Abstract
Summary Statement
The current study introduced the effects of projection errors on ankle morphological measurements using CT-based simulated radiographs by correlation analysis between 2D/3D dimensions and reliability analysis with randomised perturbations while measuring planar parameters on radiographs.
Introduction
Clinical success of total ankle arthroplasty (TAA) depends heavily on the available anatomy-based information of the morphology for using implants of precisely matched sizes. Among the clinically available medical imaging modalities, bi-planar projective radiographs are commonly used for this purpose owing to their convenience, low cost, and low radiation dose compared with other modalities such as MRI or CT. However, the intrinsic articular surface of the ankle joint is not symmetrical and oblique which implies that it is difficult to describe all the anatomical dimensions in detail with only one radiograph, thereby hindering the determination of accurate ankle morphometric parameters. The purposes of this study were to compare the measurements of ankle morphology using 3D CT images with those on planar 2D images; and to quantify the repeatability of the 2D measurements under simulated random perturbations.
Patients & Methods
Fifty-eight fresh frozen cadaveric ankle specimens were used in the current study. Each specimen was fixed in the neutral position with a plastic frame. After fixation, the specimen-fixation construct was scanned using a 16-slice spiral CT scanner (GE BrightSpeed 16, C&G Technologies, USA) with a slice thickness of 0.625 mm. A global coordinate system was embedded in the ankle specimen with the origin at the geometric center of the talus, the anteroposterior (A/P) axis in parallel to the base-plate, the superoinferior (S/I) axis perpendicular to the base-plate, and the mediolateral (M/L) axis as the line perpendicular to both the A/P and S/I axes. Fourteen 3D morphological parameters were automatically determined using a house-developed program in MATLAB R2010a (The MathWorks, Inc., USA). A simulated standard digital radiography system, in which the X-ray focus was 1 meter away from the image plane, was also introduced to determine the planar 2D morphological parameters for comparing with those determined in 3D. Reliability with randomised perturbations during measurements was also assessed in terms of the intra-class correlation coefficients using a 2-way mixed-effects average model (ICC3, k) for intra-examiner assessments. All statistical analysis was performed using SPSS 13.0 (SPSS Inc., USA).
Results
Most of the morphological parameters had high correlation and reliability, except for the maximal tibial thickness (MTiTh), distance between most vertex of tibial mortise to the level of MTiTh (MDV) and radius of trochlea tali (TaR) had moderate to low correlation which were 0.54, 0.37 and 0.09 respectively. The ICC coefficients indicated that the MDV, talus width (TaW) and inclination angle between two most vertex points of trochlea tali (MLATa) had moderate and poor reliability which were 0.59, 0.49 and 0.07 respectively.
Discussion/Conclusion
The current study introduced the effects of projection errors on ankle morphological measurements using CT-based simulated radiographs by correlation analysis between 2D/3D dimensions and reliability analysis with randomised perturbations while measuring planar parameters on radiographs. MTiTh and MDV are the important parameters to help surgeon pre-surgical decision-making. TaW is one of the critical parameters for choosing accurate sise of TAA implant. It implies that the respectively accurate pose of ankle is critical during bi-planar radiography.
