Do Not Chase Symmetric Gaps: Native Asymmetry Linked to Better Recovery in Fixed-Deformity Total Knee Arthroplasty.

BACKGROUND: Non-correctable (fixed) knee deformities in total knee arthroplasty (TKA) are technically demanding, and the intraoperative factors driving recovery in this subgroup remain unclear. We hypothesized that, in non-correctable knees, a strategy that minimizes bone resection and preserves the patient's native medio-lateral laxity asymmetry, rather than chasing symmetric gaps through additional bone removal, would be associated with superior patient-reported functional outcomes at one year, whereas no such association would be observed in correctable knees.

METHODS: We reviewed 861 consecutive primary robotic-arm-assisted TKAs (634 correctable, 227 non-correctable) at a single academic institution. Correctability was assessed intraoperatively by varus-valgus stress in extension: correctable if alignment came within 3° of neutral, non-correctable if it remained greater than 3° from neutral1-3. The robotic system quantified native and reconstructed compartment laxity in near extension (5 to 20°) and flexion (90 ± 10°) and recorded planned resection depths. Postoperative medial-lateral (M-L) laxity differential (reconstructed medial - lateral) was categorized as symmetric (less than 0.5 mm), partially corrected (0.5 to two mm), or preserved native asymmetry (two to five mm). The Knee injury and Osteoarthritis Outcome Score for Joint Replacement (KOOS JR) and PROMIS (Patient-Reported Outcomes Measurement Information System) Physical Function were collected preoperatively and at six weeks, three months, and one year, stratified by correctability.

RESULTS: The non-correctable knees showed greater PROMIS Physical Function improvement at three months (P = 0.003) and one year (7.0 ± 7.8 versus 4.9 ± 6.8; P = 0.036) despite similar preoperative scores. Among non-correctable knees, lower total bone resection volume was associated with greater KOOS JR improvement at one year (P = 0.016), strongest in the lateral flexion compartment (P = 0.039), indicating over-resection beyond implant thickness to achieve symmetric gaps is linked to poorer recovery. Greater postoperative flexion M-L laxity differential was also associated with greater KOOS JR (P = 0.009) and PROMIS (P = 0.044) improvement.

CONCLUSIONS: In non-correctable knees undergoing robotic-assisted TKA, preservation of native M-L laxity asymmetry is associated with superior patient-reported functional recovery. These findings challenge the universal goal of symmetric gaps and support an adaptive strategy that prioritizes bone preservation in fixed deformities.