Final intraoperative gaps commonly exceed resection-based reference gaps during sequential balancing in imageless robot-assisted total knee arthroplasty.

Purpose: The purposes of this study were to quantify final-minus-reference gap difference in the medial and lateral compartments in extension and flexion during passive, imageless robot-assisted total knee arthroplasty (RA-TKA) and to identify preoperative radiographic and initial intraoperative factors associated with greater final-minus-reference gap difference. Methods: A retrospective single-surgeon series of 379 primary total knee arthroplasties performed using a passive, imageless robot-assisted system for osteoarthritis was analysed. Medial and lateral compartment spaces were recorded in extension and at 90° flexion before bone resection and after trial implantation. For each compartment and knee position, the reference gap was defined from the initial gap, validated resection depth, femoral component thickness and tibial-side construct thickness. Multivariable linear regression models were adjusted for age, body mass index, sex and implant design, with cluster-robust standard errors to account for bilateral knees. Results: Mean final-minus-reference gap difference was positive in all four conditions, indicating that final gaps exceeded the calculated reference in both extension and flexion: 4.7 ± 2.1 mm medially and 3.4 ± 2.8 mm laterally in extension, and 4.2 ± 3.0 mm medially and 3.2 ± 3.7 mm laterally in flexion. Smaller initial compartment gaps and greater varus-valgus laxity range were consistently associated with greater gap difference. Posterior-stabilized design was associated with greater medial gap difference in extension and with greater medial and lateral gap difference in flexion. More valgus alignment and greater lateral osteophyte burden were associated with greater lateral gap difference. Conclusions: In passive, imageless RA-TKA, final measured gaps commonly exceeded a resection-based reference during sequential balancing. Smaller initial compartment gaps and greater varus-valgus laxity range were consistently associated with greater gap difference. These findings suggest that resection-based preresection gap estimates should be interpreted with awareness of compartment-specific tightness and coronal laxity. Level of Evidence: Level III, retrospective cohort study.