Clinical evaluation of model-based radiostereometric analysis to measure femoral head penetration and cup migration in four different cup designs.

In conventional Radiostereometric analysis (RSA) implants with attached tantalum markers are frequently used, which may be difficult to visualize. This problem can be avoided with model-based RSA (MBRSA), but it is uncertain if this method has the same precision as marker-based RSA. We evaluated the influence of cup design for the precision of MBRSA in four uncemented cups to study if the design had any influence on the precision. Stereo radiographs were analyzed postoperatively (double-examinations) and after 2 years (single examinations). The difference between the double-examinations was used to compute the precision for the methods and for each type of implant. Femoral head penetration and cup translation up to 2 years were compared using marker-based RSA as reference. The precision of proximal penetration and migration measurements did not differ between the methods for Trilogy, TMT and ABG. For Ringloc design a poorer precision was observed using MBRSA. Comparison between the methods regarding proximal penetration and cup migration at 2 years did not differ for three of the designs (p = 0.12-0.91). However, for the group with porous plasma sprayed surface (Ringloc) a significant difference between the methods was observed (ppenetration  <0.01 and pmigration  <0.01). Poorer precision, different penetration and migration values at 2 years for one of the designs indicate that the resolution of MBRSA might vary depending on surface coating and implant geometry. Therefore, we conclude that the resolution of MBRSA has to be studied for each type of basic cup design.