gms | German Medical Science

Objectives: Augmented Reality is a powerful multipurpose tool. When in conjunction with a dedicated visor, it enables users to overlay a variety of information and/or images onto their field of vision. Recently, it has been integrated as a valuable tool in surgical assistance. The objective of this single-center study, which is an expansion of a previous study [1], was to assess the intraoperative outcomes of total knee arthroplasties performed with Augmented Reality assistance. The focus was on the time required and the disparities between preplanned and achieved implant positioning, in terms of tibial cut varus and slope angles.

Methods: A total of 154 consecutive patients was selected. Pre-planning was executed following the Augmented Reality protocol, and specific varus and slope angles were established as target parameters to guide intraoperatively the device. The surgical procedures started with the tibial cut [2], and the required time for calibration, registration, and fixation of the resection block was registered. The obtained varus and slope angles were also documented and compared with the preplanned values, aiming to calculate the efficiency of this method through 2 metric. With the first metric, the outcomes of this operation were used to divide the patients into groups according to the magnitude of the error obtained for varus and slope angles. The second metric analyzed the accuracy of the AR system in the function of the target amplitude for the varus and slope angles (varus: 0°, 1° and ≥ 2°; slope: 2°, 3°, 4° and ≥ 5°), therefore detecting eventual target angles more prone to error.

Results and conclusion: The average duration of Augmented Reality tool usage was 5 ± 1 minutes. The first metric showed that in 98% of cases, the differences for varus angles were ≤ 1°. Regarding slope angles, 86% of cases exhibited differences ≤ 1° and 98% of cases had differences ≤ 2°. With the second metric, the absolute errors of the different varus groups were 0.36 ± 0.60°, 0.52 ± 0.50°, and 0.44 ± 0.56°, respectively. For the slope groups, the absolute errors were 0.45 ± 0.93°, 0.72 ± 0.79°, 0.67 ± 0.62°, and 0.88 ± 0.75°, respectively. Nevertheless, the overall error between the performed cut and the pre-planned cut was lower than 1°, with a mean difference of 0.47 ± 0.54° for varus angles and 0.75 ± 0.73° for slope angles.

The findings demonstrated exceptional accuracy in the surgical cuts, accompanied by only a slight increase in the duration of the surgery. Therefore, these results underscored the potential of this innovative technology as a valuable choice for surgical assistance.

Figure 1 [Fig. 1]