Abstract
Introduction & aims
Resurfacing of the patella is an important part of most TKA operations, usually using an onlay technique. One common practice is to medialise the patellar button and aim to recreate the patellar offset, but most systems do not well control alignment of the patella button. This study aimed to investigate for relationships between placement and outcomes and report on the accuracy of patella placement achieved with the aid of a patella Patient Specific Guide (PSG).
Method
A databse of TKR patients operated on by five surgeons from 1-Jan-2014 who had a pre-operative and post-operative CT scan and 6-month postoperative Knee Osteoarthritis and Outcome (KOOS) scores were assessed. Knees were excluded if the patella was unresurfaced or an inlay technique was used. All knee operations were performed with the Omni Apex implant range and used dome patella buttons. A sample of 40 TKRs had a patella PSG produced consisting of a replication of an inlay barrel shaped to fit flush to the patient's patella bone.
The centre of the quadriceps tendon on the superior pole of the patella bone and the patella tendon on the inferior were landmarked. 3D implant and bone models from the preoperative CT scans were registered to the post-operative CT scan. The flat plane of the implanted patella button was determined and the position of the button relative to the tendon attachments calculated. Coverage of the bone by the button and patellar offset reconstruction were also calculated. The sample of 40 TKRs for whom a patella PSG was produced had their variation in placement assessed relative to the wider population sample. All surgeries were conducted with Omni Apex implants using a domed patella.
Results
A total of 322 patients were identified in the database, and 82 were subsequently excluded as inlay rather than onlay patella. 59% (142) were female and the average age was 68.9 years (+/− 7.2).
Coverage percentage of the cut patella surface by the button was 67% (± 7%), with 83% (200) knees having greater than 60%, and 40% (96) greater than 70%. Component position was on average centralised in terms of mediolateral position (0.09mm ± 1.93 lateral). When comparing the alignment of the patients whose knees used PSG guides with those who did not, it was found there was a statistically significant reduction in the variation that both external rotation error and flexional error had (p-values 0.048 and 0.022 respectively.)
Excess medialisation of the patella button was found to weakly correlate with reduced postoperative KOOS symptoms scores (coefficient=0.14, p-value = 0.035). When subdivided into patients who reported knee clicking sometimes or more often and those who did not, patients with highly medialised buttons had a 1.5× likelihood of reporting clicking of their knee joint (p-value = 0.036).
Conclusions
The patella-femoral joint remains a crucial component in the TKA knee, but the process of resurfacing the bone is not well controlled and can negatively influence patient outcomes. PSG's are one potential mechanism of controlling patella component alignment.
