Abstract
Purpose :
To compare manual and robotic (Preceyes BV, Eindhoven, the Netherlands) manipulations in a simulated vitreoretinal peeling model (ACMIT, Wiener-Neustadt, Austria) with and without an active boundary that is updated in real-time using an OCT-based distance measurement (OPMedT and MLL, Lübeck, Germany).
Methods :
In an ex-vivo experiment, peeling was performed by three experienced surgeons on a phantom model that simulates an epiretinal membrane. We compared i) manual surgery and ii) robot-assisted surgery without and iii) with an OCT-based distance boundary. The OCT sensor measures the distance between the instrument and the target surface, automatically stopping movements beyond a predefined boundary. The boundary could be set negative, to allow penetration of the membrane. The three different methods were rated on a scale from 1 to 10 by three surgeons for i) speed, ii) safety, iii) usability and iv) ease of use/stress level. Each surgeon performed 5 to 10 experiments with each method.
Results :
On average, the surgeons rated manual surgery as insufficient for both safety (mean 5.0) and ease (mean 6.0). These were both increased (mean ≥7) by introducing robot-assisted surgery. Benefits for safety are largest: safety is strongly increased by introducing the robot (mean 7.3), and even further by adding the sensor assistance (mean 8.3). Usability and speed were considered better for manual surgery (mean 8.0 usability and mean 8.7 speed) compared with robot-assisted surgery without OCT sensor (mean 7.3 usability and speed) and with OCT sensor (mean 7.7 usability and mean 7.0 speed).
Conclusions :
Surgeons felt safer and more at ease when using robotic assistance and even more when combining this with an OCT-based boundary. This could be of benefit to improve the safety of epiretinal tissue removal. Another potential use of the sensor may be to ensure a predefined safe distance from the retina even in case of sudden patient movements. A sensor may be especially beneficial in high myopic eyes or when visibility is inadequate.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.