Introduction
CORA-based levelling osteotomy (CBLO) is a recent addition to techniques for management of cranial cruciate ligament deficiency in dogs (1,2). Preoperatively, careful identification of the proximal and distal tibial anatomic axes for correct localisation of the CORA, calculation of the required rotation to reduce the tibial plateau angle (TPA) to approximately 10°, and the chord length needed to achieve this rotation is necessary. Intraoperatively, these measurements must be judiciously transferred to the tibia to optimise outcome. The aim of this study was to model effects of inaccurate measurement transfer intraoperatively, for comparison with inter-surgeon differences during preoperative planning.
Materials and methods
Five landmarks (cranial and caudal tibial plateau, tibiotarsal joint centre, and craniocaudal diaphyseal midpoints at 1/3 and 2/3 the tibial length) necessary to define the tibial plateau, mechanical axis, and proximal and distal anatomical axes (2,3) were identified on 6 mediolateral tibial radiographs by 4 surgeons working independently. Measurements were made in ImageJ, using custom templates to aid localisation of landmarks, and copied to MS Excel. A custom macro was used to model rotational alignment of proximal and distal anatomical axes and the effects of mislocation of the angulation correction axis (ACA) by 3, 6 and 9 mm relative to the CORA at 10° steps through a full circle. The effects of rounding to nearest 0.5mm on chord length, and the effects of 0.5mm imprecision, on rotations around saw cuts of 12-30 mm radii were also simulated. The primary outcome measures were initial and final TPA, and associated interobserver errors. Secondary outcome measures were TPA error from optimal procedure TPA, and axis translation.
Results
Mean patient body mass was 32 kg (range 12-44 kg). Mean pre- and post-simulated CBLO TPA were 24.6° (SD 5.3°) and 9.3° (SD 0.7°): interobserver within-subject SD fell from 2.3° to 0.7° from pre to post. Mislocations of ACA by 3, 6 and 9 mm lead to mean TPA error ranges of 0.6°, 1.2° and 1.8°, and mean axis translation error ranges of 1.9, 3.9 and 5.9 mm, respectively. Chord length rounding produced mean absolute errors in TPA below 0.5°, whereas ±0.5mm chord length errors resulted in TPA error ranges of 1.6° to 4°, inversely related to saw radius.
Conclusion
Inter-surgeon variation in TPA measurement seems to be ameliorated by CBLO planning in this model. Modelled ACA positioning errors have relatively small effects on TPA compared to chord length errors. Larger saw radii result in smaller chord-related errors. Moving the ACA distally to accommodate a larger saw radius may reduce overall procedure errors.
1.Raske M., Hulse D., Beale B., Saunders W.B., Kishi E., Kunze C. (2013) Stabilization of the CORA based leveling osteotomy for treatment of cranial cruciate ligament injury using a bone plate augmented with a headless compression screw. Veterinary Surgery 42: 759-764
2.Piras A. (2012) Early clinical experiences with CORA based tibial plateau levelling osteotomy (CBTPLO) in toy breed dogs. Proceedings of the 16th ESVOT Congress, Bologna, Italy, 12th-15th September
3.Petazzoni M., Jaeger G.H. (2008) Atlas of Clinical Goniometry and Radiographic Measurements of the Canine Pelvic Limb, 2 ed. Milano, Merial. pp 63-67