Abstract
INTRODUCTION
An accelerometer-based portable navigation system (KneeAlign2, OrthAlign Inc., Aliso Viejo, CA) is expected to improve mechanical axis and component alignment compared to conventional instrumentation in total knee arthroplasty (TKA). However, past reports have evaluated its accuracy using only radiographic measurements. The purpose of this study was to analyze the accuracy of the KneeAlign2 system with radiography and more detailed three-dimensional (3D) CT.
METHODS
We targeted 22 patients (24 knees) with severe osteoarthritis who underwent primary TKA using the KneeAlign2 system. Cemented, fixed-bearing, cruciate-retaining prostheses were implanted in all patients. We used postoperative standing-position full-length radiographic evaluation of the lower limbs to measure the hip-knee-ankle angle (HKA), frontal femoral component angle (FFC), and frontal tibial component angle (FTC). However, lower limb rotation and knee flexion could affect radiographic measurement of HKA and the component positioning angle. We used 3D bone models reconstructed from pre- and postoperative CT images to precisely analyze the 3D component positioning. For a 3D matching bone model made from these models, a 2D projection of the pre- and postoperative component positioning planes was made, and the projection angle was measured as angle error compared to the preoperative planned position (Figure 1). Average surgery time and total blood loss on postoperative day 7 were also recorded.
RESULTS
There were 24 knees available for analysis. Mean HKA was 0.1° ± 2.2 varus; 16.7% of knees had coronal outliers exceeding 3°. Mean FFC was 0.9° ± 1.9 varus; 29.2% of femoral components were placed with coronal outliers exceeding 2°. Mean FTC was 1.2° ± 1.6 valgus; 20.8% of tibial components were placed with coronal outliers exceeding 2°. In 3D-CT evaluation, mean femoral coronal and sagittal alignment were 1.2° ± 1.7 varus (outliers exceeding ±2°: 37.5%) and 0.8° ± 2.4 flexion (outliers exceeding ±2°: 20.8%), respectively. Mean tibial coronal and sagittal alignment were 1.1° ± 1.4 valgus (outliers exceeding ±2°: 33.3%) and 0.1° ± 1.6 flexion (outliers exceeding ±2°: 20.8%), respectively. Average surgical time was 96 ± 7.7 minutes, and blood loss was 400 g ± 113 on postoperative day 7.
CONCLUSIONS
With radiographic and 3D-CT evaluation, the mean angle error values for the femoral and tibial components were less than 2° in the coronal plane, and less than 1° in the sagittal plane.
KneeAlign2 is highly accurate in positioning the femoral and tibial components in TKA.
For figures/tables, please contact authors directly.
