Healthy middle-aged Asian and Caucasian populations present with large intra- and inter-individual variations of lower limb torsion.

PURPOSE: There is a lack of standardization in the measurement of lower limb torsional alignment. Normal values published in the literature are inconsistent. A 3D-CT-scan-based method was used in a healthy population to define the femoral neck version (FNV) and the tibial torsion (TT) and their relationship with demographic parameters. The study objectives were (1) to define normal values of lower limb torsional alignment, (2) to estimate inter- and intra-individual variations of torsional deformity of healthy individuals' lower limbs. The hypothesis was that FNV and TT values would be influenced by patient characteristics such as gender, age, and ethnicity, and would have low side-to-side asymmetry. METHODS: Torsional landmarks of the lower limbs from 191 healthy subjects were automatically calculated with a 3D CT-scan-based program. The FNV was defined by the angle between the femoral neck axis and the femoral posterior condylar line. The TT angle was considered between the tibial plateau axis and the axis of the ankle. For the former, two alternatives were considered: the line connecting the more medial and lateral point of the medial and lateral plateau, respectively (method 1; TT1), or the line connecting the two more posterior points of the medial et lateral plateau (method 2; TT2). The ankle axis was defined as the line connecting the medial and lateral malleoli. These reference lines were automatically calculated. Age, gender, ethnic group, and BMI were recorded for every subject. A p value < 0.05 was considered as statistically significant. RESULTS: Overall, the mean FNV was 15.3 ± 9.5° and the mean TT was 31.6 ± 6.3°. Female hips were more anteverted than male hips. Caucasians had less anteverted hips than Asians, but more externally rotated tibias. Age and BMI were not correlated with any anatomical parameter. A substantial side-to-side asymmetry was found for FNV [absolute difference (AD) = 6.3°; percentage of asymmetry (%As) = 47%], TT1 (AD = 3°; %As = 12%), and TT2 (AD = 4.9°; %As = 9%) (p = 0.008). CONCLUSION: The findings showed that lower limb torsional parameters were highly variable from patient to patient and from one leg to the other for the same patient. The understanding of normal values concerning femoral version and external tibial torsion in the present healthy population will help surgeons to define pathological values of FNV and TT, as well as corrections to perform in case of torsional deformities. LEVEL OF EVIDENCE: Level III.

Which factors influence proximal femoral asymmetry?: a 3D CT analysis of 345 femoral pairs.

Aims: In patients where the proximal femur shows gross deformity due to degenerative changes or fracture, the contralateral femur is often used to perform preoperative templating for hip arthroplasty. However, femurs may not be symmetrical: the aim of this study was to determine the degree of variation between hips in healthy individuals and to determine whether it is affected by demographic parameters. Materials and Methods: CT-scan based modelling was used to examine the pelvis and bilateral femurs of 345 patients (211 males, 134 women; mean age 62 years (standard deviation (sd) 17), mean body mass index 27 kg/m2 (sd 5)) representing a range of ethnicities. The femoral neck-shaft angle (NSA), femoral offset (FO), femoral neck version (FNV), femoral length (FL), femoral canal flare index (fCFI), and femoral head radius (FHr) were then determined for each patient. All measurements were constructed using algorithm-calculated landmarks, resulting in reproducible and consistent constructs for each specimen. We then analyzed femoral symmetry based on absolute differences (AD) and percentage asymmetry (%AS) following a previously validated method. Results: We found an asymmetry > 2% for NSA (mean AD 2.9°, mean %AS 2.3; p = 0.03), FO (AD 3.8 mm, %AS 9.1 ; p = 0.01), FNV (AD 5.1°, %AS 46.7 ; p = 0.001) and fCFI (AD 0.2 mm, %AS 5.4 ; p = 0.7). Significant relationships were found for AD regarding NSA and ethnicity (p = 0.037), FL and height (R2 = 0.22), and fCFI and gender (R2 = 0.34). Conclusion: Our data confirm the presence of asymmetry of proximal femurs, which is mostly independent of demographic parameters. In cases where contralateral templating is used, such asymmetry may lead to inaccurate anatomical restoration of the hip if the templated sizes are routinely implanted. However, the clinical impact cannot be determined from our investigation. Cite this article: Bone Joint J 2018;100-B:839-44.