Open Access

Prevalence of clinical findings at examinations of young Swedish warmblood ridinghorses

  • Lina Jönsson1Email author,
  • Lars Roepstorff2,
  • Agneta Egenvall3,
  • Anna Näsholm1,
  • Göran Dalin2 and
  • Jan Philipsson1
Acta Veterinaria Scandinavica201355:34

DOI: 10.1186/1751-0147-55-34

Received: 22 October 2012

Accepted: 2 February 2013

Published: 18 April 2013

Abstract

Background

Soundness of an individual horse is important for animal welfare and ownereconomy. However, knowledge of health status in normal horse populations islimited due to lack of systematic health recordings. The aim of theinvestigation was to study the prevalence of veterinary clinical findings in4-5-year-old Swedish warmblood riding horses, and their influence on overallhealth scores, where associations to future longevity has beenindicated.

Results

The prevalence of clinical findings in 8,281 horses examined during1983–2005 was studied according to a standardised protocol and relatedto overall health scores in linear statistical models. Effects of sex, age,examination event and changes over time were included. In total, 49% of thehorses had clinical findings of medical health (MED), 42% in hooves (HOOF)and 74% of palpatory orthopaedic health (PALP). However, only 6%, 3% and 24%had moderate or severe findings, of MED, HOOF and PALP, respectively.Flexion test reactions were reported in 21% of the horses (5%moderate/severe), heavily influencing the overall score (H2). One fifth ofthese horses also had findings of unprovoked lameness while 83% had PALPfindings (44% with moderate/severe findings). Acute clinical signs, i.e.heat or soreness, had a large influence on the H2 score but were rare,whereas more common clinical findings had smaller effects on overall health.Large variations in recorded health results were observed among events. Adecrease in findings has occurred since 1983, in particular for PALPfindings.

Conclusions

Results of occurrence and relevance of evaluated clinical findings could beused for advice on preventive actions to keep horses sound, and possibly forbenchmarking, and genetic evaluation of health traits. The distinct effectof event on recorded clinical findings emphasises that further harmonisationof veterinary examinations are desirable.

Keywords

Medical health Orthopaedic health Flexion test Hoof Clinical signs

Introduction

Musculoskeletal disorders are the predominant cause of culling horses. In Swedenbetween 55 and 70% of cullings [1, 2] and in the German Warmblood 61% of insurance claims concerned movementrelated diseases [3]. Further, 60% of former Hanoverian auction horses had at least one periodof lameness causing a relevant interruption in training, and 35% had multipleperiods of lameness in the years following the auction [4]. At the same time, soundness is the most important trait when marketing ahorse [5]. Only if sound, talents for performance are considered. Thus, healthstatus is highly important as it affects animal welfare and the utility andsaleability of the horse. However, reports regarding prevalence of clinical findingsin normal horse populations are scarce. Thus, it is of great interest to map thegeneral health status of the sport horse.

The Swedish Riding Horse Quality Test (RHQT) was introduced in 1973 and includesevaluations of health, conformation, gaits and jumping. The RHQT includes broken-in4-year-old warmblood riding horses of both genders and 5-year-old mares that had afoal the previous year. The test is designed to fit all riding horses regardless oftalent and no prior qualifications are required. The young age of participatinghorses suggests a limited environmental effect of training on health results. Testlocations are distributed throughout the country. The RHQT performance scores havebeen found to be good early predictors of future performance and to function asbasis for breeding value estimation [6]. In 1973–1986, low overall health scores at the RHQT weresignificantly related to early future culling (n = 1,815) [7]. In 1994 the health examination was considered an important motive forparticipation at the RHQT by 66% of participants. Further, the health examinationobtained a score of 4.3 out of 5 regarding fulfillment of its objective [8]. However, specific health information, besides the overall scores, haspreviously not been evaluated.

The aim of the investigation was to study the prevalence of clinical findings andtheir influence on the overall assessment of health made by examining veterinariansin an extensively recorded population of young sport horses.

Material and methods

Material

Health information was obtained from RHQT events for horses eligible for theSwedish Warmblood (SWB) studbook during 1983–1984 and 1988–2005.Data after that period was not included as the examination regime was slightlychanged. The data from 1985–1987 had not been saved. For protocols ofhealth examinations see Figures 1 and 2, respectively. All horses at an event were examined in the sameenvironment by one veterinarian for medical health (MED) and hoof shape &quality (HOOF), constituting health examination 1. Horses were examined byanother veterinarian for palpatory orthopaedic health (PALP) and locomotionincluding flexion tests (LOCO), constituting health examination 2. Theexaminations targeted clinical findings in specified anatomical locations of thehorse, which together represent overall health status. Specific clinicalfindings were scored 0–3 (0: no clinical finding, 1: minor, 2: moderate or3 severe clinical finding). Further, overall health scores (H1 and H2) between 1(very poor) and 10 (excellent) were given by each examiner. Examiners wereexperienced horse practitioners with compulsory training in judging regimes ofthe RHQT. Examinations were performed using palpation, auscultation withstethoscopes and flexion tests. All horses were documented for health status onidentical protocols. Opportunities were given to include free text comments butwere mainly used for clarification.
https://static-content.springer.com/image/art%3A10.1186%2F1751-0147-55-34/MediaObjects/13028_2012_Article_872_Fig1_HTML.jpg
Figure 1

Protocol used for health examination 1. Including separateclinical findings examined for and scale of scoring.

https://static-content.springer.com/image/art%3A10.1186%2F1751-0147-55-34/MediaObjects/13028_2012_Article_872_Fig2_HTML.jpg
Figure 2

Protocol used for health examination 2. Including separateclinical findings examined for and scale of scoring.

Method and data structure

Available paper protocols of 9,053 horses were scanned to a digital picture, thendigitally read to classify contents using the neural network toolbox inMatlab®a, followed by manual validation. Identity wasconfirmed in the SWB database. In total, 8,281 horses had complete healthexaminations and a confirmed identity. In the available data most horses were4 years old (n = 7,788), and 493 were 5 years old. Datarepresented 3,879 mares and 4,402 males (predominantly geldings). Each year4–14 events were included, generating 193 events, each with 4 to 129evaluated horses. At these events 57 and 54 different veterinarians wereemployed for health examination 1 and 2 respectively. Approximately, a randomhalf of conducted events were included in the study. During the years1979–2001 a total of 98 000 SWBs were born. If assuming 7% loss of horseseach year due to death, export etc. [1], 23% of available horses participated in the RHQT during1983–2005. The 8,281 studied horses represented 53% of tested horses inthat time period.

Statistical analysis

Descriptive statistics and statistical analyses were performed using thestatistical package SAS®b. The overall health status of MED,HOOF, PALP and LOCO, respectively, was studied as the sum of clinical findings,including severity (0–3), at each type of examination. Thus, the sum ofclinical findings increased both with number and severity of findings. Summedvalues included both bilateral clinical findings. Corresponding summation ofPALP findings was performed in 4 groups of systemic location, i.e. joints,muscles, tendons & suspensory ligaments, and skeleton & hoof cartilageand into 5 groups of clinical sign i.e. effusion, heat, soreness, swelling,stiffness/atrophy, irrespective of whether findings were bilateral or not.Clinical signs were also grouped into primarily acute (heat, soreness) orprimarily chronic (effusion, swelling and stiffness/atrophy) findings.

High overall H1 and H2 scores were used frequently resulting in skeweddistributions (Additional file 1). However, data werekept untransformed in subsequent analyses because extensive transformationtrials showed similar fixed effect results and distribution of residualvalues.

Fixed effects of age, sex and event on health status results were analysed usingGeneral Linear Models (GLM). Effects of separate clinical findings on H1 or H2were estimated as class effects of ‘minor’ respectively‘moderate or severe’ finding compared to no finding using GLM. Theemployed model included the fixed effects of age, sex, event, and clinicalfinding. The separate clinical findings were included one at a time andtherefore 39 analyses were performed for health examination 1 and 123 analysesfor health examination 2. Non-used observations were excluded and right and leftobservations were pooled irrespectively of whether findings were bilateral ornot. Differences in H2 scores between horses with different categories ofclinical findings were evaluated with significance tests of differences betweenleast squares means using GLM.

Results

In total, 49% of studied horses had clinical findings of MED, 42% of HOOF and 74% ofPALP. However, only 6%, 3% and 24% had moderate or severe findings of MED, HOOF andPALP, respectively. Flexion test reactions were reported in 21% of horses (5%moderate/severe). The proportions of horses with clinical findings in HOOF, PALPand/or LOCO examinations are shown in Figure 3,accompanied by mean H2 scores. The H2 score gradually decreased as the number ofexamination types with reported clinical findings increased, where LOCO had thelargest separate influence. Differences in H2 scores were significant(p < 0.05) between all groups, except for the difference between‘no findings’ vs. ‘only HOOF findings’ and between‘only LOCO findings’ vs. ‘HOOF and LOCO findings’.Indicating that, for example if findings occur both during PALP and LOCOexaminations, this results in a significantly lower H2 score compared to horses withonly findings within the LOCO examination. Generally, H2 scores under 8 were foundin horses with LOCO findings. Clinical findings of MED were present in 40-60% ofhorses in each group in Figure 3 (results not shown).Additional file 1 shows that most horses (84%) werecharacterised as healthy regarding the H2 score (score 8–10) and theproportion was even higher for H1 scores (97%). Mean H1 and H2 score was 9.42 and8.78, with medians of 10 and 9 respectively.
https://static-content.springer.com/image/art%3A10.1186%2F1751-0147-55-34/MediaObjects/13028_2012_Article_872_Fig3_HTML.jpg
Figure 3

Proportion of horses with clinical findings in one or several types ofexamination. Hoof examination (HOOF) in stripe pattern (total 42%),palpatory orthopaedic examination (PALP) in green shades (total 74%), andlocomotion examination (LOCO) in red shades (total 21%), ordered after meanorthopaedic health score (H2) read clockwise, n = 8,281.

Clinical findings occurring in ≥ 2% of examined horses withinhealth examination 1 and 2 are presented in Table 1.Within health examination 1 clinical skin and mouth findings were the most frequentMED findings. The most common clinical HOOF findings concerned hoof wall quality.Findings of hoof size and shape were also quite common with emphasis onasymmetrical, small and flat hooves. Among horses with clinical MED findings themean sum of findings, including severity, was 1.7 with a maximum of 9. Thecorresponding results for HOOF were 1.9 and 7. For health examination 2 most commonclinical PALP findings generally concerned effusions or swellings in joints,accompanied by flexion test reactions.
Table 1

Most frequent clinical findings (≥2%) during health examination 1and 2, and their estimated effect on respective overall score(H1/H2)

 

Clinical finding (%)

Effect of finding on overall score

     

Minor

Moderate or severe

 
 

Minor

Moderate

Severe

Total

Estimate1

p=

Estimate1

p=

R2(%)2

Health examination 1

         

Medical health

Skin, general

13.4

0.4

0.1

14.0

−0.24

***

−1.07

***

1.60

Mouth general (e.g. wolf teeth) 3

11.2

0.2

0.0

11.4

0.02

ns.

−0.15

ns.

0.01

Local skin swelling in saddle area

7.9

1.5

0.1

9.5

−0.46

***

−0.89

***

3.56

Pododermatitis in pastern (mud fever)

6.1

1.2

0.2

7.5

−0.40

***

−1.15

***

3.61

Parrot mouth 3

5.7

0.6

0.1

6.3

−0.11

**

−0.51

***

0.30

Lymph nodes, general

3.6

0.7

0.0

4.3

−0.66

***

−1.03

***

1.86

Skin sarcoids

3.2

0.4

0.0

3.5

−0.47

***

−0.90

***

1.36

Mucous membrane hyperaemia

2.6

0.1

0.0

2.7

−0.11

*

−0.28

ns.

0.04

Dental hooks 3

2.2

0.2

0.0

2.4

−0.16

**

−0.56

**

0.14

Tilted vulva 4

2.3

0.1

0.0

2.4

0.06

ns.

−0.39

ns.

0.02

Hollow/flaccid vulva 4

2.0

0.2

0.0

2.2

−0.14

ns.

−0.32

ns.

0.07

Overweight

1.9

0.1

0.0

2.0

−0.02

ns.

−0.13

ns.

0.00

Hoof examination

         

Hoof wall cracks

11.1

0.6

0.0

11.7

−0.56

***

−0.90

***

4.33

Hooves, general

10.5

0.2

0.0

10.7

−0.23

***

−0.74

***

0.76

Asymmetrical hooves

6.8

0.2

0.0

7.1

−0.24

***

−1.04

***

0.74

Small hooves

6.3

0.3

0.0

6.6

−0.07

*

−0.04

ns.

0.03

Flat hooves

6.2

0.4

0.0

6.6

−0.47

***

−0.88

***

1.90

Underrun heels

5.2

0.2

0.0

5.4

−0.35

***

−1.26

***

1.32

Infection of the frog (thrush)

2.9

0.4

0.0

3.4

−0.39

***

−1.28

***

0.90

Contracted heels

3.3

0.1

0.0

3.4

−0.68

***

−1.00

***

2.08

Mediolateral imbalance

2.9

0.1

0.0

3.0

−0.38

***

−1.30

***

0.75

Poor hoof wall quality

2.6

0.2

0.0

2.8

−0.79

***

−1.60

***

2.68

Health examination 2

         

Palpatory orthopaedic health

Effusion

         

Hindlimb digital flexor tendon sheath

23.5

2.8

0.3

26.7

−0.17

***

−0.88

***

1.20

Tarsocrural joint

13.6

3.7

0.6

17.9

−0.40

***

−0.95

***

2.39

Metatarsophalangeal joint

12.1

3.3

0.5

15.9

−0.47

***

−1.09

***

2.66

Middle carpal/carpometacarpal joint

8.8

3.9

0.5

13.2

−0.48

***

−1.10

***

2.81

Metacarpophalangeal joint

7.3

2.6

0.2

10.2

−0.60

***

−1.23

***

2.76

Femoropatellar joint

4.2

3.5

0.6

8.3

−0.30

***

−1.17

***

2.19

Forelimb digital flexor tendon sheath

3.2

0.5

0.0

3.7

−0.40

***

−1.03

***

0.49

Forelimb distal interphalangeal joint

3.0

0.5

0.0

3.5

−0.42

***

−1.10

***

0.56

Medial part of femorotibial joint

1.7

0.5

0.1

2.3

−0.55

***

−1.07

***

0.56

Swelling

         

Metacarpus, proximal

12.6

3.0

0.4

16.0

−0.12

**

−0.46

***

0.41

Metatarsus, proximal

5.1

1.0

0.3

6.4

−0.03

ns.

−0.38

**

0.10

Metacarpus, distal

3.0

0.6

0.1

3.7

−0.16

*

−0.24

ns.

0.06

Metatarsus, distal

2.5

0.6

0.1

3.2

−0.15

ns.

−0.77

***

0.25

Metacarpophalangeal joint

1.6

1.1

0.2

2.9

−0.27

***

−1.06

***

0.75

Metatarsophalangeal joint

1.8

0.7

0.3

2.8

−0.24

*

−0.83

***

0.40

Tarsometatarsal/centrodistal joint

1.8

0.7

0.2

2.6

−0.49

***

−1.57

***

1.18

Tip of the hock

1.9

0.5

0.1

2.5

−0.41

*

−1.01

***

0.17

Atrophy/Stiffness

         

Croup/hamstrings muscles

3.3

0.8

0.1

4.2

−0.92

***

−1.69

***

2.43

Quadriceps muscles

1.6

0.4

0.1

2.1

−0.86

***

−1.62

***

1.21

Soreness

         

Back muscle/spinous processes

2.9

0.6

0.0

3.5

−0.76

***

−1.68

***

1.63

Locomotion examination

         

Hindlimb post flexion test movements in trot5

10.2

2.4

0.5

13.1

−1.73

***

−3.05

***

11.50

Forelimb post flexion test movements in trot5

9.0

1.7

0.2

10.8

−1.78

***

−3.10

***

8.97

Hindlimb unprovoked movements in trot

2.4

0.2

0.0

2.7

−2.06

***

−3.61

***

6.34

1Estimate of class effect on overall score compared to nofinding, from single trait analyses.

2R2 values from single trait analysis when thesummed R2 from fixed effects has been removed (35% for H1 and27% for H2).

3Missing information for 7 horses, n = 8,274.

4Out of 3879 examined mares.

5Missing information for 26 horses,n = 8,255.

ns = nonsign* = p < 0.05** = p < 0.01*** = p < 0.001.

Clinical finding (%) indicates prevalence of findings.N = 8,281 examined horses if nothing else indicated.

Locomotion examination (LOCO)

Few horses showed signs of lameness in unprovoked walk (1%) and trot (4%),including both fore- and hindlimbs. However, flexion test reactions were foundin 11% and 13% of horses in fore- and hindlimbs, respectively. Among horses withflexion test reactions, 20% showed unprovoked lameness. The highest sum of LOCOfindings, including severity, was 21 with a mean of 2.0 among horses with LOCOfindings. As seen in Figure 4 the number and severityof PALP findings increased with increasing degree of flexion test reactions, forall types of clinical signs.
https://static-content.springer.com/image/art%3A10.1186%2F1751-0147-55-34/MediaObjects/13028_2012_Article_872_Fig4_HTML.jpg
Figure 4

Mean sum of palpatory orthopaedic health (PALP) findings, includingseverity, in horses with different degrees of flexion testreactions.

Palpatory orthopaedic examination (PALP)

The prevalence of each clinical sign of PALP in fore- and hindlimbs are presentedin Figures 5 and 6, togetherwith the total percentage of examined horses with findings at each location.Generally, more findings were reported in hindlimbs. Additional analyses(results not shown) showed that bilateral findings (all clinical signs included)were generally found in 30% of horses with findings at each examined location.Locations with >50% bilateral findings in hindlimbs/croup were; backmuscles/spinous processes (81%), digital flexor tendon sheaths (78%), growthplates (65%), metatarsophalangeal joints (62%) and hoof cartilage (60%).Corresponding forelimb/shoulder locations were; growth plates (83%), distalinterphalangeal joint (65%), metacarpophalangeal joints (61%), digital flexortendon sheaths (58%) and shoulder muscles (57%). In examined horses mostclinical findings were evenly distributed between left and right side at eachlocation (results not shown). Among horses with PALP findings the mean sum offindings including severity was 4.4 with a maximum of 36. Further, 98% and 7% ofhorses with PALP findings had chronic and acute findings respectively; 5% showedsigns of both categories.
https://static-content.springer.com/image/art%3A10.1186%2F1751-0147-55-34/MediaObjects/13028_2012_Article_872_Fig5_HTML.jpg
Figure 5

Prevalence of clinical signs of palpatory orthopaedic health (PALP) inexamined forelimb/shoulder locations. Total prevalence of horseswith clinical findings at respective location presented below.

https://static-content.springer.com/image/art%3A10.1186%2F1751-0147-55-34/MediaObjects/13028_2012_Article_872_Fig6_HTML.jpg
Figure 6

Prevalence of clinical signs of palpatory orthopaedic health (PALP) inexamined hindlimb/croup locations. Total prevalence of horseswith clinical findings at respective location presented below.

Effects of fixed factors

A significant effect of event (p < 0.0001) and gender(p < 0.05) was present in all types of examination where mareshad slightly lower overall scores and more clinical findings, except for HOOF,where no significant gender differences were found. The age effect wasnon-significant for types of health examination but significant(p < 0.05) for clinical signs of soreness where 5-year-old mareshad more findings, in particular for soreness in back muscles/spinous processes.According to the coefficient of determination (R2), fixed effectsexplained 35% of H1 and 27% of H2 variation, primarily due to variation amongevents.

Effects of specific clinical findings on overall scores

Clinical findings with the largest effect on overall scores (≤ −1.5points) are found in Table 2, ordered bymoderate/severe finding effects on overall health scores (H1/H2) within eachgroup of examination. In particular regarding effects on H2 scores several rarebut acute PALP findings were found to have the largest influence. Additionalanalyses revealed that the H2 score decreased significantly between horses with’no finding’, ‘chronic finding’, ‘acutefinding’ and ‘both acute & chronic finding’ respectively,in decreasing order (p < 0.05 between all groups). Further, LOCOfindings were found to have a large influence on the H2 score. Among horses withany type of flexion test reaction 36% obtained an H2 score of ≤6 with amean of 7.0, compared to a mean of 9.2 among horses without flexion testreactions, p < 0.0001. For horses with severe flexion testreactions 91% had a score of ≤6 (mean: 4.8).
Table 2

Clinical findings with largest effect (≤ −1.5 points) onrespective overall score (H1/H2) at health examinations 1 and 2

 

Effect of clinical finding on overall score

 

Minor

Moderate or severe

 
 

Estimate1

p=

Prevalence (%)

Estimate1

p=

Prevalence (%)

R2(%)2

Health examination 1

Medical health

       

Spontaneous or provocable cough

−1.16

***

1.20

−2.04

***

0.20

2.80

Heart, general

−1.05

***

0.90

−1.93

***

0.02

1.50

Nasal cavity, general

−0.24

**

1.30

−1.76

***

0.05

0.27

Hoof examination

       

Poor hoof wall quality

−0.79

***

2.60

−1.60

***

0.20

2.68

Health examination 2

Palpatory orthopaedic health

       

Soreness

       

Tarsocrural joint

.

.

0.00

−6.72

***

0.01

0.30

Hindlimb superficial digital flexor tendon

−0.01

ns.

0.02

−6.49

***

0.01

0.30

Forelimb superficial digital flexor tendon

−1.27

ns.

0.04

−6.49

***

0.01

0.30

Hindlimb deep digital flexor tendon

−1.17

ns.

0.01

−6.49

***

0.01

0.30

Forelimb deep digital flexor tendon

−2.38

**

0.02

−4.08

***

0.02

0.30

Quadriceps muscles

−0.07

ns.

0.07

−2.08

ns.

0.01

0.03

Shoulder/elbow joint

−1.15

ns.

0.05

−1.83

ns.

0.01

0.05

Back muscles/spinous processes

−0.76

***

2.93

−1.68

***

0.60

1.63

Heat

       

Tarsocrural joint

2.04

ns.

0.01

−5.06

***

0.01

0.20

Metacarpus, proximal

−0.19

ns.

0.13

−3.07

*

0.01

0.10

Forelimb deep digital flexor tendon

.

.

0.00

−1.71

ns.

0.01

0.02

Metacarpophalangeal joint

−0.66

ns.

0.02

−1.50

ns.

0.02

0.03

Effusion

       

Shoulder/elbow joint

1.21

ns.

0.02

−4.61

***

0.01

0.20

Tarsometatarsal/centrodistal joint

−0.70

ns.

0.11

−1.55

*

0.04

0.07

Stiffness/Atrophy

       

Hindlimb pastern joint

0.71

ns.

0.01

−4.03

***

0.01

0.10

Femoropatellar joint

−1.49

ns.

0.01

−2.43

*

0.01

0.05

Medial part of femorotibial joint

.

.

0.00

−2.43

*

0.01

0.04

Middle carpal/carpometacarpal joint

−1.21

ns.

0.01

−2.34

*

0.01

0.04

Shoulder/elbow joint

0.83

ns.

0.02

−2.11

ns.

0.01

0.04

Metacarpophalangeal joint

−0.21

ns.

0.24

−1.90

***

0.06

0.12

Forelimb hoof cartilage

−0.27

ns.

0.25

−1.77

ns.

0.01

0.03

Croup/hamstrings muscles

−0.92

***

3.33

−1.69

***

0.86

2.43

Quadriceps muscles

−0.86

***

1.63

−1.62

***

0.51

1.21

Swelling

       

Back muscles/spinous processes

−0.90

ns.

0.06

−3.40

***

0.04

0.20

Forelimb hoof cartilage

−0.44

ns.

0.29

−2.50

***

0.06

0.23

Tarsometatarsal/centrodistal joint

−0.49

***

1.76

−1.57

***

0.81

1.18

Locomotion examination

       

Forelimb unprovoked movements in trot

−1.89

***

1.47

−4.27

***

0.09

3.30

Hindlimb unprovoked movements in trot

−2.06

***

2.40

−3.61

***

0.27

6.34

Forelimb post flexion test movements in trot3

−1.78

***

8.96

−3.10

***

1.90

8.97

Hindlimb post flexion test movements in trot3

−1.73

***

10.20

−3.05

***

2.94

11.50

Forelimb unprovoked movements in walk

−0.61

**

0.43

−3.02

***

0.03

0.20

Hindlimb unprovoked movements in walk

−1.50

***

0.60

−1.97

***

0.06

0.77

1Estimate of class effect on overall score compared to nofinding, from single trait analyses.

2R2 values from single trait analysis when thesummed R2 from fixed effects has been removed (35% for H1and 27% for H2).

3Missing information for 26 horses,n = 8,255.

ns = nonsign* = p < 0.05** = p < 0.01*** = p < 0.001.

Ordered after decreasing magnitude of estimates for moderate/severefindings within examination groups. N = 8,281 if nothingelse indicated.

Synergy effects of PALP and LOCO findings

Within horses showing flexion test reactions 83% had PALP findings (44%moderate/severe findings). The prevalence of LOCO findings increased amonghorses with acute PALP findings (30%) compared to horses with chronic findings(23%) and horses without PALP findings (13%). The most common PALP finding amonghorses with flexion test reactions were effusions in the digital flexor tendonsheath in hindlimbs (25%), metacarpo- and metatarsophalangeal joints (18% and24%, respectively), tarsocrural joint (23%), middle/lower part of carpus, i.e.middle carpal joint and carpometacarpal joint (21%) and femoropatellar joint(17%), swelling in proximal part of metacarpus (16%), atrophied croup/hamstringsmuscles (11%), atrophied quadriceps muscles (7%), and swelling intarsometatarsal/centrodistal joint (7%). All mentioned PALP prevalences weresignificantly elevated compared to horses without LOCO findings, except foreffusion in the hindlimb digital flexor tendon sheaths and swelling in proximalpart of metacarpus. Soreness in back muscles/spinous processes was elevated to5% in horses with LOCO findings.

Common clinical findings among different overall scores

Figure 7 illustrates percentages of horses withclinical findings within each H2 score, among the top 10 most common findingsassociated to one or several H2 scores (16 findings in total), from PALP andLOCO examinations. A steady overall increase in percent horses with clinicalfindings was seen as the H2 score decreased, which was also true for a majorityof presented individual clinical findings. An exception was effusion in thehindlimb digital flexor tendon sheaths that were similar for all H2 scores. Aslight trend deviation was seen for horses with H2 score ≤5 and 6 wheresome PALP findings decreased simultaneously as LOCO findings increased,indicating LOCO status to have a distinct effect on H2 regardless of PALPstatus. The most frequent clinical finding of horses with H2 scores ≤5that increased compared to other H2 scores were: flexion test reactions (fore-and hindlimbs), effusions in metacarpo- and metatarsophalangeal joints,middle/lower part of carpus, femoropatellar- and tarsocrural joint andunprovoked lameness in trot. The overall results of over 100% mean that severalclinical findings generally were present in each horse. The same trend ofincrease in findings with decreasing H2 score was seen when including all PALPand LOCO findings and their rate of severity (Figure 8).This figure also shows that effusions were the most common clinical signregardless of given H2 score, followed by LOCO findings that both increased withdecreasing H2 score. Further, most PALP findings were found at joint relatedlocations. Generally, less than one minor finding was present in horses with H2score 10 and horses with score 9 had one moderate (or two minor) finding(s). Thesame relationship was found between H1 scores and health examination 1 findings(results not shown). Clinical findings of HOOF and MED increased equally withdecreasing H1 score, where horses with H1 score ≤5 generally had moresigns of inflammation.
https://static-content.springer.com/image/art%3A10.1186%2F1751-0147-55-34/MediaObjects/13028_2012_Article_872_Fig7_HTML.jpg
Figure 7

Variation in extent of clinical findings between horses with differentorthopaedic health scores (H2). Including the 10 most commonfindings associated to each H2 score (16 findings in total) of palpatoryorthopaedic health (PALP) and locomotion (LOCO).

https://static-content.springer.com/image/art%3A10.1186%2F1751-0147-55-34/MediaObjects/13028_2012_Article_872_Fig8_HTML.jpg
Figure 8

Mean sum of clinical findings, including severity, of clinical signs(A) and systemic locations (B) of palpatory orthopaedic health(PALP). Results compare horses with each orthopaedic healthscore (H2), diagram A is complemented with locomotion examination(LOCO) results.

Trend in clinical findings and overall health scores

A decreasing trend in sum of PALP findings (including severity) was seen duringthe test period (Additional file 2). Similar trends,but not as distinct, appear for MED, HOOF and LOCO. In the beginning of the1990’s H1 and H2 scores generally increased, which to some extent wasconfirmed by a decrease in sum of clinical findings during the same time, inparticular for PALP.

Discussion

Studies on equine population health are scarce as systematic health recording is notcommon practice in the horse industry. The regime of milk cow health recording inSweden and some other countries has proved extremely useful for improving generalpopulation health and longevity. In contrast, the lack of systematic healthregistration in horses has limited the knowledge of health status in horsepopulations, and genetic evaluation of predisposition for unsoundness is absent. Theusefulness of systematic health recording of equine hospital data was alsodemonstrated regarding studies of osteochondrosis (OC) [9]. Health recordings from the RHQT are globally unique as a large number ofhealth traits have been systematically recorded during several years. Presentedresults expectedly represent the average young riding horse population in Swedenfairly well, with the exception that the worst cases will likely not participate ifowners are aware of the unsoundness, e.g. lame horses. Failure to participatebecause of physical problems at such a young age is likely to be due to acuteinjuries or contagious diseases, which are expected to occur randomly.

Equine health status has previously mostly been studied on smaller data sets, or ondata diverging from the typical riding horse population by e.g. old age, healthstatus or selection for talent prior to auctions [4, 10]. Present results generally show lower prevalences compared to studies ofolder horses, which is expected as clinical findings commonly increase with age [10, 11]. Further, differences may be due to variation among examiners orpopulations. The number and severity of clinical findings has generally decreasedover the years, both within examiner (results not shown) and overall. This mightrepresent a combination of better general health in the population and improvedhealth care. Also, increased owner awareness over time, might have influencedtowards a lower participation rate in horses with substantial health lesions.However, a recent British study found that generally horse owner awareness of healthabnormalities in their horses is low compared to veterinary examination results [12].

The large extent of reported clinical findings (Figure 3)in relation to mean H2 scores indicates that examiners aimed to include everyfinding in the protocol, also minor ones. Generally, overall scores were highsuggesting findings to have a small assessed effect on overall health, at this youngage. An important distinction was made between recording findings as objectively aspossible on one hand, and on the other hand assessing overall health status into anoverall score. Both have a unique informative value. According to Tables 1 and 2, findings with the largestseparate effects on overall health status i.e. findings of soreness or heat, wereoften serious in character but rare in the population, whereas more frequentlyoccurring findings i.e. effusions and swelling, had a smaller individual effect.During health examination 2 only 6 separate clinical findings both had aprevalence ≥ 2% and an effect of moderate/severe findings of−1.5 points or stronger: flexion test reactions in both fore- and hindlimbs,sore back muscles/spinous processes, atrophied croup/hamstrings- and quadricepsmuscles, and swelling in tarsometatarsal/centrodistal joint. For health examination1 only findings of poor hoof wall quality met the same criteria. These findingsmight represent what is most important to consider in the future, if only focusingon separate clinical findings.

The effect of event was highly significant on health results, which is also true forRQHT performance results [6], and for examiner effect in other health studies [11, 13, 14]. Factors, apart from examiner, included in the event effect were e.g.examination site and time dependent changes. Events were held within a few weeks inearly autumn each year, thus season was quite equal between events, apart fromgeographical differences. A substantial part of the event effect was due toexaminer, which had a significant effect on its own in separate analyses (resultsnot shown), however, not explaining all event variation. RHQT veterinarians attendcompulsory training to standardise their examinations, but present results suggestthat further efforts are needed. In the meantime means to analytically correctexamination results for examiner are necessary. In present analyses all systematicdifferences between events, including effects of examiner, have been adjustedfor.

A desirable improvement in the studied recording scheme would be to exclude HOOF fromthe H1 score, and rather keep it separate. Originally, HOOF was included within theH1 examination in order for both examinations to take equally long time. Horses withclinical hoof findings obtained slightly lower H2 scores, even if HOOF was notincluded in that examination (significant differences in 2 out of 4 groups,Figure 3). Radiographic equipment or endoscope wasnot used as the horses, usually about 40, have to complete the test in one day.Also, horses are judged for performance shortly after the veterinary examination andcannot be sedated. The effect of rider was not accounted for in this study. A surveyamong RHQT participants has shown that rider experience of broken-in horsessignificantly influenced the H2 score [8]. However, a majority of riders only have one horse, thus the effects ofrider and horse, respectively, are usually confounded.

Analysis of a limited RHQT data set showed that horses with H2 score ≤5generally had twice as high risk of early culling compared to those with score≥9 [7]. The present study could further reveal that the most common separatefindings that increased in horses with H2 ≤5 were LOCO results and effusionsin metacarpo-/metatarsophalangeal joints and in middle/lower part of carpus,tarsocrural- and femoropatellar joints. This information may be used in veterinarypractice when assessing relevance of clinical findings or for future preventiveactions. Commonly affected locations coincide well with statistics of insuranceclaims for morbidity in Swedish horses [15], where fetlock lesions were ranked as 1st, 11th and12th most common causes of culling. Lesions in carpal, femoropatellarand tarsocrural joints were also frequent causes of culling. Further, in HanoverianWarmbloods fetlock, tarsus, back and tendons were the most frequent locations oflocomotor apparatus disorders [4]. Present results show that horses with low health scores, previouslyshown to have lower longevity in a smaller dataset [7], have more clinical findings than other horses already at a young age.The majority of clinical signs seen are joint effusions and lameness. These types ofclinical findings are commonly also seen in horses with OC lesions or otherdevelopmental disorders. One might suspect that some of these horses coincide. Asclinical signs may be due to either local underlying lesions or, on the other hand,to complex compensatory overloading, the need for whole body health examinations isemphasised. The validity of the flexion test is presently debated due to lack ofstandardisation in amount of force applied and time of provocation [13], and because most clinically sound horses can exhibit minor flexion testreactions depending on intensity of provocation [16]. However, present flexion test results, especially moderate/severereactions, were highly related to number and severity of PALP findings. Further, ithad a large influence on the H2 score, which is related to future longevity [7], suggesting the usefulness of flexion tests.

Previous studies have found a lower risk of early culling, i.e. termination of life,in females compared to males [2; 7]. In present results, females had significantlymore clinical findings and lower overall health score, suggesting health status andculling to be two separate traits with a possible difference that mares have analternative career as broodmares. The results are partly supported by a study ofDutch Warmbloods, where mares had a higher risk of culling from basic dressage.However, no gender differences were found in elite dressage or basic/elite jumping [17]. Similar to the present study, a study of flexion test reactions of 100clinically sound horses indicated that mares obtained significantly more postflexion test reactions compared to geldings [16]. This may suggest mares to be less resistant to the provocation or moreprone to show pain reactions. Indications were found that 5-year-old broodmares hadalmost twice as high risk of exhibiting sore back muscles/spinous processes comparedto 4-year-olds, which was also true if comparing to 4-year-old mares only. Thismight be caused by the strains of pregnancy, or due to a relatively sudden increasein training after weaning the foal, where training intensity is based more on agethan on the actual training level of the horse.

Conclusion

The health examinations from the RHQT have increased the knowledge of the generalhealth status of young SWB horses, including relevance of separate clinicalfindings. Obtained results may serve as benchmarking for health status in youngriding horse populations, and for assessment of future longevity of horses inconnection to insurance and sales examinations, besides as basis for advice onpreventive actions to keep horses sound. Further, the data could be used for geneticevaluation of health traits, or as basis for further studies on clinical relevanceof separate clinical findings on longevity. However, the significant effect of eventemphasises the need for uniform regimes of examination of horses amongveterinarians, not only for research purposes but also for the general horse sectoras insurances and sales are based on these types of examinations.

Endnotes

aThe MathWorks Inc., Natick, Massachusetts, U.S.A.

bSAS institute Inc., Cary, NC, U.S.A.

Abbreviations

H1: 

Overall score, health examination 1 (medical and hooves)

H2: 

Overall score,health examination 2 (orthopaedic)

HOOF: 

Hoof examination

LOCO: 

Locomotionexamination

MED: 

Medical examination

PALP: 

Palpatory orthopaedic examination

RHQT: 

Riding horse quality test

SWB: 

Swedish Warmblood studbook

Declarations

Acknowledgements

The Swedish-Norwegian Foundation for Equine Research is gratefully acknowledgedfor funding of the study. The Swedish Warmblood Association is gratefullyacknowledged for allowing data to be used for the study.

Authors’ Affiliations

(1)
Department of Animal Breeding and Genetics
(2)
Department of Equine Studies
(3)
3Department of Clinical Sciences, Swedish University of Agricultural Sciences

References

  1. Wallin L, Strandberg E, Philipsson J, Dalin G: Estimates of longevity and causes of culling and death in Swedish warmbloodand coldblood horses. Livest Prod S. 2000, 63: 275-289. 10.1016/S0301-6226(99)00126-8.View Article
  2. Egenvall A, Penell JC, Bonnett BN, Olson P, Pringle J: Mortality of Swedish horses with complete life insurance between 1997 and2000: variations with sex, age, breed and diagnosis. Vet Rec. 2006, 158: 397-406. 10.1136/vr.158.12.397.View ArticlePubMed
  3. Clausen VM, Preisinger R, Kalm E: Analyse von Krankenheitsdaten in der Deutschen Warmblutzucht. Zuchtungskunde. 1990, 62: 167-178.
  4. Stock KF, Distl O: Survey on the development of Hanoverian Warmblood horses selected for sale atauction in 1991 to 1998. J Equine Vet Sci. 2005, 25: 210-223. 10.1016/j.jevs.2005.04.004.View Article
  5. Hennessy KD, Quinn KM, Murphy J: Producer or purchaser: Different expectations may lead to equine wastage andwelfare concerns. J Appl Anim Welf Sci. 2008, 11: 232-235. 10.1080/10888700802101023.View ArticlePubMed
  6. Viklund Å, Braam Å, Näsholm A, Strandberg E, Philipsson J: Genetic variation in competition traits at different ages and time periodsand correlations with traits at field tests of 4-year-old Swedish Warmbloodhorses. Animal. 2010, 4: 682-691. 10.1017/S1751731110000017.View ArticlePubMed
  7. Wallin L, Strandberg E, Philipsson J: Phenotypic relationship between test results of Swedish Warmblood horses as4-year-olds and longevity. Livest Prod S. 2001, 68: 97-105. 10.1016/S0301-6226(00)00244-X.View Article
  8. Eliasson C: Samband mellan resultat i kvalitetsbedömning och förberedelse avhästen samt med hästägarens ryttarerfarenhet [Correlation betweenRiding Horse Quality Test results and previous training of the horse and trainerexperience]. 1997, Master thesis No. 182. Swedish University of Agricultural Sciences.Department of Animal Breeeding and Genetics,
  9. Jönsson L, Dalin G, Egenvall A, Näsholm A, Roepstorff L, Philipsson J: Equine hospital data as a source for study of prevalence and heritability ofosteochondrosis and palmar/plantar osseous fragments of Swedish Warmbloodhorses. Equine Vet J. 2011, 43: 695-700. 10.1111/j.2042-3306.2010.00354.x.View ArticlePubMed
  10. Ireland JL, Clegg PD, McGowan CM, McKane SA, Chandler KJ, Pinchbeck GL: Disease prevalence in geriatric horses in the United Kindom: Veterinaryclinical assessment of 200 cases. Equine Vet J. 2012, 44: 101-106. 10.1111/j.2042-3306.2010.00361.x.View ArticlePubMed
  11. Egenvall A, Lönnell C, Johnston C, Roepstorff L: Orthopaedic health status of the horses from 8 riding schools - a pilotstudy. Acta Vet Scand. 2010, 52: 2-8. 10.1186/1751-0147-52-2.View Article
  12. Ireland JL, Clegg PD, McGowan CM, McKane SA, Chandler KJ, Pinchbeck GL: Comparison of owner-reported health problems with veterinary assessment ofgeriatric horses in the United Kingdom. Equine Vet J. 2012, 44: 94-100. 10.1111/j.2042-3306.2011.00394.x.View ArticlePubMed
  13. Keg PR, van Weeren PR, Schamhardt HC, Barneveld A: Variations in the force applied to flexion tests of the distal limb ofhorses. Vet Rec. 1997, 141: 435-438. 10.1136/vr.141.17.435.View ArticlePubMed
  14. Keegan KG, Dent EV, Wilson DA, Janicek J, Kramer J, Lacarrubba A, Walsh DM, Cassells MW, Esther TM, Schiltz P: Repeatability of subjective evaluation of lameness in horses. Equine Vet J. 2010, 42: 92-97. 10.2746/042516409X479568.View ArticlePubMed
  15. Penell JC, Egenvall A, Bonnett BN, Olson P, Pringle J: Specific causes of morbidity among Swedish horses insured for veterinary carebetween 1997 and 2000. Vet Rec. 2005, 157: 470-477.View ArticlePubMed
  16. Busschers E, van Weeren PR: Use of flexion test of the distal forelimb in the sound horse: repeatabilityand effect of age, gender, weight, height and fetlock joint range ofmotion. J Vet Med Assoc. 2001, 48: 413-427. 10.1046/j.1439-0442.2001.00373.x.View Article
  17. Ducro B, Gorissen B, van Eldik P, Back W: Influence of foot conformation on length of competitive life in a DutchWarmblood population. Equine Vet J. 2009, 41: 144-148. 10.2746/042516408X363800.View ArticlePubMed

Copyright

© Jonsson et al.; licensee BioMed Central Ltd. 2013

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative CommonsAttribution License (http://creativecommons.org/licenses/by/2.0), whichpermits unrestricted use, distribution, and reproduction in any medium, provided theoriginal work is properly cited.

Advertisement