Abstract
Background
Several recent reports have documented high frequency of malpositioned acetabular components, even amongst high volume arthroplasty surgeons. Robotic assisted total hip arthroplasty (THA) has the potential to improve component positioning; however, to our knowledge there are no reports examining the learning curve during the adoption of robotic assisted THA.
Purpose
The purpose of this study was to examine the learning curve of robotic assisted THA as measured by component position, operative time, intra-operative technical problems, and complications.
Methods
The first 105 robotic-assisted THAs performed by a single surgeon with a posterior approach from June 2011 to August 2013 patients were divided into three groups based on the order of surgery. Group A was cases 1–35, group B 36–70 and group C 71–105. Component position, operative time, intra-operative technical problems, and intra-operative complications were recorded.
Results
There was no significant difference between groups A, B, and C for BMI or age (Figure 1). Gender was different between groups with 20 males in group A, 9 in group B, and 16 in group C (p < 0.05). There was no difference for mean acetabular inclination, acetabular anteversion, or leg length discrepancy between groups as experience increased (p > 0.05) (Figure 2). The average operative time for groups A, B, and C was 79.8 ± 27 min, 63.2 ± 14.2 min, and 69.4 ± 16.3 min respectively (p = 0.02).
The cumulative number of outliers was two for the Lewenick safe zone and six for the Callanan safe zone. Figure 3 displays acetabular component positioning in relation to previously documented safe zones for the three groups. The risk of having an acetabular component outside of Lewenick's safe zone was not different between groups (p = 0.60). The risk of having an acetabular component outside of Callanan's safe zone decreased after group A and was statistically significant (p = 0.02).
Overall there were nine (9%) intra-operative technical problems and complications. In group A there were three complications: one loosened femoral array, one loosened pelvic array, and one cup that appeared erroneous according to the navigation system. In group B there was one femoral calcar fracture treated with a cerclage wire, one loosened femoral array, and one intra-operative delay. In group C there were three technical problems, all a loosened femoral array. There was no difference in the overall number of intra-operative complications between groups (p = 1.0).
Conclusion
A learning curve was observed, as a decreased incidence of acetabular component outliers and decreased operative time were noted with increased experience. Satisfactory acetabular component positioning and leg length matching were found throughout the learning curve of robotic assisted total hip arthroplasty, with very few outliers in either category. Based on these findings, we conclude that there is a learning curve of approximately 35 cases in robotic-assisted total hip arthroplasty.
