Fragmentation of the medial coronoid process is a common developmental and hereditary disorder in dogs, but radiographic diagnosis of this condition can sometimes be difficult, particularly in dogs with partial fragmentation, minimal fragment displacement, a small fragment, or a fragment located between the radial head and the intact portion of the medial coronoid process.1,2 Often, the x-ray beam is oriented obliquely, making it difficult or impossible to see the fragment cleavage line on radiographic images.1,3 Additionally, in some affected dogs, the medial coronoid process may only be fissured2,4,5 and not displaced, may not be mineralized,a or may be abnormally shaped,5,6 further complicating the radiographic diagnosis.
It has been suggested that a presumptive diagnosis of fragmentation of the medial coronoid process can be made on the basis of clinical signs and radiographic evidence of osteoarthrosis or sclerosis of the semilunar notch on a flexed mediolateral projection of the elbow joint.5,7,8 However, up to 22% of dogs with abnormalities of the medial coronoid process do not have typical radiographic signs of osteoarthritis,9 and the percentage of dogs in which the fragment itself is visible on radiographs may be as low as 10%.8 Overall accuracy of radiography in detecting fragmentation of the medial coronoid process has been reported to be as low as 57%,10 with sensitivity of 67% when a craniolateralcaudomedial oblique projection is obtained.1,11
Alternative imaging techniques such as computed tomography and magnetic resonance imaging have been recommended when fragmentation of the medial coronoid process is suspected and radiographic findings are inconclusive.10,12–14 Unfortunately, these imaging modalities are expensive and often only available in specialist centers. Ultrasonography may be a useful adjunct to radiography in dogs suspected to have fragmentation of the medial coronoid process, and previous studies15–17,b have shown that the condition can be identified ultrasonographically in some affected dogs, depending on the size and location of the fragment. In addition, dynamic ultrasonography may be useful in demonstrating independent movement of a free fragment,16 and secondary changes such as joint effusion and osteoarthrosis can easily be detected.15,b However, the role of ultrasonography in the diagnosis of fragmentation of the medial coronoid process in dogs remains uncertain. The purpose of the study reported here, therefore, was to determine the accuracy of ultrasonography in detecting fragmentation of the medial coronoid process in dogs suspected to have the condition on the basis of clinical and radiographic findings by comparing ultrasonographic abnormalities with intraoperative findings.
Materials and Methods
Dogs referred to the Small Animal Clinic of Justus-Liebig University with a history of thoracic limb lameness between 2002 and 2005 were candidates for inclusion in the study. All dogs were examined by an orthopedic surgeon. Dogs with clinical signs of elbow dysplasia, such as signs of pain during palpation of the medial aspect of the elbow joint, swelling of the elbow joint, a decreased range of motion, or outward rotation of the elbow joint, were examined radiographically. Mediolateral views with the elbow joint in a neutral position and in a flexed position were obtained, along with a craniolateral-caudomedial (15° lateral to the sagittal plane) view of the elbow joint. Radiographs of the shoulder joint were also obtained to rule out diseases such as osteochondrosis of the humeral head and panosteitis, and dogs with related abnormalities were excluded from the study.
Dogs with radiographic evidence of or radiographic signs suggestive of fragmentation of the medial coronoid process that subsequently underwent ultrasonography of the elbow joint followed by exploratory or therapeutic arthrotomy were included in the study. Ultrasonography was performed with an 8- to 12-MHz linear array transducer immediately before surgery while dogs were anesthetized. A standard imaging protocolc was used, and all aspects of the elbow region were examined in the transverse and longitudinal planes. The region of the medial coronoid process was imaged from the medial aspect of the limb with the elbow joint in a neutral position. The transducer was placed transversely on the medial aspect of the limb, 1 to 2 cm distal to the elbow joint. Mild pronation and supination of the antebrachium facilitated orientation and recognition of the medial coronoid process as it moved around the radial head. In this plane, the surfaces of the radius and ulna could be seen as 2 adjacent, hyperechoic, curvilinear structures with a distal shadow (Figure 1). Care was taken to avoid artifacts such as discontinuities and a rough appearance of the bony surfaces. Imaging of the region of the medial coronoid process in the longitudinal plane was more difficult and provided no additional information.
Transverse ultrasonographic image of the medial coronoid process (MCP) in a clinically normal dog. The MCP can be delineated between the contours of the radius (R) and the medial aspect of the ulna (U).
Citation: Journal of the American Veterinary Medical Association 234, 4; 10.2460/javma.234.4.480
All ultrasonographic examinations were performed by a single experienced individual (UM) who knew that an elbow joint problem was suspected but was not aware of the radiographic findings. Fragmentation or abnormalities in the shape of the medial coronoid process, thickening of the joint capsule, joint effusion, and secondary new bone formation were assessed. Discontinuities and irregularities of the bony contour of the medial coronoid process were considered suggestive of fragmentation. Fragment separation was assessed on the basis of visualization during passive movement of the limb. If the medial coronoid process was abnormally shaped but independent movement could not be documented, the coronoid process was considered to be deformed.
Because the study protocol only involved routine clinical procedures, review of the study protocol by the Animal Welfare and Research Committee was not required. However, owners of all dogs included in the study provided their consent.
Data analysis—Dogs were considered to have a free fragment if, during surgery, the medial coronoid process was not found to be attached to the underlying bone and were considered to have a nondisplaced fragment if the medial coronoid process was still attached to the underlying bone or only a fissure line was identified. Ultrasonographic findings of interest were whether a fragment was identified (yes vs no), whether a deformity of the medial coronoid process was identified (yes vs no), and whether any abnormality of the medial coronoid process (ie, a fragment, deformity, or both; yes vs no) was identified. Accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of ultrasonographic findings were calculated by comparison with intraoperative findings, with intraoperative findings considered the gold standard; exact 95% CIs were calculated by means of standard software.d The kappa statistic was used to assess agreement between ultrasonographic findings and intraoperative findings.
Results
One hundred two dogs met the criteria for inclusion in the study. Ninety-two dogs underwent unilateral arthrotomy, and 10 dogs underwent bilateral arthrotomy. Thus, information for 112 elbow joints was included in the study (58 right elbow joints and 54 left elbow joints), with each elbow considered a separate case.
There were 21 Labrador Retrievers, 21 Rottweilers, 10 German Shepherd Dogs, 7 Bernese Mountain Dogs, 6 Golden Retrievers, 2 German Longhaired Pointers, 2 Giant Schnauzers, 13 mixed-breed dogs, and 20 other dogs each representing a different breed. Seventy-three of the 102 (72%) dogs were male (8 neutered), and 29 (28%) were female (7 spayed). Thirty (30%) dogs were < 1 year old at the time of initial examination, 41 (40%) were between 1 and 2 years old, and 31 (30%) were between 3 and 10 years old. Median age at the time of initial examination was 2 years (range, 6 months to 10 years).
At surgery, 51 of the 112 (46%) joints had a free medial coronoid process fragment, 55 (49%) had a nondisplaced medial coronoid process fragment, and 6 (5%) did not have any evidence of a fragment or identifiable fissure. Thirty-six of the 112 (32%) joints had a single large fragment that was removed, and 15 (13%) had multiple fragments that were removed. Forty-one (37%) joints had erosive lesions involving the medial aspect of the humeral condyle. Six (5%) joints had osteochondrosis involving the medial aspect of the humeral condyle. In 2 of the 6 joints, the medial coronoid process was intact; in 3, a free fragment was found; and in 1, a fissure line was found.
In 34 of the 112 (30%) elbow joints, a medial coronoid process fragment was identified during ultrasonography of the joint (Figure 2), whereas in the remaining 78 (70%) joints, no fragment was seen ultrasonographically (Table 1). Fragments were identified during surgery in 33 of the 34 joints with ultrasonographic evidence of a medial coronoid process fragment, but a fragment was not identified in the remaining joint. Fragments were not seen ultrasonographically in 23 of the 51 (45%) joints in which a free fragment was found during surgery or in 50 of the 55 (91%) joints in which a nondisplaced fragment was found during surgery.
Transverse ultrasonographic image of the medial aspect of the elbow joint in a dog with fragmentation of the MCP. A small fragment of the medial coronoid process (FMCP) is evident as a short hyperechoic line with distal shadow (arrow) between the radius (R) and the ulna (U).
Citation: Journal of the American Veterinary Medical Association 234, 4; 10.2460/javma.234.4.480
Cross-classification of results of ultrasonographic imaging of the elbow joint (ie, identification of a free fragment; yes vs no) and intraoperative findings in 102 dogs (112 joints) suspected to have fragmentation of the medial coronoid process on the basis of clinical and radiographic findings.
| Ultrasonographic findings | Intraoperative findings | |||
|---|---|---|---|---|
| Free fragment | Nondisplaced fragment | No fragment | Total | |
| Fragment | 28 (55) | 5 (9) | 1 (17) | 34 (30) |
| No fragment | 23 (45) | 50 (91) | 5 (83) | 78 (70) |
| Total | 51 (46) | 55 (49) | 6 (5) | 112 (100) |
Data are given as number (%) of joints.
In 59 of the 112 (53%) elbow joints, deformation of the medial coronoid process (Figure 3) was identified during ultrasonography of the joint (Table 2). No deformation was seen ultrasonographically in 31 of the 51 (61%) joints in which a free fragment was found during surgery or in 20 of the 55 (36%) joints in which a nondisplaced fragment was found during surgery.
Transverse ultrasonographic image of the medial aspect of the elbow joint in a dog with fragmentation of the MCP. The contour of the radius (R) is smooth, but the contour of the ulna (U) is undulating with a discontinuity and step formation in the region of the MCP, indicating that the MCP has an abnormal shape. Whether there is a free fragment (arrow) can only be verified by dynamic examination.
Citation: Journal of the American Veterinary Medical Association 234, 4; 10.2460/javma.234.4.480
Cross-classification of results of ultrasonographic imaging of the elbow joint (ie, identification of a deformity of the medial coronoid process; yes vs no) and intraoperative findings in 102 dogs (112 joints) suspected to have fragmentation of the medial coronoid process on the basis of clinical and radiographic findings.
| Ultrasonographic findings | Intraoperative findings | |||
|---|---|---|---|---|
| Free fragment | Nondisplaced fragment | No fragment | Total | |
| Deformity | 20 (39) | 35 (64) | 4 (67) | 59 (53) |
| No deformity | 31 (61) | 20 (36) | 2 (33) | 53 (47) |
| Total | 51 (46) | 55 (49) | 6 (5) | 112 (100) |
Data are given as number (%) of joints.
Overall, an abnormality of the medial coronoid process (ie, a medial coronoid process fragment, deformation of the medial coronoid process, or both) was identified ultrasonographically in 90 of the 112 (80%) joints (Table 3). No abnormalities were seen ultrasonographically in 21 of the 106 (20%) joints in which abnormalities (ie, a free or nondisplaced fragment) were seen during surgery, and abnormalities were seen ultrasonographically in 5 of the 6 joints without evidence of medial coronoid process fragmentation during surgery.
Cross-classification of results of ultrasonographic imaging of the elbow joint (ie, identification of any abnormality; yes vs no) and intraoperative findings in 102 dogs (112 joints) suspected to have fragmentation of the medial coronoid process (FMCP) on the basis of clinical and radiographic findings.
| Ultrasonographic findings | Intraoperative findings | |||
|---|---|---|---|---|
| FMCP | No FMCP | Total | ||
| Abnormal | 85 (80) | 5 (83) | 90 (80) | |
| No abnormalities | 21 (20) | 1 (17) | 22 (20) | |
| Total | 106 (95) | 6 (5) | 112 (100) | |
Data are given as number (%) of joints.
Secondary lesions were identified during ultrasonographic evaluation of 96 of the 112 (86%) elbow joints and included new bone formation (30 [27%]; Figure 4), thickening of the joint capsule (26 [23%]; Figure 5), and joint effusion (50 [45%]; Figure 6). Forty-seven of the 96 (49%) joints with secondary lesions were found to have free fragments at surgery, and 43 (45%) were found to have nondisplaced fragments at surgery. The remaining 6 (6%) joints with ultrasonographic evidence of secondary lesions did not have evidence of medial coronoid process fragmentation at the time of surgery.
Longitudinal ultrasonographic image of the lateral humeral epicondyle in the right elbow joint of a dog. The proximal aspect of the humeral epicondyle (H) is smoothly delineated, but the distal aspect is irregular and rough secondary to osteophyte formation (black arrows).
Citation: Journal of the American Veterinary Medical Association 234, 4; 10.2460/javma.234.4.480
Longitudinal ultrasonographic image of the caudolateral pouch of the elbow joint in a dog. Notice the thickened joint capsule (C; arrows) and anechoic fluid consistent with mild joint effusion (J). The joint capsule ranged from 0.4 to 0.5 cm thick.
Citation: Journal of the American Veterinary Medical Association 234, 4; 10.2460/javma.234.4.480
Longitudinal ultrasonographic image of the caudolateral pouch of the elbow joint in a dog. Notice the mildly thickened joint capsule (C) and severe joint effusion (J). Echogenic material in the distal aspect of the joint space could be deposited cells or debris. The distal contour of the humeral epicondyle (H) is irregular, suggesting that there was new bone formation.
Citation: Journal of the American Veterinary Medical Association 234, 4; 10.2460/javma.234.4.480
For elbow joints with a free medial coronoid process fragment, estimated sensitivity of ultrasonographic detection of a fragment was 55% (95% CI, 40% to 69%), but for elbow joints with a nondisplaced medial coronoid process fragment, estimated sensitivity of ultrasonographic detection of a fragment was only 9% (95% CI, 3% to 20%). Estimated specificity (ie, the probability that a fragment would not be seen ultrasonographically in a joint without medial coronoid process fragmentation) was 83% (95% CI, 36% to 100%). Estimated positive predictive value of identifying a fragment ultrasonographically was 97% (95% CI, 85% to 100%), and estimated negative predictive value of not identifying a fragment ultrasonographically was 6% (95% CI, 2% to 14%). Overall, accuracy of using ultrasonographic evidence of a fragment to diagnose medial coronoid process fragmentation was only 34% (95% CI, 25% to 43%).
Estimated values for accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of using ultrasonographic evidence of any medial coronoid process abnormality (ie, a medial coronoid process fragment, deformation of the medial coronoid process, or both) for diagnosis of medial coronoid process fragmentation were 77% (95% CI, 68% to 84%), 80% (95% CI, 71% to 87%), 17% (95% CI, 0.4% to 64%), 94% (95% CI, 88% to 98%), and 5% (95% CI, 0.1% to 23%), respectively.
The kappa coefficient for the level of agreement between ultrasonographic (ie, any medial coronoid process abnormality) and surgical findings was −0.014, which was not significantly (P = 0.85) different from 0, indicating that there was no agreement.
Discussion
Results of the present study suggested that ultrasonography was of limited diagnostic value in dogs suspected, on the basis of clinical and radiographic findings, to have medial coronoid process fragmentation. In general, there was little agreement between ultrasonographic and intraoperative findings, and a fragment was seen ultrasonographically in only 33 of the 106 (31%) joints in which a fragment was identified during surgery.
During ultrasonography of dogs in the present study, we found that the medial coronoid process could be imaged most easily in a transverse plane and that ultrasonographic identification of the process was facilitated by mild pronation and supination of the antebrachium. Passive movement of the limb during ultrasonographic imaging aided in the identification of free fragments, as their movement was independent of the movement of the underlying bone.
In the present study, a medial coronoid process fragment was identified ultrasonographically in only 34 of the 112 (30%) joints, and fragments were not seen ultrasonographically in 23 of the 51 (45%) joints with a free medial coronoid process fragment or in 50 of the 55 (91%) joints with a nondisplaced medial coronoid process fragment. Possible reasons for not detecting a fragment during ultrasonography include partial fragmentation, minimal fragment displacement, and location of the fragment between the radial head and remaining portion of the medial coronoid process, where it was obscured by distal shadowing.1,2 Detection would also have been difficult if the medial coronoid process had only been fissured and was not displaced or abnormally shaped or if the fragment was not mineralized.2,4–6,a
In the present study, it was not always possible during ultrasonography of the elbow to distinguish between attached fragments, fissures, and an abnormal shape of the medial coronoid process. Therefore, we evaluated the diagnostic utility of using any ultrasonographic abnormality for diagnosing medial coronoid process fragmentation. In this instance, even though accuracy was 77% and sensitivity was 80%, calculation of the kappa coefficient indicated that there was no agreement between ultrasonographic and intraoperative findings.
In 1 elbow joint in the present study, a fragment was seen during ultrasonographic examination of the medial coronoid process, but no fragment was seen during surgery (ie, a false-positive result). The tip of the medial coronoid process appeared to be missing at the time of surgery, and it was suspected that a fragment could have been located in another part of the joint that was not accessible to the surgeon. However, the fragment could not be identified on routine postoperative radiographs.
In 47 of the 51 (92%) joints with free fragments and 43 of the 55 (78%) joints with nondisplaced fragments in the present study, secondary lesions were identified during ultrasonographic examination. Overall, joint effusion was identified in 50 (45%) joints and thickening of the joint capsule was identified in 26 (23%) joints. However, joint effusion and thickening of the joint capsule are nonspecific signs of joint abnormalities and, by themselves, are of little use in diagnosing fragmentation of the medial coronoid process.
The estimated positive predictive value of identifying a medial coronoid process fragment ultrasonographically in the present study was high (97%), in that fragments were identified surgically in 33 of the 34 joints in which a fragment was seen ultrasonographically. Similarly, the estimated positive predictive value of identifying any medial coronoid process abnormality ultrasonographically was 94%. The reason for these high positive predictive values was that we only included dogs in which medial coronoid process fragmentation was suspected on the basis of clinical and radiographic findings, with the result that 106 of the 112 (95%) joints had the condition. However, although estimated positive predictive values were high, estimated negative predictive values were low (6% and 5%), indicating that ultrasonography could not be used to determine the need for surgery and therefore would have little diagnostic impact.
An important disadvantage of ultrasonography is that it must be performed by an experienced sonographer,16,c in contrast to radiography. Orthopedic ultrasonography in particular has a long learning curve, with considerable experience required before one can confidently distinguish lesions from artifacts.c
The present study was limited by the fact that we only included dogs in which medial coronoid process fragmentation was already suspected on the basis of clinical and radiographic findings. This was not surprising because dogs in which fragmentation was not suspected would not have been expected to undergo arthrotomy. Because the ultrasonographer knew that all dogs were suspected to have an elbow problem, she may have been biased in favor of finding a fragment. However, the ultrasonographer was not aware of specific radiographic findings while performing the ultrasonographic examinations.
Abbreviations
| CI | Confidence interval |
References
- 1.
Miyabayashi T, Takiguchi M, Scrader SC, et al. Radiographic anatomy of the medial coronoid process of dogs. J Am Anim Hosp Assoc 1995;31:125–132.
- 2.
Nap RC. Pathophysiology and clinical aspects of canine elbow dysplasia. Vet Comp Orthop Traumatol 1996;9:58–59.
- 3.
Wind AP. Elbow dysplasia. In: Slatter D, ed. Textbook of small animal surgery. 2nd ed. Philadelphia: WB Saunders Co, 1993;1966–1977.
- 4.
Bardet JF. Arthroscopy of the elbow in dogs. Part II: the cranial portals in the diagnosis and treatment of lesions of the coronoid process. Vet Comp Orthop Traumatol 1997;10:60–66.
- 5.
Keller GG, Kreeger JM, Mann FA, et al. Correlation of radiography, necropsy and histologic findings in 8 dogs with elbow dysplasia. Vet Radiol Ultrasound 1997;38:272–276.
- 6.
Hornof WJ, Wind AP, Wallack ST, et al. Canine elbow dysplasia. The early radiographic detection of fragmentation of the coronoid process. Vet Clin North Am Small Anim Pract 2000;30:257–266.
- 7.
Burton NJ, Comerford EJ, Bailey M, et al. Digital analysis of ulnar trochlear notch sclerosis in Labrador retrievers. J Small Anim Pract 2007;48:220–224.
- 8.↑
Grondalen J. Arthrosis in the elbow joint of young rapidly growing dogs. VI. Interrelation between clinical, radiographical and pathoanatomical findings. Nord Vet Med 1982;34:65–75.
- 9.↑
Meyer-Lindenberg A, Langhahn A, Fehr M, et al. Prevalence of fragmented medial coronoid process of the ulna in lame adult dogs. Vet Rec 2002;151:230–234.
- 10.↑
Carpenter LG, Schwarz PD, Lowry JE, et al. Comparison of radiologic imaging techniques for diagnosis of fragmented medial coronoid process of the cubital joint in dogs. J Am Vet Med Assoc 1993;203:78–83.
- 11.
Wosar MA, Lewis DD, Neuwirth L, et al. Radiographic evaluation of elbow joints before and after surgery in dogs with possible fragmented medial coronoid process. J Am Vet Med Assoc 1999;214:52–58.
- 12.
De Rycke LM, Gielen IM, van Bree H, et al. Computed tomography of the elbow joint in clinically normal dogs. Am J Vet Res 2002;63:1400–1407.
- 13.
Reichle JK, Snaps F. The elbow. Clin Tech Small Anim Pract 1999;14:177–186.
- 14.
Snaps FR, Balligand MH, Saunders JH, et al. Comparison of radiography, magnetic resonance imaging, and surgical findings in dogs with elbow dysplasia. Am J Vet Res 1997;58:1367–1370.
- 15.↑
Knox VW IV, Sehgal CM, Wood AK. Correlation of ultrasonographic observations with anatomic features and radiography of the elbow joint in dogs. Am J Vet Res 2003;64:721–726.
- 16.↑
Kramer M, Gerwing M, Hach V, et al. Sonography of the musculoskeletal system in dogs and cats. Vet Radiol Ultrasound 1997;38:139–149.
- 17.
Lamb CR, Wong K. Ultrasonographic anatomy of the canine elbow. Vet Radiol Ultrasound 2005;46:319–325.
Snaps F, Balligand N, Saunders J, et al. Comparison of radiography and MRI for the diagnosis of fragmented medial coronoid process of cubital joint in dog (poster presentation). 4th Eur Assoc Vet Diagn Imag Conf, Cambridge, England, September 1996.
Michele U, Gerwing M, Kramer M, et al. Pathologic sonographical findings in the canine elbow (abstr). Vet Radiol Ultrasound 1999;40:559.
Michele U. Die Sonographie des Ellbogengelenkes und seines Weichteilmantels beim gesunden, erwachsenen Hund. Dr med vet thesis, Department of Surgery, Justus-Liebig University, Giessen, Germany, 2000.
BiAS für Windows, Biometrische Analyse von Stichproben, version 8.2, Epsilon-Verlag, Hochheim-Darmstadt, Germany, 2003.
