Poster 184: Assessment of Pelvic Orientation and Influence on 3D Morphological Analyses of the Hip

Abstract

Objectives:

Femoroacetabular Impingement (FAI) is the primary clinical indication for hip arthroscopy. FAI can occur due to a bony deformity on the femur, known as a cam, or on the acetabular rim, referred to as a pincer. The bony deformity causes a painful hip, and results in limited hip motion and soft tissue damage. The patients’ posture can influence impingement; specifically, forward pelvic tilt may exaggerate impingement. The patient’s natural pelvic tilt can be observed in standing radiographs. However, radiographs do not provide as much information about bony deformity as 3D imaging. 3D imaging is acquired in a supine rather than standing position, and therefore does not represent the physiological posture of the patient. Previous research focused on how changes in pelvic tilt change radiographic measurements of the hip, but how much pelvic tilt influences other 3D morphological analyses remains unclear. The objective of this study is to assess how pelvic tilt influences 3D morphological analyses of the hip through systematic simulation of various levels of pelvic tilt. Additionally, it investigated how distance/angle measurements between anatomical features are impacted by pelvic tilt.

Methods:

Twenty CT (Computed Tomography) scans of patients (14 female, 6 male, age range 18-56 years) with hip pain (3 cam, 3 pincer, 4 combined, 6 hip instabilities, 4 no cam or pincer presence) collected through an IRB (Institutional Review Boards) were used in this study. Three-dimensional surface reconstructions were virtually adjusted based on the anterior pelvic plane to obtain samples with a pelvic tilt between 20° anterior and 20° posterior inclination in steps of 5 degrees using 3-matic Medical (Materialize). Three-dimensional morphological analyses were obtained through HipMap software (Stryker) and included: center edge angle (CEA), acetabular version (at three levels: AV12, AV2, AV3), Tönnis angle (TA), alpha angles (12, 1, 2, 3), femoral torsion, acetabular coverage, combined version. Furthermore, anatomical measurements such as the vertical distance between the ASIS (Anterior Superior Iliac Spines) landmarks and the femoral head center (FHC), inclination, anteversion tilt and rotation angles between aligned and virtually adjusted orientations were calculated using 3-matic Medical (Materialize). Additionally, similar radiographic measurements of anatomical features as mentioned above were obtained by mapping landmarks of a statistical shape model onto actual hip surfaces (n=641) and simulating pelvic tilt using proprietary software (SOMA (Stryker Orthopedic Modeling and Analytics)). Statistical data analyses assess the impact of pelvic tilt on morphological and anatomical measurements including correlations, distribution, ANOVA, and intra-class correlation (ICC) using Python.

Results:

Pelvic tilt influenced all acetabular and none of the femoral measurements (Figure 1). The ICC was 1.0 for alpha angle and for femoral torsion, indicating no impact of pelvic tilt on the measurements. The ICC were 0.97, 0.96, 0.89 for TA, CEA, and acetabular coverage respectively, indicating a minor impact of pelvic tilt. In contrast, the ICC for AV12 was 0.17, AV2 0.24, AV3 0.24 showing a strong impact of pelvic tilt on acetabular version. There was a positive correlation between the pelvic tilt angle and acetabular version (R²=0,69, p<0.01 for AV12; R²=0,61, p<0.01 for AV3; R²=0,49, p<0.01 for AV12). The impact of pelvic tilt on the acetabular measurements was non-linear. For example, anterior inclination resulted in greater differences in acetabular version (on average 21° for AV12, 19° for AV2, 18° for AV3) than posterior inclination (on average 28° for AV12, 14° for AV2, 12° for AV3), but increased on average 3° per 1° of anterior inclination (Figure 2ad). For CEA the maximum change of measurement between -20° to 20° is 6.2°, with a larger difference between 20° anterior tilt and no tilt (4.9°) than between 20° posterior tilt and no tilt (1.3°) (Figure 2be). The average difference in TA was 0.2° for anterior and 1.3° for posterior tilt. For acetabular coverage the difference between -20° and 20° tilt was –11.32% (Figure 2cf). For every 1° tilt the vertical ASIS-FHC distance reduced 0.54 mm on average (Figure 3ac). The angle between the transverse plane and line between the pubic symphysis and posterior superior iliac spine was directly correlated (R2=1, p<0.01) to the pelvic tilt angle (Figure 3bd).

Conclusions:

Pelvic tilt impacts morphological hip measurements (especially acetabular version) and can lead to measurement differences between aligned CT, unaligned CT and radiographs; those differences need to be understood when planning pelvic procedures. Due to the systematic approach, this study gives insight into how morphological parameters need to be adjusted (degree changes) for patients with physiological pelvic tilt. Pelvic tilt significantly impacts acetabular version, with a maximum difference of 14.75° at 0°-10°, which is the most common range for patients, underscoring the need for patient-specific assessment. In contrast, center-edge angle changes by maximum 3.25° at a 10° tilt, and coverage by 2.4°, often clinically insignificant. For other parameters, no change, or minimal change is observed, variations in these parameters are due to differences between patients rather than pelvic tilt. These findings emphasize the importance of considering measurement sensitivity to pelvic tilt and individual patient context in clinical decision-making.

Figure 1