Custom implants promise a personalized surgical approach, with the aim of improving patient satisfaction and pain as well as lowering revision rates. However, some published research on custom TKA implants has found no clinically important improvements in postoperative validated outcomes scores, risks of revision, and implant alignment. In the interest of helping to settle this controversy, a systematic review seems warranted.
Question/purpose: Whether the use of custom implants result in clinically important improvements over conventional off-the-shelf implants for anatomically uncomplicated primary TKA in terms of (1) validated outcomes scores, (2) the risk of revision or reoperation, and (3) implant alignment.
PubMed/Medline, Embase, Web of Science, and The Cochrane Library were systematically searched to identify publications from the past 10 years relevant to this review. Publications that compared the clinical outcome measures, number of revisions, and radiological assessment of implant alignment of custom and standard implants with validated endpoints were eligible for inclusion. In the interest of capturing as much potentially relevant information as possible, we applied no requirement for minimum follow-up duration. Clinical outcomes were assessed using patient-reported outcome scores including the Knee Society Score (KSS), Forgotten Joint Score, and Knee Injury and Osteoarthritis Outcome Score. The risk for revision or reoperation were evaluated by the number of early and late manipulations, debridement procedures, and replacement of one or more components. Implant alignment was compared using postoperative deviation from the neutral (0°) mechanical axis of the limb and each component and the posterior tibial slope. All qualified studies were retrospective, and all compared custom implants with standard implants. Data on 1510 patients were reviewed (749 with custom implants and 761 with standard implants). The mean follow-up time ranged from 12 to 33 months.
There was no apparent advantage to custom implants in terms of PROM scores. Of the four studies evaluating clinical outcomes, two reported better KSS-Function scores; one reported no difference, and one found inferior KSS scores. In several studies, custom implants were associated with more frequent reoperations than standard implants. Although in general there were no differences between custom and standard implants in terms of mean coronal plane limb alignment, one of seven studies found that the proportion of patients whose alignment was outside ± 3° from the neutral axis in the coronal plane was lower in the custom group than in the standard group.
With no clear advantage in outcomes scores for pain and function, generally higher risks of reoperation and reintervention, and no overall benefit to alignment, custom implants for primary TKA for the general population currently appear to be inferior to standard implants. Whether the slight reduction in the proportion of patients with alignment outliers observed in a minority of studies will result in a substantial reduction in revision risk over time must be addressed by future studies. However, until or unless such a reduction is proven, we recommend against the routine use of custom implants in practice because of increased costs and the risks associated with their novelty.