Robotic surgery holds particular promise for complex oncologic colorectal resections, as it can overcome many limitations of the laparoscopic approach. However, similar to the situation in laparoscopic surgery, appropriate case selection (simple vs. complex) with respect to the actual robotic expertise of the team may be a critical determinant of outcome. The present study aimed to analyze the clinical outcome after robotic colorectal surgery over time based on the complexity of the surgical procedure.
All robotic colorectal resections (n = 85) performed at the Department of Surgery, Medical University of Vienna, between the beginning of the program in April 2015 until December 2019 were retrospectively analyzed. To compare surgical outcome over time, the cohort was divided into 2 time periods based on case sequence (period 1: patients 1–43, period 2: patients 44–85). Cases were assigned a complexity level (I-IV) according to the type of resection, severity of disease, sex and body mass index (BMI). Postoperative complications were classified using the Clavien-Dindo classification.
In total, 47 rectal resections (55.3%), 22 partial colectomies (25.8%), 14 abdomino-perineal resections (16.5%) and 2 proctocolectomies (2.4%) were performed. Of these, 69.4% (n = 59) were oncologic cases. The overall rate of major complications (Clavien Dindo III-V) was 16.5%. Complex cases (complexity levels III and IV) were more often followed by major complications than cases with a low to medium complexity level (I and II; 25.0 vs. 5.4%, p = 0.016). Furthermore, the rate of major complications decreased over time from 25.6% (period 1) to 7.1% (period 2, p = 0.038). Of note, the drop in major complications was associated with a learning effect, which was particularly pronounced in complex cases as well as a reduction of case complexity from 67.5% to 45.2% in the second period (p = 0.039).
The risk of major complications after robotic colorectal surgery increases significantly with escalating case complexity (levels III and IV), particularly during the initial phase of a new colorectal robotic surgery program. Before robotic proficiency has been achieved, it is therefore advisable to limit robotic colorectal resection to cases with complexity levels I and II in order to keep major complication rates at a minimum.