopaedic score, revealed that there was statistically significant improvement following treatment with NaPPS compared to treatment with carprofen at Week 8 (four weeks after treatment had ceased) in orthopaedic score (p=0.013). Improvement following carprofen treatment relative to NaPPS treatment was statistically significant for lameness at Week 2 (p<0.001), pain at Week 2 (p=0.023) and Week 3 (p<0.001) and orthopaedic score at Week 2 (p=0.041). The statistically significant difference at Week 2 for lameness in favour of carprofen was diminished by Week 8 with NaPPS the more favourable treatment (p=0.129). A summary of the efficacy results for these primary outcomes is presented in Figure 5.
Figure 5: Mean change from baseline during treatment (Weeks 2, 3 and 4) and 4 weeks after treatment ceased (Week 8) for the primary outcomes of lameness, pain and orthopaedic score (*=significant at p<0.05)
According
to the primary outcome veterinarians' impression of the
overall response to treatment, both NaPPS and carprofen
treatments were highly
effective and there was no
statistically significant difference between the two drugs
(p=0.909). There
was however, a slight advantage in favour of NaPPS
in the estimate of the magnitude of the effect.
Analysis of the activity of each treatment compared to the corresponding baseline value demonstrated that both NaPPS and carprofen were very effective treatments with statistically significant improvements in all primary outcome parameters (lameness, pain and orthopaedic score) at all weeks (p<0.05). The efficacy of each treatment was also demonstrated for secondary outcomes with significant improvements in gait stiffness and get up and go at all weeks and gait ataxia and gait weakness at Weeks 3, 4 and 8 (p<0.05). No statistically significant change to dog condition, coat condition or body weight was noted at any week for either treatment (p>0.05).
In
order to assess the effectiveness of each drug in cases of
severe OA, animals presenting with severe lameness, severe
pain and severe orthopaedic score were analysed for the mean
change from baseline (Week 1).
Although numbers in each group were small, the results
demonstrated that both NaPPS
and carprofen were effective treatments for animals suffering
severe OA. Figure
6 summarises the mean change from baseline in lameness, pain
and orthopaedic score in animals with severe OA.
Figure 6: Mean
change from baseline in lameness, pain and orthopaedic score
for animals suffering severe OA
Six
adverse events were reported during the study - four in the
NaPPS group and two in the carprofen group.
None of the four adverse events in the NaPPS group were
considered to be associated with the treatment, while both of
the adverse events reported in the carprofen group were
considered to be possibly associated with the treatment. The
incidence of 2/51 carprofen and 4/53 NaPPS adverse events is
not statistically significant (p= 0.687).
NaPPS
a disease modifying anti-osteoarthritic drug with
anti-inflammatory activities is as effective as the NSAID and
analgesic drug, carprofen, in the treatment of lameness and
pain of OA at the end of the 4 week treatment period.
NaPPS maintained its effectiveness longer than
carprofen, as indicated by the significant improvement
following NaPPS treatment compared to carprofen treatment in
orthopaedic score at Week 8 (p=0.013).
In addition NaPPS and carprofen were both found to be
effective in treating the severe clinical signs of OA with
some evidece that NaPPS treatment is superior.
Adams
M (1994) Changes in aggrecan populations in experimental
osteoarthritis. Osteoarthritis Cart 2:155-164
Alexander JW
(1979) Osteoarthritis (Degenerative Joint Disease) in the dog.
Canine Practice 6(1):31-34
Hannan N, Ghosh P, Bellenger C, Taylor T (1987) The systemic
administration of glycosaminoglycan polysulphate (Arteparon)
provides partial protection of articular cartilage from damage
produced by meniscectomy in the canine. J Orthop Res 5:47-59
Hulse
D. (1998) Treatment methods for pain in the osteoarthritic
patient. Veterinary
Clinics of
Vasseur
PB, Berry CR (1992) Progression of stifle osteoarthrosis
following reconstruction of the cranial cruciate ligament in
21 dogs. J Am Anim Hosp Assoc 28(2):129-136
To view this article in PDF format please click here