Biomechanical Models of the Hip – a Validation Study Based on 10 CT-Datasets

Consideration of the pre- and post-operative magnitude of the hip joint force R and its orientation Ɵ is of major importance for satisfactory long-term results in total hip arthroplasty. R and Ɵ can be computed by using biomechanical models with adapted geometrical/ anthropometrical parameters taken from clinical X-ray images. The objective of this study was to evaluate the models of Pauwels and Debrunner based on digital reconstructed-radiographs (central projection) from 10 CT-datasets of patients treated with telemetric hip-implants by a comparison to corresponding in-vivo measurements. R and Ɵ were computed for 10 patients with patient-specific geometric/anthropometric parameters. The model adaption was based on 28 anatomical landmarks. The root-mean-square-error of R is smaller for Debrunner (0.59/vs./0.66), and for Ɵ it is smaller for Pauwels’ (4.47/vs./7.78). Mathematical models provide potentially valuable information regarding hip joint mechanics. Regarding R, in all of the 10 patients the predictions of Pauwels’ model are consistently higher than the in-vivo measurements. Debrunner computed R in 8 cases higher and in 2 cases lower than the corresponding in-vivo forces. Pauwels’ and Debrunner showed similar tendencies: in 8 cases an overestimation of R and in 2 cases contrary results. Regarding Ɵ we found that in 5 cases the predictions of Pauwels’ are consistently higher than the in-vivo measurements and also contrary to Debrunner. As previous studies showed, an unambiguous identification of most landmarks in a 2D X-ray image is difficult. The impact of the pelvic tilt on the computational result was not considered in our study. Further investigation of this aspect is part of our ongoing work.