Two-dimensional/three-dimensional EOS™ imaging is reliable and comparable to traditional X-ray imaging assessment of knee osteoarthritis aiding surgical management

Abstract

BackgroundX-ray imaging is the gold standard for assessing lower limb conditions and preoperative planning. A novel low-radiation-dose EOS™ imaging system enables full-length weight-bearing imaging in one session and three-dimensional (3D) reconstruction. Thus, it can improve assessment of limb deformities, preoperative planning and follow-up with lower radiation exposure. The objective of this study was to measure lower limbs from EOS™ images to determine its accuracy and reproducibility in comparison with long-leg X-ray images. MethodsOver a one-year period, twenty patients (forty lower limbs) with knee osteoarthritis were recruited from clinic. Thirty-five (five excluded due to knee prosthesis) two-dimensional- (2D) EOS™, 3D EOS™ and X-ray images were measured independently by four observers, measuring lower limb angles and lengths. On average, twelve weeks later, observers repeated measurements on 2D EOS™ and X-ray images. ResultsA t-test comparing 2D EOS™ with X-ray images showed no significant difference in all angle and length measurements (P > 0.05). When analysing observers separately, all measurements showed no significant difference, apart from the femoral anatomic-mechanical angle (fAMA) from observer 2 (2D EOS™ fAMA 6.21° vs. X-ray fAMA 7.10°, P = 0.02). Intra-observer intraclass correlation coefficient (ICC) for 2D EOS™ and X-ray was 1.00 and 1.00, respectively, and inter-observer ICC was 1.00 and 0.99, respectively. A t-test comparing 2D- with 3D EOS™ images showed no significant difference in all measurements. A t-test comparing 3D EOS™ with X-ray images showed no significant difference in all measurements. ConclusionThis study showed the EOS™ imaging system to be a valid alternative method of imaging lower limbs for alignment, measurements and preoperative arthroplasty planning.