Development and clinical validation of an objective assessment tool for robotic-assisted and laparoscopic hysterectomy.

Abstract

Hysterectomy is one of most performed gynecological procedures worldwide. Recent decades have witnessed a significant uptake of laparoscopic approaches. While technical proficiency is increasingly recognized as being closely linked to patient outcomes in different surgical specialties, standardization of assessment in minimally invasive gynecological surgery remains challenging. This study aimed to develop and validate a comprehensive assessment tool for robotic-assisted and laparoscopic hysterectomy and investigate associations between surgical performance and perioperative outcomes. An international multi-center mixed-method study was conducted. Literature review was performed, and statements were proposed by a steering group of seven expert gynecologists, followed by a Delphi consensus to identify essential procedural phases, potential technical errors or near misses. The resulting assessment tool was validated using 40 unedited minimally invasive hysterectomy videos assessed by multiple raters under a national prospective multi-center observational cohort study. Inter-rater and intra-rater reliability were calculated using Cronbach's alpha and intraclass correlation. Concurrent validity was assessed through correlation with a validated error assessment tool (OCHRA), while predictive validity was established by correlating scores with clinical outcomes including operative time, blood loss, and postoperative complications according to the Clavien-Dindo classification. Consensus was achieved amongst 17 international experts from 6 countries on the composition of an objective assessment tool, composed of seven phases and four quality measures. Inter-rater and intra-rater reliability and internal consistency were considered excellent: ICC = 0.969 (CI: 0.922-0.986, p < 0.001), ICC 0.810 (CI: 0.800-0.842, p = 0.002), and Cronbach's α per phase was 0.743-0.834. When controlling potential confounders a higher STELLAR score was associated with shorter operating time, less blood loss and fewer post operative complications. Increased tool score was associated with significant lower chance of post-operative complications: r^s = -0.438, (CI: -0.173 to -0.657, p = 0.004). Operative performance during laparoscopic and robotic-assisted hysterectomy can be objectively measured and correlated with clinical outcomes. This newly developed objective assessment tool can be used for surgical quality assurance of surgical techniques during education and training sessions or surgical trials.