3-D LANDMARKING REPEATABILITY EMPHASIZES CHALLENGES IN SCAN POSITIONING DURING WEIGHT BEARING CT OF THE KNEE

INTRODUCTION

Weight bearing CT (WBCT) has shown promise in the evaluation of the knee joint instead of radiography. However, bringing weight bearing to 3-D imaging poses technical challenges that have to be overcome if repeatability is to be optimised. From prior study and experience, maintaining knee flexion angle (KFA) and centering in the vertical scan range with consistency can be difficult. One means to evaluate these distance and angle measurements from WBCT is to use a bone surface landmarking system.

OBJECTIVE

(1) To evaluate the repeatability of a manual WBCT landmarking system of the femur and tibia at the knee; and (2) from this develop a technique for evaluating repeatability of KFA and vertical scan range centering.

METHODS

14 individuals recruited and consented at the University of Kansas Medical Center had baseline and follow-up WBCT imaging suitable for analysis. Participant demographics were: mean ± SD age 61.3 ± 8.4 years, BMI 30.7 ± 4.3 kg/m² and male:female ratio 8:6. All scanning was performed on the same XFI WBCT scanner (Planmed Oy, Helsinki, Finland) with the mean ± SD interval between baseline and follow-up attendances 14.9 ± 8.1 days. A Synaflexer^TM device was used to standardise knee positioning during scanning. Imaging acquisition parameters were 96 kV tube voltage, 51.4 mA tube current, 3.5 s exposure time. A standard bone algorithm was applied for reconstruction with 0.3 mm isotropic voxels and a 21 cm vertical scan range. Both knees were included in all analyses with SD adjustments made for multiple observations from the same individual. Participant identification and scan sequence were anonymised prior to analyses. A first observer (A.B.) placed 10 femoral and 12 tibial landmarks using Stradview. These landmarks were reviewed by a second observer (T.D.T.), who placed additional landmarks at the extremes of the vertical scan within the centre of the bone medullary cavities. Bone segmentations from ScanXM were used to register landmarks from follow-up to baseline in wxRegSurf; the follow-up-to-baseline femur registration was applied to the follow-up tibial co-ordinates to assess joint positioning. Landmark repeatability was taken as the mean ± SD distance (mm) between baseline and follow-up for each landmark. A method to extract KFA and valgus alignment was developed as the angle between the lines of the extreme scan range landmarks (F00 and T00) and the centre of gravity (CoG) of the rest of the landmarks in the same bone. Valgus alignment was taken from the anterior view (<180° laterally = valgus) and KFA from lateral.

RESULTS

Landmark placement with their codes is shown in Figure 1a (in a left knee), with code definitions given in Table 1. An example of baseline and follow-up landmarking is shown on the same knee in Figure 1b with error results from all landmarks given in Table 1. Outside of the F00/T00 markers, the most notable mean error was seen at the central anterior and central posterior tibial plateau margins (T01 and T09) with mean ± SD values of 6.0 ± 4.7 mm and 5.2 ± 3.8 mm respectively, and at the medial femoral trochlear superior articular margin (F03) at 5.4 ± 3.6 mm. Error metrics were expectedly similar for F00 and T00 with mean (max.) values of 23.2 (47.0) mm and 22.8 (45.7) mm respectively, serving as a surrogate marker for variation in central placement of the knee within the vertical (z-axis) scan range. Valgus angle was consistent, showing a mean ± SD (range) difference of 0.2 ± 1.1° (-2.0 to 1.8°), whereas KFA values were less consistent at -2.5 ± 5.9° (-15.5 to 9.8°) (Table 1).

CONCLUSION

Less well anatomically defined landmarks such as the tibial plateau margins are less repeatable, with worst mean repeatability error around 5 mm. Vertical knee centring varied substantially up to a max. value of 47 mm, while KFA varied widely from -15 to 10°. Positioning of the knee consistently during WBCT remains challenging. This post hoc evaluation derived from landmarking is valuable for verification, but work needs to be done on optimising positioning protocols that can be applied prospectively to ensure positioning remains consistent for evaluation of other parameters such as meniscal extrusion and 3-D JSW.