Abstract
BACKGROUND: Robot-assisted total knee arthroplasty (r-TKA) can reportedly achieve more accurate implant positioning than conventional total knee arthroplasty (c-TKA), although its learning curve is controversial. Moreover, few studies have investigated r-TKA in Asians, who have different anatomical characteristics. This study aimed to determine the learning curve for r-TKA and compare implant positions between r-TKA and c-TKA according to the learning curve in Asian patients. METHODS: This prospective study included 50 consecutive c-TKAs (group C), followed by 50 consecutive r-TKAs conducted using the MAKO robotic system (Stryker, USA). Cumulative summation analyses were performed to assess the learning curve for operative time in r-TKA. Accordingly, the r-TKA cases were divided into the initial (group I) and proficiency cases (group P). The femoral and tibial component positions in the coronal, sagittal, and axial planes and lower limb alignment were compared among the three groups. RESULTS: r-TKA was associated with a learning curve for operative time in 18 cases. The operative time was significantly shorter in groups C and P than that in group I, with no significant difference between groups C and P. Groups I and P demonstrated fewer outliers with respect to lower limb alignment, femoral component coronal position, axial position, and tibial component sagittal position than those in group C, with no significant difference between groups I and P. CONCLUSION: The operative time did not differ significantly between r-TKA and c-TKA after the learning curve. Surgeons could expect more accurate and reproducible lower limb alignment and implant positioning with r-TKA in Asian patients, irrespective of the learning curve.