Can femoral rotation be localized and quantified using standard CT measures?

BACKGROUND: The terms "femoral anteversion" and "femoral torsion" have often been used interchangeably in the orthopaedic literature, yet they represent distinct anatomical entities. Anteversion refers to anterior tilt of the femoral neck, whereas torsion describes rotation of the femoral shaft. Together, these and other transverse plane differences describe what may be considered rotational deformities of the femur. Assessment of femoral rotation is now routinely measured by multiple axial CT methods. The most widely used radiographic technique (in which only two CT-derived axes are made, one through the femoral neck and one at the distal femoral condyles) may not accurately quantify proximal femoral anatomy nor allow identification of the anatomic locus of rotation. QUESTIONS/PURPOSES: (1) What CT methodology (a two-axis CT-derived technique, a three-axis technique adding an intertrochanteric axis--the "Kim method," or a volumetric three-dimensional reconstruction of the proximal femur) most accurately quantifies transverse plane femoral morphology; (2) localizes those deformities; and (3) is most reproducible across different observers?
METHODS: We constructed a high-definition femoral sawbones model in which osteotomies were performed at either the intertrochanteric region or femoral shaft. Transverse plane deformity was randomly introduced and CT-derived rotational profiles were constructed using three different CT methods. Accuracy and consistency of measurements of femoral rotation were calculated using p values and Fisher's exact test and intraclass correlation coefficients (ICCs).
RESULTS: All three CT methodologies accurately quantified overall transverse plane rotation (mean differences 0.69° ± 3.88°, 0.69° ± 3.88°, and -1.09° ± 4.44° for the two-plane, Kim, and volumetric methods, respectively). However, use of a single neck and single distal femoral axis does not reliably identify the anatomic locus of rotation, whereas the Kim and volumetric methods do (p < 0.0001). All three methods were highly reproducible between observers (ICCs of 0.9569, 0.9569, and 0.9359 for the traditional two-plane, Kim, and volumetric methods, respectively).
CONCLUSIONS: Only the Kim and volumetric methods can identify the anatomic location of transverse plane rotation and we recommend using one of the two techniques. Accurate anatomic localization of transverse plane rotation enables using precise anatomic terminology ("femoral torsion" versus "femoral [ante]version"). CLINICAL RELEVANCE: Current descriptions and treatment of femoral rotational deformities do not discriminate the location of rotation. The transverse plane femoral rotation requires a precise definition based on its anatomic location to maintain consistent communication between clinicians, because version of the neck and torsion of the shaft may have different treatment.