SEARCH

SEARCH BY CITATION

Keywords:

  • horse;
  • reliability;
  • rehabilitation;
  • physical therapy;
  • veterinary medicine;
  • lameness

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflicts of interest
  9. Manufacturers' addresses
  10. References

Reasons for performing study: Simple objective assessment tools are essential to monitor the clinical efficacy of therapeutic interventions used in equine orthopaedics and rehabilitation. In human medicine, goniometry is a validated tool to quantify restrictions in joint range of motion (ROM); however, the technique is not validated in horses.

Objectives: To validate 2 different goniometry techniques for the measurement of passive flexion of the fetlock, carpus and hock by examining; 1) the intra- and inter-tester reliability; 2) the differences between 2 goniometry techniques and 3) differences between standing and anaesthetised horses.

Methods: The study is composed of three parts: 1) the intra- and inter-tester reliability was examined on 10 horses, where each horse was assessed by 5 pairs of testers measuring ROM with a universal goniometer; 2) the differences between 2 goniometry techniques were examined on 14 horses, each assessed by 2 investigators (either working in pairs with one investigator holding the limb and the other measuring the joint angle, or working individually at the same time holding and measuring); 3) on 6 horses, the differences between standing and anaesthetised horses were assessed by 2 investigators with the same techniques as described above. Nonparametric tests (Mann-Whitney, Wilcoxon sign-rank) and Intraclass Correlation Coefficient (ICC) were used for statistical analysis (P<0.05).

Results: 1) The intra-tester reliability was high to excellent (ICC 0.8–1) and the inter-tester reliability low to average (ICC 0.1–0.5); 2) significant differences in joint ROM were registered in carpus and hock when measuring in pairs compared to singly and 3) significant differences in joint ROM were registered measuring anaesthetised compared to standing horses.

Conclusions: As shown in human studies, goniometry is a promising tool in documenting passive flexion of fetlock, carpus and hock, if used by the same investigator. However, additional studies are needed for further validation.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflicts of interest
  9. Manufacturers' addresses
  10. References

In horses and dogs, a restriction in joint movement can be an important factor in the development of motion asymmetries and contribute to changes in joint loading and impaired function (Khumsap et al. 2004; Jandi and Schulman 2007). Several rehabilitation modalities, such as passive joint mobilisation and muscle stretching, aim to restore joint function. Objective assessment of joint motion is essential for the assessment of therapeutic intervention and treatment outcome, as well as for the diagnosis. Noninvasive methods to assess motion asymmetry can range from conventional lameness examination to high velocity filming - the latter with the complexity of being expensive and technically advanced. For the general clinician, there is a need for simple, clinically validated and reliable instruments.

The gold standard for evaluation of joint motion in man is by radiography, which has the disadvantage of being time-consuming and exposes patients and staff to radiation. In human clinical practice, the most used noninvasive technique to measure joint range of motion (ROM) is goniometry, which is a validated and practical technique (Ekstrand et al. 1982; MacDermind et al. 1999). The goniometry is performed with a measuring device, a goniometer, which measures joint angles (in degrees) during flexion, extension, lateral flexion and rotation, respectively. Goniometry measurements are used to quantify restrictions in joint ROM, decide on appropriate therapeutic interventions and document the effectiveness of these interventions. It also allows the patient, or in veterinary medicine the animal owner, to monitor the therapeutic progress.

Several studies on goniometry in man have shown a good overall intra- and inter-tester reliability, with better intra-tester reliability (Mitchell et al. 1975; Boone et al. 1978; Rothstein et al. 1983; Goodwin et al. 1992). In a review article, Gajdosik and Bohannon (1987) conclude that the level of reliability is closely associated with joint structure and function, as well as the action measured. Measurement of a simple joint such as the elbow showed less variation than measurement of a more complex joint such as the wrist (Low 1976) and there were higher reliability measurements for larger than smaller angles (Brosseau et al. 1997).

Two studies concluded that goniometry is a reliable and objective method for determining passive joint ROM in healthy dogs (Jaegger et al. 2002; Thomas et al. 2006) and a number of studies describe the use of goniometry in canine orthopaedics (Sjöström et al. 1995; Cook et al. 2005; Dahlberg et al. 2005; Corfield et al. 2007; Jandi and Schulman 2007; Iwata et al. 2008).

To our knowledge, goniometry has not been validated in horses. To be useful in the clinic and for research, a goniometry protocol should be highly reliable, both within and between investigators, and should provide accurate data about joint ROM. Horses, compared to man and dogs, cause additional challenges in measuring joint ROM because of their large size and possible lack of cooperation in maintaining an optimal joint position. Due to these factors, it is more difficult for an investigator working singly to administer the goniometer while simultaneously holding the horse's limb. The purpose of the present study was to validate goniometry for use in horses. This was done by examining 2 different goniometry techniques, working in pairs or singly, for the measurement of passive maximum flexion of the forelimb fetlock, carpus and hock. This was performed by 1) assessing intra- and inter-tester reliability, 2) determining whether there were differences in goniometric data between measurements performed in pairs or singly and 3) determining whether there were differences in goniometric data between standing and anaesthetised horses.

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflicts of interest
  9. Manufacturers' addresses
  10. References

The experimental protocol included 2 studies on standing horses and one on anaesthetised horses; 1) standing riding horses, 2) standing Standardbred trotters and 3) anaesthetised Standardbred trotters. The investigators consisted of 11 physical therapists and one veterinarian, all experienced with using the goniometer. Table 1 summarises data for the horses and procedures investigated in each respective study. The studies, including the part on anaesthetised horses, were approved by the Ethical Committee on Animal Experiments in Uppsala, Sweden.

Table 1. Summarised data for the horses investigated in the present study
GroupNo.AgeSexProcedure
  1. Age is given as group mean with the range within parentheses. No. = numbers.

I. Standing riding horses (various breeds)1011 (7–18)7 mares, 3 geldingsFive pairs of investigators measured in pairs
II. Standing Standardbred trotters1414 (6–18)9 mares, 1 geldingOne pair of investigators, and the same 2 investigators separately, measured in pairs and singly
III. Anaesthetised Standardbred trotters66 (2–13)4 mares, 1 gelding, 1 stallionOne pair of investigators, and the same 2 investigators separately, measured in pairs and singly

Goniometry

The horses in Study 1 (standing riding horses) were measured with a universal Brodin-goniometer1 with a 1° gradation and the horses in Study 2 (standing Standardbred trotters) and 3 (anaesthetised Standardbred trotters) measured with a universal Lindon-goniometer2 with a 5° gradation. All measurements were made in triplicate. In order to blind the investigators, the reading of the gradation was made by a separate investigator.

The joint motions evaluated were the passive maximum flexion of the forelimb fetlock, carpus and hock. Maximum flexion was defined as a position where the angle between the bones of a joint was maximally decreased. The goniometer was administered to the joint according to a standardised protocol, based on previous studies on dogs (Jaegger et al. 2002).

In brief, the axis of joint rotation was determined by moving the joint and the centre of the goniometer was positioned over the joint axis. The axis of rotation is cranial to the joint in carpus and caudal in the hock. Forelimb fetlock flexion (Fig 1) was evaluated by placing the arms of the goniometer aligned with the longitudinal axis of the third metacarpal bone and cannon bone, respectively. Carpal flexion (Fig 2) was evaluated by placing the arms of the goniometer aligned with the longitudinal axis of the antebrachium, defined as the line joining the cranial to caudal midpoint at the level of the lateral styloid process and the lateral humeral condyle and the third metacarpal bone, respectively. Finally, the tarsal flexion (Fig 3) was evaluated by placing the arms of the goniometer aligned with the longitudinal axis of the third metatarsal bone and the tibial shaft, respectively.

image

Figure 1. Goniometry of forelimb fetlock flexion.

Download figure to PowerPoint

image

Figure 2. Goniometry of carpal flexion.

Download figure to PowerPoint

image

Figure 3. Goniometry of tarsal flexion.

Download figure to PowerPoint

Horses and experimental design

1. Standing riding horses: Ten healthy riding horses of various breeds were each investigated by 5 pairs of physical therapists. The horses were obtained from a riding school and the inclusion criteria were that horses were sound and served as fully functional riding horses. All horses were evaluated on the same day for a total of 75 min. The order of horse, pair of investigators and side of evaluation was randomly chosen by drawing lots. In order to explore any practical difficulties in using the goniometer on the left, respectively, the right side of the horse, 5 horses were measured on the left side and 5 on the right side.

Three joint positions, including flexion of the forelimb fetlock, carpus and hock, were evaluated 3 times by each pair of investigators, respectively. In each pair, one investigator served as the holder of the limb positioning the actual joint in passive maximum flexion. The investigator was instructed to put the actual joint in neutral position between each measurement, without putting the limb down. The other blinded investigator administered the goniometer and locked the arms of the goniometer in a fixed position, without reading the gradation. Thereafter, the gradation was registered by the first investigator.

2. Standing Standardbred trotters: Fourteen healthy Standardbred trotters were investigated by one physical therapist and one veterinarian, working in pairs and singly. The horses were obtained from the Swedish University of Agricultural Sciences. The inclusion criteria were that the horses were sound and showed no sign of lameness on basic lameness examination. Each individual was examined during the same day for a total of 25 min. The order of the 2 goniometry techniques, measured by pairs or singly, was randomly chosen by drawing lots for the first horse and alternating for each subsequent horse according to the following: 1) measured by pairs with investigator A holding the limb; 2) measured by pairs with investigator B holding the limb; 3) measured by investigator A alone and 4) measured by investigator B alone. The goniometry protocol was identical as in Study 1 (with the exception that all horses were measured on their left side) when measuring was performed in pairs. When measuring was performed by a single investigator, the limb was held and the goniometer positioned simultaneously with the use of an elastic strap that helped to fix one arm of the goniometer to the proximal part of the limb and thereafter the gradation was read by a separate investigator.

3. Acepromazine Standardbred trotters: Because of a reason outside the present study, 6 of the above mentioned Standardbred trotters were premedicated with i.m. acepromazine (0.03 mg/kg bwt) followed 30–60 min later by i.v. administration of romifidin and methadone. Anaesthesia was induced with i.v diazepam and ketamine and maintained with isofluran in oxygen in a semi-closed anaesthetic circle with the horses placed in lateral recumbency. The joints of the horses' upper left limbs were investigated on the same day, according to the same goniometry protocol and by the same investigators as in Study 2.

Dynamometer measurements

On 4 of the standing Standardbred trotters, the force the investigator used when flexing the forelimb fetlock joint was measured with a dynamometer1, by attaching the sling of the dynamometer around the hoof and then placing the joint in maximal flexion.

Radiography

Passive flexion of the forelimb fetlock joint, the carpus and hock was measured with a Lindon-goniometer on one healthy Standardbred trotter, first standing and then anaesthetised, by a single investigator. The goniometric measurements during anaesthesia were done simultaneously with radiographs on each joint (as an example, see Fig 4), to enable joint measurements and confirm positioning of the goniometer by measuring how well the goniometer was positioned according to the long bone axis.

image

Figure 4. Radiographic image of the goniometric positioning at the forelimb fetlock joint.

Download figure to PowerPoint

Statistics

Descriptive statistics (mean ± s.d.) were calculated for all goniometric data. Data from Study 1 (standing riding horses), measured with a Brodin-goniometer, were calculated with increment of 1° and data from Studies 2 and 3 (standing Standardbred trotters and anaesthetised Standardbred trotters), measured with a Lindon-goniometer, were calculated with increment of 5°. Results from the 2 different goniometers were evaluated separately. Nonparametric tests (Wilcoxon sign-rank and Mann-Whitney) and ANOVA test were used to determine whether there was a significant difference in joint ROM between horses, investigators, goniometric techniques and between standing and anaesthetised horses. To evaluate the intra- and inter-tester reliability for the 5 pairs of investigators in Studies 1 and the 2 investigators in Studies 2 and 3, as well as for the 2 goniometry techniques and the standing/anaesthetised horses, intra-class correlation coefficients (ICCs) were calculated. The ICC, a test of agreement for continuous data can range from 0.00–1.00, where values of 0.60–0.80 are regarded as evidence of good reliability and those above indicate excellent reliability (Ellasziw et al. 1994). Statistical software3,4 were used for all analyses and significance level was set at P<0.05.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflicts of interest
  9. Manufacturers' addresses
  10. References

The goniometric measurement values from all horses are shown in Table 2 and the ICCs are presented in Table 3. Mean differences for all measurements between sessions and significant differences between horses and investigators were analysed. Of the 24 horses, 2 had incomplete assessment from the measurements of hock flexion and those registrations were excluded from the analyses.

Table 2. Data from 2 types of goniometric measurements for passive flexion of the forelimb fetlock, carpus and hock in a group of Riding horses and Standardbred trotters
 Riding horsesStandardbred trotters
StandingStandingAnaesthetised
PairPairSinglyPairSingly
  1. Values are presented as mean ± s.d. and are expressed in degrees.

Forelimb fetlock55 ± 1471 ± 671 ± 777 ± 875 ± 6
Carpus156 ± 6151 ± 5147 ± 5154 ± 5154 ± 5
Hock131± 12134 ±10128 ± 8143 ± 8137 ± 9
Table 3. Intra- and inter-tester reliability of goniometric measurements in a group of riding horses and a group of Standardbred trotters
Riding horses
 Intra-tester reliabilityInter-tester reliability
ICCrmseICCrmse
Forelimb fetlock0.9430.4510
Carpus0.8220.036
Hock0.9530.439
Standardbred trotters
 Intra-tester reliablityInter-tester reliability
SinglyPairSinglyPair
ICCrmseICCrmseICCrmseICCrmse
Forelimb fetlock0.9020.8920.3950.793
Carpus0.9020.7530.3240.224
Hock0.9320.9430.7640.706
 Intra-tester reliabilityInter-tester reliability
StandingAnaesthetisedStandingAnaesthetised
ICCrmseICCrmseICCrmseICCrmse
  1. ICC, intraclass correlation coefficient; rmse, root mean squared error. Data are presented for ICC.

Forelimb fetlock0.8920.8730.5540.584
Carpus0.8120.8220.1150.454
Hock0.9330.9520.6360.893

1. Standing riding horses

The results of goniometric measurements differed significantly between horses (P<0.05). The intra-tester reliability was high to excellent and the inter-tester reliability was low to average for all joint measurements. Within test sessions, the intra-tester measurement error (defined as the root mean square error) associated with flexing the joints varied between 2 and 3° and the inter-tester varied between 6 and 10°.

2. Standing and 3. Anaesthetised Standardbred trotters

Results of goniometric measurements of the flexion of the forelimb fetlock and hock differed significantly between horses (P<0.001). Results of goniometric measurements for measuring by pairs showed significantly higher joint angles than results from measuring singly in the flexion of the carpus and hock (P<0.05, the significant differences reported were at the same level as the measurement error, defined as root mean square error). Results of goniometric measurements in flexion of the forelimb fetlock, carpus and hock showed significantly higher joint angles in anaesthetised Standardbred trotters compared to standing Standardbred trotters (P<0.05, the significant differences reported were at the same level as the measurement error, defined as root mean square error).

For working in pairs, the intra-tester reliability was high to excellent while the inter-tester reliability was low to high for all joint measurements. For investigators working singly, the intra-tester reliability was excellent and the inter-tester reliability was average to high for all joint measurements.

For measuring standing Standardbred trotters, the intra-tester reliability was high to excellent and the inter-tester reliability was low to high for all joint measurements. For measuring anaesthetised Standardbred trotters, the intra-tester reliability was high to excellent and the inter-tester reliability average to excellent for all joint measurements.

Within test sessions, the intra-tester measurement error associated with flexing the joints varied between 2 and 3° and the inter-rater between 3 and 6°.

Dynamometer data

Results of dynamometric measurements for the investigators did not differ significantly (mean force investigator A: 12 ± 3 N and investigator B: 13 ± 2 N).

Radiography

The goniometric data for measurement of maximum flexion in standing horses was for the forelimb fetlock 65°, carpus 140° and hock 135°; during anaesthesia, forelimb fetlock 70°, carpus 140° and hock 135° and from radiography, forelimb fetlock 70°, carpus 140° and hock 135° (single registrations). The arms of the goniometer were aligned with the longitudinal bone axis and reference points in all 3 joints.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflicts of interest
  9. Manufacturers' addresses
  10. References

To our knowledge, this is the first report to validate goniometer techniques in horses. The main finding of the present study is the high to excellent intra-tester reliability for the 2 goniometric techniques as well as for measuring standing and anaesthetised horses. As reported in studies in man and dogs, we found better intra-tester than inter-tester reliability scores (Ekstrand et al. 1982; MacDermind et al. 1999; Jaegger et al. 2002; Thomas et al. 2006). The data analyses suggest that goniometry in horses is a reliable tool when measuring passive flexion of the forelimb fetlock, carpus and hock and performed by the same investigator. A clinical judgement based on angular changes of <10° may not be of clinical importance, since it can be attributed to variability in goniometric assessment.

One important question is whether it is possible to accurately measure passive joint ROM in a large animal as the horse. This was tested by using 2 goniometric techniques; measuring in pairs or singly. Both techniques have their advantages. Working in pairs has the advantage of letting the investigator concentrate on measuring without holding the limb and working singly has the advantage of being able to assess the joint end-feel and the horse's reaction to the joint provocation, since a pain perception may cause a protective response and limit the range of motion. One may differentiate a possible painful joint motion in standing animals by measuring the same joint motion in anaesthetised horses. The present study showed significantly higher joint angles in anaesthetised compared to standing horses when measuring passive maximal flexion of the forelimb fetlock, carpus and hock. This is supported by a study on man (Dahlstedt and Dalen 1989). However, the significant differences between standing and anaesthetised horses, as well as between working in pairs or singly, were at the same level as the measurement errors. It is important to differentiate between statistical and clinical significance. Thus, from a clinical perspective, working in pairs should be as reliable as working singly and the measurement in standing horses matches that of anaesthetised horses when performed by the same therapist,

The sample of horses was based on a selection from 2 large groups of sport horses; horses used for riding or trotting. It is possible that there would have been more differences in goniometer data if horses of a heavier type were included. The influence of type of horse, sex, muscle mass and age on joint range of motion has been discussed. Some studies report a decrease in range of motion with increasing age (Cano et al. 2001; Butcher and Ashley-Ross 2002). In man, increased weight and muscle mass may limit joint motion (Hayes et al. 1994; Lin et al. 2005), but it is not known if this is the case in horses.

The selection of joints and joint motions was based on the clinical impression that a measurement of passive maximum extension of joints would not generate values that represent the joint ROM created during active extension, for instance the hyperextension of the forelimb fetlock created when the horse lands after a jump. Furthermore, previous studies on both man and dogs have shown a larger variability when measuring proximal joint compared to distal joint, probably due to the larger amount of soft tissue surrounding the proximal joints (Boone et al. 1978; Brousseau et al. 2001; Jaegger et al. 2002) and measurements of smaller joints in man are generally better than larger and flexion better than extension (Brousseau et al. 2001). Therefore, in the present study, measurements of joint extension and of proximal joints were not included.

In a clinical setting, or when evaluating specific therapies aimed at restoring joint function, the therapist often visually estimates the joint ROM. Based on results from human studies, it is likely that data obtained from visual estimations is less valid than goniometric measurements (Low 1976; Watkins et al. 1991). Consequently, goniometric tools can offer a more precise approach to the evaluation of joint ROM. Furthermore, it is important that the changes registered are not due to measurement errors.

The goniometric measurements in the present study were done with 2 universal goniometers; one with a 1° and the other with a 5° gradation. Results from the 2 different goniometers were evaluated separately. The root mean squared error for the intra- and inter-tester reliability varied between 2 and 10°, which suggests an excellent precision. This result is supported by a study of dogs (Jaegger et al. 2002) and several studies on man (Boone et al. 1978; Bovens et al. 1990; Cleffken et al. 2007) which suggest that a clinical judgement based on angular changes of <10° may not be of clinical importance since it can be attributed to variability in goniometric assessment. The present study provides preliminary reference goniometric measurements for sport horses, data that may be of future value when examining the clinical effects of different rehabilitation modalities.

Although this study was intended to reduce sources of error, some remaining factors could have affected the results. Despite the attempt to standardise the measurement procedure, the difficulty of simultaneous identification of anatomical reference points and positioning of the goniometer could have influenced the results. Human studies report that the patient should be positioned appropriately and be as relaxed as possible in order to minimise external factors during measurement of goniometry (Allison et al. 1996).

The reference standard for evaluation of joint motion is measurement of maximum joint excursion by radiography, which has the disadvantage of being time-consuming and exposes patients and staff to radiation. In the present study, the validation of goniometric measurements and location of the goniometer was performed with radiographs on one anaesthetised horse. The data showed good agreement between registrations done at standing, during anaesthesia and from radiographs. These results correspond to a study on Labrador retrievers (Jaegger et al. 2002), but are contradicted by a study on German shepherd dogs where a difference between goniometric measurement and radiography for 6 of 12 joint positions was observed (Thomas et al. 2006).

In conclusion, goniometric measurements taken on healthy horses were found to have a high to excellent intra-tester reliability when, working either in pairs or as a single investigator, measuring the passive flexion of the forelimb fetlock, capus and hock. We found better intra- than inter-tester reliability scores, thus the ideal evaluation of a change in joint motion is by measurements performed by the same investigator. Horses with clinical lameness were not evaluated in the present study. In other species, ROM of affected joints often decreases with lameness and this may possibly influence the goniometric measurements, since measurements of smaller angles are less valid than larger (Boone et al. 1978; Bovens et al. 1990; Jaegger et al. 2002). Therefore, it is essential that further validation studies of goniometry are conducted on horses with impaired joint ROM. Future research will need to examine and determine the sources of measurement reliability to improve the clinical assessment of joint range of motion in horses. Establishing reliability and validity of different measurement tools is an important part of both clinical practice and research, especially considering the increasing need for precise and accurate outcome measures in veterinary rehabilitation.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflicts of interest
  9. Manufacturers' addresses
  10. References

The authors wish to express their sincere gratitude to Dr Pless for her invaluable collaboration in this project.

Manufacturers' addresses

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflicts of interest
  9. Manufacturers' addresses
  10. References

1 Medema AB, Stockholm, Sweden.

2 Kebo Care Fysio, Stockholm, Sweden.

3 Microsoft Corporation, Redmond, USA.

4 Statsoft Scandinavia AB, Uppsala, Sweden.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflicts of interest
  9. Manufacturers' addresses
  10. References
  • Allison, S.C., Abraham, L.D. and Peterson, C.L. (1996) Reliability of the Modified Ashworth Scale in the assessment of plantarflexion muscle spasticity in patients with traumatic brain injury. Int. J. Rehab. Res. 19, 67-78.
  • Boone, D.C., Azen, S.P. and Lin, C.M. (1978) Reliability of goniometric measurements. Phys. Ther. 58, 1355-1390.
  • Bovens, A.M., Van Baak, M.A., Vrencken, J.G., Wijnen, J.A. and Verstappen, F.T. (1990) Variability and reliability of joint measurements. Am. J. Sports Med. 18, 58-63.
  • Brosseau, L., Tousignant, M., Budd, J., Chartier, N., Duciaume, L. and Plamondon, S. (1997) Intratester and intertester reliability and criterion validity of the parallelogram and universal goniometers for active knee flexion in healthy subjects. Physiother. Res. Int. 2, 150-166.
  • Brousseau, L., Balmer, S., Tousignant, M., O'Sullivan, J.P., Goudreault, C., Goudreault, M. and Gringas, S. (2001) Intra- and intertester reliability and criterion validity of parallelogram and universal goniometers for measuring maximum active knee flexion and extension of patients with knee restrictions. Arch. Phys. Med. Rehabil. 62, 396-402.
  • Butcher, M.T. and Ashley-Ross, M.A. (2002) Fetlock joint kinematics differ with age in Thoroughbred racehorses. J. Biomech. 35, 563-571.
  • Cano, M.R., Miro, F., Monterde, J.G., Diz, A., Martin, J. and Galisteo, A.M. (2001) Changes due to age in the kinematics of trotting Andalusian foals. Equine vet. J., Suppl. 33, 116-121.
  • Cleffken, B., Van Breulelen, G., Van Mamesen, H., Brink, P. and Olde Damink, S. (2007) Test-retest reproducibility of elbow goniometric measurements in a rigid double-blinded protocol: intervals for distinguishing between measurement error and clinical change. J. Shoulder Elbow Surg. 16, 766-794.
  • Cook, J.L., Renfro, D.C. and Tomlinson, J.L. (2005) Measurement of angles of abduction for diagnoses of shoulder instability in dogs using goniometry and digital image analysis. Vet. Surg. 34, 463-468.
  • Corfield, G.S., Read, R.A., Eastley, K.A., Richardson, J.L., Robertson, I.D. and Day, R. (2007) Assessment of the hip reduction angle for predicting osteoarthritis of the hip in the Labrador Retriever. Aust. Vet. J. 85, 212-216.
  • Dahlberg, J., Fitch, G., Evans, R.B., McClure, S.R. and Conzemius, M. (2005) The evalutaiton of extracorporeal shockwave therapy in naturally occurring osteoarthritis of the stifle joint in dogs. Vet. Comp. Orthop. Traumatol. 18, 147-152.
  • Dahlstedt, L.J. and Dalen, N. (1989) Knee laxity in cruciate ligament injury. Acta Orthop. Scand. 60, 181-184.
  • Ekstrand, J., Wiktorsson, M., Öberg, B. and Gillquist, J. (1982) Lower extremity goniometric measurements: A study to determine their reliability. Arch. Phys. Med. Rehabil. 63, 171-175.
  • Ellasziw, M., Yong, S.L. and Woodbury, J. (1994) Statistical methodology for the concurrent assessment of intrarater and interrater reliability: Using goniometric measurements as an example. Phys. Ther. 74, 777-778.
  • Gajdosik, R.L. and Bohannon, R.W. (1987) Clinical measurement of range of motion. Review of goniometry emphasizing reliability and validity. Phys. Ther. 67, 1867-1872.
  • Goodwin, J., Clark, C. and Deakes, J. (1992) Clinical methods of goniometry: a comparative study. Disabil. Rehabil. 14, 10-15.
  • Hayes, K.W., Petersen, C. and Falconer, J. (1994) An examination of Cyriax's passive motion test with patients having osteoarthritis of the knee. Phys. Ther. 74, 697-707.
  • Iwata, D., Broun, H.C., Black, A.P., Preston, C.A. and Anderson, G.I. (2008) Total hip arthroplasty outcomes assessment using functional and radiographic scores to compare canine systems. Vet. Comp. Orthop. Traumatol. 21, 221-230.
  • Jaegger, G., Marcellin-Little, D.J. and Levine, D. (2002) Reliability of goniometry in labrador retrievers. Am. J. vet. Res. 63, 979-986.
  • Jandi, A.S. and Schulman, A.J. (2007) Incidence of motion loss of the stifle joint in dogs with naturally occurring cranial cruciate ligament rupture surgically treated with tibial plateau levelling osteotomy: longitudinal clinical study of 412 cases. Vet. Surg. 36, 114-121.
  • Khumsap, S., Lanovaz, J.L. and Clayton, H.M. (2004) Three-dimensional kinematic analysis of horses with induced tarsal synovitis. Equine vet. J. 36, 659-663.
  • Lin, C.C., Ju, M.S. and Huang, H.W. (2005) Gender and age effects on elbow joint stiffness in healthy subjects. Arch. Phys. Med. Rehabil. 86, 82-85.
  • Low, J.L. (1976) The reliability of joint measurement. Phys. Ther. 62, 227-229.
  • MacDermind, J., Chesworth, B., Patterson, S. and Roth, J. (1999) Intratester and intertester reliability of goniometric measurement of passive lateral shoulder rotation. J. Hand Ther. 187-192.
  • Mitchell, W.S., Millar, J. and Sturrock, R.D. (1975) An evaluation of goniometry as an objective parameter for measuring joint motion. Scott Med. J. 20, 57-59.
  • Rothstein, J.M., Miller, P.J. and Roettger, R.F. (1983) Goniometric reliability in a clinical setting: elbow and knee mesurements. Phys. Ther. 63, 1611-1615.
  • Sjöström, L., Kasström, H. and Källberg, M. (1995) Ununited anconeal process in the dog. Pathogenesis and treatment by osteotomy of the ulna. Vet. Comp. Orthop. Traumatol. 8, 170-176.
  • Thomas, T.M., Marcellin-Little, D.J., Roe, S.C., Lascelles, B.D.X. and Brosey, B.P. (2006) Comparison of measurements obtained by use of an electrogoniometer for the assessment of joint motion in dogs. Am. J. vet. Res. 67, 1974-1979.
  • Watkins, M.A., Riddle, D.L., Lamb, R.L. and Personius, W.J. (1991) Reliability of goniometric measurements and visual estimates of knee range of motion obtained in a clinical setting. Phys. Ther. 71, 90-97.