218

Original Research

Radiographic changes of the pelvis in Labrador and Golden Retrievers after juvenile pubic symphysiodesis Objective and subjective evaluation S. Boiocchi1; L. Vezzoni1; A. Vezzoni1; V. Bronzo2; F. Rossi3 1Clinica Veterinaria Vezzoni

srl, Cremona, Italy; 2University of Milan, Department of Health, Animal Science and Food Safety, Milan, Italy; 3Clinica Veterinaria dell'Orologio, Sasso Marconi (Bologna), Italy

Keywords JPS, juvenile pubic symphysiodesis, dogs, radiographic modifications

Summary Objectives: The hypothesis of this study was that juvenile pubic symphysiodesis (JPS) results in pelvic changes that can be identified radiographically in adult dogs. Methods: The medical records at the Clinica Veterinaria Vezzoni were searched for standard ventro-dorsal views of the pelvis of adult Labrador and Golden Retrievers that had undergone JPS or had not undergone surgery. The objective assessment of radiographs included the analysis of various pelvic measurements. Subjective evaluation of radiographs was undertaken by 18 specialists and 21 general practitioners and was based on five criteria relating to 1) the acetabular fossae, 2) the pubic symphysis, 3) the margin of the cranial pubic area, 4) the pubic rami, and 5) the obturator foramen.

Correspondence to: Dr. Aldo Vezzoni, DVM Clinica Veterinaria Vezzoni srl V Massarotti, 24 Cremona 26100 Italy Phone: +39 0372 23451 Fax: +39 0372 20074 E-mail: [email protected]

Results: The radiographs of 42 Labrador Retrievers and 16 Golden Retrievers were evaluated. The most useful criteria were the radiographic measurement of the shape of the obturator foramen and two different ratios of length to width of the pubic rami; these values were significantly smaller in dogs after JPS. The pelvic canal width was the same in both groups. All objective measurements were repeatable within and between evaluators. The most reliable subjective criterion was number 4, followed by number 5 in Golden Retrievers and by 2 in Labrador Retrievers. Conclusion: Our objective and subjective evaluations were simple and yielded useful and repeatable results. There was no significant difference between general practitioners and specialists with regard to subjective evaluation, which indicates that these evaluation criteria can be used by small animal clinicians after minimal training.

Vet Comp Orthop Traumatol 2013; 26: 218–225 doi:10.3415/VCOT-12-06-0074 Received: June 10, 2012 Accepted: December 28, 2012 Pre-published online: March 22, 2013

Introduction Juvenile pubic symphysiodesis (JPS) is a minimally invasive, simple, and inexpensive surgical procedure that is commonly used to improve hip conformation in pup-

pies with early signs of hip dysplasia (1-5). The goal of JPS is to improve hip congruity by arresting ventral pelvic growth; the procedure results in premature closure of the cranial aspect of the pubic symphysis and bilateral ventrolateral rotation of the acet-

abula (acetabular ventroversion) (4, 6-9). Premature closure of the pubic symphysis using JPS depends on the efficacy of the surgical technique, whereas arresting or reducing the progression of hip dysplasia with this technique depends on proper case selection and the age of the patient (14 to 18 weeks) (1, 2, 4, 10). Radiographic changes of the pelvis after JPS have been briefly described (1, 4). These changes include complete fusion of the cranial part of the pubic symphysis, shortening and broadening of the pubic rami, widening of the area of the obturator foramen, and a more prominent fossa of the acetabular fossae. Two other studies showed that JPS resulted in a significant decrease in the diameter of the pelvic inlet; there was no significant change in pelvic inlet diameter in another study of 2001 (6, 8, 9). A simple and reliable radiographic method for identification of dogs treated with JPS would be very useful, especially for radiologists involved in official screening programmes for hip dysplasia around the world. Dogs that have undergone JPS should not be used for breeding. Although some surgeons require simultaneous surgical neutering, sterilization of these animals is not mandatory and not always done. Except for cases in which radiopaque implants are used, dogs that have undergone JPS cannot be identified (9). The presence of radiographic signs indicative of JPS would aid in identification of these animals so that they could be excluded from breeding. Moreover, the occurrence of characteristic radiographic changes could also aid in determining whether JPS does in fact cause complete premature closure of the pubic symphysis.

© Schattauer 2013

Vet Comp Orthop Traumatol 3/2013 Downloaded from www.vcot-online.com on 2015-12-11 | ID: 1000525089 | IP: 159.149.113.122 For personal or educational use only. No other uses without permission. All rights reserved.

219

S. Boiocchi et al.: Radiographic changes after JPS

The primary goal of this study was to investigate whether JPS leads to characteristic and readily recognizable radiographic changes in the pelvis that can be used to identify adult dogs that have undergone this procedure. A secondary goal was to determine whether systematic radiographic measurements obtained in ventrodorsal extended views of the pelvis would increase the rate of identification of dogs that had undergone JPS.

Materials and methods Case selection Labrador and Golden Retrievers were studied because they were the most common breeds that had undergone JPS at our hospital. Radiographs of all Labrador and Golden Retrievers that underwent evaluation for hip dysplasia at 15 to 18 weeks of age and later at 12 to 24 months of age from April 2006 to December 2009 at Clinica Veterinaria Vezzoni were assessed for symmetry and quality. Cases with suboptimal radiographs were excluded from the study. The dogs were divided into two groups. The first group consisted of all the dogs that had undergone JPS at 15 to 18 weeks of age. The operations were done by one of the authors (AV). Criteria for case selection and surgical technique have been described previously and were consistent with positive Ortolani signs, a reduction angle (AR) of 15° to 35°, a subluxation angle (AS) of 0° to 10°, a distraction index of up to 0.75, and a dorsal acetabular rim angle (DARA) of up to 12° (4). The second control group consisted of dogs with no radiographic and physical signs of hip dysplasia susceptibility based on early radiographic evaluation that had not undergone any surgery, and dogs with signs of hip dysplasia susceptibility, similar to the JPS group, that were subsequently treated nonsurgically. The JPS and control groups consisted of the same number of dogs, which were 12 to 24 months of age.

Radiographic examination For early diagnosis of susceptibility to developing hip dysplasia, radiographs were taken after sedation of the puppies with

view. Dogs that had radiographic and physical signs of susceptibility to developing hip dysplasia were identified, and JPS was recommended (4). Adult dogs were sedated using the same protocol, with the addition of propofol (2 mg/kg IV) when necessary. The standard extended VD view was the only radiographic view used for the evaluation of adult dogs in this study.

Figure 1 Ventrodorsal view of the pelvis of a 13-month-old male Labrador Retriever (juvenile pubic symphysiodesis group) showing method 1 for determination of the radiographic appearance of the obturator foramen. Line A was drawn along the greatest distance between the craniolateral aspect of the obturator foramen and its caudomedial part and line B was drawn perpendicular to line A at its midpoint.

Figure 2 Ventrodorsal view of the pelvis of a 13-month-old male Labrador Retriever (juvenile pubic symphysiodesis group) showing method 2 for determination of the the radiographic appearance of the obturator foramen. A dotted line was drawn between the ischial tuberosity and cranial pubic symphysis (α) and another dotted line was drawn from the craniolateral acetabular margin to the caudal pelvic symphysis (β) on both sides. The major diameter of the obturator foramen (F), marked in black, started at the cranial aspect of β and ended at its most caudal aspect. The diameter F was divided into four equal parts using three black lines (lines 1, 2, and 3) drawn parallel to α, from the medial to the lateral border of the obturator foramen.

medetomidine (7 μg/kg, IM) and butorphanol (0.2 mg/kg, IM). Radiographic views included a standard ventrodorsal view (VD) with extended limbs, VD view with flexed limbs (frog leg), distraction VD view, and a dorsal acetabular rim (DAR)

Radiographic evaluation and measurements Subjective assessment and objective pelvic measurements were carried out on the standard VD view in adult dogs. Radiographs were divided into a JPS group and a control group.

Objective measurements According to the described morphological changes of the pelvis, various variables were measured on VD views, including the dimension of the obturator foramen, pubic rami and pelvic inlet (1, 4). Only two of the authors (SB, LV), who were not blinded to the group assignments, carried out all the measurements on digital images using computer softwarea. The radiographic appearance of the obturator foramen was measured using three different methods to determine the most reliable one. With method 1 (▶ Figure 1), line A was drawn along the greatest distance between the craniolateral aspect of the obturator foramen and its caudomedial part and was considered the major diameter. Line B was drawn perpendicular to line A at its midpoint and was considered the minor diameter. In method 2 (▶ Figure 2), a line was drawn between the ischial tuberosity and cranial pubic symphysis (α), and a line was drawn from the craniolateral acetabular margin to the caudal pelvic symphysis (β) on both sides. The entire pubic symphysis was outlined to determine the cranial and caudal margins more precisely. Alpha (α) and beta (β) were used as guides for determining the measurements of the obturator foramen. The major diameter of the obtu-

a

OsiriX Viewer: OsiriX Imaging Software, available at: http://www.osirix-viewer.com/Downloads.html

Vet Comp Orthop Traumatol 3/2013

© Schattauer 2013 Downloaded from www.vcot-online.com on 2015-12-11 | ID: 1000525089 | IP: 159.149.113.122 For personal or educational use only. No other uses without permission. All rights reserved.

220

S. Boiocchi et al.: Radiographic changes after JPS

Figure 3 Ventrodorsal view of the pelvis of a 13-month-old male Labrador Retriever (juvenile pubic symphysiodesis group) showing method 3 for determination of the radiographic appearance of the obturator foramen: Two lines were drawn tangentially to the proximal (γ) and distal (δ) aspects of both obturator foramina. Then the distance N between these two lines on the pubic symphysis was measured. The distance N was divided into four equal parts using three black dashed lines (1, 2, 3) drawn parallel to γ across the obturator foramen.

Figure 4 Ventrodorsal view of the pelvis of a 13-month-old male Labrador Retriever (juvenile pubic symphysiodesis group) showing the radiographic assessment of the pubic rami: A circle (A) was drawn in the thinnest part of each pubic ramus and a second circle (B), marked with dotted line, was drawn in the pubic symphysis including the cranial pubic tubercle proximally. Line T, a black dashed line, was divided in four equal parts using three solid black lines (1, 2, 3) drawn perpendicular to T.

rator foramen (F) started at the cranial aspect of β and ended at its most caudal aspect. The diameter F was divided into four equal parts using three black lines (lines 1, 2 and 3) in ▶ Figure 2, drawn parallel to α. With method 3 (▶ Figure 3), two lines were drawn tangentially to the proximal (γ) and distal (δ) aspects of both obturator foramina. Then the distance N between these two lines on the pubic symphysis was measured. The distance N was divided into four equal parts using three black dashed lines (lines 1, 2, 3) in ▶ Figure 3, drawn parallel to γ across the obturator foramen. Different ratios between the various measurements were used to minimize changes related to skeletal development. For method 1, only one ratio (A:B) was calculated, and in methods 2 and 3, three different ratios were used (F:1, F:2, F:3, and N:1, N:2, N:3, respectively). The radiographic assessment of the pubic rami was measured using only one method (▶ Figure 4). A circle (A) was drawn in the thinnest part of each pubic ramus and a second circle, (B) was drawn in the pubic symphysis including the cranial pubic tubercle proximally; the circle B had to be tangential to both medial aspects of the obturator foramen. The centres of these circles (A and B) were used as guides to determine pubic

lines. Line T was drawn from the centre of the circle B (pubic symphysis) to the medial border of the acetabular fossa, passing through the centre of the circle A (pubic ramus) on each side. Line T was divided in four equal parts using three solid black lines (lines 1, 2, 3) in ▶ Figure 4, drawn perpendicular to T. The ratios T:1, T:2 and T:3 were calculated. This method was used for each pubic ramus. The pelvic canal width (PCW) was determined by measuring the distance between the right and left iliac wings (a) and the distance between the right and left craniomedial acetabular borders (b), and the ratio a:b was then calculated.

Subjective evaluation The same radiographs used for objective measurements were evaluated subjectively by two groups of veterinarians with different levels of experience. The first group consisted of 18 specialists who were either diplomates of the ECVS, ACVS or ECVDI, or were recognized experts in veterinary orthopaedic diagnostics without diplomate status; two of the authors (AV and FR) were part of the latter group. The second group consisted of 21 general small animal practitioners with different clinical inter-

A

B Figure 5 A) Ventrodorsal view of the pelvis of a 13-month-old male Labrador Retriever (juvenile pubic symphysiodesis group) showing the subjective criteria used for radiographic evaluation: 1 (visibility of the acetabular fossa), 2 (partial or complete closure of the cranial part of the pubic symphysis), 3 (irregular margin of the cranial pubic symphysis and/or increased radiodensity of the cranial pubic contour), 4 (shortening and broadening of the pubic ramus), and 5 (widening of the obturator foramen). B) Ventrodorsal view of the pelvis of a 13-month-old male Labrador Retriever (control group) showing: 1) acetabular fossa detectable in dogs with good hip conformation, 2) cranial pubic symphysis not fused, 3) normal cranial pubic contour and regular bone density, 4) thin pubic ramus, and 5) oval obturator foramen.

ests. The evaluators were blinded to any information about the dogs except the case number used for logging radiographs and the breed. Radiographs were placed in random order by one of the authors, who was not an evaluator, and examined over a period of six weeks. The evaluators in both groups were briefly coached on the radiographic changes expected in adult dogs

© Schattauer 2013

Vet Comp Orthop Traumatol 3/2013 Downloaded from www.vcot-online.com on 2015-12-11 | ID: 1000525089 | IP: 159.149.113.122 For personal or educational use only. No other uses without permission. All rights reserved.

221

S. Boiocchi et al.: Radiographic changes after JPS

after JPS. They were asked to assess the following criteria: 1) visibility of the acetabular fossae, 2) partial or complete closure of the cranial part of the pubic symphysis, 3) irregularities in the margin of the cranial pubic symphysis or increased radiodensity of the cranial pubic area or both combined, 4) shortening and broadening of the pubic rami, and 5) widening of the obturator foramina (▶ Figure 5A). It was pointed out to the evaluators that in dogs with good hip conformation, the acetabular fossae should be seen in control dogs as well (▶ Figure 5B). Evaluators were asked to complete a custom-made form for evaluation of the five radiographic criteria and to decide whether JPS had or had not been carried out in each case. Evaluators scored each criterion as present (Y), absent (N), or equivocal (Y/N) and could add comments for each case.

Statistical analysis Differences between measurements made by authors SB and LV, and between objective data from operated and non-operated Labrador and Golden Retrievers were analyzed using an analysis of variance provided that the Shapiro-Wilk test confirmed a normal distribution of the measurements. All descriptive statistics and statistical analyses were done using a statistical software programmeb. Results of the measured varib

MedCalc 12.2.1.0: MedCalc Software bvba, Ostend, Belgium. Available at: http://www.medcalc.org/ download.php

Table 1

ables were expressed as mean ± standard error of mean. A value of p 1.69; T:1 >3.15; T:2 >3.44). Labrador Retriever

Golden Retriever

A:B

F:1

F:2

F:3

T:1

T:2

A:B

F:2

T:1

T:2

Area under the curve (AUC)*

0.845 Good

0.831 Good

0.88 Good

0.69 Poor

0.929 Very good

0.825 Good

0.875 Good

0.705 Fair

0.91 0.898 Very Good Good

Standard error

0.0446

0.0442

0.0351

0.059

0.0299

0.0473

0.0644

0.0937

0.055

0.0637

Significance level p (Area = 0.5)

0.0001

0.0001

0.0001

0.0013

0.0001

0.0001

0.0001

0.0286

0.0001

0.0001

Sensitivity

66.7%

83.3%

69%

92.9%

97.6%

85.7%

81.2%

75%

100%

100%

Specificity

92.9%

69%

90.5%

42.9%

81%

78.3%

87.5%

56.2%

75%

75%

Criterion values

>1.67

>1.56

>1.46

>1.65

>3.02

>3.2

>1.71

>1.47

>3.28

>3.68

*Receiver operating characteristic = AUC. A:B = ratio of line A to line B as measured with method 1 (obturator foramen). F:1, F:2, F:3 = ratio of line F to lines 1, 2, and 3 as measured with method 2 (obturator foramen). PCW = pelvic canal width. T:1, T:2, T:3 = ratio of line T to lines 1, 2, and 3 (pubic rami).

© Schattauer 2013

Vet Comp Orthop Traumatol 3/2013 Downloaded from www.vcot-online.com on 2015-12-11 | ID: 1000525089 | IP: 159.149.113.122 For personal or educational use only. No other uses without permission. All rights reserved.

223

S. Boiocchi et al.: Radiographic changes after JPS

Table 4

Results of receiver operating characteristic (ROC) analysis of subjective evaluation in Labrador Retrievers.

Labrador Retrievers

Juvenile pubic symphysiodesis Y (present) or N (absent)

Radiographic criteria: Acetabular fossae detectable (No. 1)

Partial or complete fusion of cranial pubic symphysis (No. 2)

Irregularity of cranial pubic contour (No. 3)

Shortening and broadening of pubic rami (No. 4)

Widening of obturator foramen (No. 5)

Area under the curve (AUC)

0.881 Good

0.723 Fair

0.851 Good

0.807 Good

0.843 Good

0.806 Good

Standard error

0.0085

0.0115

0.00993

0.00993

0.00993

0.00961

Significance level p 0.0001 (Area = 0.5)

0.0001

0.0001

0.0001

0.0001

0.0001

Sensitivity

86%

79.2%

87.3%

73.5%

79.4%

71.6%

Specificity

91.8%

66.7%

87%

89.4%

90.8%

89.7%

Table 5

Results of receiver operating characteristic (ROC) analysis of subjective evaluation in Golden Retrievers.

Golden Retrievers

Juvenile pubic symphysiodesis Y (present) or N (absent)

Radiographic criteria: Acetabular fossae detectable (No. 1)

Partial or complete fusion of cranial pubic symphysis (No. 2)

Irregularity of cranial pubic contour (No. 3)

Shortening and broadening of pubic rami (No. 4)

Widening of obturator foramen (No. 5)

Area under the curve (AUC)

0.884 Good

0.725 Fair

0.832 Good

0.793 Fair

0.882 Good

0.83 Good

Standard error

0.0134

0.0194

0.0161

0.0159

0.0128

0.0153

Significance level p 0.0001 (Area = 0.5)

0.0001

0.0001

0.0001

0.0001

0.0001

Sensitivity

85.6%

86.5%

84.9%

69.2%

81.4%

76%

Specificity

92.3%

64.4%

84%

89.7%

95.2%

91.3%

probability of misdiagnosis. The results for Labrador and Golden Retrievers are shown in (▶ Figure 8, ▶ Figure 9).

Discussion Juvenile pubic symphysiodesis produced a scar at the level of the cranial pubic symphysis. Therefore, the radiolucent line, which is seen in normal dogs until several years of age, looks abnormal in dogs after JPS (14). The line is interrupted cranially by the presence of an area of increased radiopacity representing scar tissue. Similarly, damage to the cranial pelvic symphysis causes an abnormal growth of the medial part the pubic rami, resulting in deformation with shortened length and an irregular surface. The anatomical changes at the level of the pubis result in ventroflexion

of the acetabula. Thus, the shape of the obturator foramina changes and the acetabular fossae are projected medially. The primary aim of this study was to investigate whether dogs treated with JPS could be identified because some surgeons do not sterilize these dogs and do not use radiopaque implants for the procedure. One of the hypotheses of this study was that evaluation of objective and subjective radiographic criteria could be used to differentiate dogs treated with JPS and control dogs. The criteria chosen were based on anatomic changes that have been described after JPS (1, 4). Several objective measurements were used to evaluate the shape of the obturator foramen and the length of the pubic rami. All measurements were repeatable within and between the two groups of evaluators. The results of the ROC curve indicated that the best objective measure-

ments for differentiation of JPS and control dogs were the ratios A:B for the shape of the obturator foramen and T:1 and T:2 for the pubic rami. The ratio A:B was significantly smaller in the JPS group and was characterized by good AUC values. A decrease in the ratio A:B corresponded to the subjective impression of widening of the obturator foramen in JPS dogs. The ratios T:1 and T:2 were significantly smaller in the JPS group. These ratios had a high sensitivity (T:1, 100% in Golden Retrievers and 97.6% in Labrador Retrievers; T:2, 100% in Golden Retrievers and 85.7% in Labrador Retrievers) and were considered very good (T:1) and good (T:2) by the ROC analysis. The significant reduction of T ratios corresponded to the subjective evaluation of stocky pubic rami postoperatively. These variables (A:B, T:1 and T:2) were easy to measure and the authors feel that

Vet Comp Orthop Traumatol 3/2013

© Schattauer 2013 Downloaded from www.vcot-online.com on 2015-12-11 | ID: 1000525089 | IP: 159.149.113.122 For personal or educational use only. No other uses without permission. All rights reserved.

224

S. Boiocchi et al.: Radiographic changes after JPS

100

89.4

91.8

85.6

100

85.0

72.5

80

85.0

84.0

No. 1

No. 2

No. 3

Acetabular fossae detectable

Partial or complete fusion of cranial part of pubic symphysis

77.3

80

60

93.4

88.5

60

%

% 40

No. 1

No. 2

No. 3

No. 4

40

No. 5

20 0

No. 5

20 Acetabular fossae detectable

Partial or complete fusion of cranial part of pubic symphysis

Correct diagnosis

60

17.9

20

3/5

4/5

0.0

5/5

Figure 8 Number of correctly assessed subjective criteria associated with a correct diagnosis (JPS or non-operated) in Labrador Retrievers.

these measurements can be made by all practitioners regardless of their level of experience in the assessment of orthopaedic radiographs. The measurements appear to be useful in differentiating dogs that have undergone JPS and control dogs, and our recommendation is to combine objective and subjective assessment for evaluation of the pelvis. The absence of a significant difference between PCW values of JPS and control dogs showed that JPS does not alter pelvic canal width. However, the height of the pelvic canal was not measured and thus a decrease in total area could not be excluded. Ratios other than A:B, T:1, and T:2 yielded inconsistent or poor results. In this study, the ROC analysis led to different criterion values for measurements in Golden

16.7

20

10.4

2/5

47.4

40

1.1

1/5

52.6

%

0.0

0/5

100

95.8

93.1

80

42.7

40

Failed diagnosis

83.3

57.3

60

%

100

100

89.6

82.1

80

Correct diagnosis 100

98.9

Irregularity Shortening Widening of of cranial and obturator pubic broadening foramina contour and of pubic rami of bone density

Figure 7 Percentage of correct subjective assessment of various criteria associated with juvenile pubic symphysiodesis in Golden Retrievers (mean value 85.6%). No. = number.

Failed diagnosis

100

100

0

Irregularity Shortening Widening of of cranial and obturator pubic broadening foramina contour and of pubic rami of bone density

Figure 6 Percentages of correct subjective assessment of various criteria associated with juvenile pubic symphysiodesis in Labrador Retrievers (mean value 84.8%). No. = number.

0

No. 4

0

6.9

0.0

0/5

1/5

2/5

3/5

4.2

4/5

0.0

5/5

Figure 9 Number of correctly assessed subjective criteria associated with a correct diagnosis (JPS or non-operated) in Golden Retrievers.

Retrievers and Labrador Retrievers (▶ Table 3), which suggests that these measurements are breed-related and further studies are needed to evaluate breed differences. The ability of general practitioners and specialists to accurately evaluate pelvic radiographs was good, which indicated that the chosen criteria were easily understood and recognized, independent of the level of experience. Our results support the notion that this type of evaluation can be carried out by small animal practitioners after minimal training. The most reliable subjective criterion was number (No.) 4 (shortening and broadening of pubic rami) with a mean value of positive detection for both breeds of 91.6%, followed by No. 2 (closure

of pubic symphysis) for Labrador Retrievers or No. 5 (widening of the obturator foramen) for Golden Retrievers. The least reliable criterion was No. 1 (visibility of the acetabular fossa), with a mean value of positive detection for both breeds of 74.9%, because identification of the acetabular fossae was also possible in control dogs with normal hip joints and optimal acetabular ventroversion. Moreover, mild pelvic asymmetry during radiographic positioning can lead to acetabular fossa coverage on one side of the pelvis, which increases the visibility of the opposite fossa. Criterion No. 3 (cranial pubic symphysis and contour) was considered fair in Golden Retrievers (AUC 0.79) and good in Labrador Retrievers, although its AUC values were mildly elevated

© Schattauer 2013

Vet Comp Orthop Traumatol 3/2013 Downloaded from www.vcot-online.com on 2015-12-11 | ID: 1000525089 | IP: 159.149.113.122 For personal or educational use only. No other uses without permission. All rights reserved.

225

S. Boiocchi et al.: Radiographic changes after JPS

at 0.8. Superimposition of faecal material, coccygeal vertebrae, os penis and skin folds made the evaluation of criterion No. 3 difficult and affected the visibility of the cranial part of the pubic symphysis. Moving the tail and penis to one side and allowing dogs to defaecate before general anaesthesia and radiography would be very helpful. However, this is not normally done, and therefore the results of our study reflect the reality of standard routine radiography in clinical practice. Accurate evaluation of three of five subjective evaluation criteria allowed the correct assessment of 89.6% of Labrador Retrievers and 93% of Golden Retrievers. These values increased to 98.9% for Labrador Retrievers and 95.8% for Golden Retrievers when four of five criteria were correctly assessed. These results are encouraging for detection of dogs with JPS. Specialists and general practitioners were able to differentiate between JPS and control Labrador Retrievers in 86.7% of cases and Golden Retrievers in 87.5% of cases. In these animals, anatomical pelvic changes after JPS were evident on radiographs, thereby eliminating the need for an objective evaluation. In 13.3% of Labrador Retrievers and 12.5% of Golden Retrievers (mean value for both breeds 12.9%), radiographic changes were not as clear. Of the six cases that posed the most difficulty for subjective evaluation, two were correctly diagnosed after implementing criterion values; this resulted in a 1.5% increase in the rate of correct diagnoses. The four misdiagnosed cases were assessed as non-operated when in fact these dogs had undergone JPS. In these four cases, radiographic changes were less evident. One possible reason is that surgery failed to produce the expected closure of the pubic symphysis and all the associated radiographic changes. Conversely, the positivity of most or all the JPS-related radiographic signs may demonstrate that the surgery was effective independent of the final result in terms of hip dysplasia progression. Moreover, radiography has its

limitations when comparing measurements because of superimposition of various structures and other technical factors such as small differences in positioning or in centering the x-ray beam. Radiographic symmetry and accurate positioning of the patient are critical for assessment using the proposed objective criteria. The use of suboptimal radiographs is therefore not recommended. The recognition of radiographic changes after JPS is very important for official screening programs for hip dysplasia because dogs with JPS should be excluded from breeding.

Conclusions Our method for objective and subjective radiographic evaluation of signs of JPS in adult Labrador and Golden Retrievers is simple, reliable and yielded good results; however our parameters allowed differentiation of only some of the dogs evaluated. Cases with less obvious changes were not diagnosed by the specialists or the general practitioners. Further studies are required to determine the efficacy of our method in other breeds. The expected morphological changes after JPS were fusion of the pubic symphysis, arrest of the growth of the pubic rami resulting in a stocky appearance, ventroflexion of the acetabula and widening of the obturator foramina. Those modifications were observed in the standard radiographic view of the pelvis of the JPS-treated dogs in our study. Acknowledgments

We want to acknowledge all the specialists and general practitioners who agreed to perform the subjective evaluation of the cases included in this paper, for the precious time spent, and for the enthusiasm shown. We want to thank all our colleagues for their help in the execution of this study.

Conflict of interest

None declared. Online Supplementary Material Figures 1 to 5 have been made available online at www.vcot-online.com for viewing in a size larger than what was possible in print.

References 1. Dueland RT, Adams WM, Patricelli AJ, et al. Canine hip dysplasia treated by juvenile pubic Vet Comp Orthop Traumatol 2010; 23: 306-317 / Erratum in: Vet Comp Orthop Traumatol 2010; 23: 472. 2. Dueland RT, Patricelli AJ, Adams WM, et al. Canine hip dysplasia treated by juvenile pubic Vet Comp Orthop Traumatol 2010; 23: 318-325. / Erratum in: Vet Comp Orthop Traumatol 2010; 23: 472. 3. Bernardé A. Juvenile pubic Vet Surg 2010; 39: 158-164. 4. Vezzoni A, Dravelli G, Vezzoni L, et al. Comparison of conservative management and juvenile pubic Vet Comp Orthop Traumatol 2008; 21: 267-279. 5. Manley PA, Adams WM, Danielson KC, et al. Longterm outcome of juvenile pubic J Am Vet Med Assoc 2007; 230: 206-210. 6. Dueland RT, Adams WM, Fialkowski JP, et al. Effects of pubic symphysiodesis in dysplastic puppies. Vet Surg 2001; 30: 201–217. 7. Patricelli AJ, Dueland RT, Adams WM, et al. Juvenile pubic symphysiodesis in dysplastic puppies at 15 and 20 weeks of age. Vet Surg 2002; 31: 435–444. 8. Mathews KG, Stover SM, Kass PH. Effect of pubic symphysiodesis on acetabular rotation and pelvic development in guinea pigs. Am J Vet Res 1996; 57: 1427–1433. 9. Swainson SW, Conzemius MG, Riedsel EA, et al. Effect of pubic symphysiodesis on pelvic development in the skeletally immature greyhound. Vet Surg 2000; 29: 178–190. 10. Patricelli AJ, Dueland RT, Lu Y, et al. Canine pubic symphysiodesis: Investigation of electrocautery dose response by histologic examination and temperature measurement. Vet Surg 2001; 30: 261–268. 11. Metz CE. Basic principles of ROC analysis. Semin Nucl Med 1978; 8: 283–298. 12. Swets JA. ROC analysis applied to the evaluation of medical imaging techniques. Invest Radiol 1979; 14: 109–121. 13. Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic curve. Radiology 1982; 143: 29–36. 14. Smith RN. The pelvis of the young dog. Vet Rec 1964; 76: 975.

Vet Comp Orthop Traumatol 3/2013

© Schattauer 2013 Downloaded from www.vcot-online.com on 2015-12-11 | ID: 1000525089 | IP: 159.149.113.122 For personal or educational use only. No other uses without permission. All rights reserved.