3D femoral neck anteversion measurements based on the posterior femoral plane in ORTHODOC® system

Abstract

This study provides a robust measuring method of the femoral neck anteversion angle for use in a total hip replacement pre-planning program. The femora of 24 patients (69.3 +/- 6.3 years old) were CT-scanned and converted into three-dimensionally volume-rendered models in ORTHODOC (ISS Inc., CA, USA) which is the pre-planning software for ROBODOC surgery. The Mod.ISS method (the modified ISS method), designed by authors, measures the anteversion angle of the proximal-most femoral neck confluence on the plane perpendicular to the femoral mechanical axis. 3D FNC method proposed by the authors of the present study involves measurement of the anteversion angle of three-dimensional femoral neck center on a plane perpendicular to the posterior femoral plane and parallel to the posterior condylar axis. Here, we found that interobserver reproducibility was 1.8 degrees (SD = 1.3) for the Mod.ISS method and 2.4 (SD = 1.9) for the proposed 3D FNC method. The anteversion angle of the local femoral neck axis was measured as theta = 25.3(L/D) + [corrected] 5.4 in L/D = 0.1-0.6, where L/D is distance (L) from the proximal-most neck confluence along the femoral mechanical axis, normalized with respect to the diameter of the femoral head (D). At L/D = 0.5, the anteversion angle of the femoral neck axis was coincident with the average femoral neck anteversion determined by the 3D FNC method. We conclude that the 3D FNC method is a gold standard for measuring the femoral neck anteversion applicable during both pre-operative and post-operative stages, because its femoral neck center can be determined in three-dimensional space during both stages.

A semi‐automated method for measuring femoral shape to derive version and its comparison with existing methods

“…The mean and standard deviation of version angle measured using the Reikeras method (12.4°±7.4°) are comparable with some values reported in the literature (13°±7° ) although lower than others (17.8°±8.9° ). The mean value of version angle measured using the Lee 2D method on the femurs in this study (8.1°±6.0°) is comparable with values reported in the literature for 2D CT methods (8.6° and 8.8° , (no standard deviations available)).…”

“…The results were therefore compared to other similar methods reported in the literature. Version measured with the new neck fitting approach on the set of cadaveric femurs available for the study showed similar results to some other 3D methods reported in the literature on normal femurs (18.5°±7.2° (mean ± standard deviation) , 16.6°±11.4° and 19.8°±9.3° ) although higher than others (12.5°±9.0° ). The mean and standard deviation of version angle measured using the Murphy method on the femurs in this study (22.6°±8.7°) are comparable with values reported in the literature (26.0°±9.1° ).…”

Femoral offset is underestimated on anteroposterior radiographs of the pelvis but accurately assessed on anteroposterior radiographs of the hip

“…9,12 Consequently, some authors have proposed pre-operative CT as a standard procedure. 10,11 Although CT is considered to be the best method for assessing femoral offset, 13,14 its routine usage has to be questioned because of its higher radiation dose, higher cost and limited availability.…”

3D Automatic Feature Construction System for Lower Limb Alignment

“…The lower limb alignment was studied in 2D diagnostic images or 3D reconstructed bone models [7][8][9][10][11][12][13][14][15][16][17][18][19][20] based on CT and X-ray scan images. In 2D studies, Iseki et al [3] evaluated the reliability of the inter-and intra-observer assessments of the mechanical alignment.…”

“…Wright et al [17] measured the lower limb axial alignment using the long radiographs. Lee et al [10] provided a robust measuring method of the femur neck anteversion angles, which also estimated the femur mechanical axis with the semi-manual method. These studies had used a variety of joint center points and axes in 2D diagnostic images, which were difficult to represent the complex 3D features.…”