Manual Video-Based Skill Assessment Does Not Predict Postoperative Outcomes in Complex Robotic Upper Gastrointestinal
Nicholas Favazza1, Yossef Rubinov2, Alexis Benjamin3
1Department of Internal Medicine, Tufts Medical Center, Boston, USA.
Background:
The relationship between technical skill, measured through "gold standard" manual video-based review, and patient outcomes has not been clearly defined in complex robotic upper gastrointestinal (UGI) or hepatopancreatobiliary (HPB) surgery.
Methods:
Global Evaluative Assessment of Robotic Skills (GEARS) scores were prospectively captured between 2015 and 2020 across four tertiary-care institutions. Clinical, operative, and postoperative outcome data were retrospectively reviewed for all UGI (non-bariatric) and HPB surgery cases with available GEARS scores (range: 5-25). Subgroup analyses were also performed based on GEARS components and procedure length. This study has received approval from the institutional review board (IRB).
Results:
Of 4,673 cases evaluated across four tertiary-care institutions, 164 patients had UGI and HPB surgeries with evaluable GEARS; 95 UGI (esophagus, n=19; gastric, n=76) and 69 HPB (pancreas, n=22; hepatobiliary, n=47) were included. Mean age was 62.8±13.7 years; 41.5% and 56.7% identified as White and male, respectively. A total of 44 patients received neoadjuvant therapy. Mean procedure length and overall GEARS were 272.5±125.8 min and 19.9±0.86 for all procedures. Procedure length and GEARS scores demonstrated a weak inverse correlation (r = -0.275, p≤0.001). Postoperative complication rate was 51.8% (12.2% Clavien-Dindo classification of surgical complications {C-D} grade 3+) for all procedures. In multivariable analysis, each additional minute of procedure length was associated with increased odds of postoperative complications (OR=1.007, p<0.001), whereas GEARS score was not (OR=1.218, p=0.348). Findings remained consistent independent of surgery type, GEARS component score, or procedure length cohort.
Conclusion:
Procedure length is inversely associated with technical skill, as measured by GEARS. However, GEARS is not directly associated with postoperative outcomes in robotic UGI and HPB surgery. Future efforts focused on performance assessment may increase accuracy and utility by incorporating the entirety and complexity of an operation.