Management of failed UKA to TKA: conventional versus robotic-assisted conversion technique

BACKGROUND: Failure of unicompartmental knee arthroplasty (UKA) is a distressing and technically challenging complication. Conventional conversion techniques (CCT) with rods and jigs have produced varying results. A robotic-assisted conversion technique (RCT) is an unexplored, though possibly advantageous, alternative. We compare our reconstructive outcomes between conventional and robotic methods in the management of failed UKA. METHODS: Thirty-four patients with a failed UKA were retrospectively reviewed. Patients underwent conversion total knee arthroplasty (TKA) with either a CCT or RCT. Seventeen patients were included in each group. All procedures were done by a single surgeon at a single institution, with a mean time to follow-up of 3.6 years (range, 1 to 12). The primary outcome measures were the need for augments and polyethylene thickness. Secondary outcome measures were complications, need for revision, estimated blood loss (EBL), length of stay, and operative time. RESULTS: The mean polyethylene thickness was 12 mm (range, 9 to 15) in the CCT group and 10 mm (range, 9 to 14) in the RCT groups, with no statistical difference between the two groups (P = 0.07). A statistically significant difference, however, was present in the use of augments. In the CCT group, five out of 17 knees required augments, whereas none of the 17 knees in the RCT group required augments (P = 0.04). Procedurally, robotic-assisted surgery progressed uneventfully, even with metal artifact noted on the preoperative computerized tomography (CT) scans. Computer mapping of the residual bone surface after implant removal was a helpful guide in minimizing resection depth. No further revisions or reoperations were performed in either group. CONCLUSIONS: Robotic-assisted conversion TKA is technically feasible and potentially advantageous. In the absence of normal anatomic landmarks to guide conventional methods, the preoperative CT scans were unexpectedly helpful in establishing mechanical alignment and resection depth. In this limited series, RCT does not seem to be inferior to CCT. Further investigation of outcomes is warranted.