Standard radiographs are unreliable for detecting mild tibial torsion following tibial plateau leveling osteotomy in dogs.

Objective: To evaluate the diagnostic accuracy of standard postoperative radiographs for identifying mild tibial torsional malalignment (≤ 10°) following tibial plateau leveling osteotomy (TPLO) in dogs.

Methods: Cadaveric hind limbs from medium- to large-breed dogs without radiographic evidence of stifle pathology were used in this ex vivo experimental study. Each limb underwent TPLO, and a custom 3-D-printed guide was used to induce tibial torsion at 0°, 5°, and 10° in internal and external directions. For each condition, mediolateral and caudocranial postoperative radiographs were obtained. Osteotomy lines and guide pins were digitally masked. Three board-certified small animal surgeons independently reviewed each set to assess the presence, direction, and degree of torsion.

Results: 7 hind limbs were used, with 5 sets of radiographs per limb and 35 total sets. Correct identification rates were low when both direction and magnitude were considered. Two surgeons correctly classified 3 of 7 limbs without torsion, and 1 surgeon classified 4 of 7 limbs. Only 1 surgeon correctly classified 3 of 7 limbs with 5° internal torsion; for all other conditions, no surgeon correctly classified more than 2 of 7 limbs. When only direction was considered, internal torsion was correctly identified in 2, 4, and 5 of 14 radiographic sets, and external torsion was correctly identified in 3, 3, and 0 of 14 sets.

Conclusions: Torsional malalignment of up to 10° following TPLO is challenging to detect radiographically, indicating plain radiographs are unreliable for assessing torsional alignment.

Clinical relevance: Standard orthogonal radiographs are not reliable for assessing mild torsional malalignment after TPLO.