Introduction: Extracorporeal shockwave therapy (ESWT) is used as an adjunctive treatment for canine osteoarthritis (OA), but its effects in dogs with treatment-refractory advanced disease remain unclear. This study compared the efficacy of one versus two sessions of focused ESWT administered approximately 28 days apart in dogs with refractory elbow or stifle OA. Methods: In this randomized, double-blinded clinical trial, twenty-four client-owned dogs with treatment-refractory elbow (n = 12) or stifle (n = 12) osteoarthritis received ESWT using an identical per-session protocol (X-Trode, 1000 pulses at 0.14 mJ/mm²; PulseVet-Zomedica, Ann Arbor, MI, USA), once (Group L) or twice (Group E). Orthopedic examination, goniometric and limb circumference measurements, and kinetic gait analysis (peak vertical pressure [PVP], vertical impulse [VI]) were performed on days 0, 28, and 56. Owner questionnaires (Canine Brief Pain Inventory [CBPI], Client Specific Outcome Measures [CSOM]) were collected on days 0, 28, 56, and 84. Data were analyzed using chi-squared tests, t-tests, and mixed effects models in R. Results: Age, weight, BCS, and radiographic osteoarthritis severity did not differ between groups at baseline. Improvement was small and limited to selected parameters. Vertical impulse and limb circumference increased more consistently in Group E, whereas peak vertical pressure increased in both groups, including before ESWT in Group L. No sustained or treatment-associated improvement was detected in symmetry variables or joint range of motion. Owner-reported outcomes showed variable patterns without consistent treatment effects. ESWT was well tolerated, and no major adverse events occurred. Conclusion: ESWT produced modest, inconsistent improvements in dogs with treatment-refractory OA, with slightly more consistent effects following two sessions. Therapeutic efficacy appeared limited in this end-stage population.

Background: Standardized and repeatable assessment of feline gait remains challenging, highlighting the need for controlled measurement systems such as pressure-sensitive treadmills. Objectives: This study aimed to compare kinetic and temporospatial gait parameters in healthy cats walking on a treadmill operated in standing vs. running mode. We hypothesized that no significant differences would be observed between modes. Methods: Ten healthy adult cats were enrolled in a prospective, controlled crossover study. Inclusion criteria included body weight >4.0 kg, body condition score (BCS) 4–6/9, and age >24 months. Cats walked on a pressure-sensitive treadmill (FDM-T-CanidGait ^® , zebris Medical GmbH, Isny, Germany) under two conditions: standing mode (stationary belt) and running mode (moving belt). Five 2-min trials per mode were recorded. Kinetic and temporospatial gait parameters were assessed, including peak vertical force, vertical impulse, stride length, stance and swing phase duration, and step width. Right and left limb data were pooled and analyzed collectively as forelimbs (FL) and hindlimbs (HL). Data were analyzed using robust linear mixed-effects models (R 4.4.3), with significance set at p < 0.05. Results: Significant differences between treadmill modes were detected for peak vertical force and vertical impulse in both forelimbs and hindlimbs ( p < 0.01). Forelimb swing and stance phase durations also differed significantly ( p < 0.05). In contrast, stride length, step width, and hindlimb temporal parameters did not differ significantly between modes. Estimated differences were generally small in magnitude across variables. Conclusion: In healthy cats, stride length, step width, and hindlimb temporal parameters were comparable between standing and running treadmill modes. In contrast, kinetic variables (peak vertical force and vertical impulse) and forelimb stance and swing phase durations differed significantly between modes and should therefore be interpreted only within the same treadmill mode. These findings indicate that treadmill modes may be used interchangeably for selected temporospatial parameters but not for kinetic variables or forelimb temporal parameters.