Background Total ankle replacement (TAR) that can relieve pain while retaining ankle movement is a useful surgical option for patients with rheumatoid arthritis (RA). However, high complication rates and low survivorship are still problematic in TAR, as compared to total knee and hip replacements. A better understanding of ankle kinematics including three-dimentional (3D) dynamic motion after TAR should be important to explain the failures in TAR, especially those attributed to loosening and subsidence. We have developed a 3D kinematic analysis system using X-ray fluoroscopic imaging with quantitative assessment of 3D dynamic motion of artificial implants under clinical conditions.
Objectives The objective of this study was to determine in vivo kinematics of TAR in rheumatoid ankle and, in particular, to elucidate the dynamic motion of ankle after TAR by 3D evaluation of fluoroscopic imaging.
Methods We examined the evaluation of the accuracy of 2D/3D registration technique for TAR systems by computer simulation. We investigated 12ankles in 9 patients with RA implanted with a three-component mobile-bearing TAR (FINE Total Ankle System, Nakashima Medical, Okayama, Japan), which allows not only internal/external rotation but also anteroposterior (AP) translation. Fluoroscopic images were obtained while each patient was asked to perform non-weightbearing maximal plantarflexion and dorsiflexion on the implanted ankle. Thereafter tibio-talar motion was analyzed by 2D/3D registration technique; a reproduction method of the spatial position of each component in TAR, from single-view fluoroscopic images by use of computer-assisted design models. We evaluated the angles of plantar-/dorsiflexion, internal/external rotation, and inversion/eversion as well as AP translation, between the components.
Results The accuracy of 2D/3D registration technique was within 0.6° for all rotations, and 0.2 mm for in-plane translations. The average range of tibio-talar motion during the non-weightbearing activity was 19.1±4.7° (mean ±standard deviation) of plantar-/dorsiflexion. Also, as the ankles moved from maximal plantarflexion to maximal dorsiflexion, they demonstrated external rotation, eversion and anterior translation with the average range of 2.9±1.5°, 1.9±1.1°, and 2.1±0.9 mm, respectively.
Conclusions In this study, accurate evaluation of 3D dynamic motion of ankle joint after TAR was feasible. Although mobility of plantar-/dorsiflexion was unexpectedly small, and rotation and inversion/eversion was much more insignificant, kinematic patterns were similar to those of normal ankle. From these results, we think that the analysis of in vivo kinematics of TAR using 2D/3D registration technique will provide useful data for comparison not only between the implants but also between the patients.
Nishikawa M. Int Orthop 2004;28:123-126.
Shi K. J Foot Ankle Surg 2006;45:316-321.
Zuffi S. IEEE Trans Med Imaging 1999;18:981-91.
Yamazaki T. IEEE Trans Med Imaging 2004;23:602-612.
Komistek RD. Foot Ankle Int 2000;21:343-350.
Satoshi Y. J Biomech 2011;44:995-1000.
Disclosure of Interest None Declared