Tilt-adjusted Cup Anteversion in Patients with Severe Backward Pelvic Tilt is Associated with the Risk of Iliopsoas Impingement: A Three-dimensional Implantation Simulation.

backgroundAnterior overhang of the acetabular component is associated with iliopsoas impingement, which may cause groin pain and functional limitations after THA. However, little is known about the relationship between component overhang and functional alignment of the acetabular component. CT-based image simulation may be illuminating in learning more about this because CT images are more effective than radiographs for evaluating the component's overhang and position. QUESTIONS/PURPOSES: Using CT simulations based on preoperative data of nondysplastic and dysplastic hips, we asked: (1) What are the differences in the amount of component overhang, defined as the mediolateral distance from the component's edge to the native acetabular bony boundary on axial images (axial overhang), and as the AP distance on sagittal images (sagittal overhang) among pelvises with neutral and posterior tilt (in which the cephalad portion of the pelvis is more posterior than the caudad portion in the sagittal plane) in patients with dysplastic hips and those with nondysplastic hips? (2) Are increments in the amount of component overhang associated with a difference in the likelihood that the iliopsoas tendon will impinge against the edge of the acetabular component, after controlling for native acetabular abduction and anteversion and the presence of dysplasia?

methodsA total of 128 hips (dysplastic group: 73 hips; nondysplastic group: 55 hips) were evaluated. We defined a dysplastic hip as one with a lateral center-edge angle of less than 20° on AP radiographs. Pelvic models with neutral (0°) and 10° and 20° of posterior tilt were created from CT data. In simulations, acetabular component models were implanted into the true acetabulum with a tilt-adjusted orientation angle that was defined as the component's angle based on a reference for the functional pelvic plane (coronal plane of the body) in each pelvic model. Axial and sagittal component overhang were measured on CT images. Axial overhang of at least 12 mm and sagittal overhang of at least 4 mm were defined as thresholds increasing the likelihood of iliopsoas impingement according to previous studies. When determining the amount of overhang of the acetabular component, we controlled for abduction and anteversion of the native acetabulum and the presence of dysplasia by performing a multivariable logistic regression analysis.

resultsIn dysplastic hips, axial overhang increased by a mean ± SD of 5 ± 1 mm (Bonferroni adjusted p < 0.001; 95% CI, 4.7-5.1) from 0° to 10° of posterior tilt and by 5 ± 1 mm (p < 0.001; 95% CI, 4.9-5.3) from 10° to 20° of posterior tilt. Sagittal overhang increased by 1 ± 0 mm (p < 0.001; 95% CI, 1.0-1.0) from 0° to 10° of posterior tilt and by 1 ± 0 mm (p < 0.001; 95% CI, 1.0-1.0) from 10° to 20° of posterior tilt. In nondysplastic hips, axial overhang increased by a mean of 5 ± 0 mm (p < 0.001; 95% CI, 4.7-5.0) from 0° to 10° of posterior tilt and by 5 ± 1 mm (p < 0.001; 95% CI, 4.6-5.0) from 10° to 20° of posterior tilt. Sagittal overhang increased by 1 ± 0 mm (p < 0.001; 95% CI, 1.0-1.1) from 0° to 10° of posterior tilt and by 1 ± 0 mm (p < 0.001; 95% CI, 1.0-1.1) from 10° to 20° of posterior tilt. After controlling for the presence of dysplasia, we found that native acetabular abduction and anteversion and posterior pelvic tilt, presence of dysplasia (p = 0.030; adjusted odds ratio [OR], 2.2; 95% CI, 1.1-4.6), native acetabular anteversion (p < 0.001; adjusted OR, 1.4; 95% CI, 1.3-1.5), and 10° and 20° of backward tilt compared with 0° of tilt (10° of posterior tilt: p < 0.001; adjusted OR, 15; 95% CI, 5.5-41; 20° of posterior tilt: p < 0.001; adjusted OR, 333; 95% CI, 96-1157) were independently associated with axial overhang of at least 12 mm; the model showed high goodness of fit (Nagelkerke's r = 0.68). In contrast, native acetabular anteversion (p < 0.001; adjusted OR, 1.2; 95% CI, 1.1-1.2) and 20° of backward tilt compared with 0° of tilt (p = 0.015; adjusted OR, 2.2; 95% CI, 1.2-4.0) were independently associated with sagittal overhang of at least 4 mm; the model had low goodness of fit (Nagelkerke's r = 0.20).

conclusionsAcetabular component overhang is more severe when the pelvis tilts posteriorly. Moreover, posterior pelvic tilt, the presence of dysplasia, and higher native acetabular anteversion were independently associated with an increased risk of component overhang. When 20° of posterior tilt was adjusted, the risk of severe overhang was especially increased. CLINICAL RELEVANCE: Based on these results, surgeons can attempt to prevent severe overhang in patients with posterior pelvic tilt by increasing component anteversion and abduction; when component anteversion is increased by 8° and abduction is increased by 2° from the target angle of 15° of anteversion and 40° of abduction in patients with posterior tilt of 20°, the risk of severe overhang is reduced to by approximately one-twentieth. However, it is still unclear how much the degree of component anteversion should be increased when surgeons attempt to prevent anterior prosthetic dislocation at the same time. Future studies such as prospective clinical trials evaluating both prosthetic dislocation and iliopsoas impingement in patients with posterior tilt might clarify this issue.