Strengths and Limitations
To our best knowledge, this is the first study presenting a learning curve analysis during introduction of a novice robotic surgeon in the gynaecological oncology field. We used RA-CUSUM analysis, which is considered the reference standard for studying surgical learning curves29 and has an additional value in its individual risk adjustment compared to standard CUSUM analysis wherein parameters are set based on literature. Another strength is that we use oncological outcomes, i.e. survival, which is considered the foremost relevant measure of performance when treating cancer. We were able to correlate these oncological outcomes to the introduction of a novice to an experienced surgical team.
Several limitations to this study exist. Since we report a single-institution analysis of a small surgical team, our results might not be transferrable to other centres. Although the reported learning curves should be considered institutional learning curves, given the team effort of robotic procedures, individual aspects within the small surgical team could contribute to different outcomes in other centres. Not all surgeons new to robot-assisted surgery will learn at the same pace. With the current analysis we are not able to present an individual learning curve for the novice surgeon, as the learning curve fades into the institutional learning curve. To quantify the performance of a structured learning curriculum, validation in larger, preferably multicentre, datasets is warranted.
We expanded our cohort with 61 procedures as we previously reported this to be the learning phase. Inherently, the follow-up times of the latter group are shorter and additional recurrences could present in the upcoming years as not all patients in group 3 have completed the recommended five years of follow-up yet. However, the majority of patients in group 3 completed three years of follow-up, the period in which most recurrences occur. More importantly, Kaplan-Meier curves were used to correct for differences in follow-up time through censoring.