Comparison of custom to standard TKA instrumentation with computed tomography

• Published in: Knee surgery, sports traumatology, arthroscopy : official journal of the ESSKA sports traumatology, arthroscopy : official journal of the ESSKA, 2014-08, Vol.22 (8), p.1833-1842
• Main Authors: Ng, Vincent Y., Arnott, Lindsay, Li, Jia, Hopkins, Ronald, Lewis, Jamie, Sutphen, Sean, Nicholson, Lisa, Reader, Douglas, McShane, Michael A.
• Format: Article
• Language: English
• Subjects: Aged; Aged, 80 and over; Arthritis; Arthroplasty, Replacement, Knee - instrumentation; Female; Femur - diagnostic imaging; Femur - surgery; Humans; Image Processing, Computer-Assisted; Imaging, Three-Dimensional; Joint replacement surgery; Knee; Knee Joint - diagnostic imaging; Knee Joint - surgery; Male; Medical imaging; Medicine; Medicine & Public Health; Middle Aged; Orthopedics; Osteoarthritis, Knee - diagnostic imaging; Osteoarthritis, Knee - surgery; Patients; Surgeons; Surgery, Computer-Assisted; Tibia - diagnostic imaging; Tibia - surgery; Tomography; Tomography, X-Ray Computed
• ISSN: 0942-2056; 1433-7347
• DOI: 10.1007/s00167-013-2632-7

Purpose There is conflicting evidence whether custom instrumentation for total knee arthroplasty (TKA) improves component position compared to standard instrumentation. Studies have relied on long-limb radiographs limited to two-dimensional (2D) analysis and subjected to rotational inaccuracy. We used postoperative computed tomography (CT) to evaluate preoperative three-dimensional templating and CI to facilitate accurate and efficient implantation of TKA femoral and tibial components. Methods We prospectively evaluated a single-surgeon cohort of 78 TKA patients (51 custom, 27 standard) with postoperative CT scans using 3D reconstruction and contour-matching technology to preoperative imaging. Component alignment was measured in coronal, sagittal and axial planes. Results Preoperative templating for custom instrumentation was 87 and 79 % accurate for femoral and tibial component size. All custom components were within 1 size except for the tibial component in one patient (2 sizes). Tourniquet time was 5 min longer for custom (30 min) than standard (25 min). In no case was custom instrumentation aborted in favour of standard instrumentation nor was original alignment of custom instrumentation required to be adjusted intraoperatively. There were more outliers greater than 2° from intended alignment with standard instrumentation than custom for both components in all three planes. Custom instrumentation was more accurate in component position for tibial coronal alignment (custom: 1.5° ± 1.2°; standard: 3° ± 1.9°; p  = 0.0001) and both tibial (custom: 1.4° ± 1.1°; standard: 16.9° ± 6.8°; p