Objective: To report the relative sensitivity of different diagnostic imaging (DI) techniques to detect migrating foreign bodies (FB) in subcutaneous and underlying soft tissue structures of dogs and evaluate the value of intraoperative ultrasonography (intraop-US).Study design: Retrospective study. Sample population: Forty-one dogs.Methods: Medical records (2007-2019) of dogs that underwent surgery for a chronic abscess or draining tract were included when preoperative US (preop-US), computed tomography (preop-CT), or preop-MRI, and at least 6 months of follow-up information were available. Collected data included the DI techniques used, DI findings, and surgical findings. The relative sensitivity of each preoperative DI (preop-DI) technique was calculated. Dogs were divided into two groups, dogs that underwent only a preop-DI examination (group A) and dogs that underwent an additional intraop-US (group B). The success rates for FB surgical removal were compared between groups. Results: The relative sensitivity of preop-US and preop-CT/MRI was 88% (95% CI, 70%-95.8%) and 57.1% (95% CI, 32.6%-78.6%), respectively. The success rate for FB removal was higher in group B (89.5%) than in group A (59.1%; P = .038). Clinical resolution occurred in 90.2% of dogs for a median duration of follow-up of 4.2 years (6 months to 9.3 years). Conclusion: Preoperative US seems more suitable than preop-CT/MRI for detection of migrating FB in subcutaneous and underlying soft tissue structures. The use of intraop-US increases the success rate for FB surgical removal.Clinical significance: Intraoperative US should be used in combination with preop-US to increase the likelihood of migrating FB surgical removal in dogs.

The objective of this study was to evaluate the accuracy of ultrasonographic diagnosis of lesions in the canine stifle associated with cranial cruciate ligament rupture. Thirteen dogs that had a diagnosis of cranial cruciate ligament rupture were included in this prospective clinical study. Two ultrasonographers who were unaware of specific historical and clinical data performed the sonography with a high frequency (8-16 MHz) linear transducer. Surgical treatment of the affected stifle was performed within two days of ultrasonography by a surgeon who was unaware of the ultrasonographic findings. The lesions observed during ultrasonography and arthrotomy were compared at the completion of the study. Visualisation of the superficial tendons (quadriceps and long digital extensor) and ligaments (patellar ligament, collateral ligaments) of the stifle using ultrasonography was excellent. However, the detection of deep stifle ligaments (cranial cruciate ligament and caudal cruciate ligament) was extremely difficult to perform using ultrasonography. For cranial cruciate ligament rupture, the sensitivity for ultrasonographic diagnosis was 15.4%. For meniscal lesions, the sensitivity, specificity, positive and negative predictive values for ultrasonographic diagnosis were 82%, 93%, 90% and 88% respectively. High frequency ultrasonography is a non-invasive method for accurately and efficiently detecting superficial ligaments, tendons and meniscal lesions associated with cranial cruciate ligament rupture in the stifle of non-sedated dogs.

Objectives This study aimed to describe a stress radiography to detect medial glenohumeral ligament rupture and to investigate the repeatability and reproducibility of this test. Materials and Methods Six adult Beagle cadavers were used. Both shoulders of each dog were randomly assigned into two groups: a normal shoulder group (NS group) and an affected shoulder group (AS group) where the medial glenohumeral ligament was arthroscopically transected. Goniometry, image analysis and radiography (normal and stress views) were performed before and after arthroscopic procedures. An abduction angle was determined for each stress view and an evaluation of the repeatability and the reproducibility was performed. Results Abduction angles were significantly larger in the AS group as measured by goniometer (p < 0.001), image analysis (p < 0.001) and radiographic stress view (p < 0.001) than in the NS group. The abduction angle difference varied from 12° to 29.3° (median difference of 19.4°) between NS and AS groups. There were good intra-observer and very good inter-observer agreements. A very good (r = 0.74; r = 0.72, respectively) significant (p < 0.001) positive correlation between goniometric and radiographic techniques and between image analysis and radiographic techniques was observed. Clinical Significance Shoulder abduction angles measured with a stress radiograph-specific view provide objective data that may contribute to diagnosis of medial shoulder instability in dogs.

Given its poor reliability, the CT method is not recommended.