Adipocytes in synovial fluid cytology: An approach for diagnosing synovial lipomatosis.

A 2-year-old neutered male bullmastiff dog was presented with chronic left hind limb lameness. Physical examination revealed left stifle effusion and medial buttress without cranial tibial thrust. Radiographs showed joint effusion and new bone formation at the patella apex. Magnetic resonance imaging showed increased synovial fluid, widening of the joint space, abnormal infrapatellar fat body and thinning of the cranial cruciate ligament. Synoviocentesis and cytologic evaluation of synovial fluid revealed marked mononuclear inflammation with abundant fatty tissue, suggesting synovial lipomatosis in conjunction with the imaging findings. The disease was confirmed histologically after sampling the lesion during arthrotomy. Synovial lipomatosis, characterized by extensive synovial adipose tissue proliferation of the synovial membrane, is a rare "tumor-like" disorder that usually affects the stifle. Although the etiology remains unclear, joint trauma, inflammation, instability, and lipid abnormalities have been proposed as causes. Inflammatory factors may promote synoviocyte and adipocyte hyperplasia that perpetuate the process. Surgical removal may be suggested to eliminate triggers and prevent future recurrences. The report provides the first cytological description of adipocytes in synovial fluid associated with the diagnosis of synovial lipomatosis in dogs. This case report underscores the potential effectiveness of cytologic analysis of synovial fluid smears, in combination with magnetic resonance imaging (MRI), for diagnosing this condition and reducing complications associated with arthrotomy for sampling purposes. Additionally, the case highlights that synovial lipomatosis should be considered as a potential differential diagnosis for synovial masses in dogs. Further cases are needed to validate these observations in veterinary medicine.

Evaluation of subchondral bone cysts in canine elbows with radiographic osteoarthritis secondary to elbow dysplasia.

OBJECTIVE: To investigate whether subchondral bone cysts (SBCs) were present in dogs with radiographic elbow osteoarthritis (OA) and to investigate their relationship with radiographic OA severity. STUDY DESIGN: Retrospective cross-sectional study. SAMPLE POPULATION: Thirty-eight Labrador retrievers (total of 76 elbows). METHODS: Elbow computed tomography (CT) images of 18 young (≤2 years old) and 20 old (>2 years old) Labrador retrievers, which presented for elbow-associated lameness, were reviewed. Radiographic elbow OA was graded into four groups based on the largest osteophyte size on CT. The presence, number, and maximum diameter of SBCs were determined. RESULTS: Subchondral bone cysts were only identified in elbows with osteophytic new bone formation. The number and size of SBCs were associated with radiographic OA severity (p < .001 and p = .041 respectively). Specifically, the rate at which SBCs were present increased for both moderate and severe OA in comparison with the mild OA (moderate OA RR = 2.46, 95% CI 2.08-2.92, p < .001; severe OA RR = 5.60, 95% CI 4.79-6.55, p < .001). For dogs with severe OA, there was an increased likelihood that their SBCs were larger than SBCs from dogs with mild OA (OR = 1.056, 95% CI 1.012-1.101, p = .012). No SBCs were observed in elbows without radiographic evidence of OA. CONCLUSION: Subchondral bone cysts were identified as a feature of radiographic elbow osteoarthritis in Labrador retrievers, and their number and size were indicative of the presence and severity of radiographic elbow OA. CLINICAL SIGNIFICANCE: Subchondral bone cysts are a potential imaging biomarker for quantitative assessment for canine OA.

Microcomputed tomographic analysis of the ulnar trochlear notch in medium- and large-breed canine cadavers with and without medial coronoid disease.

Medial coronoid disease (MCD) is a common disease often associated with thoracic limb lameness in medium- and large-breed dogs. The term MCD includes subchondral changes of the medial coronoid process (MCP), as well as pathologies of both cartilaginous surfaces. As there are only a few comprehensive and detailed studies on the trabecular structure of the medial coronoid, the goal of this study was to compare the trabecular structure of the ulnar trochlear notch of canine cadavers with and without MCD using different micro-computed tomographic (micro-CT) parameters. Fifty-eight elbow joints from 29 canine cadavers of MCD-predisposed and non-predisposed breeds (control group) were examined radiographically, macroscopically and by microcomputed tomography. The study included elbow joints of eight Labrador Retrievers (21.6-37 kg), seven Golden Retrievers (26.3-42 kg), seven Bernese Mountain dogs (31-47 kg) and seven dogs of non-predisposed breeds (19.7-52 kg) (control group). The final diagnosis of MCD was based on necropsy and micro-computed tomographic examinations. Micro-CT examinations were performed using XtremeCT II (Scanco Medical, Zurich, Switzerland) and the following parameters were examined: bone volume fraction (BV/TV), trabecular thickness (Tb.Th), trabecular separation (Tb.Sp), trabecular number (Tb.N), connectivity density (Conn.D) and degree of anisotropy (DA). Twenty-four elbows of 44 elbows of the predisposed breeds (Labrador Retrievers, Golden Retrievers Bernese Mountain dogs) showed subchondral changes and lesions of the cartilage surfaces. The result of this study is a higher density (BV/TV) of the trabecular bone of the ulnar trochlear notch in elbows affected by MCD compared to the control group. The increased density due to trabecular reconstruction in the ulnar trochlear notch is likely the result of selectively increased loading during life.

Computed tomographic imaging and surgical management of distal insertional avulsion fragments of the caudal cruciate ligament in four horses.

OBJECTIVE: To describe cases with caudal cruciate ligament (CdCL) avulsion fragments diagnosed based on computed tomography (CT) examination and report on arthroscopic fragment removal. ANIMALS: Four Warmblood horses with hindlimb lameness and osseous fragments located in the caudal medial femorotibial joint (mFTJ). STUDY DESIGN: Short case series. METHODS: CT and arthroscopic evaluation of the caudal mFTJ were performed. The caudal mFTJ and the insertion of the CdCL on the tibia were assessed and removal of the avulsion fragments was attempted in three horses using a cranial intercondylar approach. RESULTS: The fragment was not accessible via caudomedial approaches in one horse. A cranial intercondylar approach was used in three horses, allowing removal of the intra-articular fragment in two horses, and removal of two-thirds of the proximal fragment in the last horse. Acute, profuse, arterial bleeding occurred in this horse during surgery with transient postoperative soft tissue swelling. Comorbidities included medial femoral condyle cartilage defects (3), cranial cruciate ligament lesions (2), and medial collateral ligament lesions (2). Horses were followed up for 16 months (median, range 11-28 months), at which point all were back in ridden exercise; owners' satisfaction was good. CONCLUSION: CT examination confirmed the diagnosis and allowed evaluation of the stifle joint for comorbidities. A cranial intercondylar arthroscopic approach facilitated the removal of CdCL insertional avulsion fragments, although not always complete. CLINICAL SIGNIFICANCE: A cranial intercondylar approach can allow access to CdCL avulsion fragments, but complications and incomplete removal remain possible.

An update on mobility assessment of dogs with musculoskeletal disease.

Mobility impairments associated with musculoskeletal diseases, such as osteoarthritis and degenerative joint disease, affect approximately 200,000 dogs annually and pose a notable challenge to canine health and welfare. Osteoarthritis causes the remodelling of synovial joints, alongside inflammation and impaired mechanical function which can be extremely debilitating. Secondary osteoarthritis commonly affects dogs and can be exacerbated by previous joint abnormalities, such as patellar luxation or cranial cruciate ligament rupture. Although musculoskeletal diseases can affect dogs of any age, the early subtle signs of gait abnormalities are perhaps missed by owners, thus, dogs may be in the latter stages of osteoarthritis progression when they are presented to veterinarians. Dogs showing subtle signs of gait abnormalities must be presented to veterinary practices for acute diagnosis to prevent long-term deterioration. Musculoskeletal diseases, such as osteoarthritis and degenerative joint disease, are commonly diagnosed via visible radiographic changes. However, veterinarians can use a combination of subjective and objective clinical scoring systems, such as clinical metrology instruments and gait assessment in conjunction with radiography to aid their diagnosis and longitudinal monitoring of musculoskeletal diseases. These scoring systems may be more sensitive to earlier signs of mobility impairments in dogs, ultimately, promoting increased canine health and welfare by enabling pain reduction, improvement of muscle strength and preservation of joint function. Current canine mobility scoring systems available to veterinarians will be discussed in turn throughout this review for implementation into clinical practice.

Do owner-reported changes in mobility reflect measures of activity, pain and degenerative joint disease in cats?

OBJECTIVES: The aim of this blinded, nested case-control study was to compare cats with and without early owner-reported mobility changes using subjective and objective outcome measures (owner-completed questionnaires, orthopaedic examination). METHODS: A total of 57 cats with and without early owner-reported signs of impaired mobility were allocated to the case (n = 30) and control (n = 27) groups, respectively. Participating owners completed one inclusion and two pre-visit questionnaires (Feline Musculoskeletal Pain Index, VetMetrica). Cats were then visited in their own homes, where they underwent an orthopaedic examination, an assessment of their body condition score and temperament, and the placement of an accelerometer on their collar for 2 weeks. RESULTS: There was no significant difference between groups for age category, breed, sex, temperament and body condition score. Case cats scored significantly lower for the Feline Musculoskeletal Pain Index (P = 0.003) and the VetMetrica domain of Comfort (P = 0.002), but not Vitality (P = 0.009) or Emotional Wellbeing (P = 0.018). Total pain (P <0.0001), crepitus (P = 0.002) and thickening (P = 0.003) scores were higher in case cats, as was the presence of bilateral disease (P = 0.005, odds ratio 14) and the number of bilaterally affected joints (P = 0.001). CONCLUSIONS AND RELEVANCE: Both the Feline Musculoskeletal Pain Index and orthopaedic examination were able to differentiate cats with early owner-reported signs of impaired mobility from healthy cats. VetMetrica Comfort domain scores indicated a compromised quality of life for cats with early owner-reported signs of impaired mobility compared with healthy cats. Being able to recognise signs of mobility impairment earlier would allow interventions aimed at slowing disease progression, thereby improving feline health and welfare.

Long-term outcomes in dogs with elbow dysplasia, assessed using the canine orthopaedic index.

BACKGROUND: Elbow dysplasia (ED) is an important cause of lameness in dogs. This study aimed to report long-term outcomes in dogs with elbow osteoarthritis. METHODS: Demographic data, medical management, and scores from The American College of Veterinary Surgeons' Canine Orthopaedic Index (COI) were collected from owners of dogs radiographically screened for ED, graded as normal, mild, or moderate. Telephone interviews were performed in 2017 (Q1), followed by an email survey in 2020 (Q2). The association between ED grade and deterioration in COI scores over time was evaluated with logistic regression. RESULTS: A total of 765 replies were collected for Q1 and 293 for Q2. At Q2, 222 dogs (76%) were alive, with a median age of 8 years (range 5-12 years). No association was found between ED and changes in COI score over time or between ED and survival (p = 0.071). Dogs with mild and moderate ED were treated with analgesic medications to a higher degree than dogs without ED (p < 0.05). LIMITATIONS: Only owner-assed data were assessed; no clinical orthopaedic examination or follow-up radiographic evaluation was performed. CONCLUSIONS: No association was found between the grade of ED and the worsening of clinical signs in dogs with elbow osteoarthritis.

Common Orthopedic Traits and Screening for Breeding Programs.

Orthopedic diseases are complex traits, meaning genetics and environmental factors affect risk, making identification of genetic associations difficult. In the United States, hip and elbow scores, patellar luxation scores, Legg-Calvé-Perthes disease, and shoulder osteochondrosis affectedness are available in the Orthopedic Foundation for Animals registry. Distraction indices and extended, ventrodorsal hip conformation scores are recorded by PennHIP. Application of estimated breeding values for hip and elbow dysplasia in breeder selection reduces the severity and prevalence of these traits. Genomic prediction and whole-genome sequence technologies and methods should improve knowledge of genetics underlying orthopedic diseases, leading to improved canine orthopedic genetic quality.

Comparative evaluation of radiographic and computed tomographic findings in dogs with bilateral medial coronoid disease (MCD) presenting with unilateral forelimb lameness.

OBJECTIVES: The purpose of this study was to compare the radiographic and computed tomographic (CT) findings of dogs with diagnosed bilateral medial coronoid disease, which showed clinically only unilateral lameness of the forelimbs. MATERIALS AND METHODS: Medical records, including radiographs and CT images of dogs with diagnosed bilateral MCD showing only a unilateral forelimb lameness clinically were reviewed retrospectively. Depending on the gait of each dog we established two groups to investigate their radiographs and CT data comparatively. Group I: affected non-lame limb. Group II: affected lame limb. Several evaluation systems were used to assess which factors are important for clinical decision making and a patient tailored therapeutic plan. RESULTS: Data from 84 affected elbow joints (42 dogs) diagnosed with MCD by computed tomography were included. Both the radiological and the CT analysis showed that there are significant differences between Groups I and II. Group I had a lower modified International Elbow Working Group Score (IEWG), the values of the Trochlear notch sclerosis were only slightly deviated, and this group showed less often a dislocation of the fragment compared to group II. Furthermore, the size of the fragment (both the median and the mean value) of the forelimbs from group II was almost twice as big as the one from group I. The following sizes of the fragments were calculated (group I versus (vs.) group II)-median: 0.09 cm2 vs. 0.16 cm2, mean value: 0.112 cm2 vs. 0.202 cm2. It could be shown that a larger fragment is more likely to dislocate than a smaller one. CLINICAL SIGNIFICANCE: This study provides some evidence towards a better understanding of which diagnostic parameters and findings might be important in clinical decision making. Nevertheless, a "decision tree" for the correct therapy of MCD could not be determined in this study.

Conservative management of equine tarsal collateral ligament injuries may allow return to normal performance.

OBJECTIVE: To describe type and distribution of tarsal collateral ligament (CL) injuries and to assess the long-term outcome in horses treated conservatively. ANIMALS: 78 horses (median age, 7 years [IQR, 4 to 9.75 years]) of different breeds and disciplines. PROCEDURES: Retrospective analysis (2000 through 2020) of horses with tarsal CL lesions diagnosed on ultrasound. The resting time, ability to return to work, and performance level after the injury were compared between horses having a single ligament (group S) or multiple ligaments (group M) affected and according to the case severity. RESULTS: Most of the horses (57/78) presented a single CL injury, while 21 had multiple CLs affected simultaneously, for a total of 108 CLs injured and 111 lesions. In both groups, the short lateral CL (SLCL) was the most commonly affected (44/108), followed by the long medial CL (LMCL; 27/108). Enthesopathies (72.1%) were more frequent than desmopathies alone (27.9%) and involved mostly the proximal insertion of the SLCL and the distal attachment of the LMCL. Conservative treatment (n = 62) consisted mainly of stall rest. The median resting time (120 days [IQR, 60 to 180 days]) did not significantly differ between the 2 groups (group S vs M) or according to the severity. Most horses (50/62) were able to return to work within 6 months. Horses that did not return (12/62) were more likely to have severe lesions (P = .01). Thirty-eight horses were able to perform at a level equal to or higher than before the injury. CLINICAL RELEVANCE: This study highlights the importance of thorough ultrasound assessment of tarsal CL injuries and demonstrates that conservative management is a viable option to allow these horses to return to previous performance level.

Examination of Proximodistal Patellar Position in Dogs with the Stifle at Full Extension.

OBJECTIVE: The aim of this study was to determine the factors related to functional patella alta, with which the proximodistal patellar position exceeds the reference range of healthy small dogs proximally when the stifle is at full extension. STUDY DESIGN: Mediolateral-view radiographs of dogs weighing less than 15 kg were obtained and classified into the medial patellar luxation (MPL) and control groups. The reference range of the proximodistal patellar position was determined from the control group. In both groups, a patellar position exceeding this reference range proximally was considered functional patella alta. Multiple logistic regression analysis was performed to examine the factors related to functional patella alta. A receiver operating characteristic (ROC) curve was made for each factor. RESULTS: Overall, radiographs of 127 stifles of 75 dogs were obtained. Eleven stifles in the MPL group and one stifle in the control group were determined to be functional patella alta. The factors associated with functional patella alta included a greater full extension angle of the stifle joint, longer patellar ligament, and shorter femoral trochlear length. The full extension angle of the stifle joint had the biggest area under the ROC curve. CONCLUSION: Mediolateral radiographs of the stifle joint in full extension are clinically important in dogs with MPL because some dogs might have a proximally positioned patella that is recognizable only when the stifles are extended.

Evaluation of a small-bore needle arthroscope for diagnosis and treatment of medial coronoid disease in dogs: a pilot study with short-term assessment.

CASE HISTORY: Dogs (n = 15) that were presented to a single veterinary teaching hospital with elbow dysplasia-associated lameness between September 2021 and May 2022, and were determined to require arthroscopy based on imaging results, were prospectively recruited into the study. The median duration of lameness was 4 (min 1, max 24) months. CLINICAL FINDINGS: Various breeds were represented with a median body weight of 31.6 (min 15, max 46.4) kg and median age at presentation of 14 (min 8, max 83) months. Results of imaging modalities (CT) were consistent with medial coronoid disease with fissured or fragmented medial coronoid process in all dogs. ARTHROSCOPIC FINDINGS: Feasibility of the needle arthroscopy (NA) procedure was firstly assessed in a preliminary cadaveric study in forelimbs (n = 10) collected from 10 adult dogs euthanised for reasons unrelated to the study. Elbow exploration was performed through a medial approach beginning with NA (1.9 mm 0° angle scope) followed by standard arthroscopy (SA; 2.4 mm 30° angle scope). The quality and extent of visualisation (scored through the number of anatomical structures visualised) were recorded and statistically compared. As the cadaver study indicated that NA allowed safe inspection of all structures in medial/caudal compartments, this procedure was then used in the dogs requiring treatment. In the clinical setting, elbow exploration was successful in all dogs and the treatment (removal of osteochondral fragments) was performed without requiring conversion into SA. One month after surgery, all dogs had an improvement in their lameness score (0-5) and 12/15 dogs were no longer lame. There was a reduction in Canine Orthopaedic Index scores measured a median of 99 (min 47, max 180) days after surgery (24 (IQR 19.5-31.5)) compared to the pre-operative period (49 (IQR 46.5-57); p < 0.001). CLINICAL RELEVANCE: Needle arthroscopy-assisted removal of osteochondral fragments was performed in all dogs with satisfactory short-term clinical outcome. NA is a feasible technique for diagnosis and lesion assessment in dogs with a fissured or fragmented coronoid process. Larger clinical studies with longer follow-up are necessary to validate the NanoScope operative arthroscopy system as an alternative strategy to SA for video-assisted treatment of medial coronoid disease.

Use of Ultrasonography in Diagnosis of Medial Compartment Disease of the Elbow in Dogs.

OBJECTIVE: The objective of this prospective study was to evaluate the use of ultrasonography in the diagnosis of medial coronoid process disease in unclear cases. STUDY DESIGN: Fifteen elbows (on thirteen dogs) for which radiography and computed tomography did not lead to a clear diagnosis of medial coronoid process disease were included. On each elbow, ultrasonography was performed with a high frequency linear transducer (12-18Hz). Then, arthroscopic examination of the joint was performed by a surgeon who was unaware of ultrasonographic findings to confirm medial coronoid process disease. RESULTS: At least one ultrasonographic lesion was detected in 13 out of 15 elbows. The main reported ultrasonographic lesions were joint effusion (10/15 elbows) and an abnormal shape of the medial coronoid process (irregular, ill-defined or fragmented) (9/15). CONCLUSION: Ultrasonography can be a helpful additional diagnostic tool to confirm medial coronoid process disease of the elbow joint before performing arthroscopy in unclear cases. Further studies will be needed to evaluate the use of higher frequency transducers and determine if it could improve the diagnostic value of ultrasonography.

The use of subchondral bone topography to approximate the center of rotation of the elbow joint in dogs.

OBJECTIVES: The aim of this study was to compare the approximate center of rotation in normal and diseased elbows in dogs. STUDY DESIGN: Cross-sectional study SAMPLE POPULATION: Computer tomography scans of nine dogs with unilateral fragmented medial coronoid process (FMCP). METHODS: A board certified radiologist confirmed that each dog had unilateral FMCP, and a normal contralateral elbow. Digital 3D models of all elbow joints were uploaded into a surgical planning software package. Four axes approximating the center of rotation (COR) of elbow joints were generated using five geometric shapes based on subchondral topography of the humeral condyle radius and ulna. Images showing the locations where axes exited the medial and lateral cortex of the humeral condyle were captured and imported into a second software package, for measurement of distances between exit points and the origin of a system of axes. RESULTS: In normal joints 20/27 (74%) axes exited the medial cortex, and 25/27 (93%) axes exited the lateral cortex cranial and distal to the medial and lateral epicondyles, respectively. In diseased joints 22/27 (81%) axes exited medial cortex and 19/27 (70%) axes exited the lateral cortex, caudal and distal to the medial and lateral epicondyles, respectively. CONCLUSION: Based on CT- derived geometry, the COR of elbow affected with FMCP was generally more caudal than normal. CLINICAL SIGNIFICANCE: External landmarks approximating the location of the elbow COR are provided, and while not validated, may assist in planning, creation, and assessment of procedures for FMCP.

Deep learning-based diagnosis of stifle joint diseases in dogs.

In this retrospective, analytical study, we developed a deep learning-based diagnostic model that can be applied to canine stifle joint diseases and compared its accuracy with that achieved by veterinarians to verify its potential as a reliable diagnostic method. A total of 2382 radiographs of the canine stifle joint from cooperative animal hospitals were included in a dataset. Stifle joint regions were extracted from the original images using the faster region-based convolutional neural network (R-CNN) model, and the object detection accuracy was evaluated. Four radiographic findings: patellar deviation, drawer sign, osteophyte formation, and joint effusion, were observed in the stifle joint and used to train a residual network (ResNet) classification model. Implant and growth plate groups were analyzed to compare the classification accuracy against the total dataset. All deep learning-based classification models achieved target accuracies exceeding 80%, which is comparable to or slightly less than those achieved by veterinarians. However, in the case of drawer signs, further research is necessary to improve the low sensitivity of the model. When the implant group was excluded, the classification accuracy significantly improved, indicating that the implant acted as a distraction. These results indicate that deep learning-based diagnoses can be expected to become useful diagnostic models in veterinary medicine.

Residual effects of intra-articular betamethasone and triamcinolone acetonide in an equine acute synovitis model.

BACKGROUND: Intra-articular (IA) corticosteroids are regularly used in equine athletes for the control of joint inflammation. OBJECTIVES: The goal of this study was to use an acute synovitis inflammation model to determine the residual effects of IA betamethasone and triamcinolone acetonide on various inflammatory parameters and lameness. STUDY DESIGN: Crossover randomised trial. METHODS: Five mixed-breed, 2-year-old horses were randomly allocated to an IA treatment of the radiocarpal joint with 9 mg of either betamethasone or triamcinolone acetonide. Two weeks following treatment, horses were injected with 1 µg of lipopolysaccharide (LPS) diluted in 1 ml of saline. Following LPS injection, horses were crossed-over and both sets of injections were repeated after a washout period. Blood samples were collected at multiple time points for mRNA analysis, as well as serum amyloid A (SAA) and cortisol determination. At each time point, lameness was also subjectively scored. Additional injections with saline-only or LPS-only (twice) were conducted as negative and positive controls, respectively. Two-way repeated measures analysis of variance was used to analyse all data. RESULTS: Corticosteroid-only treatments result in significant mRNA expression differences, as well as significant and prolonged cortisol suppression. Following LPS injection, there was a residual treatment effect with triamcinolone evidenced by a significant treatment effect on IL-6 and PTGS1 (cyclooxygenase-1), lameness, SAA and cortisol concentrations, while only IL-6 expression was affected by betamethasone. MAIN LIMITATIONS: The acute synovitis model used here results in significant inflammation and is not representative of the low-grade inflammation seen with typical joint disease and residual anti-inflammatory effects may be more profound in naturally occurring joint disease. CONCLUSIONS: Current regulatory guidelines may be insufficient if the concern is residual anti-inflammatory effects. Additionally, intra-articular corticosteroid administration is not without risk, as evidenced by a significant suppression of serum cortisol concentration and, as such, the benefits of their administration should be weighed against those risks.

Degree of osteophyte formation seen on radiographs provides useful prognostic information for dogs with cruciate ligament disease.

BACKGROUND: Concurrent osteoarthritis is generally present in the stifle joints of dogs with cranial cruciate ligament disease (CCLD), but it is not known if the degree of osteoarthritis-related changes affects the prognosis. Development of osteophytes is a key radiographic feature of osteoarthritis. The aim of this study was to evaluate the association between the degree of osteophyte formation seen on radiographs and CCLD-related euthanasia in dogs with CCLD. METHODS: A retrospective cohort study including 226 dogs treated for CCLD at two university animal hospitals was performed. Clinical and follow-up information was retrieved, and stifle radiographs taken during the 31 days before treatment for the CCLD were graded for osteophyte formation by three veterinary radiologists. A multivariable Cox proportional hazards model was applied to assess the association between the degree of osteophyte formation and CCLD-related euthanasia. RESULTS: An association between the degree of osteophyte formation and CCLD-related euthanasia was found, with increased osteophyte formation associated with an increased hazard of CCLD-related euthanasia (hazard ratio 1.06, 95% confidence interval 1.01-1.11, p = 0.01). LIMITATIONS: No clinical assessment of the outcome was performed. CONCLUSION: The findings suggest that evaluation of stifle radiographs for osteophyte formation could provide useful prognostic information for dogs with CCLD.