Companion dogs treated in this study resided in their caretaker’s household during and after completion of the study. Prior to any study-related procedure, caretakers signed a consent agreeing to participation, treatment and follow-up evaluations through month 12 post treatment. Study-eligible dogs were at least 1 year of age and weighed 4.54 kg (10 lbs.) or more, exhibited clinically evident weight-bearing lameness of Grade +1 on a Lameness Scoring System,29 regardless of NSAID therapy, located in one or both elbows, and had radiographic evidence of Grade 3 OA in at least one elbow. Ongoing NSAID therapy was acceptable, provided the dog was initially lame at baseline evaluation while being treated with an NSAID. Additional criteria included MRI or CT evidence of osseous osteoarthritis changes and a grade 3+ for both questions 1 and 5 of the CBPI survey. Dogs were excluded from the study if they had comorbidities likely to preclude a one-year survival post treatment, evidence of bone neoplasia, severe osteochondritis dissecans (OCD) lesions, septic arthritis, a recently ruptured synovial cyst that communicated with the articular cavity, were pregnant or lactating, or had a known hypersensitivity to the active substance or excipients. This study protocol was approved by the Louisiana State University Institutional Animal Care and Use Committee (IACUC Protocol #16-008) and all methods were carried out in accordance with relevant guidelines and regulations outlined here.
Dogs were treated with a 1.75 mCi 117mSn colloid intra-articular dose, which was normalized by body weight based on a standard intra-articular dose for a 22.72 kg (50 lb) dog (Table 8). At the time of this study dose calibration standards for 117Sn had not been determined by the National Institute of Standards and Technology. The amount of radionuclide delivered was determined by the manufacturer based on rate constants, the date of manufacture, and the date the dose was to be administered. Enrollments were assessed with respect to the delivered dose of 117mSn colloid in comparison to the intended dose. If the injected dose was >20% under or over the intended dose, the subject was excluded from the PP analysis.
Baseline assessments were obtained within 30 days of treatment. A history and physical examination were performed at baseline and Days 1, 90, 180, 270 and 365. Blood was assessed by CBC and serum chemistry at baseline, Day 180 and Day 365. Urine was collected for urinalysis via free catch or ultrasound-guided cystocentesis at baseline, daily during the post-treatment hospitalization and on Days 180 and 365.
Digital radiography, CT and MRI were each performed at baseline and Day 365. MRI was also performed at Day 180.
Standard Treatment Protocol
Dogs were anesthetized and placed into right or left lateral recumbency depending on the clinician’s approach to the target elbow. Hair overlying the elbow joint was clipped and the skin was surgically prepared by scrubbing with povidone iodine or chlorhexidine, beginning at the injection site and spiraling outward. The elbow was covered with a sterile surgical drape. The procedure was performed using either a lateral or medial approach, depending on the surgeon’s preference. For the lateral approach, with the elbow flexed at 90 degrees, a sterile 22G 1½-inch diamond-tipped spinal needle was inserted caudolaterally between the lateral condyle of the humerus and the triceps tendon, then directed distal and slightly medial along the cranio-lateral aspect of the ulnar anconeus into the supratrochlear foramen of the humerus. For a medial approach, the needle was inserted approximately 1 cm distal to the medial epicondyle and directed perpendicular into the joint. Following needle placement, 0.5 to 1.0 mL of joint fluid was aspirated into a sterile syringe. This syringe was removed and a different sterile syringe containing the 117mSn colloid was attached to the same positioned needle. The colloid was injected slowly into the joint, followed by approximately 0.3 mL of air to clear as much of the colloid from the syringe and needle as possible. The needle was removed from the joint and direct pressure was applied to the injection site for 2 minutes to prevent leaking of the radioisotope and encourage hemostasis.
A radiographic osteoarthritis grade was assigned to each elbow joint, for each dog, as previously described.30 Scoring at Day 365 was made as related to the initial score and the immediately previous examination. Scoring was: -1 (worsening), 0 (static) or +1 (improvement).
The CT scan for each patient was evaluated for specific continuous and categorical variables (Table 5). Categorical variables were assigned a grade of 0 to 3 based on subjective assessment for severity of changes, where 0 indicated an absence of the metric and 3 indicated the presence of severe changes. One and 2 indicated mild and moderate severity, respectively.
MRI studies were performed using different sequences at each site owing to the difference in equipment between sites. Each investigator followed a pre-determined quantitative morphometric protocol for MRI evaluation. Cranial and caudal JPW was measured on sagittal T2W sequences. The sagittal proton density (PD) was used to assess for the presence of hypointense synovial bodies surrounded by hyperintense synovial fluid. A positive change was considered when an increase in fluid over time and a decrease in the synovial body size occurred. The presence of synovial fluid heterogeneity that could be created by fibrin strands, was considered a negative change. Subchondral changes in bone intensity were assessed in sagittal and dorsal reformatted images using the 3D sequences optimized for each site. Sclerosis of the subchondral bone was assessed at the ulna at the base of the medial coronoid process and trochlear sulcus. Cartilage erosions were assessed by determining if the cartilage covered the joint uniformly and was assessed in the PD weighted sequences. The thickness of the joint capsule medial and lateral to the joint was measured on dorsal multiplanar reconstructed 3D scans. Images were evaluated by the radiologist investigator using eFilm viewing software (IBM Watson Health, Armonk, NY). The image for each patient was evaluated for specific continuous and categorical variables (Table 9).
The CBPI was used at each visit to assess changes in pain and activity (www.caninebpi.com).22,23 The pain severity score (PSS) was calculated as the mean of questions 1 through 4. The pain interference score (PIS) was calculated as the mean of questions 5 through 10. Improvement was assessed by comparing each visit with baseline and comparing each visit to the most recent visit. Success was defined, using the criteria for successful treatment of an individual patient23, as the improvement of one or more full integer change for the PSS and improvement of two or more full integer changes for the PIS.
Clinician assessment was used to evaluate lameness during walk and trot based on a six-point scale with Grade 0 representing no lameness and Grade 5 representing continuous non-weight-bearing lameness. The worse value (highest) of the two assessments (walk or trot) at a follow-up visit was also collected and statistically analyzed.
Safety Data Analyses
A CBC, serum chemistry, urinalysis and joint fluid analysis at baseline and on Days 180 and 365 were performed by an independent laboratory. In addition, joint fluid parameters were reviewed with respect to safety-related considerations.
Data analyses were performed with statistical analysis software (SAS, version 9.4, SAS Institute Inc, Cary, NC). For those variables that were assessed for each elbow, the elbow was the experimental unit. For those variables that could only be evaluated for the whole animal, the dog was the experimental unit. For effectiveness, tests of statistical significance were completed at a two-sided alpha level of 0.05. For the CBPI assessment, success was presented as the number and percentage of dogs meeting each criterion for baseline compared with Days 90, 180, 270 and 365, and for each visit compared with the most recent visit. For the clinician assessment of lameness, descriptive statistics (number of dogs, mean, standard error of the mean (SEM), minimum, 1st quartile median, 3rd quartile and maximum values) were compared with baseline and Days 90, 180, 270 and 365. The same descriptive statistics were also compared with change from baseline for each visit and for change from the previous visit to the current visit. Within group p-values were generated by the paired t-test or Wilcoxon signed rank test, depending on the distribution of the data, for clinical lameness assessments. The CBPI success criteria data was compared with the clinical lameness assessments of walk and trot for each dog. Lameness success was defined as an improvement from baseline of one or more integers. Two-by‐two tables for success/failure were constructed for each visit. McNemar’s test of agreement was applied to each metric. P‐values ≥ 0.05 indicate agreement. Data for the categorical variables for both CT and MRI were analyzed using the Sign Test to compare baseline with post‐treatment scores for each animal and treated elbow. The pre‐ and post‐treatment MRI data for continuous variables were analyzed using repeated measures ANOVA with dog/leg as the replicate and time as the main effect. Least‐squares means were calculated for each time point