Summary

A novel technique enables the analysis of bearing movement of mobile-bearing unicompartmental knee arthroplasty, which was not possible before.

Abstract

Introduction

Mobile-bearing (MB) unicompartmental knee arthroplasty (UKA) is an artificial knee joint based on the concept of a bearing that is highly conformable to the femoral component and moves freely on a flat tibial baseplate. MB-UKA knees are known to reproduce kinematics similar to those of normal knees, but not much is known about how the bearings move in vivo. The two-dimensional (2D)/three-dimensional (3D) registration technique is a fluoroscopic kinematic analysis methodology that uses a 3D computer model of an object to calculate its 3D coordinates. However, the MB-UKA implant could not be analyzed previously due to its size and shape. In this study, we determined the kinematics of MB-UKA implants using a novel fluoroscopic kinematic analysis technique and estimated the anteroposterior (AP) translation of the MB during non-weight-bearing knee flexion and squatting.

Method

A total of 11 unilateral MB-UKA patients were investigated. Each patient performed non-weight-bearing knee flexion in a sitting position and squatting under fluoroscopic imaging. The relative position of the bone and MB-UKA implant was calculated from postoperative CT data. We used this relative coordinate information to compute the implant position from the bone registration information and compensated for the implant registration information (double registration) to derive accurate implant kinematics. The lowest point of the femoral component relative to the tibial baseplate was assumed to coincide with the deepest point of the MB, and its AP translation was analyzed. The AP midpoint of the tibial baseplate was defined as the zero position; positive and negative values were described as anterior and posterior, respectively.

Results

On average, the MB was located at -2.8 ± 2.6 mm in non-weight-bearing knee flexion and -4.7 ± 2.6 mm in squatting at 0° of flexion. During non-weight-bearing knee flexion, it translated to -0.7 ± 3.1 mm until 40° of flexion, then moved to -6.6 ± 5.4 mm until 130° of flexion. In contrast, it translated to -5.8 ± 2.7 mm until 30° of flexion during squatting. There was no significant AP translation thereafter, with significant posterior translation only at 130° of flexion (-7.9 ± 3.4 mm). From 0° to 70° of knee flexion, the MB was significantly more posterior during squatting than during non-weight-bearing knee flexion. In squatting, there were five cases in which the MB moved anteriorly after 30° of knee flexion. In these five cases, no MB moved posteriorly more than 4 mm from 40° to 110° of knee flexion during non-weight-bearing knee flexion. On the other hand, in the remaining six cases, the MB moved posteriorly more than 4 mm in all cases from 40° to 110° of knee flexion.

Conclusion

A novel technique enables the analysis of MB-UKA implant kinematics in vivo, which was not possible before. On average, the MB moved anteriorly in the mid-flexion range during non-weight-bearing knee flexion and posteriorly during squatting. The possibility of predicting MB movement during squatting from that during non-weight-bearing knee flexion was presented.