Abstract
INTRODUCTION
The goal of the study was to perform quality control with a commercially available navigation system when introducing PST technique at our academic department. The learning curve was assessed by the Cumulative Sum (CUSUM) test. We hypothesized that the PST process for TKA was immediately under control after its introduction when analyzed with the CUSUM technique.
MATERIAL AND METHODS
The first 50 TKAs implanted with the use of PST at an academic department were scheduled to enter in a prospective, observational study. All TKAs were implanted by an experienced, high volume senior consultant with high experience in knee navigation. PSTs were carefully positioned over the bone and articular surfaces to the best fit position, without any navigated information. Then the 3D femoral and tibia PSTs positioning were recorded. The surgical procedure was then completed following the routine navigated procedure with standard navigated templates.
To assess the 3D positioning of each template individually and of both templates together as a surrogate of the final TKA positioning, one point was given for each item inside the target, giving a maximal femur and tibia scores of 4 points, and a maximal knee score of 8 points, when all items were fulfilled. Following dataset was used for CUSUM chart plotting: allowable slack = 0.5SD, acceptable limit score = 6 points for knee score and 2 points for femur and tibia scores. For each measurement Mx, two CUSUMs (upper and lower CUSUMs) were calculated. These sums were plotted against the rank of the observation i. A trend in the process results in a change in the slope of the CUSUM, whereas the values are expected to fluctuate around a horizontal line if the process is in control. The process was considered out of control if upper CUSUM or lower CUSUM is outside the acceptable deviation interval.
RESULTS
The knee score was still out of control after the 20th case (fig. 1). Both femur (fig. 2) and tibia (fig. 3) scores were still out of control after the 20th case as well. The decision was taken to interrupt the study after the 20th case as the learning curve appeared unacceptably long in comparison to the routine navigated technique.
DISCUSSION
The main result of this study is that introduction of PST in an academic center may involve a significant learning curve: the process remained out of control even after 20 procedures. The present results contradict the common belief that introduction of PSTs is easy and does not require special instruction. These results indicate that surgeons should have only a progressive confidence with the self-sitting of PSTs when introducing this technology. Introducing PSTs might involve a significant loss of accuracy, at least when comparing with a navigation controlled implantation.
In conclusion, CUSUM analysis allows monitoring the learning curve when introducing PSTs for TKA in an academic department. There may be a significant learning curve to achieve the steady state of accuracy and obtain an acceptable alignment. The decision was taken to discontinue using PSTs for TKA.
For any figures or tables, please contact the authors directly.
