A three-dimensional computed tomographic volume rendering methodology to measure the tibial torsion angle in dogs.

Abstract

To describe a three-dimensional (3D) computed tomographic (CT) methodology to measure the tibial torsion angle (TTa) and to evaluate intrarater and interrater agreements and accuracy through comparison with anatomic measurements. Ex vivo cadaveric study. Thirty-six tibiae from 18 dogs. Tibial torsion angle of each tibia was measured by using two CT techniques (axial and 3D volume rendering) by three raters who blindly measured TTa in duplicate. A semitransparent bone filter was used to enhance the visibility of the target anatomical landmarks for the 3D volume rendering CT technique. Tibial torsion angle was also quantitated in tibial specimens. Intrarater and interrater agreements were analyzed by using intraclass coefficients (ICC). Accuracy was evaluated by using adjusted R2 coefficients (R2 > 80% was considered acceptable). The 3D volume rendering CT technique had excellent intrarater and interrater agreements (ICC > 0.94) and an R2 value of 97%. The axial CT technique had good to excellent intrarater and interrater agreements (0.8 < ICC < 0.95) and an R2 of 86%. No difference was found between axial and 3D CT techniques. A mean internal TT angle of approximately -6° was found with CT and anatomic measurements. The 3D volume rendering and axial CT techniques were precise and accurate for measuring TTa in dogs unaffected by patellar luxation. Combining 3D bone manipulation with application of a semitransparent filter allows simultaneous visualization of anatomic landmarks, which may facilitate the evaluation of complex bone deformations. Internal tibial torsion may be present in nonchondrodystrophic dogs without patella luxation.