RESUMO
PURPOSE: The main objective was to analyze the computer-assisted navigation (CAN)-specific complications that forced to switch to conventional procedure in primary total knee replacement (TKR). The secondary objective was to determine the influence of those complications on TKR short-term survival. METHODS: Retrospective study of 878 primary TKR in 753 patients. Two consecutive versions of the OrthoPilot navigation system (Braun Aesculap, Germany) were used during the study time. Specific complications of CAN were defined as those due to the instrumentation (hardware or software failures), which were classified in one of two categories according to whether they occurred during the registration or tracking process. RESULTS: There were 20 (2.3%) complications related to the navigation system use that forced to switch to conventional surgery: in 11 (1.2%) knees due to loosening of the tracking pins, and in the other nine (1.0%) there were information system failures. There was a trend for a higher conversion rate to conventional surgery with the use of the first version of the software. There were no fractures, infections, or nerve injuries at the pin site. We found no differences in the distribution of baseline variables among those with or without conversion to conventional surgery. There was no significant difference (p = 0.488) in the two year survival between patients with or without conversion. CONCLUSION: CAN for primary TKR is a safe method with few specific complications that forced to switch from the navigated to the conventional procedure. Conversion to conventional surgery did not affect the short-term survival of TKR.