Abstract
Patient-specific instruments for total knee arthroplasty shift the bone landmark registration and implant positioning of computer navigation from intraoperative to the pre-operative setting. A preoperative MRI or CT scan is mandatory, with the specifications determined by the instrument manufacturer. Default implant sizing and alignment targets must be templated by the surgeon and mapped onto the virtual knee. The surgeon must also review and modify the preoperative computer plan to incorporate any clinical findings, such as flexion contracture or fixed deformity. The finalised preoperative plan is sent back to the implant vendor for fabrication of patient-specific cutting blocks in 4–6 weeks. The supposed advantages of these instruments include more accurate coronal alignment, fewer outliers, no instrumentation of intramedullary canal, decreased operative time, and decreased hospital costs to clean-sterilise instruments. There are many disadvantages of patient-specific instruments, including: cost, preoperative scheduling of imaging, the learning curve for the surgeon, and the uncalculated preoperative planning time. A set of conventional instruments should be available if the custom instruments do not fit properly. One study of 66 knees using PSI reported that frequent surgeon-directed changes were required, 2.4 per knee, implant sizes were changed in 77% of femurs and 53% of tibias, and tourniquet time was not improved. A Markov model study reported an increased cost of $4600 for 4.6 QALYs for patient-specific instruments and that the rate of revision must be reduced by 50% or more for these instruments to be cost-effective. There is little evidence to support the claims made by the manufacturers of these instruments. We advise against the widespread use of these instruments for total knee arthroplasty.
