Intra-operative fluoroscopic measurement accuracy for femoral tunnel aperture positions in anterior cruciate ligament reconstruction

초록

Purpose This study aimed to test an automated analysis software for the three-dimensional (3D) quadrant method and assess the accuracy of femoral tunnel positioning during double-bundle anterior cruciate ligament reconstruction (DB-ACLR) with fluoroscopic assistance versus conventional 3D computed tomography (3D-CT). Methods We retrospectively evaluated 32 patients who underwent DB-ACLR between 2010 and 2014. The femoral tunnel location was calculated using three methods: the quadrant fluoroscopy method, the (modified) quadrant method using conventional 3D-CT, and the automated 3D reconstruction method.Two independent observers assessed each knee position.The accuracy of tunnel location on the fluoroscopic radiograph and conventional 3D-CT was compared to the automated 3D reconstruction method. Results Femoral tunnel locations measured by the quadrant method using an automated 3D reconstruction method were 19.1 +/- 4.8 % of depth and 22.3 +/- 7.7 % of height in anteromedial (AM) femoral tunnels, 32.2 +/- 6.0% of depth and 48.4 +/- 7.7% of height in posterolateral (PL) tunnels. Using fluoroscopy, 20.3 +/- 5.0 % of depth and 26.8 +/- 8.9 % of height in AM tunnels, and 32.9 +/- 5.4% of depth and 50.6 +/- 7.9% of height in PL tunnels was achieved. Using conventional 3D CT, 21.3 +/- 4.6 % of depth and 26.0 +/- 9.1 % of height in AM tunnels, and 34.9 +/- 7.1% of depth and 50.1 +/- 8.0% of height in PL tunnels was achieved. When analyzed using the Bland-Altman plot, the Limit of Agreement (LOA) in the fluoroscopy group was closer to the equator than conventional 3D CT group in all cases, except for the height of the AM tunnel, which had a slightly lower LOA. Conclusion The fluoroscopically identified tunnel aperture location was more accurate and reproducible than the one obtained via the quadrant method using conventional 3D-CT imaging.

키워드