Objectives: To report the repeatability and reproducibility of four different anatomical proximal femoral axis (a-PFA) methods for measuring anatomical lateral distal femoral angle (a-LDFA), and to compare a-LDFA values produced by each method at three different femoral elevation angles.

Methods: Digital radiographs were obtained of seven dry canine femora at 0°, 12.5° and 25° elevations. Using image analysis software, landmarks defining four different a-PFA and the condylar axis were identified by two independent observers on two separate occasions. Corresponding a-LDFA were calculated for each femur, elevation and a-PFA. Repeatability and reproducibility parameters were calculated and compared statistically, along with the effect of technique and elevation on a-LDFA value.

Results: Interobserver repeatability coefficients were subjectively better for three of the a-PFA methods at 2° compared to the fourth at 3.1°. Median a-LDFA increased significantly (p 0.002) with increasing femoral elevation for all a-PFA methods, with a median increase of 3.3°. The median difference in a-LDFA between a-PFA methods yielding the highest and lowest measurements was 2.6° over all three elevations.

Clinical significance: The combined effects of a-PFA choice, femoral elevation and measurement reproducibility may produce typical errors of ± 2.6°, which could have implications for the selection of candidates for corrective osteotomies. Clinicians need to be aware that values obtained with one method and femoral elevation may not be equivalent to values obtained with other methods or elevations.