Accurate measurement of the TPA is critical to surgical evaluation and planning in dogs with stifle joint abnormalities. Multiple studies1–6 have examined factors potentially associated with TPA, but all of these studies involved measurements performed on radiographs obtained by means of conventional techniques. As new technology becomes more advanced and more affordable, digital methods for radiographic evaluation have become standard at many veterinary hospitals. As a result, there has been an increase in the use of digital methods for measuring TPA in dogs. To our knowledge, however, there are no published data on the applicability or reliability of digital methods for measuring TPA. Differences in the placement of measurement points associated with the ability to magnify digital images and problems associated with the use of measurement software could potentially result in differences between TPA measurements obtained by conventional and digital methods. Accordingly, the purpose of the study reported here was to compare TPA measurements obtained by use of a standard manual procedure on radiographs obtained by conventional methods with values obtained by use of a computer-assisted method on radiographs that had been obtained digitally.
Materials and Methods
Study protocol—A convenience sample of 37 client-owned dogs with stifle joint abnormalities was included in the study. Preoperative mediolateral radiographic views of both stifle joints were obtained by means of conventional and digital radiography. For conventional radiography, a standard x-ray unita and film cassettesb,c were used, and radiographs were processed with an automated processor.d For digital radiography, the same x-ray unit was used with digital cassettes,e and images were acquired digitally.f All radiographic images included the entire tibia and tarsal joint and were centered on the stifle joint with superim-position of the femoral condyles.
Dogs with radiographic signs of severe degenerative joint disease in which adequate identification of radiographic landmarks was difficult were excluded from the study. Owners of dogs included in the study signed a consent form allowing all documentation regarding their dogs to be used for scientific research and publication.
Radiographic measurements—Two viewers measured TPAs on all radiographic images on 3 separate occasions. Viewer 1 was a student with no previous experience in measuring TPA; viewer 2 was a board-certified surgeon with > 10 years of experience measuring TPA.
On conventionally acquired radiographs, TPA was measured with a protractor and fine-point pencil in accordance with standard guidelines.7 The 3 measurements were obtained on separate days, and all markings were completely removed with alcohol between measurements. Radiograph order was randomized between measurements, and viewers were blinded to their previous measurements.
On digitally acquired images, TPA was measured with standard software.g Image order was again randomized between the second and third digital measurements, and viewers were blinded to linear and angular measurements that were obtained. For measurement of the TPA, straight lines were drawn to define the tibial plateau and long axis of the tibia, and a line perpendicular to the long axis of the tibia was automatically created by the software program. The angle function of the software was then used to calculate the TPA.
Statistical analysis—Data were summarized as mean and SD, and the Kolmogorov-Smirnov test was used to determine whether data were normally distributed. One-way ANOVA followed by the least significant difference test was used to compare TPA values between viewers and between measurements for each viewer. The Pearson correlation method was used to assess correlation between values obtained by conventional and digital methods. The paired t test and 1-way ANOVA followed by the Duncan test were used to compare values obtained by means of the conventional method with values obtained by means of the digital method. All analyses were performed with standard software.h Values of P < 0.05 were considered significant.
Results
Dogs included in the study ranged from 1.5 to 13.7 years old at the time radiographs were obtained (mean ± SD, 6.8 ± 3.5 years). Nineteen were females, and 18 were males. Body weight ranged from 5.8 to 54 kg (12.8 to 118.8 lb; mean ± SD, 23.9 ± 12.4 kg [52.6 ± 27.3 lb]). There were 12 mixed-breed dogs and 25 purebred dogs representing 19 breeds.
For both viewers, both limbs, and all 3 trials, conventional TPA measurements were significantly (P < 0.001) correlated with digital measurements. There were no significant differences between conventional and digital measurements, except that for viewer 1 during the first trial, mean TPA measured conventionally was significantly lower than TPA measured digitally in the left (P = 0.004) and right (P = 0.012) hind limbs (Table 1). Measurements obtained by viewer 1 were significantly (P ≤ 0.044) different from values obtained by viewer 2.
Mean ± SDTPAs obtained in 37 dogs by means of conventional and digital methods by 2 viewers on 3 occasions.
| Viewer | Measurement | Conventional method | Digital method | ||
|---|---|---|---|---|---|
| Left hind limb | Right hind limb | Left hind limb | Right hind limb | ||
| 1 | |||||
| 1 | 25 ± 5.6* | 25 ± 5.4† | 26.60 ± 5.49* | 25.81 ± 4.86† | |
| 2 | 26 ± 5.7 | 26 ± 4.7 | 26.06 ± 6.53 | 25.75 ± 5.61 | |
| 3 | 26 ± 5.6 | 26 ± 5.7 | 25.95 ± 6.17 | 25.89 ± 5.45 | |
| 2 | |||||
| 1 | 26 ± 4.5 | 25 ± 6.2 | 25.50 ± 4.97 | 25.29 ± 5.45 | |
| 2 | 26 ± 4.7 | 26 ± 6.1 | 25.70 ± 5.22 | 26.12 ± 5.65 | |
| 3 | 26 ± 4.5 | 25 ± 5.8 | 25.74 ± 4.79 | 26.62 ± 5.76 | |
Values were signifcantly (P = 0.004) different.
Values were signifcantly (P = 0.012) different.
Discussion
Results of the present study suggested that TPA measurements obtained with a digital method were comparable to those obtained by use of the conventional method. Although significant differences were found between viewers, for each viewer, only minor differences were found between conventional and digital measurements.
The conventional method used in the present study has been described previously,7 has been shown to provide reliable and reproducible data, and can be considered the gold standard method for measuring TPA. However, the growing use of digital radiography in clinical veterinary practice has led to more widespread use of imaging software. With digital radiography, images can be altered to adjust contrast, enhance edges, and make corrections for minor over- or underexposure.8 However, there is a lack of information pertaining to the effects of its use on clinically relevant variables. In the present study, use of an architectural software program provided a practical, low-cost method for obtaining digital measurements. Subjectively, both viewers in the present study reported that the digital method for measuring TPA was easier and faster than the conventional method.
Conventional and digital radiography were performed immediately one after the other in the present study, to minimize differences in limb position that might have affected measured TPA values.6 Admittedly, it could be considered a weak point of the present study that dogs had to undergo additional radiation exposure for the second set of radiographs to be obtained.
When comparing TPAs measured on conventional and digital radiographs, we found no significant differences in values obtained by the more experienced viewer in the present study, but found significant differences for the less experienced viewer with the first set of measurements obtained. Although the cause of these significant differences was not identified, we suspect that they may have been caused by difficulties in landmark identification, random chance, and the size and type of pencil used for conventional measurement. Because there is no accepted method for determining the true TPA, we were unable to determine which measurement technique was more accurate. Measurement of TPA with the conventional method7 has been studied extensively, revealing variations between viewers.1–4 In the present study, SDs for both viewers were higher than values reported in previous studies,1,2 presumably because of the wide variation in signalment among dogs in the present study, differences in severity of degenerative joint disease, and, possibly, minor differences in positioning.
Inter-viewer and intertechnique variability in TPA measurements in the present study resulted in intraoperative rotational differences of ≤ 0.73 mm, depending on which saw blade was used. Thus, although significant differences were detected between viewers and between methods, the clinical importance of those differences was minimal.
Abbreviations
| TPA | Tibial plateau angle |
References
- 1.
Morris E, Lipowitz AL. Comparison of tibial plateau angles in dogs with and without cranial cruciate ligament injuries. J Am Vet Med Assoc 2001;218:363–366.
- 2.
Wilke VL, Conzemius MG, Besancon MF, et al. Comparison of tibial plateau angle between clinically normal Greyhounds and Labrador Retrievers with and without rupture of the cranial cruciate ligament. J Am Vet Med Assoc 2002;221:1426–1429.
- 3.
Caylor KB, Zumpano CA, Evans LM, et al. Intra- and interobserver measurement variability of tibial plateau slope from lateral radiographs in dogs. J Am Anim Hosp Assoc 2001;37:263–268.
- 4.
Fettig AA, Rand WM, Sato AF, et al. Observer variability of tibial plateau slope measurement in 40 dogs with cranial cruciate ligament-deficient stifle joints. Vet Surg 2003;32:471–480.
- 5.
Baroni E, Matthias RR, Marcellin-Little DJ, et al. Comparison of radiographic assessments of the tibial plateau slope in dogs. Am J Vet Res 2003;64:586–589.
- 6.↑
Reif U, Dejardin LM, Probst CW, et al. Influence of limb positioning and measurement method on the magnitude of the tibial plateau angle. Vet Surg 2004;33:368–375.
- 7.↑
Slocum B, Devine T. Cranial tibial thrust: a primary force in the canine stifle. J Am Vet Med Assoc 1983;183:456–459.
Gierth HF 80/15, Gierth X-Ray International GmbH, Riesa, Germany.
Konica KR DIN 6832 (30 × 40 cm), Konica Corp, Tokyo, Japan.
KO 400 Mediphot (24 × 30 cm and 18 × 24 cm), Mediphot Inc, North Wales, Pa.
MI 5 x-ray film processor, Medical Index GmbH, Bad Rappenau, Germany.
Fuji IP cassettes (35.4 × 43 cm, 24 × 30 cm, and 18 × 24 cm), Fuji Film Co, Tokyo, Japan.
Fuji computed radiography model CR-IR 347P, Fuji Film Co, Tokyo, Japan.
ArchiCAD, version 9, Graphisoft US Inc, Newton, Mass.
SPSS, version 14.0 for Windows, SPSS Inc, Chicago, Ill.
