Precision of a Novel Computed Tomographic Method for Quantification of Femoral Varus in Dogs and an Assessment of the Effect of Femoral Malpositioning

Abstract

To assess the precision of a novel protocol for determination of femoral varus angle (FVA) using computed tomography (CT) in dogs, and to quantify the effect of femoral rotational and sagittal plane malpositioning on measured FVA. Cross-sectional study. Femora (n = 66) from dogs that had pelvic limb CT examination for patellar instability (26) or other reasons (10). Three observers measured FVA of each of 66 femora on three separate occasions. Standardized orientation of a volume rendered image was achieved by superimposition of the caudal and distal aspects of the femoral condyles on a lateral projection, definition of a sagittal plane axis, and finally rotation through 90° to yield a cranial projection. Intra- and inter-observer variability were estimated using the intra-class correlation coefficient. The effect of variation in rotational and sagittal plane orientation on measured FVA was subsequently quantified using 6 femora with FVAs between -0.4° and 19°. Intra-class correlation coefficients for the 3 observers, indicating intra-observer variation, were 0.982, 0.937, and 0.974. The intra-class correlation coefficient of the means of the results from each observer, indicating inter-observer variation, was 0.976. Consistent linear variations in measured FVA occurred as a result of rotational malpositioning in all 6 tested femora, and as a result of sagittal plane malpositioning in femora with FVAs ≥ 7.9°. The reported protocol for the measurement of FVA in dogs is repeatable and reproducible. Small variations in femoral orientation, as might be expected with conventional radiography, lead to clinically significant alterations in measured FVA.