This article appeared in a journal published by Elsevier.

The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright
 Author's personal copy

Journal of Equine Veterinary Science 31 (2011) 625-629

Journal of Equine Veterinary Science

journal homepage: www.j-evs.com

Original Research

Thermographic Assessment of Saddles Used on Jumping Horses

Tiago Z. Arruda DVM a, Karin Erica Brass DVM, PhD b , Flavio D. De La Corte DVM, PhD b
a
Federal University of Santa Maria, Rio Grande do Sul, Brazil
b
Department of Large Animal Clinics, Federal University of Santa Maria, Santa Maria, Rio Grande do Sul, Brazil

a r t i c l e i n f o a b s r t a c t

Article history: In this study, 62 saddles used on 129 jump horses were evaluated. Data such as age,
Received 21 December 2010 breed, gender, body score, competition level, changes in performance and reaction at
Received in revised form saddling, resistance in performing exercises, and back pain were obtained from the
08 March 2011
horses. Years of use, number of horses and riders using the saddle, and riders’ level of
Accepted 02 May 2011
experience were also registered. Before and after a training session, horses were sub-
jected to thermography of their thoracolumbar region. Saddles were evaluated by ther-
mography soon after their withdrawal from the animal after training. The image
Keywords:
Saddle
obtained identified the area of interaction between the saddle panels and the back of the
Back pain horse, asymmetry between panels, and the possible presence of contact with the column.
Thermography Each saddle was used on an average of 4.6
3.7 horses. Asymmetry between panels was
Horse identified on 62.8% of saddles and central contact with the thoracolumbar column was
found on 37.2% of the saddles. Heat points on the column and withers were found on
28.7% and 33.3% of the horses, respectively. The thermographic image of the thor-
acolumbar column after the training session was asymmetric on 55.8% of the animals.
Only the panels of 51.2% of the saddles had an interaction of 76% to 100% with the back of
the horse. At rest, 39.5% had a heat point compatible with a pressure area from the
saddle. Thermography proved to be a useful tool in evaluating the adjustment and
position of saddles used on jumpers.
Ó 2011 Elsevier Inc. All rights reserved.

1. Introduction emission from the body by infrared radiation, indicating

Poor fitting or incorrect positioning of the saddle is
considered a frequent cause of back pain in horses, in
addition to lack of balance and/or poor qualification of the
rider [1]. The primary source of pain in these cases is the
irritation of dorsal nerve roots and branches of the spinal
nerves. Muscle injuries are usually associated with loss of
contraction power and with pain [2,3]. Variable clinical
signs, long course (weeks or months), and a tendency to
spontaneous recovery (around 65%) render the diagnosis
and treatment evaluation of back pain more difficult [4].
Thermography presents itself as a noninvasive diag-
nostic imaging technique that detects the superficial heat
Corresponding author at: Karin Erica Brass, DVM, PhD, Department of
Large Animal Clinics, Federal University of Santa Maria, Avenue Roraima
1000, 97.105-900 Santa Maria, RS, Brazil.
E-mail address: kbrass@terra.com.br (K.E. Brass).
the temperature of the body surface. The local skin
temperature is of little diagnostic value unless it can be
compared with adjacent tissues and with the opposite side
of the body. Therefore, absolute temperature values cannot
be compared between horses because skin temperature
does not depend only on vascular activity, but also on the
metabolic state and hair coat length. In addition to man-
made artifacts, such as the use of blankets and support
bandages, other factors affecting skin temperature are
movement, external energy radiation (sunlight), envi-
ronmental temperature, air flow, and calibration of the
thermographic camera [5].
In studies on back pain in human beings, thermography
presented a 95% specificity and 99% sensitivity in identi-
fying pathologies [6,7]. In race horse preventive sports
medicine, thermography has shown good association
between problem perception by the trainer and the diag-
nosis of the attending veterinarian later on. In one study,

0737-0806/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.jevs.2011.05.011
 Author's personal copy
626 T.Z. Arruda et al. / Journal of Equine Veterinary Science 31 (2011) 625-629
Fig. 1. Contact area (%) between saddle panels and thoracolumbar region of horses identified by thermography. Contact until (A) 25%, (B) 50%, (C) 75%, and
(D) 100%.
increased heat 2 weeks before the occurrence of clinical
signs related to the same region were noticed in most
cases [8].
The straightforward documentation of the area of
interaction between saddle and thoracic/lumbar region,
attributable to the adjustment of the saddle, can possibly be
obtained by means of thermography. Considering the basic
principle of thermal symmetry, image assessment of the
dynamic interaction between saddle and back of the horse
by thermography could show not only the heat generated
in areas of greater interaction with the saddle, but also the
physiological effects of riding on the back of the horse [9].
The aim of this study was to determine the frequency of
abnormal saddle contact using thermography as a potential
imaging technique to assess proper/improper saddle
fitting. Considering possible negative effects on perfor-
mance, factors related to the jumping horse, rider, and
saddle that could affect the interaction area between saddle
and back of the horse were also evaluated.
2. Materials and Methods
In this study, 129 jumping horses were randomly chosen
and their 62 saddles were evaluated immediately before
and after a training session. The horses’ mean age was 8.0

3.4 years. Their breed (63% Warmbloods, 23% mixed breed,
9% Thoroughbreds, 3% Arabians, and 2% Quarter horses),
gender (10 stallions, 54 mares, and 65 castrated horses),
level of competition (51.9% experienced horses jumping 1
to 1.5 m; 38.8% young horses starting training and jumping
up to 1.1 m; 9.3% riding school horses jumping up to 0.9 m),
time of training with the current trainer, body score
(emaciated ¼ 1; poor ¼ 1.5; thin ¼ 2, moderately thin ¼ 2.5;
below the weight ¼ 3; athletic ¼ 3.5; moderately fat ¼ 4;
fat¼ 4.5; and obese ¼ 5) [10], number of people who ride
the same horse during the week, reaction of the horse
when saddled and reported loss of performance, resistance
to perform specific tasks at training, and history of back
pain (diagnosed by the attending veterinarian) were
tabulated for each horse. Riders were classified according
to their competition level as follows: professional, amateur,
co-rider (only rides horses for owners and not for compe-
titions), and beginners (tournaments up to 0.9 m). Saddle
characteristics such as years of use, number of riders that
use them, and number of horses that were saddled with
them were also registered.
Later, horses were submitted to a thermographic eval-
uation of their thoracolumbar region. The examination was
carried out at a quiet place, protected from sun radiation
and wind, after 15 minutes of acclimatization to the envi-
ronmental temperature, as recommended [5]. With
a noncontact thermographic camera (FLIR ThermaCAM
E25, FLIR Systems Brasil, Sorocaba, São Paulo, Brazil),
images of the thoracolumbar region were captured with an
emissivity (e) between 0.99 and 1.0 and a temperature
range between 14 C and 37 C for later evaluation. After
thermography at rest, the saddled horses started their
training session. At the end of the training session, duration
and type of work (dressage or jumping) were registered.
A thermographic image was obtained immediately after
removal of the saddle to evaluate the area of interaction
with the back of the horse. On the thermographic image of
the saddle panels, areas with increased temperature
indicative of more interaction with the animal were eval-
uated for asymmetry, contact with the dorsal midline,
and skin interaction area of the panels using a grid to divide
the images in four areas (up to 25%, 50%, 75%, and 100%
area of interaction) (Fig. 1). After another 15 minutes of
acclimatization, the thermographic examination of the
thoracolumbar region of the horses was repeated (to
 Author's personal copy
T.Z. Arruda et al. / Journal of Equine Veterinary Science 31 (2011) 625-629
Fig. 2. Thermographic images of saddles: (A) Asymmetric pressure on the right; (B) Asymmetric pressure on the left; (C, D) Contact point on the dorsal midline.
compensate for the insulating effect of sweating areas) for
comparison with the image obtained before training. On
images taken after exercise, heat pattern was evaluated as
symmetric or asymmetric heat emission throughout the
thoracolumbar surface. Comparing this image with the one
obtained before training, the appearance of heat points on
the vertebral column and/or withers was identified. The
image of the horse was compared with the image of the
saddle panels after work, looking for a possible point of
chronic heat associated with inadequate saddle fit on the
thoracic and lumbar spine. All thermographic images were
obtained by the same observer.
Descriptive statistics of data included frequency distri-
bution. Chi-square test or Fisher exact test analysis was
used to evaluate possible interaction between asymmetry
of the saddle and level of the rider, type of work, and
lumbar pressure distribution; focal heat point and changes
in performance, lumbar pain, and resistance to perform
certain exercises; and central contact on the saddle and
pressure on the column. Variance analysis was used to
analyze body score, age, exercise duration, time of use of
the saddle (years), number of horses that used the same
saddle, and number of riders that used the same saddle.
Asymmetric heat distribution on the saddle and the
nonparametric variables were submitted to correlation
analysis by the Spearman’s correlation test. Statistical
significance was set at P < .05.
3. Results and Discussion
Asymmetric interaction between saddle panels and
horse’s back was observed on 62.8% of the saddles evalu-
ated (Figs. 2A, B). According to Turner et al. [9], symmetry is
the most important aspect of the thermographic evaluation
of a saddle, representing an objective documentation of the
distribution of different pressures on the back of the horse.
This pressure is the result of the interaction and force
627
distribution between horse and rider, including the rider’s
position and riding technique [11]. A correctly fitted saddle
enhances both the horse’s and the rider’s ability to perform
with accuracy, lightness, and freedom of movement [12].
Conversely, 55.8% of the horses presented an asymmetric
thermographic image of the thoracolumbar region after
a 15-minute exercise. In 2003, De La Corte and Mikail [13]
observed that 38% of a group of 51 jumping horses
showed an asymmetric heat distribution in the thor-
acolumbar region, suggesting the presence of physical
problems possibly related to the saddles. However, their
study did not include an evaluation of the saddle for
asymmetries nor were the horses examined to detect pain
over the same area. The authors concluded that those
asymmetries could indicate problems such as bursitis of
the supraspinal ligament bursae, muscle lesions, or atro-
phies and recommended further evaluation of any horse
with asymmetric thermographic patterns.
An interaction area of 76% to 100% between the saddle
panels and back was observed on 51.2% of the saddles
(Fig. 3). This large area favors the absorption of the rider’s
weight and impact, distributing them over the withers and
along the back of the horse [4,9]. Saddles with a panel
interaction area of up to 75%, 50%, and 25% were less
frequent (30.2%, 13.2%, and 5.4%, respectively), but not less
important. In addition to inappropriate adjustment to the
horses’ back, this lack of adequate contact may also be
affected by the riding ability of the rider [1,4,11].
Among riders, the majority were amateurs (48.8%),
theoretically showing less ability and balance than profes-
sionals (31.8%). Co-riders (13.2%) and beginners (6.2%) were
less represented. No association was found between asym-
metry of the saddle and the rider’s experience level (P ¼ .46)
(Table 1), but a correlation was found between panel
asymmetry and individual riders (r2 ¼ 0.22; P ¼ .013),
suggesting that some characteristics of the rider may be
responsible for asymmetry. Dorsopathies can be induced by
 Author's personal copy

628 T.Z. Arruda et al. / Journal of Equine Veterinary Science 31 (2011) 625-629

Table 2
Number (n) and percentage (%) of the symmetric and asymmetric heat
distribution on the saddles according to the horse’s level (r2 ¼ 0.19, P ¼ .02)

Horses Symmetric Asymmetric Total

n % n % n %

Experienced 31 46.3 36 53.7 67 51.9

Young 14 28.0 36 72.0 50 38.8
Riding school 3 25.0 9 75.0 12 9.3
Total 48 37.2 81 62.8 129 100.0

more reliable when evaluated on a population of similar

Fig. 3. Area of contact (%) between the saddle panels and thoracolumbar
region identified by thermography on 62 saddles used on 129 jumpers.
imbalance or bad qualification of the horse’s rider [1,4]. In
contrast, Peham and Schobesberger [14] showed on
a computer model simulation that stress caused by a rider is
lower than that of pathological processes such as increased
stiffness of the back. Moreover, Peham et al. [15] demon-
strated the beneficial effect of a rider on the horse’s motion
pattern variability. Considering the level of competition of
the horses, the riding school horses showed a higher
incidence of saddles with asymmetry (r2 ¼ 0.19; P ¼ .029)
(Table 2). Characteristic changes of the pressure center occur
while horses move at the walk, sitting trot, rising trot, and
canter [16,17], thus the heat distribution can change
according to the gait used for the exercise.
Pressure is not only relevant to muscular function but
also to the comfort of the horse and its back health [18,19].
On 51 (39.5%) horses examined before exercise, a heat point
observed on the thermographic image of the saddle and,
compatible with an isolated pressure area, suggested
a possible chronic lesion caused by a poor saddle fit. This
mapping, already described by several authors [2,3,5,20],
allows the identification of hot areas caused by local
temperature increase owing to an inflammatory process. Of
these, 33 horses (64.7%) showed a heat point over the
withers and 18 (35.73%) over the thoracolumbar column.
This demonstrates that the most common mistake is
setting the saddle too far ahead, thus promoting an
unbalanced position of the rider and restricting the
movements of the horses’ forelimbs.
Mean body score of the horses was 3.7
0.4, with
a variation coefficient of 12%. Assessment of saddle fitting is
Table 1
Number (n) and percentage (%) of symmetric and asymmetric heat
distribution patterns on the saddle according to the rider’s level
conformation, or with a more homogeneous body score,
because it minimizes an important variable, that is, the
conformation of the horse’s back. The correlation between
saddle panel asymmetry and the horse’s body score (r2 ¼
0.24; P ¼ .005) (Table 3) shows that body condition can
also influence saddle fit. Loss of body condition and dorsal
muscle atrophy can lead to inappropriate saddle fitting [9].
The training time with the current trainer varied
between 1 and 120 months. In all, 80 (62%) of the horses
had been trained by the actual trainer for 1 year or more,
increasing reliability on the clinical history of the horses,
with regard to changes in behavior or performance.
Considering the type of exercise performed during the
training session, 57.4% of the horses performed dressage
exercises, whereas 42.6% jumped (Table 4). The training
sessions lasted 33.2
15.3 minutes on average. There was
no association between type of exercise and percentage of
asymmetric saddles.
Horse reactions, such as spinal dorsoflexion, biting, or
kicking when a saddle is placed on the horse’s back, were
observed for only two horses (1.6%). Jeffcott and Haussler
[4] termed the behavior where horses show transient
ventroflexion without changing performance “cold back.”
This is not associated to any lumbar pain. Also, the riders
reported that 24.0% of the horses showed resistance to
perform certain exercises during the training session that
could be related to lumbar pain, because clinical signals are
variable and may not be connected to a specific location of
the thoracolumbar column, thus making the diagnosis
more difficult, as described by several authors [1,2,4,8,9].
Back pain previously diagnosed by the attending veteri-
narian was present in 5.4 % of the horses. Alves et al. [20]
reported that 4.35% of the skeletal muscle disorders were
related to a diagnosed back pain. Loss of performance
without any apparent cause, described by riders, was
observed for 10 horses (7.7%). Larger the asymmetry of the
Table 3
Number (n) and percentage (%) of symmetric and asymmetric heat
distribution patterns on the saddle according to the horse’s body score
(r2 ¼ 0.244; P ¼ .005)

Rider's Level Symmetric Asymmetric Total Body Score Symmetric Asymmetric Total

n % n % n % n % n % n %

Professionals 20 48.8 21 51.2 41 31.8 3.0 6 22.2 21 77.8 27 21.0

Amateurs 19 30.2 44 69.8 63 48.8 3.5 9 24.3 28 75.7 37 28.7
Co-riders 8 47.1 9 52.9 17 13.2 4.0 32 52.5 29 47.5 61 47.3
Beginners 2 25.0 6 75.0 8 6.2 4.5 0 0.0 2 100.0 2 1.5
Total 48 37.2 81 62.8 129 100.0 5.0 1 50.0 1 50.0 2 1.5
Total 48 37.2 81 62.8 129 100.0
P > .05.
 Author's personal copy
T.Z. Arruda et al. / Journal of Equine Veterinary Science 31 (2011) 625-629
Table 4
Number (n) and percentage (%) of the area of heat distribution on the
saddles of dressage and jumping exercise
Heat Distribution Area Dressage Jumping
on the Saddle
n % n %
Asymmetric contact 47 63.5 34
Midline contact 27 36.4 21
Contact area 25% 4 5.4 3 5.45
Contact area 50% 13 17.6 4 7.3
Contact area 75% 25 33.8 16
Contact area 100% 34 45.9 32
Total 74 57.37 55
P > .05.
heat distribution on the saddle panels, worst was the
performance of the horse (r2 ¼ 0.26; P ¼ .003). Many
authors [1,2,9,21] believe that dorsopathy is the main cause
for changes in gait and performance. This is because back
problems may be represented by signs such as difficulty
in performing training exercises, refusal to jump, or
changes in jumping style, and even reluctance to trot or
gallop [20-22].
The saddles were used for 4.5
4.3 years on 4.6
3.7
horses by 2.1
1.0 different riders. As the time of use of the
saddle (r2 ¼ 0.18; P ¼ .044) or the number of horses on
which the same saddle was used (r2 ¼ 0.17; P ¼ .045)
increased, the frequency of saddle asymmetry also
increased. The large number of horses being ridden with
the same saddle clearly shows that the correct adjustment
to the back was completely ignored by the riders and
trainers in this study. Many authors [1,4,18,19,21] have
shown that conformation differences between individuals
may result in an adjustment of the saddle that does not fit
all horses properly. This can cause disease, gait changes, or
loss of performance. As a consequence, the thermographic
images obtained from the saddles indicate that 37.2% of the
saddles presented ill-fitting points on the dorsal midline
(Figs. 2C, D). This can be considered an important pressure
point on the back, which can cause serious consequences
over time. Of the images obtained from the horses’ back
after work, 28.7% presented a heat point on the thor-
acolumbar column and 33.3% on the withers. These results
can also be attributed to an inadequate adjustment of the
saddle on the horse, already demonstrated by other
researchers [4,9,18,19]. In addition, Haussler et al. [23]
showed that jumpers have longer spinal processes and
consequently are more susceptible to kissing spines.
Excessive pressure of the saddle on the dorsal midline may
be causing desmitis or bursitis on the lumbar vertebrae.
4. Conclusion
The high percentage of asymmetric saddles might be an
indication that the vast majority of saddles used in jumping
are not adjusted for the horse on which they are used.
Thermography has shown to be a useful tool for diagnosing
interaction asymmetry on the saddle. The use of saddles
629
unsuitable for a given animal indicates the lack of attention
to the equipment used in horse riding.
References
61.8
38.2 [1] Harman JC. Alternative medicine in equine practice: part II.
Proceedings of the North American Veterinary Conference; 2005
[Internet]. Available at: http://www.ivis.org/proceedings/navc/
2005/LA/083.pdf?LA¼1. Accessed March 25, 2009.
29.1
[2] Boffi FM, Cittar JS, Balskus G, Muriel MG. Patologias que afectan el
582
rendimiento deportivo. In: Boffi FM, editor. Fisiologia del ejercicio
42.63 en equinos. Buenos Aires, Argential: Inter-Médica; 2006. p. 255-66.
[3] Piercy RJ, Rivero JL. Muscle disorders of equine athletes. In:
Hinchcliff KW, Kaneps AJ, Geor RJ, editors. Equine medicine and
surgery. Edinburgh, Scotland: Saunders; 2005. p. 77-110.
[4] Jeffcott LB, Haussler KK. Back and pelvis. In: Hinchcliff KW,
Kaneps AJ, Geor RJ, editors. Equine medicine and surgery. Edin-
burgh, Scotland: Saunders; 2005. p. 433-74.
[5] Turner TA. Thermography as an aid to the clinical lameness evalu-
ation. Vet Clin North Am Equine Pract 1991;7:311-38.
[6] Weinstein AS, Weinstein G. A review of 500 patients with low back
complaints: comparison of five clinically accepted diagnostic
modalities. In: Proceedings of the 1st Meeting of the Academy of
Neuromuscular Thermography 1985; Postgrad Med; special ed
1986:40-3.
[7] Green J, Coyle M, Becker C, Reilly A. Abnormal thermographic
findings in asymptomatic volunteers. Thermology 1986;2:13-5.
[8] Turner TA. Diagnostic thermography. Vet Clin North Am Equine
Pract 2001;17:95-113.
[9] Turner TA, Waldsmith JK, Wilson JH. How to assess saddle fit in
horses. Proceedings Am Assoc Equine Pract 2004;50:196-201.
[10] Wright B, Rietveld G, Lawlis P. Body condition scoring of horses.
OMAFRA Factsheet 98-093, Jan 1999 1-5. Available at: http://www.
omafra.gov.on.ca/english/livestock/horses/facts/98-101.pdf.
[11] De Cocq P, Clayton HM, Terada K, Muller M, van Leeuwen JL.
Usability of normal force distribution measurements to evaluate
asymmetrical loading of the back of the horse and different rider
positions on a standing horse. Vet J 2009;181:266-73.
[12] Harman J. Tack and saddle fit. Vet Clin North Am Equine Pract
1999;15:247-61.
[13] De La Corte FD, Mikail S. Thermographic patterns of the thoracic-
lumbar regions in 51 Warmblood horses. In: Proceedings of the
8th Congress of the World Equine Veterinary Association. Buenos
Aires, Argentina: Asociación Argentina de Veterinaria Equina; 2003.
p. 230.
[14] Peham C, Schobesberger H. Influence of the load of a rider or of
a region with increased stiffness on the equine back: a modeling
study. Equine Vet J 2004;36:703-5.
[15] Peham C, Licka T, Schobesberger H, Meschan E. Influence of the rider
on the variability of the equine gait. Hum Mov Sci 2004;23:663-71.
[16] Fruehwirth B, Peham C, Scheidl M, Schobesberger H. Evaluation of
pressure distribution under an English saddle at walk, trot and
canter. Equine Vet J 2004;36:754-7.
[17] Jeffcott LB, Holmes MA, Townsend HG. Validity of saddle pressure
measurements using force sensing array technologydpreliminary
studies. Vet J 1999;158:113-9.
[18] Meschan EM, Peham C, Schobesberger H, Licka TF. The influence of
the width of the saddle tree on the forces and the pressure distri-
bution under the saddle. Vet J 2007;173:578-84.
[19] Kotschwar AB, Baltacis A, Peham C. The influence of different saddle
pads on force and pressure changes beneath saddles with excessively
wide trees. Vet J 2010;184:322-5. doi:10.1016/j.tvjl.2009.02.018.
[20] Alves AL, Fonseca BP, Thomassian A, Nicoletti JL, Hussni CA,
Silveira AB. Lombalgia em eqüinos. Bras J Vet Res Anim Sci
2007;44:191-9.
[21] Turner TA. Back problems in horses. Proc Am Assoc Equine Pract
2003;49:71-4.
[22] Martin BB, Klide A. Diagnosis and treatment of chronic back pain in
horses. Proc Am Assoc Equine Pract 1997;43:310-1.
[23] Haussler KK, Stover SM, Willits NH. Pathologic changes in the
lumbosacral vertebrae and pelvis in Thoroughbred horses. Am J Vet
Res 1999;60:143-53.