Abstract
Purpose
Proper positioning of the components of a knee prosthesis for obtaining post-operative knee joint alignment is vital to obtain good and long term performance of a knee replacement. Although the reasons for failure of knee arthroplasty have not been studied in depth, the few studies that have been published claim that as much as 25% of knee replacement failures are related to malpositioning or malalignment [x].
The use of patient-matched cutting blocks is a recent development in orthopaedics. In contrast to the standard cutting blocks, they are designed to fit the individual anatomy based on 3D medical images. Thus, landmarks and reference axes can be identified with higher accuracy and precision. Moreover, stable positioning of the blocks with respect to the defined axes is easier to achieve. Both may contribute to better alignment of the components.
The objective of this study was to check the accuracy of femoral component orientation in a cadaver study using specimen-matched cutting blocks in six specimens; first for a bi-compartmental replacement, and then for a tri-compartmental replacement in the same specimen.
Materials and Methods
Frames with infrared reflective spherical markers were fixed to six cadaveric femurs and helical CT scans were made. A bone surface reconstruction was created and the relevant landmarks for describing alignment were marked using 3D visualisation software (Mimics). The centres of the spherical markers were also determined. Based on the geometry of the articular surface and the position of the landmarks, custom-made cutting blocks were designed. One cutting block was prepared to guide implantation of a bi-compartmental device and another one to guide implantation of the femoral component of a total knee replacement.
The knee was opened and the custom-made cutting block for the bi-compartmental implant was seated onto the surface. The block was used to make the anterior cut, after which it was removed and replaced with the conventional cutting block using the same pinning holes to ensure the same axial rotational alignment. The other cuts were made using the conventional cutting block and the bi-compartmental femoral component was implanted. Afterwards, a similar procedure was used to make the extra cuts for the total knee component.
The position of the components with respect to the reflective markers was measured by locating three reference points and “painting” the articular surface with a wand with reflective markers. The position of all marker spheres was continuously recorded with four infrared cameras and Nexus software.
Results
Average alignment for the bi-compartmental component in transverse and frontal planes were 0.2° (standard deviation: 2.4°) and 0.4° (standard deviation: 2.8°), respectively. Average alignment for the tri-compartmental component in transverse and frontal planes were 0.6° (standard deviation: 3.2°) and 0.9° (standard deviation: 5.5°), respectively.
Conclusions
The specimen matched cutting blocks, designed based on CT scan data, achieved a similar level of alignment accuracy as reported for navigation systems.
