Abstract
INTRODUCTION
During total knee arthroplasty (TKA), the pursuit of accurate alignment, proper bone cuts, and good soft tissue balancing sometimes can result in the overhang of the femoral component, especially in smaller-sized Asian knees. As size and shape of the distal femur are highly variable, component designs that offer increased shape and size offerings may be desirable to fit the distal femur. This study tested the hypothesis that increased shape and size offerings in TKA femoral designs may improve their fit to the Japanese femur compared to designs that offer only one shape and limited sizes.
METHODS
Five contemporary femoral component designs were evaluated (Designs A-E). Design A has multiple mediolateral (ML) size offerings for a specific component anteroposterior (AP) size, and the finest increment (2mm) in AP sizes among all the designs. Designs B-E have single ML offerings across component AP sizes. For each design, virtual TKA resections were performed on the digital surfaces of 82 Japanese distal femora, each sized by selecting the component AP size that most closely matched but did not exceed the femoral AP dimension (Fig 1A,B). The aspect ratio (ML/AP) of the resected femora was regressed against the aspect ratio of their properly sized components per design. The closeness of each design to the perfect shape match was evaluated by the root-mean-square deviation (RMSD) of the deviations between the femoral bone and components. Differences in ML dimensions (overhang/underhang) between component and resected femora were calculated (Fig1C,D). The incidence of clinically significant femoral overhang (>3mm), in which component downsizing is required, were analyzed.
RESULTS
Design A captured the shape variability in the resected femur, with component aspect ratios being the closest to the anatomy of the resected femur among the five designs (RMSD=0.04) (Fig 2). In contrast, Designs B-E had greater deviation from the shape of the resected femur (RMSD=0.08–0.12), indicating higher incidence of shape mismatch that may lead to surgical compromise. Designs C and E had the highest incidence and severity for clinically significant overhang, followed by Designs B and D (Fig 3A). Design A exhibited the lowest incidence and severity of clinically significant overhang and had the least variability in ML width differences (standard deviation=2.4mm) compared to the other designs (Designs B-E, standard deviation=4.0–4.9mm). In all the designs investigated, the percentage of bones that required downsizing was the highest in Designs C (48%) and E (39%), followed by Designs B and D (17% and 22%). In contrast, minimal downsizing was required in Design A (4%). The highest incidences of downsizing were generally observed in mid-sized components (Fig 3B).
DISCUSSION
The design family with multiple ML offerings per AP size (Design A) provides the closest match to the shape of the distal femur compared to those with single ML offerings (Designs B-E). Additionally, increased AP size offerings in Design A (12 sizes) further improve component fit compared to Designs B-E (7–9 sizes). Among all five design families investigated, Design A exhibited minimal incidence of downsizing due to clinically significant overhang in the Japanese patients.
