Measurements of tunnel placements after anterior cruciate ligament reconstruction — A comparison between CT, radiographs and MRI

Abstract

BackgroundA non-anatomic placement of the femoral and tibial tunnels may affect outcome in anterior cruciate ligament (ACL) reconstructions. Tunnel placements are validated with varying imaging modalities. We compared measurements of tunnel placements between radiographs, computed tomography (CT) and magnetic resonance imaging (MRI) in a clinical setting, assessed the reliability and aimed to decide on a possible “gold standard”. MethodsAll patients who had undergone at least two of three modalities, radiographs, MRI and CT, after ACL reconstruction between January 2011 and June 2013 were included. Two radiologists measured tunnel placements according to a standardized protocol. Interobserver agreement was assessed with intraclass correlation coefficients (ICC), the intermodality differences with Bland–Atman plots. Radiation data for CT studies were collected. ResultsForty-six CTs, 45 radiographs and 30 MRIs were reviewed. Femoral inter-observer agreement for radiographs was ICC=0.64, for CT ICC=0.86 and for MRI ICC=0.75. Tibial inter-observer agreement for radiographs was ICC=0.92, for CT-mip ICC=0.91, for CT and MRI ICC=0.87. No intermodality differences between the femoral measurements were observed. In the tibia, there were differences between radiographs and CT (−3.9%), radiographs-MRI (−3.6%), CT–CTmip (3.2%) and CTmip–MRI (−3.1%). The effective radiation doses varied between 0.025 and 0.045mSv, mean and median was 0.033mSv. ConclusionThere were differences in the tibial measurements between summation and single slice images. Only 3D–CT depicted the femoral tunnel in both directions. CT was consistently reliable in both femoral and tibial measurements. Effective radiation dose from CT was lower than previously reported. CT can safely be used in routine clinical practice to evaluate tunnel placements after ACL reconstruction. Level of evidenceLevel III — study of diagnostic test without a universally applied “gold standard”.